diff options
238 files changed, 6402 insertions, 4103 deletions
@@ -2365,8 +2365,6 @@ E: acme@redhat.com W: http://oops.ghostprotocols.net:81/blog/ P: 1024D/9224DF01 D5DF E3BB E3C8 BCBB F8AD 841A B6AB 4681 9224 DF01 D: IPX, LLC, DCCP, cyc2x, wl3501_cs, net/ hacks -S: R. Brasílio Itiberê, 4270/1010 - Água Verde -S: 80240-060 - Curitiba - Paraná S: Brazil N: Karsten Merker diff --git a/Documentation/RCU/trace.txt b/Documentation/RCU/trace.txt index a851118775d8..6a8c73f55b80 100644 --- a/Documentation/RCU/trace.txt +++ b/Documentation/RCU/trace.txt @@ -1,18 +1,22 @@ CONFIG_RCU_TRACE debugfs Files and Formats -The rcutree implementation of RCU provides debugfs trace output that -summarizes counters and state. This information is useful for debugging -RCU itself, and can sometimes also help to debug abuses of RCU. -The following sections describe the debugfs files and formats. +The rcutree and rcutiny implementations of RCU provide debugfs trace +output that summarizes counters and state. This information is useful for +debugging RCU itself, and can sometimes also help to debug abuses of RCU. +The following sections describe the debugfs files and formats, first +for rcutree and next for rcutiny. -Hierarchical RCU debugfs Files and Formats +CONFIG_TREE_RCU and CONFIG_TREE_PREEMPT_RCU debugfs Files and Formats -This implementation of RCU provides three debugfs files under the +These implementations of RCU provides five debugfs files under the top-level directory RCU: rcu/rcudata (which displays fields in struct -rcu_data), rcu/rcugp (which displays grace-period counters), and -rcu/rcuhier (which displays the struct rcu_node hierarchy). +rcu_data), rcu/rcudata.csv (which is a .csv spreadsheet version of +rcu/rcudata), rcu/rcugp (which displays grace-period counters), +rcu/rcuhier (which displays the struct rcu_node hierarchy), and +rcu/rcu_pending (which displays counts of the reasons that the +rcu_pending() function decided that there was core RCU work to do). The output of "cat rcu/rcudata" looks as follows: @@ -130,7 +134,8 @@ o "ci" is the number of RCU callbacks that have been invoked for been registered in absence of CPU-hotplug activity. o "co" is the number of RCU callbacks that have been orphaned due to - this CPU going offline. + this CPU going offline. These orphaned callbacks have been moved + to an arbitrarily chosen online CPU. o "ca" is the number of RCU callbacks that have been adopted due to other CPUs going offline. Note that ci+co-ca+ql is the number of @@ -168,12 +173,12 @@ o "gpnum" is the number of grace periods that have started. It is The output of "cat rcu/rcuhier" looks as follows, with very long lines: -c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6 oqlen=0 +c=6902 g=6903 s=2 jfq=3 j=72c7 nfqs=13142/nfqsng=0(13142) fqlh=6 1/1 .>. 0:127 ^0 3/3 .>. 0:35 ^0 0/0 .>. 36:71 ^1 0/0 .>. 72:107 ^2 0/0 .>. 108:127 ^3 3/3f .>. 0:5 ^0 2/3 .>. 6:11 ^1 0/0 .>. 12:17 ^2 0/0 .>. 18:23 ^3 0/0 .>. 24:29 ^4 0/0 .>. 30:35 ^5 0/0 .>. 36:41 ^0 0/0 .>. 42:47 ^1 0/0 .>. 48:53 ^2 0/0 .>. 54:59 ^3 0/0 .>. 60:65 ^4 0/0 .>. 66:71 ^5 0/0 .>. 72:77 ^0 0/0 .>. 78:83 ^1 0/0 .>. 84:89 ^2 0/0 .>. 90:95 ^3 0/0 .>. 96:101 ^4 0/0 .>. 102:107 ^5 0/0 .>. 108:113 ^0 0/0 .>. 114:119 ^1 0/0 .>. 120:125 ^2 0/0 .>. 126:127 ^3 rcu_bh: -c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0 oqlen=0 +c=-226 g=-226 s=1 jfq=-5701 j=72c7 nfqs=88/nfqsng=0(88) fqlh=0 0/1 .>. 0:127 ^0 0/3 .>. 0:35 ^0 0/0 .>. 36:71 ^1 0/0 .>. 72:107 ^2 0/0 .>. 108:127 ^3 0/3f .>. 0:5 ^0 0/3 .>. 6:11 ^1 0/0 .>. 12:17 ^2 0/0 .>. 18:23 ^3 0/0 .>. 24:29 ^4 0/0 .>. 30:35 ^5 0/0 .>. 36:41 ^0 0/0 .>. 42:47 ^1 0/0 .>. 48:53 ^2 0/0 .>. 54:59 ^3 0/0 .>. 60:65 ^4 0/0 .>. 66:71 ^5 0/0 .>. 72:77 ^0 0/0 .>. 78:83 ^1 0/0 .>. 84:89 ^2 0/0 .>. 90:95 ^3 0/0 .>. 96:101 ^4 0/0 .>. 102:107 ^5 0/0 .>. 108:113 ^0 0/0 .>. 114:119 ^1 0/0 .>. 120:125 ^2 0/0 .>. 126:127 ^3 @@ -212,11 +217,6 @@ o "fqlh" is the number of calls to force_quiescent_state() that exited immediately (without even being counted in nfqs above) due to contention on ->fqslock. -o "oqlen" is the number of callbacks on the "orphan" callback - list. RCU callbacks are placed on this list by CPUs going - offline, and are "adopted" either by the CPU helping the outgoing - CPU or by the next rcu_barrier*() call, whichever comes first. - o Each element of the form "1/1 0:127 ^0" represents one struct rcu_node. Each line represents one level of the hierarchy, from root to leaves. It is best to think of the rcu_data structures @@ -326,3 +326,115 @@ o "nn" is the number of times that this CPU needed nothing. Alert readers will note that the rcu "nn" number for a given CPU very closely matches the rcu_bh "np" number for that same CPU. This is due to short-circuit evaluation in rcu_pending(). + + +CONFIG_TINY_RCU and CONFIG_TINY_PREEMPT_RCU debugfs Files and Formats + +These implementations of RCU provides a single debugfs file under the +top-level directory RCU, namely rcu/rcudata, which displays fields in +rcu_bh_ctrlblk, rcu_sched_ctrlblk and, for CONFIG_TINY_PREEMPT_RCU, +rcu_preempt_ctrlblk. + +The output of "cat rcu/rcudata" is as follows: + +rcu_preempt: qlen=24 gp=1097669 g197/p197/c197 tasks=... + ttb=. btg=no ntb=184 neb=0 nnb=183 j=01f7 bt=0274 + normal balk: nt=1097669 gt=0 bt=371 b=0 ny=25073378 nos=0 + exp balk: bt=0 nos=0 +rcu_sched: qlen: 0 +rcu_bh: qlen: 0 + +This is split into rcu_preempt, rcu_sched, and rcu_bh sections, with the +rcu_preempt section appearing only in CONFIG_TINY_PREEMPT_RCU builds. +The last three lines of the rcu_preempt section appear only in +CONFIG_RCU_BOOST kernel builds. The fields are as follows: + +o "qlen" is the number of RCU callbacks currently waiting either + for an RCU grace period or waiting to be invoked. This is the + only field present for rcu_sched and rcu_bh, due to the + short-circuiting of grace period in those two cases. + +o "gp" is the number of grace periods that have completed. + +o "g197/p197/c197" displays the grace-period state, with the + "g" number being the number of grace periods that have started + (mod 256), the "p" number being the number of grace periods + that the CPU has responded to (also mod 256), and the "c" + number being the number of grace periods that have completed + (once again mode 256). + + Why have both "gp" and "g"? Because the data flowing into + "gp" is only present in a CONFIG_RCU_TRACE kernel. + +o "tasks" is a set of bits. The first bit is "T" if there are + currently tasks that have recently blocked within an RCU + read-side critical section, the second bit is "N" if any of the + aforementioned tasks are blocking the current RCU grace period, + and the third bit is "E" if any of the aforementioned tasks are + blocking the current expedited grace period. Each bit is "." + if the corresponding condition does not hold. + +o "ttb" is a single bit. It is "B" if any of the blocked tasks + need to be priority boosted and "." otherwise. + +o "btg" indicates whether boosting has been carried out during + the current grace period, with "exp" indicating that boosting + is in progress for an expedited grace period, "no" indicating + that boosting has not yet started for a normal grace period, + "begun" indicating that boosting has bebug for a normal grace + period, and "done" indicating that boosting has completed for + a normal grace period. + +o "ntb" is the total number of tasks subjected to RCU priority boosting + periods since boot. + +o "neb" is the number of expedited grace periods that have had + to resort to RCU priority boosting since boot. + +o "nnb" is the number of normal grace periods that have had + to resort to RCU priority boosting since boot. + +o "j" is the low-order 12 bits of the jiffies counter in hexadecimal. + +o "bt" is the low-order 12 bits of the value that the jiffies counter + will have at the next time that boosting is scheduled to begin. + +o In the line beginning with "normal balk", the fields are as follows: + + o "nt" is the number of times that the system balked from + boosting because there were no blocked tasks to boost. + Note that the system will balk from boosting even if the + grace period is overdue when the currently running task + is looping within an RCU read-side critical section. + There is no point in boosting in this case, because + boosting a running task won't make it run any faster. + + o "gt" is the number of times that the system balked + from boosting because, although there were blocked tasks, + none of them were preventing the current grace period + from completing. + + o "bt" is the number of times that the system balked + from boosting because boosting was already in progress. + + o "b" is the number of times that the system balked from + boosting because boosting had already completed for + the grace period in question. + + o "ny" is the number of times that the system balked from + boosting because it was not yet time to start boosting + the grace period in question. + + o "nos" is the number of times that the system balked from + boosting for inexplicable ("not otherwise specified") + reasons. This can actually happen due to races involving + increments of the jiffies counter. + +o In the line beginning with "exp balk", the fields are as follows: + + o "bt" is the number of times that the system balked from + boosting because there were no blocked tasks to boost. + + o "nos" is the number of times that the system balked from + boosting for inexplicable ("not otherwise specified") + reasons. diff --git a/Documentation/dontdiff b/Documentation/dontdiff index d9bcffd59433..470d3dba1a69 100644 --- a/Documentation/dontdiff +++ b/Documentation/dontdiff @@ -62,6 +62,10 @@ aic7*reg_print.c* aic7*seq.h* aicasm aicdb.h* +altivec1.c +altivec2.c +altivec4.c +altivec8.c asm-offsets.h asm_offsets.h autoconf.h* @@ -76,6 +80,7 @@ btfixupprep build bvmlinux bzImage* +capflags.c classlist.h* comp*.log compile.h* @@ -94,6 +99,7 @@ devlist.h* docproc elf2ecoff elfconfig.h* +evergreen_reg_safe.h fixdep flask.h fore200e_mkfirm @@ -108,9 +114,16 @@ genksyms *_gray256.c ihex2fw ikconfig.h* +inat-tables.c initramfs_data.cpio initramfs_data.cpio.gz initramfs_list +int16.c +int1.c +int2.c +int32.c +int4.c +int8.c kallsyms kconfig keywords.c @@ -140,6 +153,7 @@ mkprep mktables mktree modpost +modules.builtin modules.order modversions.h* ncscope.* @@ -153,14 +167,23 @@ pca200e.bin pca200e_ecd.bin2 piggy.gz piggyback +piggy.S pnmtologo ppc_defs.h* pss_boot.h qconf +r100_reg_safe.h +r200_reg_safe.h +r300_reg_safe.h +r420_reg_safe.h +r600_reg_safe.h raid6altivec*.c raid6int*.c raid6tables.c relocs +rn50_reg_safe.h +rs600_reg_safe.h +rv515_reg_safe.h series setup setup.bin @@ -169,6 +192,7 @@ sImage sm_tbl* split-include syscalltab.h +tables.c tags tftpboot.img timeconst.h @@ -190,6 +214,7 @@ vmlinux vmlinux-* vmlinux.aout vmlinux.lds +voffset.h vsyscall.lds vsyscall_32.lds wanxlfw.inc @@ -200,3 +225,4 @@ wakeup.elf wakeup.lds zImage* zconf.hash.c +zoffset.h diff --git a/Documentation/kernel-docs.txt b/Documentation/kernel-docs.txt index 715eaaf1519d..9a8674629a07 100644 --- a/Documentation/kernel-docs.txt +++ b/Documentation/kernel-docs.txt @@ -537,7 +537,7 @@ Notes: Further information in http://www.oreilly.com/catalog/linuxdrive2/ - * Title: "Linux Device Drivers, 3nd Edition" + * Title: "Linux Device Drivers, 3rd Edition" Authors: Jonathan Corbet, Alessandro Rubini, and Greg Kroah-Hartman Publisher: O'Reilly & Associates. Date: 2005. @@ -592,14 +592,6 @@ Pages: 600. ISBN: 0-13-101908-2 - * Title: "The Design and Implementation of the 4.4 BSD UNIX - Operating System" - Author: Marshall Kirk McKusick, Keith Bostic, Michael J. Karels, - John S. Quarterman. - Publisher: Addison-Wesley. - Date: 1996. - ISBN: 0-201-54979-4 - * Title: "Programming for the real world - POSIX.4" Author: Bill O. Gallmeister. Publisher: O'Reilly & Associates, Inc.. @@ -610,28 +602,13 @@ POSIX. Good reference. * Title: "UNIX Systems for Modern Architectures: Symmetric - Multiprocesssing and Caching for Kernel Programmers" + Multiprocessing and Caching for Kernel Programmers" Author: Curt Schimmel. Publisher: Addison Wesley. Date: June, 1994. Pages: 432. ISBN: 0-201-63338-8 - * Title: "The Design and Implementation of the 4.3 BSD UNIX - Operating System" - Author: Samuel J. Leffler, Marshall Kirk McKusick, Michael J. - Karels, John S. Quarterman. - Publisher: Addison-Wesley. - Date: 1989 (reprinted with corrections on October, 1990). - ISBN: 0-201-06196-1 - - * Title: "The Design of the UNIX Operating System" - Author: Maurice J. Bach. - Publisher: Prentice Hall. - Date: 1986. - Pages: 471. - ISBN: 0-13-201757-1 - MISCELLANEOUS: * Name: linux/Documentation diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 1031923f5254..d6496fde6180 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1579,20 +1579,12 @@ and is between 256 and 4096 characters. It is defined in the file nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels Format: [panic,][num] - Valid num: 0,1,2 + Valid num: 0 0 - turn nmi_watchdog off - 1 - use the IO-APIC timer for the NMI watchdog - 2 - use the local APIC for the NMI watchdog using - a performance counter. Note: This will use one - performance counter and the local APIC's performance - vector. When panic is specified, panic when an NMI watchdog timeout occurs. This is useful when you use a panic=... timeout and need the box quickly up again. - Instead of 1 and 2 it is possible to use the following - symbolic names: lapic and ioapic - Example: nmi_watchdog=2 or nmi_watchdog=panic,lapic netpoll.carrier_timeout= [NET] Specifies amount of time (in seconds) that diff --git a/Documentation/trace/events-power.txt b/Documentation/trace/events-power.txt new file mode 100644 index 000000000000..96d87b67fe37 --- /dev/null +++ b/Documentation/trace/events-power.txt @@ -0,0 +1,90 @@ + + Subsystem Trace Points: power + +The power tracing system captures events related to power transitions +within the kernel. Broadly speaking there are three major subheadings: + + o Power state switch which reports events related to suspend (S-states), + cpuidle (C-states) and cpufreq (P-states) + o System clock related changes + o Power domains related changes and transitions + +This document describes what each of the tracepoints is and why they +might be useful. + +Cf. include/trace/events/power.h for the events definitions. + +1. Power state switch events +============================ + +1.1 New trace API +----------------- + +A 'cpu' event class gathers the CPU-related events: cpuidle and +cpufreq. + +cpu_idle "state=%lu cpu_id=%lu" +cpu_frequency "state=%lu cpu_id=%lu" + +A suspend event is used to indicate the system going in and out of the +suspend mode: + +machine_suspend "state=%lu" + + +Note: the value of '-1' or '4294967295' for state means an exit from the current state, +i.e. trace_cpu_idle(4, smp_processor_id()) means that the system +enters the idle state 4, while trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id()) +means that the system exits the previous idle state. + +The event which has 'state=4294967295' in the trace is very important to the user +space tools which are using it to detect the end of the current state, and so to +correctly draw the states diagrams and to calculate accurate statistics etc. + +1.2 DEPRECATED trace API +------------------------ + +A new Kconfig option CONFIG_EVENT_POWER_TRACING_DEPRECATED with the default value of +'y' has been created. This allows the legacy trace power API to be used conjointly +with the new trace API. +The Kconfig option, the old trace API (in include/trace/events/power.h) and the +old trace points will disappear in a future release (namely 2.6.41). + +power_start "type=%lu state=%lu cpu_id=%lu" +power_frequency "type=%lu state=%lu cpu_id=%lu" +power_end "cpu_id=%lu" + +The 'type' parameter takes one of those macros: + . POWER_NONE = 0, + . POWER_CSTATE = 1, /* C-State */ + . POWER_PSTATE = 2, /* Fequency change or DVFS */ + +The 'state' parameter is set depending on the type: + . Target C-state for type=POWER_CSTATE, + . Target frequency for type=POWER_PSTATE, + +power_end is used to indicate the exit of a state, corresponding to the latest +power_start event. + +2. Clocks events +================ +The clock events are used for clock enable/disable and for +clock rate change. + +clock_enable "%s state=%lu cpu_id=%lu" +clock_disable "%s state=%lu cpu_id=%lu" +clock_set_rate "%s state=%lu cpu_id=%lu" + +The first parameter gives the clock name (e.g. "gpio1_iclk"). +The second parameter is '1' for enable, '0' for disable, the target +clock rate for set_rate. + +3. Power domains events +======================= +The power domain events are used for power domains transitions + +power_domain_target "%s state=%lu cpu_id=%lu" + +The first parameter gives the power domain name (e.g. "mpu_pwrdm"). +The second parameter is the power domain target state. + diff --git a/MAINTAINERS b/MAINTAINERS index 7585e9dd835b..b1dda78a1e75 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4627,7 +4627,7 @@ PERFORMANCE EVENTS SUBSYSTEM M: Peter Zijlstra <a.p.zijlstra@chello.nl> M: Paul Mackerras <paulus@samba.org> M: Ingo Molnar <mingo@elte.hu> -M: Arnaldo Carvalho de Melo <acme@redhat.com> +M: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> S: Supported F: kernel/perf_event*.c F: include/linux/perf_event.h diff --git a/arch/alpha/include/asm/perf_event.h b/arch/alpha/include/asm/perf_event.h index fe792ca818f6..5996e7a6757e 100644 --- a/arch/alpha/include/asm/perf_event.h +++ b/arch/alpha/include/asm/perf_event.h @@ -1,10 +1,4 @@ #ifndef __ASM_ALPHA_PERF_EVENT_H #define __ASM_ALPHA_PERF_EVENT_H -#ifdef CONFIG_PERF_EVENTS -extern void init_hw_perf_events(void); -#else -static inline void init_hw_perf_events(void) { } -#endif - #endif /* __ASM_ALPHA_PERF_EVENT_H */ diff --git a/arch/alpha/kernel/irq_alpha.c b/arch/alpha/kernel/irq_alpha.c index 5f77afb88e89..4c8bb374eb0a 100644 --- a/arch/alpha/kernel/irq_alpha.c +++ b/arch/alpha/kernel/irq_alpha.c @@ -112,8 +112,6 @@ init_IRQ(void) wrent(entInt, 0); alpha_mv.init_irq(); - - init_hw_perf_events(); } /* diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 1cc49683fb69..90561c45e7d8 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -14,6 +14,7 @@ #include <linux/kernel.h> #include <linux/kdebug.h> #include <linux/mutex.h> +#include <linux/init.h> #include <asm/hwrpb.h> #include <asm/atomic.h> @@ -863,13 +864,13 @@ static void alpha_perf_event_irq_handler(unsigned long la_ptr, /* * Init call to initialise performance events at kernel startup. */ -void __init init_hw_perf_events(void) +int __init init_hw_perf_events(void) { pr_info("Performance events: "); if (!supported_cpu()) { pr_cont("No support for your CPU.\n"); - return; + return 0; } pr_cont("Supported CPU type!\n"); @@ -881,6 +882,8 @@ void __init init_hw_perf_events(void) /* And set up PMU specification */ alpha_pmu = &ev67_pmu; - perf_pmu_register(&pmu); -} + perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW); + return 0; +} +early_initcall(init_hw_perf_events); diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 07a50357492a..fdfa4976b0bf 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -3034,11 +3034,11 @@ init_hw_perf_events(void) pr_info("no hardware support available\n"); } - perf_pmu_register(&pmu); + perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW); return 0; } -arch_initcall(init_hw_perf_events); +early_initcall(init_hw_perf_events); /* * Callchain handling code. diff --git a/arch/mips/kernel/perf_event_mipsxx.c b/arch/mips/kernel/perf_event_mipsxx.c index 5c7c6fc07565..183e0d226669 100644 --- a/arch/mips/kernel/perf_event_mipsxx.c +++ b/arch/mips/kernel/perf_event_mipsxx.c @@ -1047,6 +1047,6 @@ init_hw_perf_events(void) return 0; } -arch_initcall(init_hw_perf_events); +early_initcall(init_hw_perf_events); #endif /* defined(CONFIG_CPU_MIPS32)... */ diff --git a/arch/powerpc/kernel/e500-pmu.c b/arch/powerpc/kernel/e500-pmu.c index 7c07de0d8943..b150b510510f 100644 --- a/arch/powerpc/kernel/e500-pmu.c +++ b/arch/powerpc/kernel/e500-pmu.c @@ -126,4 +126,4 @@ static int init_e500_pmu(void) return register_fsl_emb_pmu(&e500_pmu); } -arch_initcall(init_e500_pmu); +early_initcall(init_e500_pmu); diff --git a/arch/powerpc/kernel/mpc7450-pmu.c b/arch/powerpc/kernel/mpc7450-pmu.c index 09d72028f317..2cc5e0301d0b 100644 --- a/arch/powerpc/kernel/mpc7450-pmu.c +++ b/arch/powerpc/kernel/mpc7450-pmu.c @@ -414,4 +414,4 @@ static int init_mpc7450_pmu(void) return register_power_pmu(&mpc7450_pmu); } -arch_initcall(init_mpc7450_pmu); +early_initcall(init_mpc7450_pmu); diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index 3129c855933c..567480705789 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -1379,7 +1379,7 @@ int register_power_pmu(struct power_pmu *pmu) freeze_events_kernel = MMCR0_FCHV; #endif /* CONFIG_PPC64 */ - perf_pmu_register(&power_pmu); + perf_pmu_register(&power_pmu, "cpu", PERF_TYPE_RAW); perf_cpu_notifier(power_pmu_notifier); return 0; diff --git a/arch/powerpc/kernel/perf_event_fsl_emb.c b/arch/powerpc/kernel/perf_event_fsl_emb.c index 7ecca59ddf77..4dcf5f831e9d 100644 --- a/arch/powerpc/kernel/perf_event_fsl_emb.c +++ b/arch/powerpc/kernel/perf_event_fsl_emb.c @@ -681,7 +681,7 @@ int register_fsl_emb_pmu(struct fsl_emb_pmu *pmu) pr_info("%s performance monitor hardware support registered\n", pmu->name); - perf_pmu_register(&fsl_emb_pmu); + perf_pmu_register(&fsl_emb_pmu, "cpu", PERF_TYPE_RAW); return 0; } diff --git a/arch/powerpc/kernel/power4-pmu.c b/arch/powerpc/kernel/power4-pmu.c index 2a361cdda635..ead8b3c2649e 100644 --- a/arch/powerpc/kernel/power4-pmu.c +++ b/arch/powerpc/kernel/power4-pmu.c @@ -613,4 +613,4 @@ static int init_power4_pmu(void) return register_power_pmu(&power4_pmu); } -arch_initcall(init_power4_pmu); +early_initcall(init_power4_pmu); diff --git a/arch/powerpc/kernel/power5+-pmu.c b/arch/powerpc/kernel/power5+-pmu.c index 199de527d411..eca0ac595cb6 100644 --- a/arch/powerpc/kernel/power5+-pmu.c +++ b/arch/powerpc/kernel/power5+-pmu.c @@ -682,4 +682,4 @@ static int init_power5p_pmu(void) return register_power_pmu(&power5p_pmu); } -arch_initcall(init_power5p_pmu); +early_initcall(init_power5p_pmu); diff --git a/arch/powerpc/kernel/power5-pmu.c b/arch/powerpc/kernel/power5-pmu.c index 98b6a729a9dd..d5ff0f64a5e6 100644 --- a/arch/powerpc/kernel/power5-pmu.c +++ b/arch/powerpc/kernel/power5-pmu.c @@ -621,4 +621,4 @@ static int init_power5_pmu(void) return register_power_pmu(&power5_pmu); } -arch_initcall(init_power5_pmu); +early_initcall(init_power5_pmu); diff --git a/arch/powerpc/kernel/power6-pmu.c b/arch/powerpc/kernel/power6-pmu.c index 84a607bda8fb..31603927e376 100644 --- a/arch/powerpc/kernel/power6-pmu.c +++ b/arch/powerpc/kernel/power6-pmu.c @@ -544,4 +544,4 @@ static int init_power6_pmu(void) return register_power_pmu(&power6_pmu); } -arch_initcall(init_power6_pmu); +early_initcall(init_power6_pmu); diff --git a/arch/powerpc/kernel/power7-pmu.c b/arch/powerpc/kernel/power7-pmu.c index 852f7b7f6b40..593740fcb799 100644 --- a/arch/powerpc/kernel/power7-pmu.c +++ b/arch/powerpc/kernel/power7-pmu.c @@ -369,4 +369,4 @@ static int init_power7_pmu(void) return register_power_pmu(&power7_pmu); } -arch_initcall(init_power7_pmu); +early_initcall(init_power7_pmu); diff --git a/arch/powerpc/kernel/ppc970-pmu.c b/arch/powerpc/kernel/ppc970-pmu.c index 3fee685de4df..9a6e093858fe 100644 --- a/arch/powerpc/kernel/ppc970-pmu.c +++ b/arch/powerpc/kernel/ppc970-pmu.c @@ -494,4 +494,4 @@ static int init_ppc970_pmu(void) return register_power_pmu(&ppc970_pmu); } -arch_initcall(init_ppc970_pmu); +early_initcall(init_ppc970_pmu); diff --git a/arch/sh/kernel/cpu/sh4/perf_event.c b/arch/sh/kernel/cpu/sh4/perf_event.c index dbf3b4bb71fe..748955df018d 100644 --- a/arch/sh/kernel/cpu/sh4/perf_event.c +++ b/arch/sh/kernel/cpu/sh4/perf_event.c @@ -250,4 +250,4 @@ static int __init sh7750_pmu_init(void) return register_sh_pmu(&sh7750_pmu); } -arch_initcall(sh7750_pmu_init); +early_initcall(sh7750_pmu_init); diff --git a/arch/sh/kernel/cpu/sh4a/perf_event.c b/arch/sh/kernel/cpu/sh4a/perf_event.c index 580276525731..17e6bebfede0 100644 --- a/arch/sh/kernel/cpu/sh4a/perf_event.c +++ b/arch/sh/kernel/cpu/sh4a/perf_event.c @@ -284,4 +284,4 @@ static int __init sh4a_pmu_init(void) return register_sh_pmu(&sh4a_pmu); } -arch_initcall(sh4a_pmu_init); +early_initcall(sh4a_pmu_init); diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c index 5a4b33435650..2ee21a47b5af 100644 --- a/arch/sh/kernel/perf_event.c +++ b/arch/sh/kernel/perf_event.c @@ -389,7 +389,7 @@ int __cpuinit register_sh_pmu(struct sh_pmu *_pmu) WARN_ON(_pmu->num_events > MAX_HWEVENTS); - perf_pmu_register(&pmu); + perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW); perf_cpu_notifier(sh_pmu_notifier); return 0; } diff --git a/arch/sparc/include/asm/perf_event.h b/arch/sparc/include/asm/perf_event.h index 6e8bfa1786da..4d3dbe3703e9 100644 --- a/arch/sparc/include/asm/perf_event.h +++ b/arch/sparc/include/asm/perf_event.h @@ -4,8 +4,6 @@ #ifdef CONFIG_PERF_EVENTS #include <asm/ptrace.h> -extern void init_hw_perf_events(void); - #define perf_arch_fetch_caller_regs(regs, ip) \ do { \ unsigned long _pstate, _asi, _pil, _i7, _fp; \ @@ -26,8 +24,6 @@ do { \ (regs)->u_regs[UREG_I6] = _fp; \ (regs)->u_regs[UREG_I7] = _i7; \ } while (0) -#else -static inline void init_hw_perf_events(void) { } #endif #endif diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index a4bd7ba74c89..300f810142f5 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -270,8 +270,6 @@ int __init nmi_init(void) atomic_set(&nmi_active, -1); } } - if (!err) - init_hw_perf_events(); return err; } diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 0d6deb55a2ae..760578687e7c 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1307,20 +1307,23 @@ static bool __init supported_pmu(void) return false; } -void __init init_hw_perf_events(void) +int __init init_hw_perf_events(void) { pr_info("Performance events: "); if (!supported_pmu()) { pr_cont("No support for PMU type '%s'\n", sparc_pmu_type); - return; + return 0; } pr_cont("Supported PMU type is '%s'\n", sparc_pmu_type); - perf_pmu_register(&pmu); + perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW); register_die_notifier(&perf_event_nmi_notifier); + + return 0; } +early_initcall(init_hw_perf_events); void perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 76561d20ea2f..4a2adaa9aefc 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -180,8 +180,15 @@ extern void *text_poke_early(void *addr, const void *opcode, size_t len); * On the local CPU you need to be protected again NMI or MCE handlers seeing an * inconsistent instruction while you patch. */ +struct text_poke_param { + void *addr; + const void *opcode; + size_t len; +}; + extern void *text_poke(void *addr, const void *opcode, size_t len); extern void *text_poke_smp(void *addr, const void *opcode, size_t len); +extern void text_poke_smp_batch(struct text_poke_param *params, int n); #if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) #define IDEAL_NOP_SIZE_5 5 diff --git a/arch/x86/include/asm/irq.h b/arch/x86/include/asm/irq.h index 13b0ebaa512f..ba870bb6dd8e 100644 --- a/arch/x86/include/asm/irq.h +++ b/arch/x86/include/asm/irq.h @@ -15,10 +15,6 @@ static inline int irq_canonicalize(int irq) return ((irq == 2) ? 9 : irq); } -#ifdef CONFIG_X86_LOCAL_APIC -# define ARCH_HAS_NMI_WATCHDOG -#endif - #ifdef CONFIG_X86_32 extern void irq_ctx_init(int cpu); #else diff --git a/arch/x86/include/asm/kdebug.h b/arch/x86/include/asm/kdebug.h index 5bdfca86581b..f23eb2528464 100644 --- a/arch/x86/include/asm/kdebug.h +++ b/arch/x86/include/asm/kdebug.h @@ -28,7 +28,7 @@ extern void die(const char *, struct pt_regs *,long); extern int __must_check __die(const char *, struct pt_regs *, long); extern void show_registers(struct pt_regs *regs); extern void show_trace(struct task_struct *t, struct pt_regs *regs, - unsigned long *sp, unsigned long bp); + unsigned long *sp); extern void __show_regs(struct pt_regs *regs, int all); extern void show_regs(struct pt_regs *regs); extern unsigned long oops_begin(void); diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 6b89f5e86021..86030f63ba02 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -123,6 +123,10 @@ #define MSR_AMD64_IBSCTL 0xc001103a #define MSR_AMD64_IBSBRTARGET 0xc001103b +/* Fam 15h MSRs */ +#define MSR_F15H_PERF_CTL 0xc0010200 +#define MSR_F15H_PERF_CTR 0xc0010201 + /* Fam 10h MSRs */ #define MSR_FAM10H_MMIO_CONF_BASE 0xc0010058 #define FAM10H_MMIO_CONF_ENABLE (1<<0) diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index 932f0f86b4b7..c4021b953510 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -5,41 +5,15 @@ #include <asm/irq.h> #include <asm/io.h> -#ifdef ARCH_HAS_NMI_WATCHDOG - -/** - * do_nmi_callback - * - * Check to see if a callback exists and execute it. Return 1 - * if the handler exists and was handled successfully. - */ -int do_nmi_callback(struct pt_regs *regs, int cpu); +#ifdef CONFIG_X86_LOCAL_APIC extern void die_nmi(char *str, struct pt_regs *regs, int do_panic); -extern int check_nmi_watchdog(void); -#if !defined(CONFIG_LOCKUP_DETECTOR) -extern int nmi_watchdog_enabled; -#endif extern int avail_to_resrv_perfctr_nmi_bit(unsigned int); extern int reserve_perfctr_nmi(unsigned int); extern void release_perfctr_nmi(unsigned int); extern int reserve_evntsel_nmi(unsigned int); extern void release_evntsel_nmi(unsigned int); -extern void setup_apic_nmi_watchdog(void *); -extern void stop_apic_nmi_watchdog(void *); -extern void disable_timer_nmi_watchdog(void); -extern void enable_timer_nmi_watchdog(void); -extern int nmi_watchdog_tick(struct pt_regs *regs, unsigned reason); -extern void cpu_nmi_set_wd_enabled(void); - -extern atomic_t nmi_active; -extern unsigned int nmi_watchdog; -#define NMI_NONE 0 -#define NMI_IO_APIC 1 -#define NMI_LOCAL_APIC 2 -#define NMI_INVALID 3 - struct ctl_table; extern int proc_nmi_enabled(struct ctl_table *, int , void __user *, size_t *, loff_t *); @@ -47,33 +21,8 @@ extern int unknown_nmi_panic; void arch_trigger_all_cpu_backtrace(void); #define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace - -static inline void localise_nmi_watchdog(void) -{ - if (nmi_watchdog == NMI_IO_APIC) - nmi_watchdog = NMI_LOCAL_APIC; -} - -/* check if nmi_watchdog is active (ie was specified at boot) */ -static inline int nmi_watchdog_active(void) -{ - /* - * actually it should be: - * return (nmi_watchdog == NMI_LOCAL_APIC || - * nmi_watchdog == NMI_IO_APIC) - * but since they are power of two we could use a - * cheaper way --cvg - */ - return nmi_watchdog & (NMI_LOCAL_APIC | NMI_IO_APIC); -} #endif -void lapic_watchdog_stop(void); -int lapic_watchdog_init(unsigned nmi_hz); -int lapic_wd_event(unsigned nmi_hz); -unsigned lapic_adjust_nmi_hz(unsigned hz); -void disable_lapic_nmi_watchdog(void); -void enable_lapic_nmi_watchdog(void); void stop_nmi(void); void restart_nmi(void); diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index 550e26b1dbb3..d9d4dae305f6 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -125,7 +125,6 @@ union cpuid10_edx { #define IBS_OP_MAX_CNT_EXT 0x007FFFFFULL /* not a register bit mask */ #ifdef CONFIG_PERF_EVENTS -extern void init_hw_perf_events(void); extern void perf_events_lapic_init(void); #define PERF_EVENT_INDEX_OFFSET 0 @@ -156,7 +155,6 @@ extern unsigned long perf_misc_flags(struct pt_regs *regs); } #else -static inline void init_hw_perf_events(void) { } static inline void perf_events_lapic_init(void) { } #endif diff --git a/arch/x86/include/asm/perf_event_p4.h b/arch/x86/include/asm/perf_event_p4.h index a70cd216be5d..295e2ff18a6a 100644 --- a/arch/x86/include/asm/perf_event_p4.h +++ b/arch/x86/include/asm/perf_event_p4.h @@ -744,14 +744,6 @@ enum P4_ESCR_EMASKS { }; /* - * P4 PEBS specifics (Replay Event only) - * - * Format (bits): - * 0-6: metric from P4_PEBS_METRIC enum - * 7 : reserved - * 8 : reserved - * 9-11 : reserved - * * Note we have UOP and PEBS bits reserved for now * just in case if we will need them once */ @@ -788,5 +780,60 @@ enum P4_PEBS_METRIC { P4_PEBS_METRIC__max }; +/* + * Notes on internal configuration of ESCR+CCCR tuples + * + * Since P4 has quite the different architecture of + * performance registers in compare with "architectural" + * once and we have on 64 bits to keep configuration + * of performance event, the following trick is used. + * + * 1) Since both ESCR and CCCR registers have only low + * 32 bits valuable, we pack them into a single 64 bit + * configuration. Low 32 bits of such config correspond + * to low 32 bits of CCCR register and high 32 bits + * correspond to low 32 bits of ESCR register. + * + * 2) The meaning of every bit of such config field can + * be found in Intel SDM but it should be noted that + * we "borrow" some reserved bits for own usage and + * clean them or set to a proper value when we do + * a real write to hardware registers. + * + * 3) The format of bits of config is the following + * and should be either 0 or set to some predefined + * values: + * + * Low 32 bits + * ----------- + * 0-6: P4_PEBS_METRIC enum + * 7-11: reserved + * 12: reserved (Enable) + * 13-15: reserved (ESCR select) + * 16-17: Active Thread + * 18: Compare + * 19: Complement + * 20-23: Threshold + * 24: Edge + * 25: reserved (FORCE_OVF) + * 26: reserved (OVF_PMI_T0) + * 27: reserved (OVF_PMI_T1) + * 28-29: reserved + * 30: reserved (Cascade) + * 31: reserved (OVF) + * + * High 32 bits + * ------------ + * 0: reserved (T1_USR) + * 1: reserved (T1_OS) + * 2: reserved (T0_USR) + * 3: reserved (T0_OS) + * 4: Tag Enable + * 5-8: Tag Value + * 9-24: Event Mask (may use P4_ESCR_EMASK_BIT helper) + * 25-30: enum P4_EVENTS + * 31: reserved (HT thread) + */ + #endif /* PERF_EVENT_P4_H */ diff --git a/arch/x86/include/asm/smpboot_hooks.h b/arch/x86/include/asm/smpboot_hooks.h index 1def60114906..6c22bf353f26 100644 --- a/arch/x86/include/asm/smpboot_hooks.h +++ b/arch/x86/include/asm/smpboot_hooks.h @@ -48,7 +48,6 @@ static inline void __init smpboot_setup_io_apic(void) setup_IO_APIC(); else { nr_ioapics = 0; - localise_nmi_watchdog(); } #endif } diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h index 2b16a2ad23dc..52b5c7ed3608 100644 --- a/arch/x86/include/asm/stacktrace.h +++ b/arch/x86/include/asm/stacktrace.h @@ -7,6 +7,7 @@ #define _ASM_X86_STACKTRACE_H #include <linux/uaccess.h> +#include <linux/ptrace.h> extern int kstack_depth_to_print; @@ -46,7 +47,7 @@ struct stacktrace_ops { }; void dump_trace(struct task_struct *tsk, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, + unsigned long *stack, const struct stacktrace_ops *ops, void *data); #ifdef CONFIG_X86_32 @@ -57,13 +58,39 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, #define get_bp(bp) asm("movq %%rbp, %0" : "=r" (bp) :) #endif +#ifdef CONFIG_FRAME_POINTER +static inline unsigned long +stack_frame(struct task_struct *task, struct pt_regs *regs) +{ + unsigned long bp; + + if (regs) + return regs->bp; + + if (task == current) { + /* Grab bp right from our regs */ + get_bp(bp); + return bp; + } + + /* bp is the last reg pushed by switch_to */ + return *(unsigned long *)task->thread.sp; +} +#else +static inline unsigned long +stack_frame(struct task_struct *task, struct pt_regs *regs) +{ + return 0; +} +#endif + extern void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl); + unsigned long *stack, char *log_lvl); extern void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl); + unsigned long *sp, char *log_lvl); extern unsigned int code_bytes; diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h index 5469630b27f5..fa7b9176b76c 100644 --- a/arch/x86/include/asm/timer.h +++ b/arch/x86/include/asm/timer.h @@ -10,12 +10,6 @@ unsigned long long native_sched_clock(void); extern int recalibrate_cpu_khz(void); -#if defined(CONFIG_X86_32) && defined(CONFIG_X86_IO_APIC) -extern int timer_ack; -#else -# define timer_ack (0) -#endif - extern int no_timer_check; /* Accelerators for sched_clock() diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 5079f24c955a..553d0b0d639b 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -591,17 +591,21 @@ static atomic_t stop_machine_first; static int wrote_text; struct text_poke_params { - void *addr; - const void *opcode; - size_t len; + struct text_poke_param *params; + int nparams; }; static int __kprobes stop_machine_text_poke(void *data) { struct text_poke_params *tpp = data; + struct text_poke_param *p; + int i; if (atomic_dec_and_test(&stop_machine_first)) { - text_poke(tpp->addr, tpp->opcode, tpp->len); + for (i = 0; i < tpp->nparams; i++) { + p = &tpp->params[i]; + text_poke(p->addr, p->opcode, p->len); + } smp_wmb(); /* Make sure other cpus see that this has run */ wrote_text = 1; } else { @@ -610,8 +614,12 @@ static int __kprobes stop_machine_text_poke(void *data) smp_mb(); /* Load wrote_text before following execution */ } - flush_icache_range((unsigned long)tpp->addr, - (unsigned long)tpp->addr + tpp->len); + for (i = 0; i < tpp->nparams; i++) { + p = &tpp->params[i]; + flush_icache_range((unsigned long)p->addr, + (unsigned long)p->addr + p->len); + } + return 0; } @@ -631,10 +639,13 @@ static int __kprobes stop_machine_text_poke(void *data) void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) { struct text_poke_params tpp; + struct text_poke_param p; - tpp.addr = addr; - tpp.opcode = opcode; - tpp.len = len; + p.addr = addr; + p.opcode = opcode; + p.len = len; + tpp.params = &p; + tpp.nparams = 1; atomic_set(&stop_machine_first, 1); wrote_text = 0; /* Use __stop_machine() because the caller already got online_cpus. */ @@ -642,6 +653,26 @@ void *__kprobes text_poke_smp(void *addr, const void *opcode, size_t len) return addr; } +/** + * text_poke_smp_batch - Update instructions on a live kernel on SMP + * @params: an array of text_poke parameters + * @n: the number of elements in params. + * + * Modify multi-byte instruction by using stop_machine() on SMP. Since the + * stop_machine() is heavy task, it is better to aggregate text_poke requests + * and do it once if possible. + * + * Note: Must be called under get_online_cpus() and text_mutex. + */ +void __kprobes text_poke_smp_batch(struct text_poke_param *params, int n) +{ + struct text_poke_params tpp = {.params = params, .nparams = n}; + + atomic_set(&stop_machine_first, 1); + wrote_text = 0; + stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); +} + #if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) #ifdef CONFIG_X86_64 diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile index 910f20b457c4..3966b564ea47 100644 --- a/arch/x86/kernel/apic/Makefile +++ b/arch/x86/kernel/apic/Makefile @@ -3,10 +3,7 @@ # obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o -ifneq ($(CONFIG_HARDLOCKUP_DETECTOR),y) -obj-$(CONFIG_X86_LOCAL_APIC) += nmi.o -endif -obj-$(CONFIG_HARDLOCKUP_DETECTOR) += hw_nmi.o +obj-y += hw_nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_SMP) += ipi.o diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 78218135b48e..fb7657822aad 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -31,7 +31,6 @@ #include <linux/init.h> #include <linux/cpu.h> #include <linux/dmi.h> -#include <linux/nmi.h> #include <linux/smp.h> #include <linux/mm.h> @@ -799,11 +798,7 @@ void __init setup_boot_APIC_clock(void) * PIT/HPET going. Otherwise register lapic as a dummy * device. */ - if (nmi_watchdog != NMI_IO_APIC) - lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; - else - pr_warning("APIC timer registered as dummy," - " due to nmi_watchdog=%d!\n", nmi_watchdog); + lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; /* Setup the lapic or request the broadcast */ setup_APIC_timer(); @@ -1387,7 +1382,6 @@ void __cpuinit end_local_APIC_setup(void) } #endif - setup_apic_nmi_watchdog(NULL); apic_pm_activate(); /* @@ -1758,17 +1752,10 @@ int __init APIC_init_uniprocessor(void) setup_IO_APIC(); else { nr_ioapics = 0; - localise_nmi_watchdog(); } -#else - localise_nmi_watchdog(); #endif x86_init.timers.setup_percpu_clockev(); -#ifdef CONFIG_X86_64 - check_nmi_watchdog(); -#endif - return 0; } diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c index 62f6e1e55b90..72ec29e1ae06 100644 --- a/arch/x86/kernel/apic/hw_nmi.c +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -17,20 +17,31 @@ #include <linux/nmi.h> #include <linux/module.h> +#ifdef CONFIG_HARDLOCKUP_DETECTOR u64 hw_nmi_get_sample_period(void) { return (u64)(cpu_khz) * 1000 * 60; } +#endif -#ifdef ARCH_HAS_NMI_WATCHDOG - +#ifdef arch_trigger_all_cpu_backtrace /* For reliability, we're prepared to waste bits here. */ static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly; +/* "in progress" flag of arch_trigger_all_cpu_backtrace */ +static unsigned long backtrace_flag; + void arch_trigger_all_cpu_backtrace(void) { int i; + if (test_and_set_bit(0, &backtrace_flag)) + /* + * If there is already a trigger_all_cpu_backtrace() in progress + * (backtrace_flag == 1), don't output double cpu dump infos. + */ + return; + cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask); printk(KERN_INFO "sending NMI to all CPUs:\n"); @@ -42,6 +53,9 @@ void arch_trigger_all_cpu_backtrace(void) break; mdelay(1); } + + clear_bit(0, &backtrace_flag); + smp_mb__after_clear_bit(); } static int __kprobes @@ -50,7 +64,7 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self, { struct die_args *args = __args; struct pt_regs *regs; - int cpu = smp_processor_id(); + int cpu; switch (cmd) { case DIE_NMI: @@ -62,6 +76,7 @@ arch_trigger_all_cpu_backtrace_handler(struct notifier_block *self, } regs = args->regs; + cpu = smp_processor_id(); if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { static arch_spinlock_t lock = __ARCH_SPIN_LOCK_UNLOCKED; @@ -91,18 +106,3 @@ static int __init register_trigger_all_cpu_backtrace(void) } early_initcall(register_trigger_all_cpu_backtrace); #endif - -/* STUB calls to mimic old nmi_watchdog behaviour */ -#if defined(CONFIG_X86_LOCAL_APIC) -unsigned int nmi_watchdog = NMI_NONE; -EXPORT_SYMBOL(nmi_watchdog); -void acpi_nmi_enable(void) { return; } -void acpi_nmi_disable(void) { return; } -#endif -atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ -EXPORT_SYMBOL(nmi_active); -int unknown_nmi_panic; -void cpu_nmi_set_wd_enabled(void) { return; } -void stop_apic_nmi_watchdog(void *unused) { return; } -void setup_apic_nmi_watchdog(void *unused) { return; } -int __init check_nmi_watchdog(void) { return 0; } diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index fadcd743a74f..16c2db8750a2 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -54,7 +54,6 @@ #include <asm/dma.h> #include <asm/timer.h> #include <asm/i8259.h> -#include <asm/nmi.h> #include <asm/msidef.h> #include <asm/hypertransport.h> #include <asm/setup.h> @@ -2642,24 +2641,6 @@ static void lapic_register_intr(int irq) "edge"); } -static void __init setup_nmi(void) -{ - /* - * Dirty trick to enable the NMI watchdog ... - * We put the 8259A master into AEOI mode and - * unmask on all local APICs LVT0 as NMI. - * - * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire') - * is from Maciej W. Rozycki - so we do not have to EOI from - * the NMI handler or the timer interrupt. - */ - apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ..."); - - enable_NMI_through_LVT0(); - - apic_printk(APIC_VERBOSE, " done.\n"); -} - /* * This looks a bit hackish but it's about the only one way of sending * a few INTA cycles to 8259As and any associated glue logic. ICR does @@ -2765,15 +2746,6 @@ static inline void __init check_timer(void) */ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT); legacy_pic->init(1); -#ifdef CONFIG_X86_32 - { - unsigned int ver; - - ver = apic_read(APIC_LVR); - ver = GET_APIC_VERSION(ver); - timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver)); - } -#endif pin1 = find_isa_irq_pin(0, mp_INT); apic1 = find_isa_irq_apic(0, mp_INT); @@ -2821,10 +2793,6 @@ static inline void __init check_timer(void) unmask_ioapic(cfg); } if (timer_irq_works()) { - if (nmi_watchdog == NMI_IO_APIC) { - setup_nmi(); - legacy_pic->unmask(0); - } if (disable_timer_pin_1 > 0) clear_IO_APIC_pin(0, pin1); goto out; @@ -2850,11 +2818,6 @@ static inline void __init check_timer(void) if (timer_irq_works()) { apic_printk(APIC_QUIET, KERN_INFO "....... works.\n"); timer_through_8259 = 1; - if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->mask(0); - setup_nmi(); - legacy_pic->unmask(0); - } goto out; } /* @@ -2866,15 +2829,6 @@ static inline void __init check_timer(void) apic_printk(APIC_QUIET, KERN_INFO "....... failed.\n"); } - if (nmi_watchdog == NMI_IO_APIC) { - apic_printk(APIC_QUIET, KERN_WARNING "timer doesn't work " - "through the IO-APIC - disabling NMI Watchdog!\n"); - nmi_watchdog = NMI_NONE; - } -#ifdef CONFIG_X86_32 - timer_ack = 0; -#endif - apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...\n"); diff --git a/arch/x86/kernel/apic/nmi.c b/arch/x86/kernel/apic/nmi.c deleted file mode 100644 index c90041ccb742..000000000000 --- a/arch/x86/kernel/apic/nmi.c +++ /dev/null @@ -1,567 +0,0 @@ -/* - * NMI watchdog support on APIC systems - * - * Started by Ingo Molnar <mingo@redhat.com> - * - * Fixes: - * Mikael Pettersson : AMD K7 support for local APIC NMI watchdog. - * Mikael Pettersson : Power Management for local APIC NMI watchdog. - * Mikael Pettersson : Pentium 4 support for local APIC NMI watchdog. - * Pavel Machek and - * Mikael Pettersson : PM converted to driver model. Disable/enable API. - */ - -#include <asm/apic.h> - -#include <linux/nmi.h> -#include <linux/mm.h> -#include <linux/delay.h> -#include <linux/interrupt.h> -#include <linux/module.h> -#include <linux/slab.h> -#include <linux/sysdev.h> -#include <linux/sysctl.h> -#include <linux/percpu.h> -#include <linux/kprobes.h> -#include <linux/cpumask.h> -#include <linux/kernel_stat.h> -#include <linux/kdebug.h> -#include <linux/smp.h> - -#include <asm/i8259.h> -#include <asm/io_apic.h> -#include <asm/proto.h> -#include <asm/timer.h> - -#include <asm/mce.h> - -#include <asm/mach_traps.h> - -int unknown_nmi_panic; -int nmi_watchdog_enabled; - -/* For reliability, we're prepared to waste bits here. */ -static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly; - -/* nmi_active: - * >0: the lapic NMI watchdog is active, but can be disabled - * <0: the lapic NMI watchdog has not been set up, and cannot - * be enabled - * 0: the lapic NMI watchdog is disabled, but can be enabled - */ -atomic_t nmi_active = ATOMIC_INIT(0); /* oprofile uses this */ -EXPORT_SYMBOL(nmi_active); - -unsigned int nmi_watchdog = NMI_NONE; -EXPORT_SYMBOL(nmi_watchdog); - -static int panic_on_timeout; - -static unsigned int nmi_hz = HZ; -static DEFINE_PER_CPU(short, wd_enabled); -static int endflag __initdata; - -static inline unsigned int get_nmi_count(int cpu) -{ - return per_cpu(irq_stat, cpu).__nmi_count; -} - -static inline int mce_in_progress(void) -{ -#if defined(CONFIG_X86_MCE) - return atomic_read(&mce_entry) > 0; -#endif - return 0; -} - -/* - * Take the local apic timer and PIT/HPET into account. We don't - * know which one is active, when we have highres/dyntick on - */ -static inline unsigned int get_timer_irqs(int cpu) -{ - return per_cpu(irq_stat, cpu).apic_timer_irqs + - per_cpu(irq_stat, cpu).irq0_irqs; -} - -#ifdef CONFIG_SMP -/* - * The performance counters used by NMI_LOCAL_APIC don't trigger when - * the CPU is idle. To make sure the NMI watchdog really ticks on all - * CPUs during the test make them busy. - */ -static __init void nmi_cpu_busy(void *data) -{ - local_irq_enable_in_hardirq(); - /* - * Intentionally don't use cpu_relax here. This is - * to make sure that the performance counter really ticks, - * even if there is a simulator or similar that catches the - * pause instruction. On a real HT machine this is fine because - * all other CPUs are busy with "useless" delay loops and don't - * care if they get somewhat less cycles. - */ - while (endflag == 0) - mb(); -} -#endif - -static void report_broken_nmi(int cpu, unsigned int *prev_nmi_count) -{ - printk(KERN_CONT "\n"); - - printk(KERN_WARNING - "WARNING: CPU#%d: NMI appears to be stuck (%d->%d)!\n", - cpu, prev_nmi_count[cpu], get_nmi_count(cpu)); - - printk(KERN_WARNING - "Please report this to bugzilla.kernel.org,\n"); - printk(KERN_WARNING - "and attach the output of the 'dmesg' command.\n"); - - per_cpu(wd_enabled, cpu) = 0; - atomic_dec(&nmi_active); -} - -static void __acpi_nmi_disable(void *__unused) -{ - apic_write(APIC_LVT0, APIC_DM_NMI | APIC_LVT_MASKED); -} - -int __init check_nmi_watchdog(void) -{ - unsigned int *prev_nmi_count; - int cpu; - - if (!nmi_watchdog_active() || !atomic_read(&nmi_active)) - return 0; - - prev_nmi_count = kmalloc(nr_cpu_ids * sizeof(int), GFP_KERNEL); - if (!prev_nmi_count) - goto error; - - printk(KERN_INFO "Testing NMI watchdog ... "); - -#ifdef CONFIG_SMP - if (nmi_watchdog == NMI_LOCAL_APIC) - smp_call_function(nmi_cpu_busy, (void *)&endflag, 0); -#endif - - for_each_possible_cpu(cpu) - prev_nmi_count[cpu] = get_nmi_count(cpu); - local_irq_enable(); - mdelay((20 * 1000) / nmi_hz); /* wait 20 ticks */ - - for_each_online_cpu(cpu) { - if (!per_cpu(wd_enabled, cpu)) - continue; - if (get_nmi_count(cpu) - prev_nmi_count[cpu] <= 5) - report_broken_nmi(cpu, prev_nmi_count); - } - endflag = 1; - if (!atomic_read(&nmi_active)) { - kfree(prev_nmi_count); - atomic_set(&nmi_active, -1); - goto error; - } - printk("OK.\n"); - - /* - * now that we know it works we can reduce NMI frequency to - * something more reasonable; makes a difference in some configs - */ - if (nmi_watchdog == NMI_LOCAL_APIC) - nmi_hz = lapic_adjust_nmi_hz(1); - - kfree(prev_nmi_count); - return 0; -error: - if (nmi_watchdog == NMI_IO_APIC) { - if (!timer_through_8259) - legacy_pic->mask(0); - on_each_cpu(__acpi_nmi_disable, NULL, 1); - } - -#ifdef CONFIG_X86_32 - timer_ack = 0; -#endif - return -1; -} - -static int __init setup_nmi_watchdog(char *str) -{ - unsigned int nmi; - - if (!strncmp(str, "panic", 5)) { - panic_on_timeout = 1; - str = strchr(str, ','); - if (!str) - return 1; - ++str; - } - - if (!strncmp(str, "lapic", 5)) - nmi_watchdog = NMI_LOCAL_APIC; - else if (!strncmp(str, "ioapic", 6)) - nmi_watchdog = NMI_IO_APIC; - else { - get_option(&str, &nmi); - if (nmi >= NMI_INVALID) - return 0; - nmi_watchdog = nmi; - } - - return 1; -} -__setup("nmi_watchdog=", setup_nmi_watchdog); - -/* - * Suspend/resume support - */ -#ifdef CONFIG_PM - -static int nmi_pm_active; /* nmi_active before suspend */ - -static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state) -{ - /* only CPU0 goes here, other CPUs should be offline */ - nmi_pm_active = atomic_read(&nmi_active); - stop_apic_nmi_watchdog(NULL); - BUG_ON(atomic_read(&nmi_active) != 0); - return 0; -} - -static int lapic_nmi_resume(struct sys_device *dev) -{ - /* only CPU0 goes here, other CPUs should be offline */ - if (nmi_pm_active > 0) { - setup_apic_nmi_watchdog(NULL); - touch_nmi_watchdog(); - } - return 0; -} - -static struct sysdev_class nmi_sysclass = { - .name = "lapic_nmi", - .resume = lapic_nmi_resume, - .suspend = lapic_nmi_suspend, -}; - -static struct sys_device device_lapic_nmi = { - .id = 0, - .cls = &nmi_sysclass, -}; - -static int __init init_lapic_nmi_sysfs(void) -{ - int error; - - /* - * should really be a BUG_ON but b/c this is an - * init call, it just doesn't work. -dcz - */ - if (nmi_watchdog != NMI_LOCAL_APIC) - return 0; - - if (atomic_read(&nmi_active) < 0) - return 0; - - error = sysdev_class_register(&nmi_sysclass); - if (!error) - error = sysdev_register(&device_lapic_nmi); - return error; -} - -/* must come after the local APIC's device_initcall() */ -late_initcall(init_lapic_nmi_sysfs); - -#endif /* CONFIG_PM */ - -static void __acpi_nmi_enable(void *__unused) -{ - apic_write(APIC_LVT0, APIC_DM_NMI); -} - -/* - * Enable timer based NMIs on all CPUs: - */ -void acpi_nmi_enable(void) -{ - if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC) - on_each_cpu(__acpi_nmi_enable, NULL, 1); -} - -/* - * Disable timer based NMIs on all CPUs: - */ -void acpi_nmi_disable(void) -{ - if (atomic_read(&nmi_active) && nmi_watchdog == NMI_IO_APIC) - on_each_cpu(__acpi_nmi_disable, NULL, 1); -} - -/* - * This function is called as soon the LAPIC NMI watchdog driver has everything - * in place and it's ready to check if the NMIs belong to the NMI watchdog - */ -void cpu_nmi_set_wd_enabled(void) -{ - __get_cpu_var(wd_enabled) = 1; -} - -void setup_apic_nmi_watchdog(void *unused) -{ - if (__get_cpu_var(wd_enabled)) - return; - - /* cheap hack to support suspend/resume */ - /* if cpu0 is not active neither should the other cpus */ - if (smp_processor_id() != 0 && atomic_read(&nmi_active) <= 0) - return; - - switch (nmi_watchdog) { - case NMI_LOCAL_APIC: - if (lapic_watchdog_init(nmi_hz) < 0) { - __get_cpu_var(wd_enabled) = 0; - return; - } - /* FALL THROUGH */ - case NMI_IO_APIC: - __get_cpu_var(wd_enabled) = 1; - atomic_inc(&nmi_active); - } -} - -void stop_apic_nmi_watchdog(void *unused) -{ - /* only support LOCAL and IO APICs for now */ - if (!nmi_watchdog_active()) - return; - if (__get_cpu_var(wd_enabled) == 0) - return; - if (nmi_watchdog == NMI_LOCAL_APIC) - lapic_watchdog_stop(); - else - __acpi_nmi_disable(NULL); - __get_cpu_var(wd_enabled) = 0; - atomic_dec(&nmi_active); -} - -/* - * the best way to detect whether a CPU has a 'hard lockup' problem - * is to check it's local APIC timer IRQ counts. If they are not - * changing then that CPU has some problem. - * - * as these watchdog NMI IRQs are generated on every CPU, we only - * have to check the current processor. - * - * since NMIs don't listen to _any_ locks, we have to be extremely - * careful not to rely on unsafe variables. The printk might lock - * up though, so we have to break up any console locks first ... - * [when there will be more tty-related locks, break them up here too!] - */ - -static DEFINE_PER_CPU(unsigned, last_irq_sum); -static DEFINE_PER_CPU(long, alert_counter); -static DEFINE_PER_CPU(int, nmi_touch); - -void touch_nmi_watchdog(void) -{ - if (nmi_watchdog_active()) { - unsigned cpu; - - /* - * Tell other CPUs to reset their alert counters. We cannot - * do it ourselves because the alert count increase is not - * atomic. - */ - for_each_present_cpu(cpu) { - if (per_cpu(nmi_touch, cpu) != 1) - per_cpu(nmi_touch, cpu) = 1; - } - } - - /* - * Tickle the softlockup detector too: - */ - touch_softlockup_watchdog(); -} -EXPORT_SYMBOL(touch_nmi_watchdog); - -notrace __kprobes int -nmi_watchdog_tick(struct pt_regs *regs, unsigned reason) -{ - /* - * Since current_thread_info()-> is always on the stack, and we - * always switch the stack NMI-atomically, it's safe to use - * smp_processor_id(). - */ - unsigned int sum; - int touched = 0; - int cpu = smp_processor_id(); - int rc = 0; - - sum = get_timer_irqs(cpu); - - if (__get_cpu_var(nmi_touch)) { - __get_cpu_var(nmi_touch) = 0; - touched = 1; - } - - /* We can be called before check_nmi_watchdog, hence NULL check. */ - if (cpumask_test_cpu(cpu, to_cpumask(backtrace_mask))) { - static DEFINE_RAW_SPINLOCK(lock); /* Serialise the printks */ - - raw_spin_lock(&lock); - printk(KERN_WARNING "NMI backtrace for cpu %d\n", cpu); - show_regs(regs); - dump_stack(); - raw_spin_unlock(&lock); - cpumask_clear_cpu(cpu, to_cpumask(backtrace_mask)); - - rc = 1; - } - - /* Could check oops_in_progress here too, but it's safer not to */ - if (mce_in_progress()) - touched = 1; - - /* if the none of the timers isn't firing, this cpu isn't doing much */ - if (!touched && __get_cpu_var(last_irq_sum) == sum) { - /* - * Ayiee, looks like this CPU is stuck ... - * wait a few IRQs (5 seconds) before doing the oops ... - */ - __this_cpu_inc(alert_counter); - if (__this_cpu_read(alert_counter) == 5 * nmi_hz) - /* - * die_nmi will return ONLY if NOTIFY_STOP happens.. - */ - die_nmi("BUG: NMI Watchdog detected LOCKUP", - regs, panic_on_timeout); - } else { - __get_cpu_var(last_irq_sum) = sum; - __this_cpu_write(alert_counter, 0); - } - - /* see if the nmi watchdog went off */ - if (!__get_cpu_var(wd_enabled)) - return rc; - switch (nmi_watchdog) { - case NMI_LOCAL_APIC: - rc |= lapic_wd_event(nmi_hz); - break; - case NMI_IO_APIC: - /* - * don't know how to accurately check for this. - * just assume it was a watchdog timer interrupt - * This matches the old behaviour. - */ - rc = 1; - break; - } - return rc; -} - -#ifdef CONFIG_SYSCTL - -static void enable_ioapic_nmi_watchdog_single(void *unused) -{ - __get_cpu_var(wd_enabled) = 1; - atomic_inc(&nmi_active); - __acpi_nmi_enable(NULL); -} - -static void enable_ioapic_nmi_watchdog(void) -{ - on_each_cpu(enable_ioapic_nmi_watchdog_single, NULL, 1); - touch_nmi_watchdog(); -} - -static void disable_ioapic_nmi_watchdog(void) -{ - on_each_cpu(stop_apic_nmi_watchdog, NULL, 1); -} - -static int __init setup_unknown_nmi_panic(char *str) -{ - unknown_nmi_panic = 1; - return 1; -} -__setup("unknown_nmi_panic", setup_unknown_nmi_panic); - -static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) -{ - unsigned char reason = get_nmi_reason(); - char buf[64]; - - sprintf(buf, "NMI received for unknown reason %02x\n", reason); - die_nmi(buf, regs, 1); /* Always panic here */ - return 0; -} - -/* - * proc handler for /proc/sys/kernel/nmi - */ -int proc_nmi_enabled(struct ctl_table *table, int write, - void __user *buffer, size_t *length, loff_t *ppos) -{ - int old_state; - - nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0; - old_state = nmi_watchdog_enabled; - proc_dointvec(table, write, buffer, length, ppos); - if (!!old_state == !!nmi_watchdog_enabled) - return 0; - - if (atomic_read(&nmi_active) < 0 || !nmi_watchdog_active()) { - printk(KERN_WARNING - "NMI watchdog is permanently disabled\n"); - return -EIO; - } - - if (nmi_watchdog == NMI_LOCAL_APIC) { - if (nmi_watchdog_enabled) - enable_lapic_nmi_watchdog(); - else - disable_lapic_nmi_watchdog(); - } else if (nmi_watchdog == NMI_IO_APIC) { - if (nmi_watchdog_enabled) - enable_ioapic_nmi_watchdog(); - else - disable_ioapic_nmi_watchdog(); - } else { - printk(KERN_WARNING - "NMI watchdog doesn't know what hardware to touch\n"); - return -EIO; - } - return 0; -} - -#endif /* CONFIG_SYSCTL */ - -int do_nmi_callback(struct pt_regs *regs, int cpu) -{ -#ifdef CONFIG_SYSCTL - if (unknown_nmi_panic) - return unknown_nmi_panic_callback(regs, cpu); -#endif - return 0; -} - -void arch_trigger_all_cpu_backtrace(void) -{ - int i; - - cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask); - - printk(KERN_INFO "sending NMI to all CPUs:\n"); - apic->send_IPI_all(NMI_VECTOR); - - /* Wait for up to 10 seconds for all CPUs to do the backtrace */ - for (i = 0; i < 10 * 1000; i++) { - if (cpumask_empty(to_cpumask(backtrace_mask))) - break; - mdelay(1); - } -} diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 4b68bda30938..1d59834396bd 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -894,7 +894,6 @@ void __init identify_boot_cpu(void) #else vgetcpu_set_mode(); #endif - init_hw_perf_events(); } void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 6d75b9145b13..0a360d146596 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -330,9 +330,6 @@ static bool reserve_pmc_hardware(void) { int i; - if (nmi_watchdog == NMI_LOCAL_APIC) - disable_lapic_nmi_watchdog(); - for (i = 0; i < x86_pmu.num_counters; i++) { if (!reserve_perfctr_nmi(x86_pmu.perfctr + i)) goto perfctr_fail; @@ -355,9 +352,6 @@ perfctr_fail: for (i--; i >= 0; i--) release_perfctr_nmi(x86_pmu.perfctr + i); - if (nmi_watchdog == NMI_LOCAL_APIC) - enable_lapic_nmi_watchdog(); - return false; } @@ -369,9 +363,6 @@ static void release_pmc_hardware(void) release_perfctr_nmi(x86_pmu.perfctr + i); release_evntsel_nmi(x86_pmu.eventsel + i); } - - if (nmi_watchdog == NMI_LOCAL_APIC) - enable_lapic_nmi_watchdog(); } #else @@ -384,15 +375,53 @@ static void release_pmc_hardware(void) {} static bool check_hw_exists(void) { u64 val, val_new = 0; - int ret = 0; + int i, reg, ret = 0; + + /* + * Check to see if the BIOS enabled any of the counters, if so + * complain and bail. + */ + for (i = 0; i < x86_pmu.num_counters; i++) { + reg = x86_pmu.eventsel + i; + ret = rdmsrl_safe(reg, &val); + if (ret) + goto msr_fail; + if (val & ARCH_PERFMON_EVENTSEL_ENABLE) + goto bios_fail; + } + if (x86_pmu.num_counters_fixed) { + reg = MSR_ARCH_PERFMON_FIXED_CTR_CTRL; + ret = rdmsrl_safe(reg, &val); + if (ret) + goto msr_fail; + for (i = 0; i < x86_pmu.num_counters_fixed; i++) { + if (val & (0x03 << i*4)) + goto bios_fail; + } + } + + /* + * Now write a value and read it back to see if it matches, + * this is needed to detect certain hardware emulators (qemu/kvm) + * that don't trap on the MSR access and always return 0s. + */ val = 0xabcdUL; - ret |= checking_wrmsrl(x86_pmu.perfctr, val); + ret = checking_wrmsrl(x86_pmu.perfctr, val); ret |= rdmsrl_safe(x86_pmu.perfctr, &val_new); if (ret || val != val_new) - return false; + goto msr_fail; return true; + +bios_fail: + printk(KERN_CONT "Broken BIOS detected, using software events only.\n"); + printk(KERN_ERR FW_BUG "the BIOS has corrupted hw-PMU resources (MSR %x is %Lx)\n", reg, val); + return false; + +msr_fail: + printk(KERN_CONT "Broken PMU hardware detected, using software events only.\n"); + return false; } static void reserve_ds_buffers(void); @@ -451,7 +480,7 @@ static int x86_setup_perfctr(struct perf_event *event) struct hw_perf_event *hwc = &event->hw; u64 config; - if (!hwc->sample_period) { + if (!is_sampling_event(event)) { hwc->sample_period = x86_pmu.max_period; hwc->last_period = hwc->sample_period; local64_set(&hwc->period_left, hwc->sample_period); @@ -1362,7 +1391,7 @@ static void __init pmu_check_apic(void) pr_info("no hardware sampling interrupt available.\n"); } -void __init init_hw_perf_events(void) +int __init init_hw_perf_events(void) { struct event_constraint *c; int err; @@ -1377,20 +1406,18 @@ void __init init_hw_perf_events(void) err = amd_pmu_init(); break; default: - return; + return 0; } if (err != 0) { pr_cont("no PMU driver, software events only.\n"); - return; + return 0; } pmu_check_apic(); /* sanity check that the hardware exists or is emulated */ - if (!check_hw_exists()) { - pr_cont("Broken PMU hardware detected, software events only.\n"); - return; - } + if (!check_hw_exists()) + return 0; pr_cont("%s PMU driver.\n", x86_pmu.name); @@ -1438,9 +1465,12 @@ void __init init_hw_perf_events(void) pr_info("... fixed-purpose events: %d\n", x86_pmu.num_counters_fixed); pr_info("... event mask: %016Lx\n", x86_pmu.intel_ctrl); - perf_pmu_register(&pmu); + perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW); perf_cpu_notifier(x86_pmu_notifier); + + return 0; } +early_initcall(init_hw_perf_events); static inline void x86_pmu_read(struct perf_event *event) { @@ -1686,7 +1716,7 @@ perf_callchain_kernel(struct perf_callchain_entry *entry, struct pt_regs *regs) perf_callchain_store(entry, regs->ip); - dump_trace(NULL, regs, NULL, regs->bp, &backtrace_ops, entry); + dump_trace(NULL, regs, NULL, &backtrace_ops, entry); } #ifdef CONFIG_COMPAT diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index e421b8cd6944..67e2202a6039 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -1,7 +1,5 @@ #ifdef CONFIG_CPU_SUP_AMD -static DEFINE_RAW_SPINLOCK(amd_nb_lock); - static __initconst const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] @@ -275,7 +273,7 @@ done: return &emptyconstraint; } -static struct amd_nb *amd_alloc_nb(int cpu, int nb_id) +static struct amd_nb *amd_alloc_nb(int cpu) { struct amd_nb *nb; int i; @@ -285,7 +283,7 @@ static struct amd_nb *amd_alloc_nb(int cpu, int nb_id) if (!nb) return NULL; - nb->nb_id = nb_id; + nb->nb_id = -1; /* * initialize all possible NB constraints @@ -306,7 +304,7 @@ static int amd_pmu_cpu_prepare(int cpu) if (boot_cpu_data.x86_max_cores < 2) return NOTIFY_OK; - cpuc->amd_nb = amd_alloc_nb(cpu, -1); + cpuc->amd_nb = amd_alloc_nb(cpu); if (!cpuc->amd_nb) return NOTIFY_BAD; @@ -325,8 +323,6 @@ static void amd_pmu_cpu_starting(int cpu) nb_id = amd_get_nb_id(cpu); WARN_ON_ONCE(nb_id == BAD_APICID); - raw_spin_lock(&amd_nb_lock); - for_each_online_cpu(i) { nb = per_cpu(cpu_hw_events, i).amd_nb; if (WARN_ON_ONCE(!nb)) @@ -341,8 +337,6 @@ static void amd_pmu_cpu_starting(int cpu) cpuc->amd_nb->nb_id = nb_id; cpuc->amd_nb->refcnt++; - - raw_spin_unlock(&amd_nb_lock); } static void amd_pmu_cpu_dead(int cpu) @@ -354,8 +348,6 @@ static void amd_pmu_cpu_dead(int cpu) cpuhw = &per_cpu(cpu_hw_events, cpu); - raw_spin_lock(&amd_nb_lock); - if (cpuhw->amd_nb) { struct amd_nb *nb = cpuhw->amd_nb; @@ -364,8 +356,6 @@ static void amd_pmu_cpu_dead(int cpu) cpuhw->amd_nb = NULL; } - - raw_spin_unlock(&amd_nb_lock); } static __initconst const struct x86_pmu amd_pmu = { diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index c8f5c088cad1..24e390e40f2e 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -816,6 +816,32 @@ static int intel_pmu_hw_config(struct perf_event *event) if (ret) return ret; + if (event->attr.precise_ip && + (event->hw.config & X86_RAW_EVENT_MASK) == 0x003c) { + /* + * Use an alternative encoding for CPU_CLK_UNHALTED.THREAD_P + * (0x003c) so that we can use it with PEBS. + * + * The regular CPU_CLK_UNHALTED.THREAD_P event (0x003c) isn't + * PEBS capable. However we can use INST_RETIRED.ANY_P + * (0x00c0), which is a PEBS capable event, to get the same + * count. + * + * INST_RETIRED.ANY_P counts the number of cycles that retires + * CNTMASK instructions. By setting CNTMASK to a value (16) + * larger than the maximum number of instructions that can be + * retired per cycle (4) and then inverting the condition, we + * count all cycles that retire 16 or less instructions, which + * is every cycle. + * + * Thereby we gain a PEBS capable cycle counter. + */ + u64 alt_config = 0x108000c0; /* INST_RETIRED.TOTAL_CYCLES */ + + alt_config |= (event->hw.config & ~X86_RAW_EVENT_MASK); + event->hw.config = alt_config; + } + if (event->attr.type != PERF_TYPE_RAW) return 0; diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index d9f4ff8fcd69..d5a236615501 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -16,32 +16,12 @@ #include <linux/kernel.h> #include <linux/bitops.h> #include <linux/smp.h> -#include <linux/nmi.h> +#include <asm/nmi.h> #include <linux/kprobes.h> #include <asm/apic.h> #include <asm/perf_event.h> -struct nmi_watchdog_ctlblk { - unsigned int cccr_msr; - unsigned int perfctr_msr; /* the MSR to reset in NMI handler */ - unsigned int evntsel_msr; /* the MSR to select the events to handle */ -}; - -/* Interface defining a CPU specific perfctr watchdog */ -struct wd_ops { - int (*reserve)(void); - void (*unreserve)(void); - int (*setup)(unsigned nmi_hz); - void (*rearm)(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz); - void (*stop)(void); - unsigned perfctr; - unsigned evntsel; - u64 checkbit; -}; - -static const struct wd_ops *wd_ops; - /* * this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's * offset from MSR_P4_BSU_ESCR0. @@ -60,8 +40,6 @@ static const struct wd_ops *wd_ops; static DECLARE_BITMAP(perfctr_nmi_owner, NMI_MAX_COUNTER_BITS); static DECLARE_BITMAP(evntsel_nmi_owner, NMI_MAX_COUNTER_BITS); -static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk); - /* converts an msr to an appropriate reservation bit */ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr) { @@ -172,623 +150,3 @@ void release_evntsel_nmi(unsigned int msr) clear_bit(counter, evntsel_nmi_owner); } EXPORT_SYMBOL(release_evntsel_nmi); - -void disable_lapic_nmi_watchdog(void) -{ - BUG_ON(nmi_watchdog != NMI_LOCAL_APIC); - - if (atomic_read(&nmi_active) <= 0) - return; - - on_each_cpu(stop_apic_nmi_watchdog, NULL, 1); - - if (wd_ops) - wd_ops->unreserve(); - - BUG_ON(atomic_read(&nmi_active) != 0); -} - -void enable_lapic_nmi_watchdog(void) -{ - BUG_ON(nmi_watchdog != NMI_LOCAL_APIC); - - /* are we already enabled */ - if (atomic_read(&nmi_active) != 0) - return; - - /* are we lapic aware */ - if (!wd_ops) - return; - if (!wd_ops->reserve()) { - printk(KERN_ERR "NMI watchdog: cannot reserve perfctrs\n"); - return; - } - - on_each_cpu(setup_apic_nmi_watchdog, NULL, 1); - touch_nmi_watchdog(); -} - -/* - * Activate the NMI watchdog via the local APIC. - */ - -static unsigned int adjust_for_32bit_ctr(unsigned int hz) -{ - u64 counter_val; - unsigned int retval = hz; - - /* - * On Intel CPUs with P6/ARCH_PERFMON only 32 bits in the counter - * are writable, with higher bits sign extending from bit 31. - * So, we can only program the counter with 31 bit values and - * 32nd bit should be 1, for 33.. to be 1. - * Find the appropriate nmi_hz - */ - counter_val = (u64)cpu_khz * 1000; - do_div(counter_val, retval); - if (counter_val > 0x7fffffffULL) { - u64 count = (u64)cpu_khz * 1000; - do_div(count, 0x7fffffffUL); - retval = count + 1; - } - return retval; -} - -static void write_watchdog_counter(unsigned int perfctr_msr, - const char *descr, unsigned nmi_hz) -{ - u64 count = (u64)cpu_khz * 1000; - - do_div(count, nmi_hz); - if (descr) - pr_debug("setting %s to -0x%08Lx\n", descr, count); - wrmsrl(perfctr_msr, 0 - count); -} - -static void write_watchdog_counter32(unsigned int perfctr_msr, - const char *descr, unsigned nmi_hz) -{ - u64 count = (u64)cpu_khz * 1000; - - do_div(count, nmi_hz); - if (descr) - pr_debug("setting %s to -0x%08Lx\n", descr, count); - wrmsr(perfctr_msr, (u32)(-count), 0); -} - -/* - * AMD K7/K8/Family10h/Family11h support. - * AMD keeps this interface nicely stable so there is not much variety - */ -#define K7_EVNTSEL_ENABLE (1 << 22) -#define K7_EVNTSEL_INT (1 << 20) -#define K7_EVNTSEL_OS (1 << 17) -#define K7_EVNTSEL_USR (1 << 16) -#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING 0x76 -#define K7_NMI_EVENT K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING - -static int setup_k7_watchdog(unsigned nmi_hz) -{ - unsigned int perfctr_msr, evntsel_msr; - unsigned int evntsel; - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - - perfctr_msr = wd_ops->perfctr; - evntsel_msr = wd_ops->evntsel; - - wrmsrl(perfctr_msr, 0UL); - - evntsel = K7_EVNTSEL_INT - | K7_EVNTSEL_OS - | K7_EVNTSEL_USR - | K7_NMI_EVENT; - - /* setup the timer */ - wrmsr(evntsel_msr, evntsel, 0); - write_watchdog_counter(perfctr_msr, "K7_PERFCTR0", nmi_hz); - - /* initialize the wd struct before enabling */ - wd->perfctr_msr = perfctr_msr; - wd->evntsel_msr = evntsel_msr; - wd->cccr_msr = 0; /* unused */ - - /* ok, everything is initialized, announce that we're set */ - cpu_nmi_set_wd_enabled(); - - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= K7_EVNTSEL_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); - - return 1; -} - -static void single_msr_stop_watchdog(void) -{ - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - - wrmsr(wd->evntsel_msr, 0, 0); -} - -static int single_msr_reserve(void) -{ - if (!reserve_perfctr_nmi(wd_ops->perfctr)) - return 0; - - if (!reserve_evntsel_nmi(wd_ops->evntsel)) { - release_perfctr_nmi(wd_ops->perfctr); - return 0; - } - return 1; -} - -static void single_msr_unreserve(void) -{ - release_evntsel_nmi(wd_ops->evntsel); - release_perfctr_nmi(wd_ops->perfctr); -} - -static void __kprobes -single_msr_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) -{ - /* start the cycle over again */ - write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz); -} - -static const struct wd_ops k7_wd_ops = { - .reserve = single_msr_reserve, - .unreserve = single_msr_unreserve, - .setup = setup_k7_watchdog, - .rearm = single_msr_rearm, - .stop = single_msr_stop_watchdog, - .perfctr = MSR_K7_PERFCTR0, - .evntsel = MSR_K7_EVNTSEL0, - .checkbit = 1ULL << 47, -}; - -/* - * Intel Model 6 (PPro+,P2,P3,P-M,Core1) - */ -#define P6_EVNTSEL0_ENABLE (1 << 22) -#define P6_EVNTSEL_INT (1 << 20) -#define P6_EVNTSEL_OS (1 << 17) -#define P6_EVNTSEL_USR (1 << 16) -#define P6_EVENT_CPU_CLOCKS_NOT_HALTED 0x79 -#define P6_NMI_EVENT P6_EVENT_CPU_CLOCKS_NOT_HALTED - -static int setup_p6_watchdog(unsigned nmi_hz) -{ - unsigned int perfctr_msr, evntsel_msr; - unsigned int evntsel; - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - - perfctr_msr = wd_ops->perfctr; - evntsel_msr = wd_ops->evntsel; - - /* KVM doesn't implement this MSR */ - if (wrmsr_safe(perfctr_msr, 0, 0) < 0) - return 0; - - evntsel = P6_EVNTSEL_INT - | P6_EVNTSEL_OS - | P6_EVNTSEL_USR - | P6_NMI_EVENT; - - /* setup the timer */ - wrmsr(evntsel_msr, evntsel, 0); - nmi_hz = adjust_for_32bit_ctr(nmi_hz); - write_watchdog_counter32(perfctr_msr, "P6_PERFCTR0", nmi_hz); - - /* initialize the wd struct before enabling */ - wd->perfctr_msr = perfctr_msr; - wd->evntsel_msr = evntsel_msr; - wd->cccr_msr = 0; /* unused */ - - /* ok, everything is initialized, announce that we're set */ - cpu_nmi_set_wd_enabled(); - - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= P6_EVNTSEL0_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); - - return 1; -} - -static void __kprobes p6_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) -{ - /* - * P6 based Pentium M need to re-unmask - * the apic vector but it doesn't hurt - * other P6 variant. - * ArchPerfom/Core Duo also needs this - */ - apic_write(APIC_LVTPC, APIC_DM_NMI); - - /* P6/ARCH_PERFMON has 32 bit counter write */ - write_watchdog_counter32(wd->perfctr_msr, NULL, nmi_hz); -} - -static const struct wd_ops p6_wd_ops = { - .reserve = single_msr_reserve, - .unreserve = single_msr_unreserve, - .setup = setup_p6_watchdog, - .rearm = p6_rearm, - .stop = single_msr_stop_watchdog, - .perfctr = MSR_P6_PERFCTR0, - .evntsel = MSR_P6_EVNTSEL0, - .checkbit = 1ULL << 39, -}; - -/* - * Intel P4 performance counters. - * By far the most complicated of all. - */ -#define MSR_P4_MISC_ENABLE_PERF_AVAIL (1 << 7) -#define P4_ESCR_EVENT_SELECT(N) ((N) << 25) -#define P4_ESCR_OS (1 << 3) -#define P4_ESCR_USR (1 << 2) -#define P4_CCCR_OVF_PMI0 (1 << 26) -#define P4_CCCR_OVF_PMI1 (1 << 27) -#define P4_CCCR_THRESHOLD(N) ((N) << 20) -#define P4_CCCR_COMPLEMENT (1 << 19) -#define P4_CCCR_COMPARE (1 << 18) -#define P4_CCCR_REQUIRED (3 << 16) -#define P4_CCCR_ESCR_SELECT(N) ((N) << 13) -#define P4_CCCR_ENABLE (1 << 12) -#define P4_CCCR_OVF (1 << 31) - -#define P4_CONTROLS 18 -static unsigned int p4_controls[18] = { - MSR_P4_BPU_CCCR0, - MSR_P4_BPU_CCCR1, - MSR_P4_BPU_CCCR2, - MSR_P4_BPU_CCCR3, - MSR_P4_MS_CCCR0, - MSR_P4_MS_CCCR1, - MSR_P4_MS_CCCR2, - MSR_P4_MS_CCCR3, - MSR_P4_FLAME_CCCR0, - MSR_P4_FLAME_CCCR1, - MSR_P4_FLAME_CCCR2, - MSR_P4_FLAME_CCCR3, - MSR_P4_IQ_CCCR0, - MSR_P4_IQ_CCCR1, - MSR_P4_IQ_CCCR2, - MSR_P4_IQ_CCCR3, - MSR_P4_IQ_CCCR4, - MSR_P4_IQ_CCCR5, -}; -/* - * Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter - * CRU_ESCR0 (with any non-null event selector) through a complemented - * max threshold. [IA32-Vol3, Section 14.9.9] - */ -static int setup_p4_watchdog(unsigned nmi_hz) -{ - unsigned int perfctr_msr, evntsel_msr, cccr_msr; - unsigned int evntsel, cccr_val; - unsigned int misc_enable, dummy; - unsigned int ht_num; - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - - rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy); - if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL)) - return 0; - -#ifdef CONFIG_SMP - /* detect which hyperthread we are on */ - if (smp_num_siblings == 2) { - unsigned int ebx, apicid; - - ebx = cpuid_ebx(1); - apicid = (ebx >> 24) & 0xff; - ht_num = apicid & 1; - } else -#endif - ht_num = 0; - - /* - * performance counters are shared resources - * assign each hyperthread its own set - * (re-use the ESCR0 register, seems safe - * and keeps the cccr_val the same) - */ - if (!ht_num) { - /* logical cpu 0 */ - perfctr_msr = MSR_P4_IQ_PERFCTR0; - evntsel_msr = MSR_P4_CRU_ESCR0; - cccr_msr = MSR_P4_IQ_CCCR0; - cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4); - - /* - * If we're on the kdump kernel or other situation, we may - * still have other performance counter registers set to - * interrupt and they'll keep interrupting forever because - * of the P4_CCCR_OVF quirk. So we need to ACK all the - * pending interrupts and disable all the registers here, - * before reenabling the NMI delivery. Refer to p4_rearm() - * about the P4_CCCR_OVF quirk. - */ - if (reset_devices) { - unsigned int low, high; - int i; - - for (i = 0; i < P4_CONTROLS; i++) { - rdmsr(p4_controls[i], low, high); - low &= ~(P4_CCCR_ENABLE | P4_CCCR_OVF); - wrmsr(p4_controls[i], low, high); - } - } - } else { - /* logical cpu 1 */ - perfctr_msr = MSR_P4_IQ_PERFCTR1; - evntsel_msr = MSR_P4_CRU_ESCR0; - cccr_msr = MSR_P4_IQ_CCCR1; - - /* Pentium 4 D processors don't support P4_CCCR_OVF_PMI1 */ - if (boot_cpu_data.x86_model == 4 && boot_cpu_data.x86_mask == 4) - cccr_val = P4_CCCR_OVF_PMI0; - else - cccr_val = P4_CCCR_OVF_PMI1; - cccr_val |= P4_CCCR_ESCR_SELECT(4); - } - - evntsel = P4_ESCR_EVENT_SELECT(0x3F) - | P4_ESCR_OS - | P4_ESCR_USR; - - cccr_val |= P4_CCCR_THRESHOLD(15) - | P4_CCCR_COMPLEMENT - | P4_CCCR_COMPARE - | P4_CCCR_REQUIRED; - - wrmsr(evntsel_msr, evntsel, 0); - wrmsr(cccr_msr, cccr_val, 0); - write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0", nmi_hz); - - wd->perfctr_msr = perfctr_msr; - wd->evntsel_msr = evntsel_msr; - wd->cccr_msr = cccr_msr; - - /* ok, everything is initialized, announce that we're set */ - cpu_nmi_set_wd_enabled(); - - apic_write(APIC_LVTPC, APIC_DM_NMI); - cccr_val |= P4_CCCR_ENABLE; - wrmsr(cccr_msr, cccr_val, 0); - return 1; -} - -static void stop_p4_watchdog(void) -{ - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - wrmsr(wd->cccr_msr, 0, 0); - wrmsr(wd->evntsel_msr, 0, 0); -} - -static int p4_reserve(void) -{ - if (!reserve_perfctr_nmi(MSR_P4_IQ_PERFCTR0)) - return 0; -#ifdef CONFIG_SMP - if (smp_num_siblings > 1 && !reserve_perfctr_nmi(MSR_P4_IQ_PERFCTR1)) - goto fail1; -#endif - if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0)) - goto fail2; - /* RED-PEN why is ESCR1 not reserved here? */ - return 1; - fail2: -#ifdef CONFIG_SMP - if (smp_num_siblings > 1) - release_perfctr_nmi(MSR_P4_IQ_PERFCTR1); - fail1: -#endif - release_perfctr_nmi(MSR_P4_IQ_PERFCTR0); - return 0; -} - -static void p4_unreserve(void) -{ -#ifdef CONFIG_SMP - if (smp_num_siblings > 1) - release_perfctr_nmi(MSR_P4_IQ_PERFCTR1); -#endif - release_evntsel_nmi(MSR_P4_CRU_ESCR0); - release_perfctr_nmi(MSR_P4_IQ_PERFCTR0); -} - -static void __kprobes p4_rearm(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz) -{ - unsigned dummy; - /* - * P4 quirks: - * - An overflown perfctr will assert its interrupt - * until the OVF flag in its CCCR is cleared. - * - LVTPC is masked on interrupt and must be - * unmasked by the LVTPC handler. - */ - rdmsrl(wd->cccr_msr, dummy); - dummy &= ~P4_CCCR_OVF; - wrmsrl(wd->cccr_msr, dummy); - apic_write(APIC_LVTPC, APIC_DM_NMI); - /* start the cycle over again */ - write_watchdog_counter(wd->perfctr_msr, NULL, nmi_hz); -} - -static const struct wd_ops p4_wd_ops = { - .reserve = p4_reserve, - .unreserve = p4_unreserve, - .setup = setup_p4_watchdog, - .rearm = p4_rearm, - .stop = stop_p4_watchdog, - /* RED-PEN this is wrong for the other sibling */ - .perfctr = MSR_P4_BPU_PERFCTR0, - .evntsel = MSR_P4_BSU_ESCR0, - .checkbit = 1ULL << 39, -}; - -/* - * Watchdog using the Intel architected PerfMon. - * Used for Core2 and hopefully all future Intel CPUs. - */ -#define ARCH_PERFMON_NMI_EVENT_SEL ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL -#define ARCH_PERFMON_NMI_EVENT_UMASK ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK - -static struct wd_ops intel_arch_wd_ops; - -static int setup_intel_arch_watchdog(unsigned nmi_hz) -{ - unsigned int ebx; - union cpuid10_eax eax; - unsigned int unused; - unsigned int perfctr_msr, evntsel_msr; - unsigned int evntsel; - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - - /* - * Check whether the Architectural PerfMon supports - * Unhalted Core Cycles Event or not. - * NOTE: Corresponding bit = 0 in ebx indicates event present. - */ - cpuid(10, &(eax.full), &ebx, &unused, &unused); - if ((eax.split.mask_length < - (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) || - (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT)) - return 0; - - perfctr_msr = wd_ops->perfctr; - evntsel_msr = wd_ops->evntsel; - - wrmsrl(perfctr_msr, 0UL); - - evntsel = ARCH_PERFMON_EVENTSEL_INT - | ARCH_PERFMON_EVENTSEL_OS - | ARCH_PERFMON_EVENTSEL_USR - | ARCH_PERFMON_NMI_EVENT_SEL - | ARCH_PERFMON_NMI_EVENT_UMASK; - - /* setup the timer */ - wrmsr(evntsel_msr, evntsel, 0); - nmi_hz = adjust_for_32bit_ctr(nmi_hz); - write_watchdog_counter32(perfctr_msr, "INTEL_ARCH_PERFCTR0", nmi_hz); - - wd->perfctr_msr = perfctr_msr; - wd->evntsel_msr = evntsel_msr; - wd->cccr_msr = 0; /* unused */ - - /* ok, everything is initialized, announce that we're set */ - cpu_nmi_set_wd_enabled(); - - apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= ARCH_PERFMON_EVENTSEL_ENABLE; - wrmsr(evntsel_msr, evntsel, 0); - intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1); - return 1; -} - -static struct wd_ops intel_arch_wd_ops __read_mostly = { - .reserve = single_msr_reserve, - .unreserve = single_msr_unreserve, - .setup = setup_intel_arch_watchdog, - .rearm = p6_rearm, - .stop = single_msr_stop_watchdog, - .perfctr = MSR_ARCH_PERFMON_PERFCTR1, - .evntsel = MSR_ARCH_PERFMON_EVENTSEL1, -}; - -static void probe_nmi_watchdog(void) -{ - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_AMD: - if (boot_cpu_data.x86 == 6 || - (boot_cpu_data.x86 >= 0xf && boot_cpu_data.x86 <= 0x15)) - wd_ops = &k7_wd_ops; - return; - case X86_VENDOR_INTEL: - /* Work around where perfctr1 doesn't have a working enable - * bit as described in the following errata: - * AE49 Core Duo and Intel Core Solo 65 nm - * AN49 Intel Pentium Dual-Core - * AF49 Dual-Core Intel Xeon Processor LV - */ - if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) || - ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 15 && - boot_cpu_data.x86_mask == 4))) { - intel_arch_wd_ops.perfctr = MSR_ARCH_PERFMON_PERFCTR0; - intel_arch_wd_ops.evntsel = MSR_ARCH_PERFMON_EVENTSEL0; - } - if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { - wd_ops = &intel_arch_wd_ops; - break; - } - switch (boot_cpu_data.x86) { - case 6: - if (boot_cpu_data.x86_model > 13) - return; - - wd_ops = &p6_wd_ops; - break; - case 15: - wd_ops = &p4_wd_ops; - break; - default: - return; - } - break; - } -} - -/* Interface to nmi.c */ - -int lapic_watchdog_init(unsigned nmi_hz) -{ - if (!wd_ops) { - probe_nmi_watchdog(); - if (!wd_ops) { - printk(KERN_INFO "NMI watchdog: CPU not supported\n"); - return -1; - } - - if (!wd_ops->reserve()) { - printk(KERN_ERR - "NMI watchdog: cannot reserve perfctrs\n"); - return -1; - } - } - - if (!(wd_ops->setup(nmi_hz))) { - printk(KERN_ERR "Cannot setup NMI watchdog on CPU %d\n", - raw_smp_processor_id()); - return -1; - } - - return 0; -} - -void lapic_watchdog_stop(void) -{ - if (wd_ops) - wd_ops->stop(); -} - -unsigned lapic_adjust_nmi_hz(unsigned hz) -{ - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - if (wd->perfctr_msr == MSR_P6_PERFCTR0 || - wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR1) - hz = adjust_for_32bit_ctr(hz); - return hz; -} - -int __kprobes lapic_wd_event(unsigned nmi_hz) -{ - struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk); - u64 ctr; - - rdmsrl(wd->perfctr_msr, ctr); - if (ctr & wd_ops->checkbit) /* perfctr still running? */ - return 0; - - wd_ops->rearm(wd, nmi_hz); - return 1; -} diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 6e8752c1bd52..8474c998cbd4 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -175,21 +175,21 @@ static const struct stacktrace_ops print_trace_ops = { void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, char *log_lvl) + unsigned long *stack, char *log_lvl) { printk("%sCall Trace:\n", log_lvl); - dump_trace(task, regs, stack, bp, &print_trace_ops, log_lvl); + dump_trace(task, regs, stack, &print_trace_ops, log_lvl); } void show_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp) + unsigned long *stack) { - show_trace_log_lvl(task, regs, stack, bp, ""); + show_trace_log_lvl(task, regs, stack, ""); } void show_stack(struct task_struct *task, unsigned long *sp) { - show_stack_log_lvl(task, NULL, sp, 0, ""); + show_stack_log_lvl(task, NULL, sp, ""); } /* @@ -210,7 +210,7 @@ void dump_stack(void) init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); - show_trace(NULL, NULL, &stack, bp); + show_trace(NULL, NULL, &stack); } EXPORT_SYMBOL(dump_stack); diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index 1bc7f75a5bda..74cc1eda384b 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -17,11 +17,12 @@ #include <asm/stacktrace.h> -void dump_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, +void dump_trace(struct task_struct *task, + struct pt_regs *regs, unsigned long *stack, const struct stacktrace_ops *ops, void *data) { int graph = 0; + unsigned long bp; if (!task) task = current; @@ -34,18 +35,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, stack = (unsigned long *)task->thread.sp; } -#ifdef CONFIG_FRAME_POINTER - if (!bp) { - if (task == current) { - /* Grab bp right from our regs */ - get_bp(bp); - } else { - /* bp is the last reg pushed by switch_to */ - bp = *(unsigned long *) task->thread.sp; - } - } -#endif - + bp = stack_frame(task, regs); for (;;) { struct thread_info *context; @@ -65,7 +55,7 @@ EXPORT_SYMBOL(dump_trace); void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) + unsigned long *sp, char *log_lvl) { unsigned long *stack; int i; @@ -87,7 +77,7 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, touch_nmi_watchdog(); } printk(KERN_CONT "\n"); - show_trace_log_lvl(task, regs, sp, bp, log_lvl); + show_trace_log_lvl(task, regs, sp, log_lvl); } @@ -112,8 +102,7 @@ void show_registers(struct pt_regs *regs) u8 *ip; printk(KERN_EMERG "Stack:\n"); - show_stack_log_lvl(NULL, regs, ®s->sp, - 0, KERN_EMERG); + show_stack_log_lvl(NULL, regs, ®s->sp, KERN_EMERG); printk(KERN_EMERG "Code: "); diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index 6a340485249a..64101335de19 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -139,8 +139,8 @@ fixup_bp_irq_link(unsigned long bp, unsigned long *stack, * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack */ -void dump_trace(struct task_struct *task, struct pt_regs *regs, - unsigned long *stack, unsigned long bp, +void dump_trace(struct task_struct *task, + struct pt_regs *regs, unsigned long *stack, const struct stacktrace_ops *ops, void *data) { const unsigned cpu = get_cpu(); @@ -149,6 +149,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned used = 0; struct thread_info *tinfo; int graph = 0; + unsigned long bp; if (!task) task = current; @@ -160,18 +161,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs, stack = (unsigned long *)task->thread.sp; } -#ifdef CONFIG_FRAME_POINTER - if (!bp) { - if (task == current) { - /* Grab bp right from our regs */ - get_bp(bp); - } else { - /* bp is the last reg pushed by switch_to */ - bp = *(unsigned long *) task->thread.sp; - } - } -#endif - + bp = stack_frame(task, regs); /* * Print function call entries in all stacks, starting at the * current stack address. If the stacks consist of nested @@ -235,7 +225,7 @@ EXPORT_SYMBOL(dump_trace); void show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, - unsigned long *sp, unsigned long bp, char *log_lvl) + unsigned long *sp, char *log_lvl) { unsigned long *irq_stack_end; unsigned long *irq_stack; @@ -279,7 +269,7 @@ show_stack_log_lvl(struct task_struct *task, struct pt_regs *regs, preempt_enable(); printk(KERN_CONT "\n"); - show_trace_log_lvl(task, regs, sp, bp, log_lvl); + show_trace_log_lvl(task, regs, sp, log_lvl); } void show_registers(struct pt_regs *regs) @@ -308,7 +298,7 @@ void show_registers(struct pt_regs *regs) printk(KERN_EMERG "Stack:\n"); show_stack_log_lvl(NULL, regs, (unsigned long *)sp, - regs->bp, KERN_EMERG); + KERN_EMERG); printk(KERN_EMERG "Code: "); diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 1cbd54c0df99..5940282bd2f9 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -1184,6 +1184,10 @@ static void __kprobes optimized_callback(struct optimized_kprobe *op, { struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + /* This is possible if op is under delayed unoptimizing */ + if (kprobe_disabled(&op->kp)) + return; + preempt_disable(); if (kprobe_running()) { kprobes_inc_nmissed_count(&op->kp); @@ -1401,10 +1405,16 @@ int __kprobes arch_prepare_optimized_kprobe(struct optimized_kprobe *op) return 0; } -/* Replace a breakpoint (int3) with a relative jump. */ -int __kprobes arch_optimize_kprobe(struct optimized_kprobe *op) +#define MAX_OPTIMIZE_PROBES 256 +static struct text_poke_param *jump_poke_params; +static struct jump_poke_buffer { + u8 buf[RELATIVEJUMP_SIZE]; +} *jump_poke_bufs; + +static void __kprobes setup_optimize_kprobe(struct text_poke_param *tprm, + u8 *insn_buf, + struct optimized_kprobe *op) { - unsigned char jmp_code[RELATIVEJUMP_SIZE]; s32 rel = (s32)((long)op->optinsn.insn - ((long)op->kp.addr + RELATIVEJUMP_SIZE)); @@ -1412,16 +1422,79 @@ int __kprobes arch_optimize_kprobe(struct optimized_kprobe *op) memcpy(op->optinsn.copied_insn, op->kp.addr + INT3_SIZE, RELATIVE_ADDR_SIZE); - jmp_code[0] = RELATIVEJUMP_OPCODE; - *(s32 *)(&jmp_code[1]) = rel; + insn_buf[0] = RELATIVEJUMP_OPCODE; + *(s32 *)(&insn_buf[1]) = rel; + + tprm->addr = op->kp.addr; + tprm->opcode = insn_buf; + tprm->len = RELATIVEJUMP_SIZE; +} + +/* + * Replace breakpoints (int3) with relative jumps. + * Caller must call with locking kprobe_mutex and text_mutex. + */ +void __kprobes arch_optimize_kprobes(struct list_head *oplist) +{ + struct optimized_kprobe *op, *tmp; + int c = 0; + + list_for_each_entry_safe(op, tmp, oplist, list) { + WARN_ON(kprobe_disabled(&op->kp)); + /* Setup param */ + setup_optimize_kprobe(&jump_poke_params[c], + jump_poke_bufs[c].buf, op); + list_del_init(&op->list); + if (++c >= MAX_OPTIMIZE_PROBES) + break; + } /* * text_poke_smp doesn't support NMI/MCE code modifying. * However, since kprobes itself also doesn't support NMI/MCE * code probing, it's not a problem. */ - text_poke_smp(op->kp.addr, jmp_code, RELATIVEJUMP_SIZE); - return 0; + text_poke_smp_batch(jump_poke_params, c); +} + +static void __kprobes setup_unoptimize_kprobe(struct text_poke_param *tprm, + u8 *insn_buf, + struct optimized_kprobe *op) +{ + /* Set int3 to first byte for kprobes */ + insn_buf[0] = BREAKPOINT_INSTRUCTION; + memcpy(insn_buf + 1, op->optinsn.copied_insn, RELATIVE_ADDR_SIZE); + + tprm->addr = op->kp.addr; + tprm->opcode = insn_buf; + tprm->len = RELATIVEJUMP_SIZE; +} + +/* + * Recover original instructions and breakpoints from relative jumps. + * Caller must call with locking kprobe_mutex. + */ +extern void arch_unoptimize_kprobes(struct list_head *oplist, + struct list_head *done_list) +{ + struct optimized_kprobe *op, *tmp; + int c = 0; + + list_for_each_entry_safe(op, tmp, oplist, list) { + /* Setup param */ + setup_unoptimize_kprobe(&jump_poke_params[c], + jump_poke_bufs[c].buf, op); + list_move(&op->list, done_list); + if (++c >= MAX_OPTIMIZE_PROBES) + break; + } + + /* + * text_poke_smp doesn't support NMI/MCE code modifying. + * However, since kprobes itself also doesn't support NMI/MCE + * code probing, it's not a problem. + */ + text_poke_smp_batch(jump_poke_params, c); } /* Replace a relative jump with a breakpoint (int3). */ @@ -1453,11 +1526,35 @@ static int __kprobes setup_detour_execution(struct kprobe *p, } return 0; } + +static int __kprobes init_poke_params(void) +{ + /* Allocate code buffer and parameter array */ + jump_poke_bufs = kmalloc(sizeof(struct jump_poke_buffer) * + MAX_OPTIMIZE_PROBES, GFP_KERNEL); + if (!jump_poke_bufs) + return -ENOMEM; + + jump_poke_params = kmalloc(sizeof(struct text_poke_param) * + MAX_OPTIMIZE_PROBES, GFP_KERNEL); + if (!jump_poke_params) { + kfree(jump_poke_bufs); + jump_poke_bufs = NULL; + return -ENOMEM; + } + + return 0; +} +#else /* !CONFIG_OPTPROBES */ +static int __kprobes init_poke_params(void) +{ + return 0; +} #endif int __init arch_init_kprobes(void) { - return 0; + return init_poke_params(); } int __kprobes arch_trampoline_kprobe(struct kprobe *p) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 57d1868a86aa..c852041bfc3d 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -91,8 +91,7 @@ void exit_thread(void) void show_regs(struct pt_regs *regs) { show_registers(regs); - show_trace(NULL, regs, (unsigned long *)kernel_stack_pointer(regs), - regs->bp); + show_trace(NULL, regs, (unsigned long *)kernel_stack_pointer(regs)); } void show_regs_common(void) @@ -374,6 +373,7 @@ void default_idle(void) { if (hlt_use_halt()) { trace_power_start(POWER_CSTATE, 1, smp_processor_id()); + trace_cpu_idle(1, smp_processor_id()); current_thread_info()->status &= ~TS_POLLING; /* * TS_POLLING-cleared state must be visible before we @@ -444,6 +444,7 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait); void mwait_idle_with_hints(unsigned long ax, unsigned long cx) { trace_power_start(POWER_CSTATE, (ax>>4)+1, smp_processor_id()); + trace_cpu_idle((ax>>4)+1, smp_processor_id()); if (!need_resched()) { if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); @@ -460,6 +461,7 @@ static void mwait_idle(void) { if (!need_resched()) { trace_power_start(POWER_CSTATE, 1, smp_processor_id()); + trace_cpu_idle(1, smp_processor_id()); if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); @@ -481,10 +483,12 @@ static void mwait_idle(void) static void poll_idle(void) { trace_power_start(POWER_CSTATE, 0, smp_processor_id()); + trace_cpu_idle(0, smp_processor_id()); local_irq_enable(); while (!need_resched()) cpu_relax(); - trace_power_end(0); + trace_power_end(smp_processor_id()); + trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id()); } /* diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index 96586c3cbbbf..4b9befa0e347 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -113,8 +113,8 @@ void cpu_idle(void) stop_critical_timings(); pm_idle(); start_critical_timings(); - trace_power_end(smp_processor_id()); + trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id()); } tick_nohz_restart_sched_tick(); preempt_enable_no_resched(); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index b3d7a3a04f38..4c818a738396 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -142,6 +142,8 @@ void cpu_idle(void) start_critical_timings(); trace_power_end(smp_processor_id()); + trace_cpu_idle(PWR_EVENT_EXIT, + smp_processor_id()); /* In many cases the interrupt that ended idle has already called exit_idle. But some idle diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 083e99d1b7df..68f61ac632e1 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -281,6 +281,13 @@ static void __cpuinit smp_callin(void) */ smp_store_cpu_info(cpuid); + /* + * This must be done before setting cpu_online_mask + * or calling notify_cpu_starting. + */ + set_cpu_sibling_map(raw_smp_processor_id()); + wmb(); + notify_cpu_starting(cpuid); /* @@ -316,16 +323,6 @@ notrace static void __cpuinit start_secondary(void *unused) */ check_tsc_sync_target(); - if (nmi_watchdog == NMI_IO_APIC) { - legacy_pic->mask(0); - enable_NMI_through_LVT0(); - legacy_pic->unmask(0); - } - - /* This must be done before setting cpu_online_mask */ - set_cpu_sibling_map(raw_smp_processor_id()); - wmb(); - /* * We need to hold call_lock, so there is no inconsistency * between the time smp_call_function() determines number of @@ -1061,8 +1058,6 @@ static int __init smp_sanity_check(unsigned max_cpus) printk(KERN_INFO "SMP mode deactivated.\n"); smpboot_clear_io_apic(); - localise_nmi_watchdog(); - connect_bsp_APIC(); setup_local_APIC(); end_local_APIC_setup(); @@ -1196,7 +1191,6 @@ void __init native_smp_cpus_done(unsigned int max_cpus) #ifdef CONFIG_X86_IO_APIC setup_ioapic_dest(); #endif - check_nmi_watchdog(); mtrr_aps_init(); } @@ -1341,8 +1335,6 @@ int native_cpu_disable(void) if (cpu == 0) return -EBUSY; - if (nmi_watchdog == NMI_LOCAL_APIC) - stop_apic_nmi_watchdog(NULL); clear_local_APIC(); cpu_disable_common(); diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index b53c525368a7..938c8e10a19a 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -73,22 +73,22 @@ static const struct stacktrace_ops save_stack_ops_nosched = { */ void save_stack_trace(struct stack_trace *trace) { - dump_trace(current, NULL, NULL, 0, &save_stack_ops, trace); + dump_trace(current, NULL, NULL, &save_stack_ops, trace); if (trace->nr_entries < trace->max_entries) trace->entries[trace->nr_entries++] = ULONG_MAX; } EXPORT_SYMBOL_GPL(save_stack_trace); -void save_stack_trace_bp(struct stack_trace *trace, unsigned long bp) +void save_stack_trace_regs(struct stack_trace *trace, struct pt_regs *regs) { - dump_trace(current, NULL, NULL, bp, &save_stack_ops, trace); + dump_trace(current, regs, NULL, &save_stack_ops, trace); if (trace->nr_entries < trace->max_entries) trace->entries[trace->nr_entries++] = ULONG_MAX; } void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace) { - dump_trace(tsk, NULL, NULL, 0, &save_stack_ops_nosched, trace); + dump_trace(tsk, NULL, NULL, &save_stack_ops_nosched, trace); if (trace->nr_entries < trace->max_entries) trace->entries[trace->nr_entries++] = ULONG_MAX; } diff --git a/arch/x86/kernel/time.c b/arch/x86/kernel/time.c index fb5cc5e14cfa..25a28a245937 100644 --- a/arch/x86/kernel/time.c +++ b/arch/x86/kernel/time.c @@ -22,10 +22,6 @@ #include <asm/hpet.h> #include <asm/time.h> -#if defined(CONFIG_X86_32) && defined(CONFIG_X86_IO_APIC) -int timer_ack; -#endif - #ifdef CONFIG_X86_64 volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES; #endif @@ -63,20 +59,6 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id) /* Keep nmi watchdog up to date */ inc_irq_stat(irq0_irqs); - /* Optimized out for !IO_APIC and x86_64 */ - if (timer_ack) { - /* - * Subtle, when I/O APICs are used we have to ack timer IRQ - * manually to deassert NMI lines for the watchdog if run - * on an 82489DX-based system. - */ - raw_spin_lock(&i8259A_lock); - outb(0x0c, PIC_MASTER_OCW3); - /* Ack the IRQ; AEOI will end it automatically. */ - inb(PIC_MASTER_POLL); - raw_spin_unlock(&i8259A_lock); - } - global_clock_event->event_handler(global_clock_event); /* MCA bus quirk: Acknowledge irq0 by setting bit 7 in port 0x61 */ diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index cb838ca42c96..c76aaca5694d 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -83,6 +83,8 @@ EXPORT_SYMBOL_GPL(used_vectors); static int ignore_nmis; +int unknown_nmi_panic; + static inline void conditional_sti(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) @@ -300,6 +302,13 @@ gp_in_kernel: die("general protection fault", regs, error_code); } +static int __init setup_unknown_nmi_panic(char *str) +{ + unknown_nmi_panic = 1; + return 1; +} +__setup("unknown_nmi_panic", setup_unknown_nmi_panic); + static notrace __kprobes void mem_parity_error(unsigned char reason, struct pt_regs *regs) { @@ -342,9 +351,11 @@ io_check_error(unsigned char reason, struct pt_regs *regs) reason = (reason & 0xf) | 8; outb(reason, 0x61); - i = 2000; - while (--i) - udelay(1000); + i = 20000; + while (--i) { + touch_nmi_watchdog(); + udelay(100); + } reason &= ~8; outb(reason, 0x61); @@ -371,7 +382,7 @@ unknown_nmi_error(unsigned char reason, struct pt_regs *regs) reason, smp_processor_id()); printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n"); - if (panic_on_unrecovered_nmi) + if (unknown_nmi_panic || panic_on_unrecovered_nmi) panic("NMI: Not continuing"); printk(KERN_EMERG "Dazed and confused, but trying to continue\n"); @@ -397,20 +408,8 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) return; - -#ifndef CONFIG_LOCKUP_DETECTOR - /* - * Ok, so this is none of the documented NMI sources, - * so it must be the NMI watchdog. - */ - if (nmi_watchdog_tick(regs, reason)) - return; - if (!do_nmi_callback(regs, cpu)) -#endif /* !CONFIG_LOCKUP_DETECTOR */ - unknown_nmi_error(reason, regs); -#else - unknown_nmi_error(reason, regs); #endif + unknown_nmi_error(reason, regs); return; } @@ -446,14 +445,12 @@ do_nmi(struct pt_regs *regs, long error_code) void stop_nmi(void) { - acpi_nmi_disable(); ignore_nmis++; } void restart_nmi(void) { ignore_nmis--; - acpi_nmi_enable(); } /* May run on IST stack. */ diff --git a/arch/x86/mm/kmemcheck/error.c b/arch/x86/mm/kmemcheck/error.c index af3b6c8a436f..704a37cedddb 100644 --- a/arch/x86/mm/kmemcheck/error.c +++ b/arch/x86/mm/kmemcheck/error.c @@ -185,7 +185,7 @@ void kmemcheck_error_save(enum kmemcheck_shadow state, e->trace.entries = e->trace_entries; e->trace.max_entries = ARRAY_SIZE(e->trace_entries); e->trace.skip = 0; - save_stack_trace_bp(&e->trace, regs->bp); + save_stack_trace_regs(&e->trace, regs); /* Round address down to nearest 16 bytes */ shadow_copy = kmemcheck_shadow_lookup(address diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c index 2d49d4e19a36..72cbec14d783 100644 --- a/arch/x86/oprofile/backtrace.c +++ b/arch/x86/oprofile/backtrace.c @@ -126,7 +126,7 @@ x86_backtrace(struct pt_regs * const regs, unsigned int depth) if (!user_mode_vm(regs)) { unsigned long stack = kernel_stack_pointer(regs); if (depth) - dump_trace(NULL, regs, (unsigned long *)stack, 0, + dump_trace(NULL, regs, (unsigned long *)stack, &backtrace_ops, &depth); return; } diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 4e8baad36d37..358c8b9c96a7 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -732,6 +732,9 @@ int __init op_nmi_init(struct oprofile_operations *ops) case 0x14: cpu_type = "x86-64/family14h"; break; + case 0x15: + cpu_type = "x86-64/family15h"; + break; default: return -ENODEV; } diff --git a/arch/x86/oprofile/nmi_timer_int.c b/arch/x86/oprofile/nmi_timer_int.c index e3ecb71b5790..0636dd93cef8 100644 --- a/arch/x86/oprofile/nmi_timer_int.c +++ b/arch/x86/oprofile/nmi_timer_int.c @@ -58,9 +58,6 @@ static void timer_stop(void) int __init op_nmi_timer_init(struct oprofile_operations *ops) { - if ((nmi_watchdog != NMI_IO_APIC) || (atomic_read(&nmi_active) <= 0)) - return -ENODEV; - ops->start = timer_start; ops->stop = timer_stop; ops->cpu_type = "timer"; diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index 7d90d47655ba..51104b33fd51 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -29,11 +29,12 @@ #include "op_x86_model.h" #include "op_counter.h" -#define NUM_COUNTERS 4 +#define NUM_COUNTERS 4 +#define NUM_COUNTERS_F15H 6 #ifdef CONFIG_OPROFILE_EVENT_MULTIPLEX -#define NUM_VIRT_COUNTERS 32 +#define NUM_VIRT_COUNTERS 32 #else -#define NUM_VIRT_COUNTERS NUM_COUNTERS +#define NUM_VIRT_COUNTERS 0 #endif #define OP_EVENT_MASK 0x0FFF @@ -41,7 +42,8 @@ #define MSR_AMD_EVENTSEL_RESERVED ((0xFFFFFCF0ULL<<32)|(1ULL<<21)) -static unsigned long reset_value[NUM_VIRT_COUNTERS]; +static int num_counters; +static unsigned long reset_value[OP_MAX_COUNTER]; #define IBS_FETCH_SIZE 6 #define IBS_OP_SIZE 12 @@ -387,7 +389,7 @@ static void op_mux_switch_ctrl(struct op_x86_model_spec const *model, int i; /* enable active counters */ - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { int virt = op_x86_phys_to_virt(i); if (!reset_value[virt]) continue; @@ -406,7 +408,7 @@ static void op_amd_shutdown(struct op_msrs const * const msrs) { int i; - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { if (!msrs->counters[i].addr) continue; release_perfctr_nmi(MSR_K7_PERFCTR0 + i); @@ -418,7 +420,7 @@ static int op_amd_fill_in_addresses(struct op_msrs * const msrs) { int i; - for (i = 0; i < NUM_COUNTERS; i++) { + for (i = 0; i < num_counters; i++) { if (!reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i)) goto fail; if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i)) { @@ -426,8 +428,13 @@ static int op_amd_fill_in_addresses(struct op_msrs * const msrs) goto fail; } /* both registers must be reserved */ - msrs->counters[i].addr = MSR_K7_PERFCTR0 + i; - msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i; + if (num_counters == NUM_COUNTERS_F15H) { + msrs->counters[i].addr = MSR_F15H_PERF_CTR + (i << 1); + msrs->controls[i].addr = MSR_F15H_PERF_CTL + (i << 1); + } else { + msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i; + msrs->counters[i].addr = MSR_K7_PERFCTR0 + i; + } continue; fail: if (!counter_config[i].enabled) @@ -447,7 +454,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, int i; /* setup reset_value */ - for (i = 0; i < NUM_VIRT_COUNTERS; ++i) { + for (i = 0; i < OP_MAX_COUNTER; ++i) { if (counter_config[i].enabled && msrs->counters[op_x86_virt_to_phys(i)].addr) reset_value[i] = counter_config[i].count; @@ -456,7 +463,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, } /* clear all counters */ - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { if (!msrs->controls[i].addr) continue; rdmsrl(msrs->controls[i].addr, val); @@ -472,7 +479,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, } /* enable active counters */ - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { int virt = op_x86_phys_to_virt(i); if (!reset_value[virt]) continue; @@ -503,7 +510,7 @@ static int op_amd_check_ctrs(struct pt_regs * const regs, u64 val; int i; - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { int virt = op_x86_phys_to_virt(i); if (!reset_value[virt]) continue; @@ -526,7 +533,7 @@ static void op_amd_start(struct op_msrs const * const msrs) u64 val; int i; - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { if (!reset_value[op_x86_phys_to_virt(i)]) continue; rdmsrl(msrs->controls[i].addr, val); @@ -546,7 +553,7 @@ static void op_amd_stop(struct op_msrs const * const msrs) * Subtle: stop on all counters to avoid race with setting our * pm callback */ - for (i = 0; i < NUM_COUNTERS; ++i) { + for (i = 0; i < num_counters; ++i) { if (!reset_value[op_x86_phys_to_virt(i)]) continue; rdmsrl(msrs->controls[i].addr, val); @@ -706,18 +713,29 @@ static int setup_ibs_files(struct super_block *sb, struct dentry *root) return 0; } +struct op_x86_model_spec op_amd_spec; + static int op_amd_init(struct oprofile_operations *ops) { init_ibs(); create_arch_files = ops->create_files; ops->create_files = setup_ibs_files; + + if (boot_cpu_data.x86 == 0x15) { + num_counters = NUM_COUNTERS_F15H; + } else { + num_counters = NUM_COUNTERS; + } + + op_amd_spec.num_counters = num_counters; + op_amd_spec.num_controls = num_counters; + op_amd_spec.num_virt_counters = max(num_counters, NUM_VIRT_COUNTERS); + return 0; } struct op_x86_model_spec op_amd_spec = { - .num_counters = NUM_COUNTERS, - .num_controls = NUM_COUNTERS, - .num_virt_counters = NUM_VIRT_COUNTERS, + /* num_counters/num_controls filled in at runtime */ .reserved = MSR_AMD_EVENTSEL_RESERVED, .event_mask = OP_EVENT_MASK, .init = op_amd_init, diff --git a/arch/x86/oprofile/op_model_p4.c b/arch/x86/oprofile/op_model_p4.c index 182558dd5515..9fadec074142 100644 --- a/arch/x86/oprofile/op_model_p4.c +++ b/arch/x86/oprofile/op_model_p4.c @@ -11,7 +11,7 @@ #include <linux/oprofile.h> #include <linux/smp.h> #include <linux/ptrace.h> -#include <linux/nmi.h> +#include <asm/nmi.h> #include <asm/msr.h> #include <asm/fixmap.h> #include <asm/apic.h> diff --git a/drivers/acpi/acpica/nsinit.c b/drivers/acpi/acpica/nsinit.c index 660a2728908d..0cac7ec0d2ec 100644 --- a/drivers/acpi/acpica/nsinit.c +++ b/drivers/acpi/acpica/nsinit.c @@ -577,9 +577,7 @@ acpi_ns_init_one_device(acpi_handle obj_handle, * as possible (without an NMI being received in the middle of * this) - so disable NMIs and initialize the device: */ - acpi_nmi_disable(); status = acpi_ns_evaluate(info); - acpi_nmi_enable(); if (ACPI_SUCCESS(status)) { walk_info->num_INI++; diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index c63a43823744..1109f6848a43 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -355,6 +355,7 @@ void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) dprintk("FREQ: %lu - CPU: %lu", (unsigned long)freqs->new, (unsigned long)freqs->cpu); trace_power_frequency(POWER_PSTATE, freqs->new, freqs->cpu); + trace_cpu_frequency(freqs->new, freqs->cpu); srcu_notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_POSTCHANGE, freqs); if (likely(policy) && likely(policy->cpu == freqs->cpu)) diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index a50710843378..08d5f05378d9 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -107,6 +107,7 @@ static void cpuidle_idle_call(void) if (cpuidle_curr_governor->reflect) cpuidle_curr_governor->reflect(dev); trace_power_end(smp_processor_id()); + trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id()); } /** diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index c131d58bcb50..56ac09d6c930 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -220,9 +220,8 @@ static int intel_idle(struct cpuidle_device *dev, struct cpuidle_state *state) kt_before = ktime_get_real(); stop_critical_timings(); -#ifndef MODULE trace_power_start(POWER_CSTATE, (eax >> 4) + 1, cpu); -#endif + trace_cpu_idle((eax >> 4) + 1, cpu); if (!need_resched()) { __monitor((void *)¤t_thread_info()->flags, 0, 0); diff --git a/drivers/watchdog/hpwdt.c b/drivers/watchdog/hpwdt.c index 3d77116e4634..dea7b5bf6e2c 100644 --- a/drivers/watchdog/hpwdt.c +++ b/drivers/watchdog/hpwdt.c @@ -642,19 +642,14 @@ static struct notifier_block die_notifier = { */ #ifdef CONFIG_HPWDT_NMI_DECODING -#ifdef ARCH_HAS_NMI_WATCHDOG +#ifdef CONFIG_X86_LOCAL_APIC static void __devinit hpwdt_check_nmi_decoding(struct pci_dev *dev) { /* * If nmi_watchdog is turned off then we can turn on * our nmi decoding capability. */ - if (!nmi_watchdog_active()) - hpwdt_nmi_decoding = 1; - else - dev_warn(&dev->dev, "NMI decoding is disabled. To enable this " - "functionality you must reboot with nmi_watchdog=0 " - "and load the hpwdt driver with priority=1.\n"); + hpwdt_nmi_decoding = 1; } #else static void __devinit hpwdt_check_nmi_decoding(struct pci_dev *dev) @@ -662,7 +657,7 @@ static void __devinit hpwdt_check_nmi_decoding(struct pci_dev *dev) dev_warn(&dev->dev, "NMI decoding is disabled. " "Your kernel does not support a NMI Watchdog.\n"); } -#endif /* ARCH_HAS_NMI_WATCHDOG */ +#endif /* CONFIG_X86_LOCAL_APIC */ static int __devinit hpwdt_init_nmi_decoding(struct pci_dev *dev) { diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 5476c066d4ee..3c4039d5eef1 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -763,7 +763,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, int metadata; unsigned int revokes = 0; int x; - int error; + int error = 0; if (!*top) sm->sm_first = 0; @@ -780,7 +780,11 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, if (metadata) revokes = (height) ? sdp->sd_inptrs : sdp->sd_diptrs; - error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); + if (ip != GFS2_I(sdp->sd_rindex)) + error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); + else if (!sdp->sd_rgrps) + error = gfs2_ri_update(ip); + if (error) return error; @@ -879,7 +883,8 @@ out_rg_gunlock: out_rlist: gfs2_rlist_free(&rlist); out: - gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); + if (ip != GFS2_I(sdp->sd_rindex)) + gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); return error; } diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index f92c17704169..08a8beb152e6 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -541,21 +541,6 @@ out_locked: spin_unlock(&gl->gl_spin); } -static unsigned int gfs2_lm_lock(struct gfs2_sbd *sdp, void *lock, - unsigned int req_state, - unsigned int flags) -{ - int ret = LM_OUT_ERROR; - - if (!sdp->sd_lockstruct.ls_ops->lm_lock) - return req_state == LM_ST_UNLOCKED ? 0 : req_state; - - if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) - ret = sdp->sd_lockstruct.ls_ops->lm_lock(lock, - req_state, flags); - return ret; -} - /** * do_xmote - Calls the DLM to change the state of a lock * @gl: The lock state @@ -575,13 +560,14 @@ __acquires(&gl->gl_spin) lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP | LM_FLAG_PRIORITY); - BUG_ON(gl->gl_state == target); - BUG_ON(gl->gl_state == gl->gl_target); + GLOCK_BUG_ON(gl, gl->gl_state == target); + GLOCK_BUG_ON(gl, gl->gl_state == gl->gl_target); if ((target == LM_ST_UNLOCKED || target == LM_ST_DEFERRED) && glops->go_inval) { set_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); do_error(gl, 0); /* Fail queued try locks */ } + gl->gl_req = target; spin_unlock(&gl->gl_spin); if (glops->go_xmote_th) glops->go_xmote_th(gl); @@ -594,15 +580,17 @@ __acquires(&gl->gl_spin) gl->gl_state == LM_ST_DEFERRED) && !(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) lck_flags |= LM_FLAG_TRY_1CB; - ret = gfs2_lm_lock(sdp, gl, target, lck_flags); - if (!(ret & LM_OUT_ASYNC)) { - finish_xmote(gl, ret); + if (sdp->sd_lockstruct.ls_ops->lm_lock) { + /* lock_dlm */ + ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); + GLOCK_BUG_ON(gl, ret); + } else { /* lock_nolock */ + finish_xmote(gl, target); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); - } else { - GLOCK_BUG_ON(gl, ret != LM_OUT_ASYNC); } + spin_lock(&gl->gl_spin); } @@ -951,17 +939,22 @@ int gfs2_glock_wait(struct gfs2_holder *gh) void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...) { + struct va_format vaf; va_list args; va_start(args, fmt); + if (seq) { struct gfs2_glock_iter *gi = seq->private; vsprintf(gi->string, fmt, args); seq_printf(seq, gi->string); } else { - printk(KERN_ERR " "); - vprintk(fmt, args); + vaf.fmt = fmt; + vaf.va = &args; + + printk(KERN_ERR " %pV", &vaf); } + va_end(args); } @@ -1361,24 +1354,28 @@ static int gfs2_should_freeze(const struct gfs2_glock *gl) * @gl: Pointer to the glock * @ret: The return value from the dlm * + * The gl_reply field is under the gl_spin lock so that it is ok + * to use a bitfield shared with other glock state fields. */ void gfs2_glock_complete(struct gfs2_glock *gl, int ret) { struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct; + spin_lock(&gl->gl_spin); gl->gl_reply = ret; if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) { - spin_lock(&gl->gl_spin); if (gfs2_should_freeze(gl)) { set_bit(GLF_FROZEN, &gl->gl_flags); spin_unlock(&gl->gl_spin); return; } - spin_unlock(&gl->gl_spin); } + + spin_unlock(&gl->gl_spin); set_bit(GLF_REPLY_PENDING, &gl->gl_flags); + smp_wmb(); gfs2_glock_hold(gl); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); @@ -1626,18 +1623,17 @@ static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags) static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh) { struct task_struct *gh_owner = NULL; - char buffer[KSYM_SYMBOL_LEN]; char flags_buf[32]; - sprint_symbol(buffer, gh->gh_ip); if (gh->gh_owner_pid) gh_owner = pid_task(gh->gh_owner_pid, PIDTYPE_PID); - gfs2_print_dbg(seq, " H: s:%s f:%s e:%d p:%ld [%s] %s\n", - state2str(gh->gh_state), - hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags), - gh->gh_error, - gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1, - gh_owner ? gh_owner->comm : "(ended)", buffer); + gfs2_print_dbg(seq, " H: s:%s f:%s e:%d p:%ld [%s] %pS\n", + state2str(gh->gh_state), + hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags), + gh->gh_error, + gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1, + gh_owner ? gh_owner->comm : "(ended)", + (void *)gh->gh_ip); return 0; } @@ -1782,12 +1778,13 @@ int __init gfs2_glock_init(void) } #endif - glock_workqueue = alloc_workqueue("glock_workqueue", WQ_RESCUER | + glock_workqueue = alloc_workqueue("glock_workqueue", WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_FREEZEABLE, 0); if (IS_ERR(glock_workqueue)) return PTR_ERR(glock_workqueue); - gfs2_delete_workqueue = alloc_workqueue("delete_workqueue", WQ_RESCUER | - WQ_FREEZEABLE, 0); + gfs2_delete_workqueue = alloc_workqueue("delete_workqueue", + WQ_MEM_RECLAIM | WQ_FREEZEABLE, + 0); if (IS_ERR(gfs2_delete_workqueue)) { destroy_workqueue(glock_workqueue); return PTR_ERR(gfs2_delete_workqueue); diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index db1c26d6d220..691851ceb615 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -87,11 +87,10 @@ enum { #define GL_ASYNC 0x00000040 #define GL_EXACT 0x00000080 #define GL_SKIP 0x00000100 -#define GL_ATIME 0x00000200 #define GL_NOCACHE 0x00000400 /* - * lm_lock() and lm_async_cb return flags + * lm_async_cb return flags * * LM_OUT_ST_MASK * Masks the lower two bits of lock state in the returned value. @@ -99,15 +98,11 @@ enum { * LM_OUT_CANCELED * The lock request was canceled. * - * LM_OUT_ASYNC - * The result of the request will be returned in an LM_CB_ASYNC callback. - * */ #define LM_OUT_ST_MASK 0x00000003 #define LM_OUT_CANCELED 0x00000008 -#define LM_OUT_ASYNC 0x00000080 -#define LM_OUT_ERROR 0x00000100 +#define LM_OUT_ERROR 0x00000004 /* * lm_recovery_done() messages @@ -124,25 +119,12 @@ struct lm_lockops { void (*lm_unmount) (struct gfs2_sbd *sdp); void (*lm_withdraw) (struct gfs2_sbd *sdp); void (*lm_put_lock) (struct kmem_cache *cachep, struct gfs2_glock *gl); - unsigned int (*lm_lock) (struct gfs2_glock *gl, - unsigned int req_state, unsigned int flags); + int (*lm_lock) (struct gfs2_glock *gl, unsigned int req_state, + unsigned int flags); void (*lm_cancel) (struct gfs2_glock *gl); const match_table_t *lm_tokens; }; -#define LM_FLAG_TRY 0x00000001 -#define LM_FLAG_TRY_1CB 0x00000002 -#define LM_FLAG_NOEXP 0x00000004 -#define LM_FLAG_ANY 0x00000008 -#define LM_FLAG_PRIORITY 0x00000010 - -#define GL_ASYNC 0x00000040 -#define GL_EXACT 0x00000080 -#define GL_SKIP 0x00000100 -#define GL_NOCACHE 0x00000400 - -#define GLR_TRYFAILED 13 - extern struct workqueue_struct *gfs2_delete_workqueue; static inline struct gfs2_holder *gfs2_glock_is_locked_by_me(struct gfs2_glock *gl) { @@ -212,6 +194,8 @@ int gfs2_glock_nq_num(struct gfs2_sbd *sdp, int gfs2_glock_nq_m(unsigned int num_gh, struct gfs2_holder *ghs); void gfs2_glock_dq_m(unsigned int num_gh, struct gfs2_holder *ghs); void gfs2_glock_dq_uninit_m(unsigned int num_gh, struct gfs2_holder *ghs); + +__attribute__ ((format(printf, 2, 3))) void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...); /** diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 0d149dcc04e5..263561bf1a50 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -325,7 +325,6 @@ static void trans_go_sync(struct gfs2_glock *gl) if (gl->gl_state != LM_ST_UNLOCKED && test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { - flush_workqueue(gfs2_delete_workqueue); gfs2_meta_syncfs(sdp); gfs2_log_shutdown(sdp); } diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 764fbb49efc8..8d3d2b4a0a7d 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -207,12 +207,14 @@ struct gfs2_glock { spinlock_t gl_spin; - unsigned int gl_state; - unsigned int gl_target; - unsigned int gl_reply; + /* State fields protected by gl_spin */ + unsigned int gl_state:2, /* Current state */ + gl_target:2, /* Target state */ + gl_demote_state:2, /* State requested by remote node */ + gl_req:2, /* State in last dlm request */ + gl_reply:8; /* Last reply from the dlm */ + unsigned int gl_hash; - unsigned int gl_req; - unsigned int gl_demote_state; /* state requested by remote node */ unsigned long gl_demote_time; /* time of first demote request */ struct list_head gl_holders; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index e1213f7f9217..14e682dbe8bf 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -916,17 +916,8 @@ static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) if (error) return error; - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - error = vmtruncate(inode, attr->ia_size); - if (error) - return error; - } - setattr_copy(inode, attr); mark_inode_dirty(inode); - - gfs2_assert_warn(GFS2_SB(inode), !error); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index 1c09425b45fd..6e493aee28f8 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -146,15 +146,13 @@ static u32 make_flags(const u32 lkid, const unsigned int gfs_flags, return lkf; } -static unsigned int gdlm_lock(struct gfs2_glock *gl, - unsigned int req_state, unsigned int flags) +static int gdlm_lock(struct gfs2_glock *gl, unsigned int req_state, + unsigned int flags) { struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct; - int error; int req; u32 lkf; - gl->gl_req = req_state; req = make_mode(req_state); lkf = make_flags(gl->gl_lksb.sb_lkid, flags, req); @@ -162,13 +160,8 @@ static unsigned int gdlm_lock(struct gfs2_glock *gl, * Submit the actual lock request. */ - error = dlm_lock(ls->ls_dlm, req, &gl->gl_lksb, lkf, gl->gl_strname, - GDLM_STRNAME_BYTES - 1, 0, gdlm_ast, gl, gdlm_bast); - if (error == -EAGAIN) - return 0; - if (error) - return LM_OUT_ERROR; - return LM_OUT_ASYNC; + return dlm_lock(ls->ls_dlm, req, &gl->gl_lksb, lkf, gl->gl_strname, + GDLM_STRNAME_BYTES - 1, 0, gdlm_ast, gl, gdlm_bast); } static void gdlm_put_lock(struct kmem_cache *cachep, struct gfs2_glock *gl) diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 12cbea7502c2..1db6b7343229 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -1069,7 +1069,6 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); - struct buffer_head *dibh; u32 ouid, ogid, nuid, ngid; int error; @@ -1100,25 +1099,10 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) if (error) goto out_gunlock_q; - error = gfs2_meta_inode_buffer(ip, &dibh); + error = gfs2_setattr_simple(ip, attr); if (error) goto out_end_trans; - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - int error; - - error = vmtruncate(inode, attr->ia_size); - gfs2_assert_warn(sdp, !error); - } - - setattr_copy(inode, attr); - mark_inode_dirty(inode); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); gfs2_quota_change(ip, -blocks, ouid, ogid); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index f606baf9ba72..a689901963de 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -666,6 +666,10 @@ static int gfs2_adjust_quota(struct gfs2_inode *ip, loff_t loc, qp->qu_limit = cpu_to_be64(fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift); qd->qd_qb.qb_limit = qp->qu_limit; } + if (fdq->d_fieldmask & FS_DQ_BCOUNT) { + qp->qu_value = cpu_to_be64(fdq->d_bcount >> sdp->sd_fsb2bb_shift); + qd->qd_qb.qb_value = qp->qu_value; + } } /* Write the quota into the quota file on disk */ @@ -1509,7 +1513,7 @@ out: } /* GFS2 only supports a subset of the XFS fields */ -#define GFS2_FIELDMASK (FS_DQ_BSOFT|FS_DQ_BHARD) +#define GFS2_FIELDMASK (FS_DQ_BSOFT|FS_DQ_BHARD|FS_DQ_BCOUNT) static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, struct fs_disk_quota *fdq) @@ -1569,9 +1573,15 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, if ((fdq->d_fieldmask & FS_DQ_BSOFT) && ((fdq->d_blk_softlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_warn))) fdq->d_fieldmask ^= FS_DQ_BSOFT; + if ((fdq->d_fieldmask & FS_DQ_BHARD) && ((fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_limit))) fdq->d_fieldmask ^= FS_DQ_BHARD; + + if ((fdq->d_fieldmask & FS_DQ_BCOUNT) && + ((fdq->d_bcount >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_value))) + fdq->d_fieldmask ^= FS_DQ_BCOUNT; + if (fdq->d_fieldmask == 0) goto out_i; @@ -1620,4 +1630,3 @@ const struct quotactl_ops gfs2_quotactl_ops = { .get_dqblk = gfs2_get_dqblk, .set_dqblk = gfs2_set_dqblk, }; - diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 33c8407b876f..7293ea27020c 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -500,7 +500,7 @@ u64 gfs2_ri_total(struct gfs2_sbd *sdp) for (rgrps = 0;; rgrps++) { loff_t pos = rgrps * sizeof(struct gfs2_rindex); - if (pos + sizeof(struct gfs2_rindex) >= i_size_read(inode)) + if (pos + sizeof(struct gfs2_rindex) > i_size_read(inode)) break; error = gfs2_internal_read(ip, &ra_state, buf, &pos, sizeof(struct gfs2_rindex)); @@ -583,7 +583,7 @@ static int read_rindex_entry(struct gfs2_inode *ip, * Returns: 0 on successful update, error code otherwise */ -static int gfs2_ri_update(struct gfs2_inode *ip) +int gfs2_ri_update(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct inode *inode = &ip->i_inode; @@ -614,46 +614,6 @@ static int gfs2_ri_update(struct gfs2_inode *ip) } /** - * gfs2_ri_update_special - Pull in a new resource index from the disk - * - * This is a special version that's safe to call from gfs2_inplace_reserve_i. - * In this case we know that we don't have any resource groups in memory yet. - * - * @ip: pointer to the rindex inode - * - * Returns: 0 on successful update, error code otherwise - */ -static int gfs2_ri_update_special(struct gfs2_inode *ip) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct inode *inode = &ip->i_inode; - struct file_ra_state ra_state; - struct gfs2_rgrpd *rgd; - unsigned int max_data = 0; - int error; - - file_ra_state_init(&ra_state, inode->i_mapping); - for (sdp->sd_rgrps = 0;; sdp->sd_rgrps++) { - /* Ignore partials */ - if ((sdp->sd_rgrps + 1) * sizeof(struct gfs2_rindex) > - i_size_read(inode)) - break; - error = read_rindex_entry(ip, &ra_state); - if (error) { - clear_rgrpdi(sdp); - return error; - } - } - list_for_each_entry(rgd, &sdp->sd_rindex_list, rd_list) - if (rgd->rd_data > max_data) - max_data = rgd->rd_data; - sdp->sd_max_rg_data = max_data; - - sdp->sd_rindex_uptodate = 1; - return 0; -} - -/** * gfs2_rindex_hold - Grab a lock on the rindex * @sdp: The GFS2 superblock * @ri_gh: the glock holder @@ -1226,16 +1186,25 @@ int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex, error = gfs2_rindex_hold(sdp, &al->al_ri_gh); else if (!sdp->sd_rgrps) /* We may not have the rindex read in, so: */ - error = gfs2_ri_update_special(ip); + error = gfs2_ri_update(ip); if (error) return error; } +try_again: do { error = get_local_rgrp(ip, &last_unlinked); /* If there is no space, flushing the log may release some */ - if (error) + if (error) { + if (ip == GFS2_I(sdp->sd_rindex) && + !sdp->sd_rindex_uptodate) { + error = gfs2_ri_update(ip); + if (error) + return error; + goto try_again; + } gfs2_log_flush(sdp, NULL); + } } while (error && tries++ < 3); if (error) { diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 0e35c0466f9a..50c2bb04369c 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -48,6 +48,7 @@ extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex, extern void gfs2_inplace_release(struct gfs2_inode *ip); +extern int gfs2_ri_update(struct gfs2_inode *ip); extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n); extern int gfs2_alloc_di(struct gfs2_inode *ip, u64 *bn, u64 *generation); diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 30b58f07c8a6..439b61c03262 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -1296,10 +1296,8 @@ fail: int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data) { - struct inode *inode = &ip->i_inode; struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_ea_location el; - struct buffer_head *dibh; int error; error = gfs2_ea_find(ip, GFS2_EATYPE_SYS, GFS2_POSIX_ACL_ACCESS, &el); @@ -1321,26 +1319,7 @@ int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data) if (error) return error; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto out_trans_end; - - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - int error; - - error = vmtruncate(inode, attr->ia_size); - gfs2_assert_warn(GFS2_SB(inode), !error); - } - - setattr_copy(inode, attr); - mark_inode_dirty(inode); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - -out_trans_end: + error = gfs2_setattr_simple(ip, attr); gfs2_trans_end(sdp); return error; } diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 8beabb958f61..47e3997f7b5c 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -154,12 +154,14 @@ enum { TRACE_EVENT_FL_ENABLED_BIT, TRACE_EVENT_FL_FILTERED_BIT, TRACE_EVENT_FL_RECORDED_CMD_BIT, + TRACE_EVENT_FL_CAP_ANY_BIT, }; enum { TRACE_EVENT_FL_ENABLED = (1 << TRACE_EVENT_FL_ENABLED_BIT), TRACE_EVENT_FL_FILTERED = (1 << TRACE_EVENT_FL_FILTERED_BIT), TRACE_EVENT_FL_RECORDED_CMD = (1 << TRACE_EVENT_FL_RECORDED_CMD_BIT), + TRACE_EVENT_FL_CAP_ANY = (1 << TRACE_EVENT_FL_CAP_ANY_BIT), }; struct ftrace_event_call { @@ -196,6 +198,14 @@ struct ftrace_event_call { #endif }; +#define __TRACE_EVENT_FLAGS(name, value) \ + static int __init trace_init_flags_##name(void) \ + { \ + event_##name.flags = value; \ + return 0; \ + } \ + early_initcall(trace_init_flags_##name); + #define PERF_MAX_TRACE_SIZE 2048 #define MAX_FILTER_PRED 32 @@ -215,6 +225,10 @@ enum { FILTER_PTR_STRING, }; +#define EVENT_STORAGE_SIZE 128 +extern struct mutex event_storage_mutex; +extern char event_storage[EVENT_STORAGE_SIZE]; + extern int trace_event_raw_init(struct ftrace_event_call *call); extern int trace_define_field(struct ftrace_event_call *call, const char *type, const char *name, int offset, int size, diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 6ed8812bfe2d..caa151fbebb7 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -90,6 +90,12 @@ extern struct group_info init_groups; */ # define CAP_INIT_BSET CAP_FULL_SET +#ifdef CONFIG_RCU_BOOST +#define INIT_TASK_RCU_BOOST() \ + .rcu_boost_mutex = NULL, +#else +#define INIT_TASK_RCU_BOOST() +#endif #ifdef CONFIG_TREE_PREEMPT_RCU #define INIT_TASK_RCU_TREE_PREEMPT() \ .rcu_blocked_node = NULL, @@ -101,7 +107,8 @@ extern struct group_info init_groups; .rcu_read_lock_nesting = 0, \ .rcu_read_unlock_special = 0, \ .rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry), \ - INIT_TASK_RCU_TREE_PREEMPT() + INIT_TASK_RCU_TREE_PREEMPT() \ + INIT_TASK_RCU_BOOST() #else #define INIT_TASK_RCU_PREEMPT(tsk) #endif diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h index e7d1b2e0070d..b78edb58ee66 100644 --- a/include/linux/kprobes.h +++ b/include/linux/kprobes.h @@ -275,7 +275,9 @@ extern int arch_prepared_optinsn(struct arch_optimized_insn *optinsn); extern int arch_check_optimized_kprobe(struct optimized_kprobe *op); extern int arch_prepare_optimized_kprobe(struct optimized_kprobe *op); extern void arch_remove_optimized_kprobe(struct optimized_kprobe *op); -extern int arch_optimize_kprobe(struct optimized_kprobe *op); +extern void arch_optimize_kprobes(struct list_head *oplist); +extern void arch_unoptimize_kprobes(struct list_head *oplist, + struct list_head *done_list); extern void arch_unoptimize_kprobe(struct optimized_kprobe *op); extern kprobe_opcode_t *get_optinsn_slot(void); extern void free_optinsn_slot(kprobe_opcode_t *slot, int dirty); diff --git a/include/linux/nmi.h b/include/linux/nmi.h index 06aab5eee134..c536f8545f74 100644 --- a/include/linux/nmi.h +++ b/include/linux/nmi.h @@ -14,22 +14,14 @@ * may be used to reset the timeout - for code which intentionally * disables interrupts for a long time. This call is stateless. */ -#ifdef ARCH_HAS_NMI_WATCHDOG +#if defined(ARCH_HAS_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR) #include <asm/nmi.h> extern void touch_nmi_watchdog(void); -extern void acpi_nmi_disable(void); -extern void acpi_nmi_enable(void); #else -#ifndef CONFIG_HARDLOCKUP_DETECTOR static inline void touch_nmi_watchdog(void) { touch_softlockup_watchdog(); } -#else -extern void touch_nmi_watchdog(void); -#endif -static inline void acpi_nmi_disable(void) { } -static inline void acpi_nmi_enable(void) { } #endif /* diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 4f1279e105ee..dda5b0a3ff60 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -215,8 +215,9 @@ struct perf_event_attr { */ precise_ip : 2, /* skid constraint */ mmap_data : 1, /* non-exec mmap data */ + sample_id_all : 1, /* sample_type all events */ - __reserved_1 : 46; + __reserved_1 : 45; union { __u32 wakeup_events; /* wakeup every n events */ @@ -327,6 +328,15 @@ struct perf_event_header { enum perf_event_type { /* + * If perf_event_attr.sample_id_all is set then all event types will + * have the sample_type selected fields related to where/when + * (identity) an event took place (TID, TIME, ID, CPU, STREAM_ID) + * described in PERF_RECORD_SAMPLE below, it will be stashed just after + * the perf_event_header and the fields already present for the existing + * fields, i.e. at the end of the payload. That way a newer perf.data + * file will be supported by older perf tools, with these new optional + * fields being ignored. + * * The MMAP events record the PROT_EXEC mappings so that we can * correlate userspace IPs to code. They have the following structure: * @@ -578,6 +588,10 @@ struct perf_event; struct pmu { struct list_head entry; + struct device *dev; + char *name; + int type; + int * __percpu pmu_disable_count; struct perf_cpu_context * __percpu pmu_cpu_context; int task_ctx_nr; @@ -758,6 +772,9 @@ struct perf_event { u64 shadow_ctx_time; struct perf_event_attr attr; + u16 header_size; + u16 id_header_size; + u16 read_size; struct hw_perf_event hw; struct perf_event_context *ctx; @@ -903,7 +920,7 @@ struct perf_output_handle { #ifdef CONFIG_PERF_EVENTS -extern int perf_pmu_register(struct pmu *pmu); +extern int perf_pmu_register(struct pmu *pmu, char *name, int type); extern void perf_pmu_unregister(struct pmu *pmu); extern int perf_num_counters(void); @@ -970,6 +987,11 @@ extern int perf_event_overflow(struct perf_event *event, int nmi, struct perf_sample_data *data, struct pt_regs *regs); +static inline bool is_sampling_event(struct perf_event *event) +{ + return event->attr.sample_period != 0; +} + /* * Return 1 for a software event, 0 for a hardware event */ diff --git a/include/linux/rculist.h b/include/linux/rculist.h index f31ef61f1c65..2dea94fc4402 100644 --- a/include/linux/rculist.h +++ b/include/linux/rculist.h @@ -241,11 +241,6 @@ static inline void list_splice_init_rcu(struct list_head *list, #define list_first_entry_rcu(ptr, type, member) \ list_entry_rcu((ptr)->next, type, member) -#define __list_for_each_rcu(pos, head) \ - for (pos = rcu_dereference_raw(list_next_rcu(head)); \ - pos != (head); \ - pos = rcu_dereference_raw(list_next_rcu((pos))) - /** * list_for_each_entry_rcu - iterate over rcu list of given type * @pos: the type * to use as a loop cursor. diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 03cda7bed985..af5614856285 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -47,6 +47,8 @@ extern int rcutorture_runnable; /* for sysctl */ #endif /* #ifdef CONFIG_RCU_TORTURE_TEST */ +#define UINT_CMP_GE(a, b) (UINT_MAX / 2 >= (a) - (b)) +#define UINT_CMP_LT(a, b) (UINT_MAX / 2 < (a) - (b)) #define ULONG_CMP_GE(a, b) (ULONG_MAX / 2 >= (a) - (b)) #define ULONG_CMP_LT(a, b) (ULONG_MAX / 2 < (a) - (b)) @@ -66,7 +68,6 @@ extern void call_rcu_sched(struct rcu_head *head, extern void synchronize_sched(void); extern void rcu_barrier_bh(void); extern void rcu_barrier_sched(void); -extern void synchronize_sched_expedited(void); extern int sched_expedited_torture_stats(char *page); static inline void __rcu_read_lock_bh(void) @@ -118,7 +119,6 @@ static inline int rcu_preempt_depth(void) #endif /* #else #ifdef CONFIG_PREEMPT_RCU */ /* Internal to kernel */ -extern void rcu_init(void); extern void rcu_sched_qs(int cpu); extern void rcu_bh_qs(int cpu); extern void rcu_check_callbacks(int cpu, int user); diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index 13877cb93a60..30ebd7c8d874 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -27,7 +27,9 @@ #include <linux/cache.h> -#define rcu_init_sched() do { } while (0) +static inline void rcu_init(void) +{ +} #ifdef CONFIG_TINY_RCU @@ -58,6 +60,11 @@ static inline void synchronize_rcu_bh_expedited(void) synchronize_sched(); } +static inline void synchronize_sched_expedited(void) +{ + synchronize_sched(); +} + #ifdef CONFIG_TINY_RCU static inline void rcu_preempt_note_context_switch(void) @@ -125,16 +132,12 @@ static inline void rcu_cpu_stall_reset(void) } #ifdef CONFIG_DEBUG_LOCK_ALLOC - extern int rcu_scheduler_active __read_mostly; extern void rcu_scheduler_starting(void); - #else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ - static inline void rcu_scheduler_starting(void) { } - #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */ #endif /* __LINUX_RCUTINY_H */ diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index 95518e628794..3a933482734a 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -30,6 +30,7 @@ #ifndef __LINUX_RCUTREE_H #define __LINUX_RCUTREE_H +extern void rcu_init(void); extern void rcu_note_context_switch(int cpu); extern int rcu_needs_cpu(int cpu); extern void rcu_cpu_stall_reset(void); @@ -47,6 +48,7 @@ static inline void exit_rcu(void) #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ extern void synchronize_rcu_bh(void); +extern void synchronize_sched_expedited(void); extern void synchronize_rcu_expedited(void); static inline void synchronize_rcu_bh_expedited(void) diff --git a/include/linux/sched.h b/include/linux/sched.h index 883ad10eadbf..777cd01e240e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -316,6 +316,7 @@ extern int proc_dowatchdog_thresh(struct ctl_table *table, int write, size_t *lenp, loff_t *ppos); extern unsigned int softlockup_panic; extern int softlockup_thresh; +void lockup_detector_init(void); #else static inline void touch_softlockup_watchdog(void) { @@ -326,6 +327,9 @@ static inline void touch_softlockup_watchdog_sync(void) static inline void touch_all_softlockup_watchdogs(void) { } +static inline void lockup_detector_init(void) +{ +} #endif #ifdef CONFIG_DETECT_HUNG_TASK @@ -1234,6 +1238,9 @@ struct task_struct { #ifdef CONFIG_TREE_PREEMPT_RCU struct rcu_node *rcu_blocked_node; #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifdef CONFIG_RCU_BOOST + struct rt_mutex *rcu_boost_mutex; +#endif /* #ifdef CONFIG_RCU_BOOST */ #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) struct sched_info sched_info; @@ -1766,7 +1773,8 @@ extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t * #ifdef CONFIG_PREEMPT_RCU #define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */ -#define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */ +#define RCU_READ_UNLOCK_BOOSTED (1 << 1) /* boosted while in RCU read-side. */ +#define RCU_READ_UNLOCK_NEED_QS (1 << 2) /* RCU core needs CPU response. */ static inline void rcu_copy_process(struct task_struct *p) { @@ -1774,7 +1782,10 @@ static inline void rcu_copy_process(struct task_struct *p) p->rcu_read_unlock_special = 0; #ifdef CONFIG_TREE_PREEMPT_RCU p->rcu_blocked_node = NULL; -#endif +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifdef CONFIG_RCU_BOOST + p->rcu_boost_mutex = NULL; +#endif /* #ifdef CONFIG_RCU_BOOST */ INIT_LIST_HEAD(&p->rcu_node_entry); } diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h index 51efbef38fb0..25310f1d7f37 100644 --- a/include/linux/stacktrace.h +++ b/include/linux/stacktrace.h @@ -2,6 +2,7 @@ #define __LINUX_STACKTRACE_H struct task_struct; +struct pt_regs; #ifdef CONFIG_STACKTRACE struct task_struct; @@ -13,7 +14,8 @@ struct stack_trace { }; extern void save_stack_trace(struct stack_trace *trace); -extern void save_stack_trace_bp(struct stack_trace *trace, unsigned long bp); +extern void save_stack_trace_regs(struct stack_trace *trace, + struct pt_regs *regs); extern void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace); diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index cacc27a0e285..18cd0684fc4e 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -127,8 +127,6 @@ extern struct trace_event_functions exit_syscall_print_funcs; #define SYSCALL_TRACE_ENTER_EVENT(sname) \ static struct syscall_metadata \ __attribute__((__aligned__(4))) __syscall_meta_##sname; \ - static struct ftrace_event_call \ - __attribute__((__aligned__(4))) event_enter_##sname; \ static struct ftrace_event_call __used \ __attribute__((__aligned__(4))) \ __attribute__((section("_ftrace_events"))) \ @@ -137,13 +135,12 @@ extern struct trace_event_functions exit_syscall_print_funcs; .class = &event_class_syscall_enter, \ .event.funcs = &enter_syscall_print_funcs, \ .data = (void *)&__syscall_meta_##sname,\ - } + }; \ + __TRACE_EVENT_FLAGS(enter_##sname, TRACE_EVENT_FL_CAP_ANY) #define SYSCALL_TRACE_EXIT_EVENT(sname) \ static struct syscall_metadata \ __attribute__((__aligned__(4))) __syscall_meta_##sname; \ - static struct ftrace_event_call \ - __attribute__((__aligned__(4))) event_exit_##sname; \ static struct ftrace_event_call __used \ __attribute__((__aligned__(4))) \ __attribute__((section("_ftrace_events"))) \ @@ -152,7 +149,8 @@ extern struct trace_event_functions exit_syscall_print_funcs; .class = &event_class_syscall_exit, \ .event.funcs = &exit_syscall_print_funcs, \ .data = (void *)&__syscall_meta_##sname,\ - } + }; \ + __TRACE_EVENT_FLAGS(exit_##sname, TRACE_EVENT_FL_CAP_ANY) #define SYSCALL_METADATA(sname, nb) \ SYSCALL_TRACE_ENTER_EVENT(sname); \ diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index a4a90b6726ce..d3e4f87e95c0 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -106,6 +106,7 @@ static inline void tracepoint_update_probe_range(struct tracepoint *begin, #define TP_PROTO(args...) args #define TP_ARGS(args...) args +#define TP_CONDITION(args...) args #ifdef CONFIG_TRACEPOINTS @@ -119,12 +120,14 @@ static inline void tracepoint_update_probe_range(struct tracepoint *begin, * as "(void *, void)". The DECLARE_TRACE_NOARGS() will pass in just * "void *data", where as the DECLARE_TRACE() will pass in "void *data, proto". */ -#define __DO_TRACE(tp, proto, args) \ +#define __DO_TRACE(tp, proto, args, cond) \ do { \ struct tracepoint_func *it_func_ptr; \ void *it_func; \ void *__data; \ \ + if (!(cond)) \ + return; \ rcu_read_lock_sched_notrace(); \ it_func_ptr = rcu_dereference_sched((tp)->funcs); \ if (it_func_ptr) { \ @@ -142,7 +145,7 @@ static inline void tracepoint_update_probe_range(struct tracepoint *begin, * not add unwanted padding between the beginning of the section and the * structure. Force alignment to the same alignment as the section start. */ -#define __DECLARE_TRACE(name, proto, args, data_proto, data_args) \ +#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \ extern struct tracepoint __tracepoint_##name; \ static inline void trace_##name(proto) \ { \ @@ -151,7 +154,8 @@ static inline void tracepoint_update_probe_range(struct tracepoint *begin, do_trace: \ __DO_TRACE(&__tracepoint_##name, \ TP_PROTO(data_proto), \ - TP_ARGS(data_args)); \ + TP_ARGS(data_args), \ + TP_CONDITION(cond)); \ } \ static inline int \ register_trace_##name(void (*probe)(data_proto), void *data) \ @@ -186,7 +190,7 @@ do_trace: \ EXPORT_SYMBOL(__tracepoint_##name) #else /* !CONFIG_TRACEPOINTS */ -#define __DECLARE_TRACE(name, proto, args, data_proto, data_args) \ +#define __DECLARE_TRACE(name, proto, args, cond, data_proto, data_args) \ static inline void trace_##name(proto) \ { } \ static inline int \ @@ -227,13 +231,20 @@ do_trace: \ * "void *__data, proto" as the callback prototype. */ #define DECLARE_TRACE_NOARGS(name) \ - __DECLARE_TRACE(name, void, , void *__data, __data) + __DECLARE_TRACE(name, void, , 1, void *__data, __data) #define DECLARE_TRACE(name, proto, args) \ - __DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), \ + __DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), 1, \ PARAMS(void *__data, proto), \ PARAMS(__data, args)) +#define DECLARE_TRACE_CONDITION(name, proto, args, cond) \ + __DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), PARAMS(cond), \ + PARAMS(void *__data, proto), \ + PARAMS(__data, args)) + +#define TRACE_EVENT_FLAGS(event, flag) + #endif /* DECLARE_TRACE */ #ifndef TRACE_EVENT @@ -347,11 +358,21 @@ do_trace: \ DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) #define DEFINE_EVENT_PRINT(template, name, proto, args, print) \ DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) +#define DEFINE_EVENT_CONDITION(template, name, proto, \ + args, cond) \ + DECLARE_TRACE_CONDITION(name, PARAMS(proto), \ + PARAMS(args), PARAMS(cond)) #define TRACE_EVENT(name, proto, args, struct, assign, print) \ DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) #define TRACE_EVENT_FN(name, proto, args, struct, \ assign, print, reg, unreg) \ DECLARE_TRACE(name, PARAMS(proto), PARAMS(args)) +#define TRACE_EVENT_CONDITION(name, proto, args, cond, \ + struct, assign, print) \ + DECLARE_TRACE_CONDITION(name, PARAMS(proto), \ + PARAMS(args), PARAMS(cond)) + +#define TRACE_EVENT_FLAGS(event, flag) #endif /* ifdef TRACE_EVENT (see note above) */ diff --git a/include/trace/define_trace.h b/include/trace/define_trace.h index 1dfab5401511..b0b4eb24d592 100644 --- a/include/trace/define_trace.h +++ b/include/trace/define_trace.h @@ -26,6 +26,15 @@ #define TRACE_EVENT(name, proto, args, tstruct, assign, print) \ DEFINE_TRACE(name) +#undef TRACE_EVENT_CONDITION +#define TRACE_EVENT_CONDITION(name, proto, args, cond, tstruct, assign, print) \ + TRACE_EVENT(name, \ + PARAMS(proto), \ + PARAMS(args), \ + PARAMS(tstruct), \ + PARAMS(assign), \ + PARAMS(print)) + #undef TRACE_EVENT_FN #define TRACE_EVENT_FN(name, proto, args, tstruct, \ assign, print, reg, unreg) \ @@ -39,6 +48,10 @@ #define DEFINE_EVENT_PRINT(template, name, proto, args, print) \ DEFINE_TRACE(name) +#undef DEFINE_EVENT_CONDITION +#define DEFINE_EVENT_CONDITION(template, name, proto, args, cond) \ + DEFINE_EVENT(template, name, PARAMS(proto), PARAMS(args)) + #undef DECLARE_TRACE #define DECLARE_TRACE(name, proto, args) \ DEFINE_TRACE(name) @@ -75,9 +88,11 @@ #undef TRACE_EVENT #undef TRACE_EVENT_FN +#undef TRACE_EVENT_CONDITION #undef DECLARE_EVENT_CLASS #undef DEFINE_EVENT #undef DEFINE_EVENT_PRINT +#undef DEFINE_EVENT_CONDITION #undef TRACE_HEADER_MULTI_READ #undef DECLARE_TRACE diff --git a/include/trace/events/power.h b/include/trace/events/power.h index 286784d69b8f..1bcc2a8c00e2 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -7,16 +7,67 @@ #include <linux/ktime.h> #include <linux/tracepoint.h> -#ifndef _TRACE_POWER_ENUM_ -#define _TRACE_POWER_ENUM_ -enum { - POWER_NONE = 0, - POWER_CSTATE = 1, /* C-State */ - POWER_PSTATE = 2, /* Fequency change or DVFS */ - POWER_SSTATE = 3, /* Suspend */ -}; +DECLARE_EVENT_CLASS(cpu, + + TP_PROTO(unsigned int state, unsigned int cpu_id), + + TP_ARGS(state, cpu_id), + + TP_STRUCT__entry( + __field( u32, state ) + __field( u32, cpu_id ) + ), + + TP_fast_assign( + __entry->state = state; + __entry->cpu_id = cpu_id; + ), + + TP_printk("state=%lu cpu_id=%lu", (unsigned long)__entry->state, + (unsigned long)__entry->cpu_id) +); + +DEFINE_EVENT(cpu, cpu_idle, + + TP_PROTO(unsigned int state, unsigned int cpu_id), + + TP_ARGS(state, cpu_id) +); + +/* This file can get included multiple times, TRACE_HEADER_MULTI_READ at top */ +#ifndef _PWR_EVENT_AVOID_DOUBLE_DEFINING +#define _PWR_EVENT_AVOID_DOUBLE_DEFINING + +#define PWR_EVENT_EXIT -1 #endif +DEFINE_EVENT(cpu, cpu_frequency, + + TP_PROTO(unsigned int frequency, unsigned int cpu_id), + + TP_ARGS(frequency, cpu_id) +); + +TRACE_EVENT(machine_suspend, + + TP_PROTO(unsigned int state), + + TP_ARGS(state), + + TP_STRUCT__entry( + __field( u32, state ) + ), + + TP_fast_assign( + __entry->state = state; + ), + + TP_printk("state=%lu", (unsigned long)__entry->state) +); + +/* This code will be removed after deprecation time exceeded (2.6.41) */ +#ifdef CONFIG_EVENT_POWER_TRACING_DEPRECATED + /* * The power events are used for cpuidle & suspend (power_start, power_end) * and for cpufreq (power_frequency) @@ -75,6 +126,36 @@ TRACE_EVENT(power_end, ); +/* Deprecated dummy functions must be protected against multi-declartion */ +#ifndef _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED +#define _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED + +enum { + POWER_NONE = 0, + POWER_CSTATE = 1, + POWER_PSTATE = 2, +}; +#endif /* _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED */ + +#else /* CONFIG_EVENT_POWER_TRACING_DEPRECATED */ + +#ifndef _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED +#define _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED +enum { + POWER_NONE = 0, + POWER_CSTATE = 1, + POWER_PSTATE = 2, +}; + +/* These dummy declaration have to be ripped out when the deprecated + events get removed */ +static inline void trace_power_start(u64 type, u64 state, u64 cpuid) {}; +static inline void trace_power_end(u64 cpuid) {}; +static inline void trace_power_frequency(u64 type, u64 state, u64 cpuid) {}; +#endif /* _PWR_EVENT_AVOID_DOUBLE_DEFINING_DEPRECATED */ + +#endif /* CONFIG_EVENT_POWER_TRACING_DEPRECATED */ + /* * The clock events are used for clock enable/disable and for * clock rate change @@ -153,7 +234,6 @@ DEFINE_EVENT(power_domain, power_domain_target, TP_ARGS(name, state, cpu_id) ); - #endif /* _TRACE_POWER_H */ /* This part must be outside protection */ diff --git a/include/trace/events/syscalls.h b/include/trace/events/syscalls.h index fb726ac7caee..5a4c04a75b3d 100644 --- a/include/trace/events/syscalls.h +++ b/include/trace/events/syscalls.h @@ -40,6 +40,8 @@ TRACE_EVENT_FN(sys_enter, syscall_regfunc, syscall_unregfunc ); +TRACE_EVENT_FLAGS(sys_enter, TRACE_EVENT_FL_CAP_ANY) + TRACE_EVENT_FN(sys_exit, TP_PROTO(struct pt_regs *regs, long ret), @@ -62,6 +64,8 @@ TRACE_EVENT_FN(sys_exit, syscall_regfunc, syscall_unregfunc ); +TRACE_EVENT_FLAGS(sys_exit, TRACE_EVENT_FL_CAP_ANY) + #endif /* CONFIG_HAVE_SYSCALL_TRACEPOINTS */ #endif /* _TRACE_EVENTS_SYSCALLS_H */ diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index a9377c0083ad..e16610c208c9 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -82,6 +82,10 @@ TRACE_EVENT(name, PARAMS(proto), PARAMS(args), \ PARAMS(tstruct), PARAMS(assign), PARAMS(print)) \ +#undef TRACE_EVENT_FLAGS +#define TRACE_EVENT_FLAGS(name, value) \ + __TRACE_EVENT_FLAGS(name, value) + #include TRACE_INCLUDE(TRACE_INCLUDE_FILE) @@ -129,6 +133,9 @@ #define DEFINE_EVENT_PRINT(template, name, proto, args, print) \ DEFINE_EVENT(template, name, PARAMS(proto), PARAMS(args)) +#undef TRACE_EVENT_FLAGS +#define TRACE_EVENT_FLAGS(event, flag) + #include TRACE_INCLUDE(TRACE_INCLUDE_FILE) /* @@ -289,13 +296,19 @@ static struct trace_event_functions ftrace_event_type_funcs_##call = { \ #undef __array #define __array(type, item, len) \ - BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ - ret = trace_define_field(event_call, #type "[" #len "]", #item, \ + do { \ + mutex_lock(&event_storage_mutex); \ + BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ + snprintf(event_storage, sizeof(event_storage), \ + "%s[%d]", #type, len); \ + ret = trace_define_field(event_call, event_storage, #item, \ offsetof(typeof(field), item), \ sizeof(field.item), \ is_signed_type(type), FILTER_OTHER); \ - if (ret) \ - return ret; + mutex_unlock(&event_storage_mutex); \ + if (ret) \ + return ret; \ + } while (0); #undef __dynamic_array #define __dynamic_array(type, item, len) \ diff --git a/init/Kconfig b/init/Kconfig index 62b71491dde5..8dfd094e6875 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -393,7 +393,6 @@ config PREEMPT_RCU config RCU_TRACE bool "Enable tracing for RCU" - depends on TREE_RCU || TREE_PREEMPT_RCU help This option provides tracing in RCU which presents stats in debugfs for debugging RCU implementation. @@ -459,6 +458,60 @@ config TREE_RCU_TRACE TREE_PREEMPT_RCU implementations, permitting Makefile to trivially select kernel/rcutree_trace.c. +config RCU_BOOST + bool "Enable RCU priority boosting" + depends on RT_MUTEXES && TINY_PREEMPT_RCU + default n + help + This option boosts the priority of preempted RCU readers that + block the current preemptible RCU grace period for too long. + This option also prevents heavy loads from blocking RCU + callback invocation for all flavors of RCU. + + Say Y here if you are working with real-time apps or heavy loads + Say N here if you are unsure. + +config RCU_BOOST_PRIO + int "Real-time priority to boost RCU readers to" + range 1 99 + depends on RCU_BOOST + default 1 + help + This option specifies the real-time priority to which preempted + RCU readers are to be boosted. If you are working with CPU-bound + real-time applications, you should specify a priority higher then + the highest-priority CPU-bound application. + + Specify the real-time priority, or take the default if unsure. + +config RCU_BOOST_DELAY + int "Milliseconds to delay boosting after RCU grace-period start" + range 0 3000 + depends on RCU_BOOST + default 500 + help + This option specifies the time to wait after the beginning of + a given grace period before priority-boosting preempted RCU + readers blocking that grace period. Note that any RCU reader + blocking an expedited RCU grace period is boosted immediately. + + Accept the default if unsure. + +config SRCU_SYNCHRONIZE_DELAY + int "Microseconds to delay before waiting for readers" + range 0 20 + default 10 + help + This option controls how long SRCU delays before entering its + loop waiting on SRCU readers. The purpose of this loop is + to avoid the unconditional context-switch penalty that would + otherwise be incurred if there was an active SRCU reader, + in a manner similar to adaptive locking schemes. This should + be set to be a bit longer than the common-case SRCU read-side + critical-section overhead. + + Accept the default if unsure. + endmenu # "RCU Subsystem" config IKCONFIG diff --git a/init/main.c b/init/main.c index 8646401f7a0e..ea51770c0170 100644 --- a/init/main.c +++ b/init/main.c @@ -67,6 +67,7 @@ #include <linux/sfi.h> #include <linux/shmem_fs.h> #include <linux/slab.h> +#include <linux/perf_event.h> #include <asm/io.h> #include <asm/bugs.h> @@ -603,6 +604,8 @@ asmlinkage void __init start_kernel(void) "enabled *very* early, fixing it\n"); local_irq_disable(); } + idr_init_cache(); + perf_event_init(); rcu_init(); radix_tree_init(); /* init some links before init_ISA_irqs() */ @@ -658,7 +661,6 @@ asmlinkage void __init start_kernel(void) enable_debug_pagealloc(); kmemleak_init(); debug_objects_mem_init(); - idr_init_cache(); setup_per_cpu_pageset(); numa_policy_init(); if (late_time_init) @@ -882,6 +884,7 @@ static int __init kernel_init(void * unused) smp_prepare_cpus(setup_max_cpus); do_pre_smp_initcalls(); + lockup_detector_init(); smp_init(); sched_init_smp(); diff --git a/kernel/futex.c b/kernel/futex.c index 40a8777a27d0..3019b92e6917 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -69,6 +69,14 @@ int __read_mostly futex_cmpxchg_enabled; #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) /* + * Futex flags used to encode options to functions and preserve them across + * restarts. + */ +#define FLAGS_SHARED 0x01 +#define FLAGS_CLOCKRT 0x02 +#define FLAGS_HAS_TIMEOUT 0x04 + +/* * Priority Inheritance state: */ struct futex_pi_state { @@ -123,6 +131,12 @@ struct futex_q { u32 bitset; }; +static const struct futex_q futex_q_init = { + /* list gets initialized in queue_me()*/ + .key = FUTEX_KEY_INIT, + .bitset = FUTEX_BITSET_MATCH_ANY +}; + /* * Hash buckets are shared by all the futex_keys that hash to the same * location. Each key may have multiple futex_q structures, one for each task @@ -283,8 +297,7 @@ again: return 0; } -static inline -void put_futex_key(int fshared, union futex_key *key) +static inline void put_futex_key(union futex_key *key) { drop_futex_key_refs(key); } @@ -870,7 +883,8 @@ double_unlock_hb(struct futex_hash_bucket *hb1, struct futex_hash_bucket *hb2) /* * Wake up waiters matching bitset queued on this futex (uaddr). */ -static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) +static int +futex_wake(u32 __user *uaddr, unsigned int flags, int nr_wake, u32 bitset) { struct futex_hash_bucket *hb; struct futex_q *this, *next; @@ -881,7 +895,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) if (!bitset) return -EINVAL; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key); if (unlikely(ret != 0)) goto out; @@ -907,7 +921,7 @@ static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) } spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); out: return ret; } @@ -917,7 +931,7 @@ out: * to this virtual address: */ static int -futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, +futex_wake_op(u32 __user *uaddr1, unsigned int flags, u32 __user *uaddr2, int nr_wake, int nr_wake2, int op) { union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; @@ -927,10 +941,10 @@ futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int ret, op_ret; retry: - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); if (unlikely(ret != 0)) goto out_put_key1; @@ -962,11 +976,11 @@ retry_private: if (ret) goto out_put_keys; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); goto retry; } @@ -996,9 +1010,9 @@ retry_private: double_unlock_hb(hb1, hb2); out_put_keys: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out_put_key1: - put_futex_key(fshared, &key1); + put_futex_key(&key1); out: return ret; } @@ -1133,13 +1147,13 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, /** * futex_requeue() - Requeue waiters from uaddr1 to uaddr2 * @uaddr1: source futex user address - * @fshared: 0 for a PROCESS_PRIVATE futex, 1 for PROCESS_SHARED + * @flags: futex flags (FLAGS_SHARED, etc.) * @uaddr2: target futex user address * @nr_wake: number of waiters to wake (must be 1 for requeue_pi) * @nr_requeue: number of waiters to requeue (0-INT_MAX) * @cmpval: @uaddr1 expected value (or %NULL) * @requeue_pi: if we are attempting to requeue from a non-pi futex to a - * pi futex (pi to pi requeue is not supported) + * pi futex (pi to pi requeue is not supported) * * Requeue waiters on uaddr1 to uaddr2. In the requeue_pi case, try to acquire * uaddr2 atomically on behalf of the top waiter. @@ -1148,9 +1162,9 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, * >=0 - on success, the number of tasks requeued or woken * <0 - on error */ -static int futex_requeue(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, - int nr_wake, int nr_requeue, u32 *cmpval, - int requeue_pi) +static int futex_requeue(u32 __user *uaddr1, unsigned int flags, + u32 __user *uaddr2, int nr_wake, int nr_requeue, + u32 *cmpval, int requeue_pi) { union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; int drop_count = 0, task_count = 0, ret; @@ -1191,10 +1205,10 @@ retry: pi_state = NULL; } - ret = get_futex_key(uaddr1, fshared, &key1); + ret = get_futex_key(uaddr1, flags & FLAGS_SHARED, &key1); if (unlikely(ret != 0)) goto out; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); if (unlikely(ret != 0)) goto out_put_key1; @@ -1216,11 +1230,11 @@ retry_private: if (ret) goto out_put_keys; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); goto retry; } if (curval != *cmpval) { @@ -1260,8 +1274,8 @@ retry_private: break; case -EFAULT: double_unlock_hb(hb1, hb2); - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); ret = fault_in_user_writeable(uaddr2); if (!ret) goto retry; @@ -1269,8 +1283,8 @@ retry_private: case -EAGAIN: /* The owner was exiting, try again. */ double_unlock_hb(hb1, hb2); - put_futex_key(fshared, &key2); - put_futex_key(fshared, &key1); + put_futex_key(&key2); + put_futex_key(&key1); cond_resched(); goto retry; default: @@ -1352,9 +1366,9 @@ out_unlock: drop_futex_key_refs(&key1); out_put_keys: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out_put_key1: - put_futex_key(fshared, &key1); + put_futex_key(&key1); out: if (pi_state != NULL) free_pi_state(pi_state); @@ -1494,7 +1508,7 @@ static void unqueue_me_pi(struct futex_q *q) * private futexes. */ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, - struct task_struct *newowner, int fshared) + struct task_struct *newowner) { u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; @@ -1587,20 +1601,11 @@ handle_fault: goto retry; } -/* - * In case we must use restart_block to restart a futex_wait, - * we encode in the 'flags' shared capability - */ -#define FLAGS_SHARED 0x01 -#define FLAGS_CLOCKRT 0x02 -#define FLAGS_HAS_TIMEOUT 0x04 - static long futex_wait_restart(struct restart_block *restart); /** * fixup_owner() - Post lock pi_state and corner case management * @uaddr: user address of the futex - * @fshared: whether the futex is shared (1) or not (0) * @q: futex_q (contains pi_state and access to the rt_mutex) * @locked: if the attempt to take the rt_mutex succeeded (1) or not (0) * @@ -1613,8 +1618,7 @@ static long futex_wait_restart(struct restart_block *restart); * 0 - success, lock not taken * <0 - on error (-EFAULT) */ -static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, - int locked) +static int fixup_owner(u32 __user *uaddr, struct futex_q *q, int locked) { struct task_struct *owner; int ret = 0; @@ -1625,7 +1629,7 @@ static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, * did a lock-steal - fix up the PI-state in that case: */ if (q->pi_state->owner != current) - ret = fixup_pi_state_owner(uaddr, q, current, fshared); + ret = fixup_pi_state_owner(uaddr, q, current); goto out; } @@ -1652,7 +1656,7 @@ static int fixup_owner(u32 __user *uaddr, int fshared, struct futex_q *q, * lock. Fix the state up. */ owner = rt_mutex_owner(&q->pi_state->pi_mutex); - ret = fixup_pi_state_owner(uaddr, q, owner, fshared); + ret = fixup_pi_state_owner(uaddr, q, owner); goto out; } @@ -1715,7 +1719,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * futex_wait_setup() - Prepare to wait on a futex * @uaddr: the futex userspace address * @val: the expected value - * @fshared: whether the futex is shared (1) or not (0) + * @flags: futex flags (FLAGS_SHARED, etc.) * @q: the associated futex_q * @hb: storage for hash_bucket pointer to be returned to caller * @@ -1728,7 +1732,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * 0 - uaddr contains val and hb has been locked * <1 - -EFAULT or -EWOULDBLOCK (uaddr does not contain val) and hb is unlcoked */ -static int futex_wait_setup(u32 __user *uaddr, u32 val, int fshared, +static int futex_wait_setup(u32 __user *uaddr, u32 val, unsigned int flags, struct futex_q *q, struct futex_hash_bucket **hb) { u32 uval; @@ -1752,8 +1756,7 @@ static int futex_wait_setup(u32 __user *uaddr, u32 val, int fshared, * rare, but normal. */ retry: - q->key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q->key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q->key); if (unlikely(ret != 0)) return ret; @@ -1769,10 +1772,10 @@ retry_private: if (ret) goto out; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &q->key); + put_futex_key(&q->key); goto retry; } @@ -1783,32 +1786,29 @@ retry_private: out: if (ret) - put_futex_key(fshared, &q->key); + put_futex_key(&q->key); return ret; } -static int futex_wait(u32 __user *uaddr, int fshared, - u32 val, ktime_t *abs_time, u32 bitset, int clockrt) +static int futex_wait(u32 __user *uaddr, unsigned int flags, u32 val, + ktime_t *abs_time, u32 bitset) { struct hrtimer_sleeper timeout, *to = NULL; struct restart_block *restart; struct futex_hash_bucket *hb; - struct futex_q q; + struct futex_q q = futex_q_init; int ret; if (!bitset) return -EINVAL; - - q.pi_state = NULL; q.bitset = bitset; - q.rt_waiter = NULL; - q.requeue_pi_key = NULL; if (abs_time) { to = &timeout; - hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : - CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_on_stack(&to->timer, (flags & FLAGS_CLOCKRT) ? + CLOCK_REALTIME : CLOCK_MONOTONIC, + HRTIMER_MODE_ABS); hrtimer_init_sleeper(to, current); hrtimer_set_expires_range_ns(&to->timer, *abs_time, current->timer_slack_ns); @@ -1819,7 +1819,7 @@ retry: * Prepare to wait on uaddr. On success, holds hb lock and increments * q.key refs. */ - ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); + ret = futex_wait_setup(uaddr, val, flags, &q, &hb); if (ret) goto out; @@ -1852,12 +1852,7 @@ retry: restart->futex.val = val; restart->futex.time = abs_time->tv64; restart->futex.bitset = bitset; - restart->futex.flags = FLAGS_HAS_TIMEOUT; - - if (fshared) - restart->futex.flags |= FLAGS_SHARED; - if (clockrt) - restart->futex.flags |= FLAGS_CLOCKRT; + restart->futex.flags = flags; ret = -ERESTART_RESTARTBLOCK; @@ -1873,7 +1868,6 @@ out: static long futex_wait_restart(struct restart_block *restart) { u32 __user *uaddr = restart->futex.uaddr; - int fshared = 0; ktime_t t, *tp = NULL; if (restart->futex.flags & FLAGS_HAS_TIMEOUT) { @@ -1881,11 +1875,9 @@ static long futex_wait_restart(struct restart_block *restart) tp = &t; } restart->fn = do_no_restart_syscall; - if (restart->futex.flags & FLAGS_SHARED) - fshared = 1; - return (long)futex_wait(uaddr, fshared, restart->futex.val, tp, - restart->futex.bitset, - restart->futex.flags & FLAGS_CLOCKRT); + + return (long)futex_wait(uaddr, restart->futex.flags, + restart->futex.val, tp, restart->futex.bitset); } @@ -1895,12 +1887,12 @@ static long futex_wait_restart(struct restart_block *restart) * if there are waiters then it will block, it does PI, etc. (Due to * races the kernel might see a 0 value of the futex too.) */ -static int futex_lock_pi(u32 __user *uaddr, int fshared, - int detect, ktime_t *time, int trylock) +static int futex_lock_pi(u32 __user *uaddr, unsigned int flags, int detect, + ktime_t *time, int trylock) { struct hrtimer_sleeper timeout, *to = NULL; struct futex_hash_bucket *hb; - struct futex_q q; + struct futex_q q = futex_q_init; int res, ret; if (refill_pi_state_cache()) @@ -1914,12 +1906,8 @@ static int futex_lock_pi(u32 __user *uaddr, int fshared, hrtimer_set_expires(&to->timer, *time); } - q.pi_state = NULL; - q.rt_waiter = NULL; - q.requeue_pi_key = NULL; retry: - q.key = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr, fshared, &q.key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key); if (unlikely(ret != 0)) goto out; @@ -1941,7 +1929,7 @@ retry_private: * exit to complete. */ queue_unlock(&q, hb); - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); cond_resched(); goto retry; default: @@ -1971,7 +1959,7 @@ retry_private: * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. */ - res = fixup_owner(uaddr, fshared, &q, !ret); + res = fixup_owner(uaddr, &q, !ret); /* * If fixup_owner() returned an error, proprogate that. If it acquired * the lock, clear our -ETIMEDOUT or -EINTR. @@ -1995,7 +1983,7 @@ out_unlock_put_key: queue_unlock(&q, hb); out_put_key: - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); out: if (to) destroy_hrtimer_on_stack(&to->timer); @@ -2008,10 +1996,10 @@ uaddr_faulted: if (ret) goto out_put_key; - if (!fshared) + if (!(flags & FLAGS_SHARED)) goto retry_private; - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); goto retry; } @@ -2020,7 +2008,7 @@ uaddr_faulted: * This is the in-kernel slowpath: we look up the PI state (if any), * and do the rt-mutex unlock. */ -static int futex_unlock_pi(u32 __user *uaddr, int fshared) +static int futex_unlock_pi(u32 __user *uaddr, unsigned int flags) { struct futex_hash_bucket *hb; struct futex_q *this, *next; @@ -2038,7 +2026,7 @@ retry: if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current)) return -EPERM; - ret = get_futex_key(uaddr, fshared, &key); + ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &key); if (unlikely(ret != 0)) goto out; @@ -2093,14 +2081,14 @@ retry: out_unlock: spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); out: return ret; pi_faulted: spin_unlock(&hb->lock); - put_futex_key(fshared, &key); + put_futex_key(&key); ret = fault_in_user_writeable(uaddr); if (!ret) @@ -2160,7 +2148,7 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, /** * futex_wait_requeue_pi() - Wait on uaddr and take uaddr2 * @uaddr: the futex we initially wait on (non-pi) - * @fshared: whether the futexes are shared (1) or not (0). They must be + * @flags: futex flags (FLAGS_SHARED, FLAGS_CLOCKRT, etc.), they must be * the same type, no requeueing from private to shared, etc. * @val: the expected value of uaddr * @abs_time: absolute timeout @@ -2198,16 +2186,16 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * 0 - On success * <0 - On error */ -static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, +static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, u32 val, ktime_t *abs_time, u32 bitset, - int clockrt, u32 __user *uaddr2) + u32 __user *uaddr2) { struct hrtimer_sleeper timeout, *to = NULL; struct rt_mutex_waiter rt_waiter; struct rt_mutex *pi_mutex = NULL; struct futex_hash_bucket *hb; - union futex_key key2; - struct futex_q q; + union futex_key key2 = FUTEX_KEY_INIT; + struct futex_q q = futex_q_init; int res, ret; if (!bitset) @@ -2215,8 +2203,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, if (abs_time) { to = &timeout; - hrtimer_init_on_stack(&to->timer, clockrt ? CLOCK_REALTIME : - CLOCK_MONOTONIC, HRTIMER_MODE_ABS); + hrtimer_init_on_stack(&to->timer, (flags & FLAGS_CLOCKRT) ? + CLOCK_REALTIME : CLOCK_MONOTONIC, + HRTIMER_MODE_ABS); hrtimer_init_sleeper(to, current); hrtimer_set_expires_range_ns(&to->timer, *abs_time, current->timer_slack_ns); @@ -2229,12 +2218,10 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, debug_rt_mutex_init_waiter(&rt_waiter); rt_waiter.task = NULL; - key2 = FUTEX_KEY_INIT; - ret = get_futex_key(uaddr2, fshared, &key2); + ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2); if (unlikely(ret != 0)) goto out; - q.pi_state = NULL; q.bitset = bitset; q.rt_waiter = &rt_waiter; q.requeue_pi_key = &key2; @@ -2243,7 +2230,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, * Prepare to wait on uaddr. On success, increments q.key (key1) ref * count. */ - ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); + ret = futex_wait_setup(uaddr, val, flags, &q, &hb); if (ret) goto out_key2; @@ -2273,8 +2260,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, */ if (q.pi_state && (q.pi_state->owner != current)) { spin_lock(q.lock_ptr); - ret = fixup_pi_state_owner(uaddr2, &q, current, - fshared); + ret = fixup_pi_state_owner(uaddr2, &q, current); spin_unlock(q.lock_ptr); } } else { @@ -2293,7 +2279,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, * Fixup the pi_state owner and possibly acquire the lock if we * haven't already. */ - res = fixup_owner(uaddr2, fshared, &q, !ret); + res = fixup_owner(uaddr2, &q, !ret); /* * If fixup_owner() returned an error, proprogate that. If it * acquired the lock, clear -ETIMEDOUT or -EINTR. @@ -2324,9 +2310,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, int fshared, } out_put_keys: - put_futex_key(fshared, &q.key); + put_futex_key(&q.key); out_key2: - put_futex_key(fshared, &key2); + put_futex_key(&key2); out: if (to) { @@ -2551,58 +2537,57 @@ void exit_robust_list(struct task_struct *curr) long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, u32 __user *uaddr2, u32 val2, u32 val3) { - int clockrt, ret = -ENOSYS; - int cmd = op & FUTEX_CMD_MASK; - int fshared = 0; + int ret = -ENOSYS, cmd = op & FUTEX_CMD_MASK; + unsigned int flags = 0; if (!(op & FUTEX_PRIVATE_FLAG)) - fshared = 1; + flags |= FLAGS_SHARED; - clockrt = op & FUTEX_CLOCK_REALTIME; - if (clockrt && cmd != FUTEX_WAIT_BITSET && cmd != FUTEX_WAIT_REQUEUE_PI) - return -ENOSYS; + if (op & FUTEX_CLOCK_REALTIME) { + flags |= FLAGS_CLOCKRT; + if (cmd != FUTEX_WAIT_BITSET && cmd != FUTEX_WAIT_REQUEUE_PI) + return -ENOSYS; + } switch (cmd) { case FUTEX_WAIT: val3 = FUTEX_BITSET_MATCH_ANY; case FUTEX_WAIT_BITSET: - ret = futex_wait(uaddr, fshared, val, timeout, val3, clockrt); + ret = futex_wait(uaddr, flags, val, timeout, val3); break; case FUTEX_WAKE: val3 = FUTEX_BITSET_MATCH_ANY; case FUTEX_WAKE_BITSET: - ret = futex_wake(uaddr, fshared, val, val3); + ret = futex_wake(uaddr, flags, val, val3); break; case FUTEX_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, NULL, 0); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, NULL, 0); break; case FUTEX_CMP_REQUEUE: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, - 0); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, &val3, 0); break; case FUTEX_WAKE_OP: - ret = futex_wake_op(uaddr, fshared, uaddr2, val, val2, val3); + ret = futex_wake_op(uaddr, flags, uaddr2, val, val2, val3); break; case FUTEX_LOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_lock_pi(uaddr, fshared, val, timeout, 0); + ret = futex_lock_pi(uaddr, flags, val, timeout, 0); break; case FUTEX_UNLOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_unlock_pi(uaddr, fshared); + ret = futex_unlock_pi(uaddr, flags); break; case FUTEX_TRYLOCK_PI: if (futex_cmpxchg_enabled) - ret = futex_lock_pi(uaddr, fshared, 0, timeout, 1); + ret = futex_lock_pi(uaddr, flags, 0, timeout, 1); break; case FUTEX_WAIT_REQUEUE_PI: val3 = FUTEX_BITSET_MATCH_ANY; - ret = futex_wait_requeue_pi(uaddr, fshared, val, timeout, val3, - clockrt, uaddr2); + ret = futex_wait_requeue_pi(uaddr, flags, val, timeout, val3, + uaddr2); break; case FUTEX_CMP_REQUEUE_PI: - ret = futex_requeue(uaddr, fshared, uaddr2, val, val2, &val3, - 1); + ret = futex_requeue(uaddr, flags, uaddr2, val, val2, &val3, 1); break; default: ret = -ENOSYS; diff --git a/kernel/hw_breakpoint.c b/kernel/hw_breakpoint.c index e5325825aeb6..086adf25a55e 100644 --- a/kernel/hw_breakpoint.c +++ b/kernel/hw_breakpoint.c @@ -641,7 +641,7 @@ int __init init_hw_breakpoint(void) constraints_initialized = 1; - perf_pmu_register(&perf_breakpoint); + perf_pmu_register(&perf_breakpoint, "breakpoint", PERF_TYPE_BREAKPOINT); return register_die_notifier(&hw_breakpoint_exceptions_nb); diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 9737a76e106f..7663e5df0e6f 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -354,13 +354,20 @@ static inline int kprobe_aggrprobe(struct kprobe *p) return p->pre_handler == aggr_pre_handler; } +/* Return true(!0) if the kprobe is unused */ +static inline int kprobe_unused(struct kprobe *p) +{ + return kprobe_aggrprobe(p) && kprobe_disabled(p) && + list_empty(&p->list); +} + /* * Keep all fields in the kprobe consistent */ -static inline void copy_kprobe(struct kprobe *old_p, struct kprobe *p) +static inline void copy_kprobe(struct kprobe *ap, struct kprobe *p) { - memcpy(&p->opcode, &old_p->opcode, sizeof(kprobe_opcode_t)); - memcpy(&p->ainsn, &old_p->ainsn, sizeof(struct arch_specific_insn)); + memcpy(&p->opcode, &ap->opcode, sizeof(kprobe_opcode_t)); + memcpy(&p->ainsn, &ap->ainsn, sizeof(struct arch_specific_insn)); } #ifdef CONFIG_OPTPROBES @@ -384,6 +391,17 @@ void __kprobes opt_pre_handler(struct kprobe *p, struct pt_regs *regs) } } +/* Free optimized instructions and optimized_kprobe */ +static __kprobes void free_aggr_kprobe(struct kprobe *p) +{ + struct optimized_kprobe *op; + + op = container_of(p, struct optimized_kprobe, kp); + arch_remove_optimized_kprobe(op); + arch_remove_kprobe(p); + kfree(op); +} + /* Return true(!0) if the kprobe is ready for optimization. */ static inline int kprobe_optready(struct kprobe *p) { @@ -397,6 +415,33 @@ static inline int kprobe_optready(struct kprobe *p) return 0; } +/* Return true(!0) if the kprobe is disarmed. Note: p must be on hash list */ +static inline int kprobe_disarmed(struct kprobe *p) +{ + struct optimized_kprobe *op; + + /* If kprobe is not aggr/opt probe, just return kprobe is disabled */ + if (!kprobe_aggrprobe(p)) + return kprobe_disabled(p); + + op = container_of(p, struct optimized_kprobe, kp); + + return kprobe_disabled(p) && list_empty(&op->list); +} + +/* Return true(!0) if the probe is queued on (un)optimizing lists */ +static int __kprobes kprobe_queued(struct kprobe *p) +{ + struct optimized_kprobe *op; + + if (kprobe_aggrprobe(p)) { + op = container_of(p, struct optimized_kprobe, kp); + if (!list_empty(&op->list)) + return 1; + } + return 0; +} + /* * Return an optimized kprobe whose optimizing code replaces * instructions including addr (exclude breakpoint). @@ -422,30 +467,23 @@ static struct kprobe *__kprobes get_optimized_kprobe(unsigned long addr) /* Optimization staging list, protected by kprobe_mutex */ static LIST_HEAD(optimizing_list); +static LIST_HEAD(unoptimizing_list); static void kprobe_optimizer(struct work_struct *work); static DECLARE_DELAYED_WORK(optimizing_work, kprobe_optimizer); +static DECLARE_COMPLETION(optimizer_comp); #define OPTIMIZE_DELAY 5 -/* Kprobe jump optimizer */ -static __kprobes void kprobe_optimizer(struct work_struct *work) +/* + * Optimize (replace a breakpoint with a jump) kprobes listed on + * optimizing_list. + */ +static __kprobes void do_optimize_kprobes(void) { - struct optimized_kprobe *op, *tmp; - - /* Lock modules while optimizing kprobes */ - mutex_lock(&module_mutex); - mutex_lock(&kprobe_mutex); - if (kprobes_all_disarmed || !kprobes_allow_optimization) - goto end; - - /* - * Wait for quiesence period to ensure all running interrupts - * are done. Because optprobe may modify multiple instructions - * there is a chance that Nth instruction is interrupted. In that - * case, running interrupt can return to 2nd-Nth byte of jump - * instruction. This wait is for avoiding it. - */ - synchronize_sched(); + /* Optimization never be done when disarmed */ + if (kprobes_all_disarmed || !kprobes_allow_optimization || + list_empty(&optimizing_list)) + return; /* * The optimization/unoptimization refers online_cpus via @@ -459,17 +497,111 @@ static __kprobes void kprobe_optimizer(struct work_struct *work) */ get_online_cpus(); mutex_lock(&text_mutex); - list_for_each_entry_safe(op, tmp, &optimizing_list, list) { - WARN_ON(kprobe_disabled(&op->kp)); - if (arch_optimize_kprobe(op) < 0) - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; - list_del_init(&op->list); + arch_optimize_kprobes(&optimizing_list); + mutex_unlock(&text_mutex); + put_online_cpus(); +} + +/* + * Unoptimize (replace a jump with a breakpoint and remove the breakpoint + * if need) kprobes listed on unoptimizing_list. + */ +static __kprobes void do_unoptimize_kprobes(struct list_head *free_list) +{ + struct optimized_kprobe *op, *tmp; + + /* Unoptimization must be done anytime */ + if (list_empty(&unoptimizing_list)) + return; + + /* Ditto to do_optimize_kprobes */ + get_online_cpus(); + mutex_lock(&text_mutex); + arch_unoptimize_kprobes(&unoptimizing_list, free_list); + /* Loop free_list for disarming */ + list_for_each_entry_safe(op, tmp, free_list, list) { + /* Disarm probes if marked disabled */ + if (kprobe_disabled(&op->kp)) + arch_disarm_kprobe(&op->kp); + if (kprobe_unused(&op->kp)) { + /* + * Remove unused probes from hash list. After waiting + * for synchronization, these probes are reclaimed. + * (reclaiming is done by do_free_cleaned_kprobes.) + */ + hlist_del_rcu(&op->kp.hlist); + } else + list_del_init(&op->list); } mutex_unlock(&text_mutex); put_online_cpus(); -end: +} + +/* Reclaim all kprobes on the free_list */ +static __kprobes void do_free_cleaned_kprobes(struct list_head *free_list) +{ + struct optimized_kprobe *op, *tmp; + + list_for_each_entry_safe(op, tmp, free_list, list) { + BUG_ON(!kprobe_unused(&op->kp)); + list_del_init(&op->list); + free_aggr_kprobe(&op->kp); + } +} + +/* Start optimizer after OPTIMIZE_DELAY passed */ +static __kprobes void kick_kprobe_optimizer(void) +{ + if (!delayed_work_pending(&optimizing_work)) + schedule_delayed_work(&optimizing_work, OPTIMIZE_DELAY); +} + +/* Kprobe jump optimizer */ +static __kprobes void kprobe_optimizer(struct work_struct *work) +{ + LIST_HEAD(free_list); + + /* Lock modules while optimizing kprobes */ + mutex_lock(&module_mutex); + mutex_lock(&kprobe_mutex); + + /* + * Step 1: Unoptimize kprobes and collect cleaned (unused and disarmed) + * kprobes before waiting for quiesence period. + */ + do_unoptimize_kprobes(&free_list); + + /* + * Step 2: Wait for quiesence period to ensure all running interrupts + * are done. Because optprobe may modify multiple instructions + * there is a chance that Nth instruction is interrupted. In that + * case, running interrupt can return to 2nd-Nth byte of jump + * instruction. This wait is for avoiding it. + */ + synchronize_sched(); + + /* Step 3: Optimize kprobes after quiesence period */ + do_optimize_kprobes(); + + /* Step 4: Free cleaned kprobes after quiesence period */ + do_free_cleaned_kprobes(&free_list); + mutex_unlock(&kprobe_mutex); mutex_unlock(&module_mutex); + + /* Step 5: Kick optimizer again if needed */ + if (!list_empty(&optimizing_list) || !list_empty(&unoptimizing_list)) + kick_kprobe_optimizer(); + else + /* Wake up all waiters */ + complete_all(&optimizer_comp); +} + +/* Wait for completing optimization and unoptimization */ +static __kprobes void wait_for_kprobe_optimizer(void) +{ + if (delayed_work_pending(&optimizing_work)) + wait_for_completion(&optimizer_comp); } /* Optimize kprobe if p is ready to be optimized */ @@ -495,42 +627,99 @@ static __kprobes void optimize_kprobe(struct kprobe *p) /* Check if it is already optimized. */ if (op->kp.flags & KPROBE_FLAG_OPTIMIZED) return; - op->kp.flags |= KPROBE_FLAG_OPTIMIZED; - list_add(&op->list, &optimizing_list); - if (!delayed_work_pending(&optimizing_work)) - schedule_delayed_work(&optimizing_work, OPTIMIZE_DELAY); + + if (!list_empty(&op->list)) + /* This is under unoptimizing. Just dequeue the probe */ + list_del_init(&op->list); + else { + list_add(&op->list, &optimizing_list); + kick_kprobe_optimizer(); + } +} + +/* Short cut to direct unoptimizing */ +static __kprobes void force_unoptimize_kprobe(struct optimized_kprobe *op) +{ + get_online_cpus(); + arch_unoptimize_kprobe(op); + put_online_cpus(); + if (kprobe_disabled(&op->kp)) + arch_disarm_kprobe(&op->kp); } /* Unoptimize a kprobe if p is optimized */ -static __kprobes void unoptimize_kprobe(struct kprobe *p) +static __kprobes void unoptimize_kprobe(struct kprobe *p, bool force) { struct optimized_kprobe *op; - if ((p->flags & KPROBE_FLAG_OPTIMIZED) && kprobe_aggrprobe(p)) { - op = container_of(p, struct optimized_kprobe, kp); - if (!list_empty(&op->list)) - /* Dequeue from the optimization queue */ + if (!kprobe_aggrprobe(p) || kprobe_disarmed(p)) + return; /* This is not an optprobe nor optimized */ + + op = container_of(p, struct optimized_kprobe, kp); + if (!kprobe_optimized(p)) { + /* Unoptimized or unoptimizing case */ + if (force && !list_empty(&op->list)) { + /* + * Only if this is unoptimizing kprobe and forced, + * forcibly unoptimize it. (No need to unoptimize + * unoptimized kprobe again :) + */ list_del_init(&op->list); - else - /* Replace jump with break */ - arch_unoptimize_kprobe(op); - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + force_unoptimize_kprobe(op); + } + return; + } + + op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + if (!list_empty(&op->list)) { + /* Dequeue from the optimization queue */ + list_del_init(&op->list); + return; + } + /* Optimized kprobe case */ + if (force) + /* Forcibly update the code: this is a special case */ + force_unoptimize_kprobe(op); + else { + list_add(&op->list, &unoptimizing_list); + kick_kprobe_optimizer(); } } +/* Cancel unoptimizing for reusing */ +static void reuse_unused_kprobe(struct kprobe *ap) +{ + struct optimized_kprobe *op; + + BUG_ON(!kprobe_unused(ap)); + /* + * Unused kprobe MUST be on the way of delayed unoptimizing (means + * there is still a relative jump) and disabled. + */ + op = container_of(ap, struct optimized_kprobe, kp); + if (unlikely(list_empty(&op->list))) + printk(KERN_WARNING "Warning: found a stray unused " + "aggrprobe@%p\n", ap->addr); + /* Enable the probe again */ + ap->flags &= ~KPROBE_FLAG_DISABLED; + /* Optimize it again (remove from op->list) */ + BUG_ON(!kprobe_optready(ap)); + optimize_kprobe(ap); +} + /* Remove optimized instructions */ static void __kprobes kill_optimized_kprobe(struct kprobe *p) { struct optimized_kprobe *op; op = container_of(p, struct optimized_kprobe, kp); - if (!list_empty(&op->list)) { - /* Dequeue from the optimization queue */ + if (!list_empty(&op->list)) + /* Dequeue from the (un)optimization queue */ list_del_init(&op->list); - op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; - } - /* Don't unoptimize, because the target code will be freed. */ + + op->kp.flags &= ~KPROBE_FLAG_OPTIMIZED; + /* Don't touch the code, because it is already freed. */ arch_remove_optimized_kprobe(op); } @@ -543,16 +732,6 @@ static __kprobes void prepare_optimized_kprobe(struct kprobe *p) arch_prepare_optimized_kprobe(op); } -/* Free optimized instructions and optimized_kprobe */ -static __kprobes void free_aggr_kprobe(struct kprobe *p) -{ - struct optimized_kprobe *op; - - op = container_of(p, struct optimized_kprobe, kp); - arch_remove_optimized_kprobe(op); - kfree(op); -} - /* Allocate new optimized_kprobe and try to prepare optimized instructions */ static __kprobes struct kprobe *alloc_aggr_kprobe(struct kprobe *p) { @@ -587,7 +766,8 @@ static __kprobes void try_to_optimize_kprobe(struct kprobe *p) op = container_of(ap, struct optimized_kprobe, kp); if (!arch_prepared_optinsn(&op->optinsn)) { /* If failed to setup optimizing, fallback to kprobe */ - free_aggr_kprobe(ap); + arch_remove_optimized_kprobe(op); + kfree(op); return; } @@ -631,21 +811,16 @@ static void __kprobes unoptimize_all_kprobes(void) return; kprobes_allow_optimization = false; - printk(KERN_INFO "Kprobes globally unoptimized\n"); - get_online_cpus(); /* For avoiding text_mutex deadlock */ - mutex_lock(&text_mutex); for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) { if (!kprobe_disabled(p)) - unoptimize_kprobe(p); + unoptimize_kprobe(p, false); } } - - mutex_unlock(&text_mutex); - put_online_cpus(); - /* Allow all currently running kprobes to complete */ - synchronize_sched(); + /* Wait for unoptimizing completion */ + wait_for_kprobe_optimizer(); + printk(KERN_INFO "Kprobes globally unoptimized\n"); } int sysctl_kprobes_optimization; @@ -669,44 +844,60 @@ int proc_kprobes_optimization_handler(struct ctl_table *table, int write, } #endif /* CONFIG_SYSCTL */ +/* Put a breakpoint for a probe. Must be called with text_mutex locked */ static void __kprobes __arm_kprobe(struct kprobe *p) { - struct kprobe *old_p; + struct kprobe *_p; /* Check collision with other optimized kprobes */ - old_p = get_optimized_kprobe((unsigned long)p->addr); - if (unlikely(old_p)) - unoptimize_kprobe(old_p); /* Fallback to unoptimized kprobe */ + _p = get_optimized_kprobe((unsigned long)p->addr); + if (unlikely(_p)) + /* Fallback to unoptimized kprobe */ + unoptimize_kprobe(_p, true); arch_arm_kprobe(p); optimize_kprobe(p); /* Try to optimize (add kprobe to a list) */ } -static void __kprobes __disarm_kprobe(struct kprobe *p) +/* Remove the breakpoint of a probe. Must be called with text_mutex locked */ +static void __kprobes __disarm_kprobe(struct kprobe *p, bool reopt) { - struct kprobe *old_p; + struct kprobe *_p; - unoptimize_kprobe(p); /* Try to unoptimize */ - arch_disarm_kprobe(p); + unoptimize_kprobe(p, false); /* Try to unoptimize */ - /* If another kprobe was blocked, optimize it. */ - old_p = get_optimized_kprobe((unsigned long)p->addr); - if (unlikely(old_p)) - optimize_kprobe(old_p); + if (!kprobe_queued(p)) { + arch_disarm_kprobe(p); + /* If another kprobe was blocked, optimize it. */ + _p = get_optimized_kprobe((unsigned long)p->addr); + if (unlikely(_p) && reopt) + optimize_kprobe(_p); + } + /* TODO: reoptimize others after unoptimized this probe */ } #else /* !CONFIG_OPTPROBES */ #define optimize_kprobe(p) do {} while (0) -#define unoptimize_kprobe(p) do {} while (0) +#define unoptimize_kprobe(p, f) do {} while (0) #define kill_optimized_kprobe(p) do {} while (0) #define prepare_optimized_kprobe(p) do {} while (0) #define try_to_optimize_kprobe(p) do {} while (0) #define __arm_kprobe(p) arch_arm_kprobe(p) -#define __disarm_kprobe(p) arch_disarm_kprobe(p) +#define __disarm_kprobe(p, o) arch_disarm_kprobe(p) +#define kprobe_disarmed(p) kprobe_disabled(p) +#define wait_for_kprobe_optimizer() do {} while (0) + +/* There should be no unused kprobes can be reused without optimization */ +static void reuse_unused_kprobe(struct kprobe *ap) +{ + printk(KERN_ERR "Error: There should be no unused kprobe here.\n"); + BUG_ON(kprobe_unused(ap)); +} static __kprobes void free_aggr_kprobe(struct kprobe *p) { + arch_remove_kprobe(p); kfree(p); } @@ -732,11 +923,10 @@ static void __kprobes arm_kprobe(struct kprobe *kp) /* Disarm a kprobe with text_mutex */ static void __kprobes disarm_kprobe(struct kprobe *kp) { - get_online_cpus(); /* For avoiding text_mutex deadlock */ + /* Ditto */ mutex_lock(&text_mutex); - __disarm_kprobe(kp); + __disarm_kprobe(kp, true); mutex_unlock(&text_mutex); - put_online_cpus(); } /* @@ -942,7 +1132,7 @@ static int __kprobes add_new_kprobe(struct kprobe *ap, struct kprobe *p) BUG_ON(kprobe_gone(ap) || kprobe_gone(p)); if (p->break_handler || p->post_handler) - unoptimize_kprobe(ap); /* Fall back to normal kprobe */ + unoptimize_kprobe(ap, true); /* Fall back to normal kprobe */ if (p->break_handler) { if (ap->break_handler) @@ -993,19 +1183,21 @@ static void __kprobes init_aggr_kprobe(struct kprobe *ap, struct kprobe *p) * This is the second or subsequent kprobe at the address - handle * the intricacies */ -static int __kprobes register_aggr_kprobe(struct kprobe *old_p, +static int __kprobes register_aggr_kprobe(struct kprobe *orig_p, struct kprobe *p) { int ret = 0; - struct kprobe *ap = old_p; + struct kprobe *ap = orig_p; - if (!kprobe_aggrprobe(old_p)) { - /* If old_p is not an aggr_kprobe, create new aggr_kprobe. */ - ap = alloc_aggr_kprobe(old_p); + if (!kprobe_aggrprobe(orig_p)) { + /* If orig_p is not an aggr_kprobe, create new aggr_kprobe. */ + ap = alloc_aggr_kprobe(orig_p); if (!ap) return -ENOMEM; - init_aggr_kprobe(ap, old_p); - } + init_aggr_kprobe(ap, orig_p); + } else if (kprobe_unused(ap)) + /* This probe is going to die. Rescue it */ + reuse_unused_kprobe(ap); if (kprobe_gone(ap)) { /* @@ -1039,23 +1231,6 @@ static int __kprobes register_aggr_kprobe(struct kprobe *old_p, return add_new_kprobe(ap, p); } -/* Try to disable aggr_kprobe, and return 1 if succeeded.*/ -static int __kprobes try_to_disable_aggr_kprobe(struct kprobe *p) -{ - struct kprobe *kp; - - list_for_each_entry_rcu(kp, &p->list, list) { - if (!kprobe_disabled(kp)) - /* - * There is an active probe on the list. - * We can't disable aggr_kprobe. - */ - return 0; - } - p->flags |= KPROBE_FLAG_DISABLED; - return 1; -} - static int __kprobes in_kprobes_functions(unsigned long addr) { struct kprobe_blackpoint *kb; @@ -1098,34 +1273,33 @@ static kprobe_opcode_t __kprobes *kprobe_addr(struct kprobe *p) /* Check passed kprobe is valid and return kprobe in kprobe_table. */ static struct kprobe * __kprobes __get_valid_kprobe(struct kprobe *p) { - struct kprobe *old_p, *list_p; + struct kprobe *ap, *list_p; - old_p = get_kprobe(p->addr); - if (unlikely(!old_p)) + ap = get_kprobe(p->addr); + if (unlikely(!ap)) return NULL; - if (p != old_p) { - list_for_each_entry_rcu(list_p, &old_p->list, list) + if (p != ap) { + list_for_each_entry_rcu(list_p, &ap->list, list) if (list_p == p) /* kprobe p is a valid probe */ goto valid; return NULL; } valid: - return old_p; + return ap; } /* Return error if the kprobe is being re-registered */ static inline int check_kprobe_rereg(struct kprobe *p) { int ret = 0; - struct kprobe *old_p; mutex_lock(&kprobe_mutex); - old_p = __get_valid_kprobe(p); - if (old_p) + if (__get_valid_kprobe(p)) ret = -EINVAL; mutex_unlock(&kprobe_mutex); + return ret; } @@ -1229,67 +1403,121 @@ fail_with_jump_label: } EXPORT_SYMBOL_GPL(register_kprobe); +/* Check if all probes on the aggrprobe are disabled */ +static int __kprobes aggr_kprobe_disabled(struct kprobe *ap) +{ + struct kprobe *kp; + + list_for_each_entry_rcu(kp, &ap->list, list) + if (!kprobe_disabled(kp)) + /* + * There is an active probe on the list. + * We can't disable this ap. + */ + return 0; + + return 1; +} + +/* Disable one kprobe: Make sure called under kprobe_mutex is locked */ +static struct kprobe *__kprobes __disable_kprobe(struct kprobe *p) +{ + struct kprobe *orig_p; + + /* Get an original kprobe for return */ + orig_p = __get_valid_kprobe(p); + if (unlikely(orig_p == NULL)) + return NULL; + + if (!kprobe_disabled(p)) { + /* Disable probe if it is a child probe */ + if (p != orig_p) + p->flags |= KPROBE_FLAG_DISABLED; + + /* Try to disarm and disable this/parent probe */ + if (p == orig_p || aggr_kprobe_disabled(orig_p)) { + disarm_kprobe(orig_p); + orig_p->flags |= KPROBE_FLAG_DISABLED; + } + } + + return orig_p; +} + /* * Unregister a kprobe without a scheduler synchronization. */ static int __kprobes __unregister_kprobe_top(struct kprobe *p) { - struct kprobe *old_p, *list_p; + struct kprobe *ap, *list_p; - old_p = __get_valid_kprobe(p); - if (old_p == NULL) + /* Disable kprobe. This will disarm it if needed. */ + ap = __disable_kprobe(p); + if (ap == NULL) return -EINVAL; - if (old_p == p || - (kprobe_aggrprobe(old_p) && - list_is_singular(&old_p->list))) { + if (ap == p) /* - * Only probe on the hash list. Disarm only if kprobes are - * enabled and not gone - otherwise, the breakpoint would - * already have been removed. We save on flushing icache. + * This probe is an independent(and non-optimized) kprobe + * (not an aggrprobe). Remove from the hash list. */ - if (!kprobes_all_disarmed && !kprobe_disabled(old_p)) - disarm_kprobe(old_p); - hlist_del_rcu(&old_p->hlist); - } else { + goto disarmed; + + /* Following process expects this probe is an aggrprobe */ + WARN_ON(!kprobe_aggrprobe(ap)); + + if (list_is_singular(&ap->list) && kprobe_disarmed(ap)) + /* + * !disarmed could be happen if the probe is under delayed + * unoptimizing. + */ + goto disarmed; + else { + /* If disabling probe has special handlers, update aggrprobe */ if (p->break_handler && !kprobe_gone(p)) - old_p->break_handler = NULL; + ap->break_handler = NULL; if (p->post_handler && !kprobe_gone(p)) { - list_for_each_entry_rcu(list_p, &old_p->list, list) { + list_for_each_entry_rcu(list_p, &ap->list, list) { if ((list_p != p) && (list_p->post_handler)) goto noclean; } - old_p->post_handler = NULL; + ap->post_handler = NULL; } noclean: + /* + * Remove from the aggrprobe: this path will do nothing in + * __unregister_kprobe_bottom(). + */ list_del_rcu(&p->list); - if (!kprobe_disabled(old_p)) { - try_to_disable_aggr_kprobe(old_p); - if (!kprobes_all_disarmed) { - if (kprobe_disabled(old_p)) - disarm_kprobe(old_p); - else - /* Try to optimize this probe again */ - optimize_kprobe(old_p); - } - } + if (!kprobe_disabled(ap) && !kprobes_all_disarmed) + /* + * Try to optimize this probe again, because post + * handler may have been changed. + */ + optimize_kprobe(ap); } return 0; + +disarmed: + BUG_ON(!kprobe_disarmed(ap)); + hlist_del_rcu(&ap->hlist); + return 0; } static void __kprobes __unregister_kprobe_bottom(struct kprobe *p) { - struct kprobe *old_p; + struct kprobe *ap; if (list_empty(&p->list)) + /* This is an independent kprobe */ arch_remove_kprobe(p); else if (list_is_singular(&p->list)) { - /* "p" is the last child of an aggr_kprobe */ - old_p = list_entry(p->list.next, struct kprobe, list); + /* This is the last child of an aggrprobe */ + ap = list_entry(p->list.next, struct kprobe, list); list_del(&p->list); - arch_remove_kprobe(old_p); - free_aggr_kprobe(old_p); + free_aggr_kprobe(ap); } + /* Otherwise, do nothing. */ } int __kprobes register_kprobes(struct kprobe **kps, int num) @@ -1607,29 +1835,13 @@ static void __kprobes kill_kprobe(struct kprobe *p) int __kprobes disable_kprobe(struct kprobe *kp) { int ret = 0; - struct kprobe *p; mutex_lock(&kprobe_mutex); - /* Check whether specified probe is valid. */ - p = __get_valid_kprobe(kp); - if (unlikely(p == NULL)) { + /* Disable this kprobe */ + if (__disable_kprobe(kp) == NULL) ret = -EINVAL; - goto out; - } - /* If the probe is already disabled (or gone), just return */ - if (kprobe_disabled(kp)) - goto out; - - kp->flags |= KPROBE_FLAG_DISABLED; - if (p != kp) - /* When kp != p, p is always enabled. */ - try_to_disable_aggr_kprobe(p); - - if (!kprobes_all_disarmed && kprobe_disabled(p)) - disarm_kprobe(p); -out: mutex_unlock(&kprobe_mutex); return ret; } @@ -1927,36 +2139,27 @@ static void __kprobes disarm_all_kprobes(void) mutex_lock(&kprobe_mutex); /* If kprobes are already disarmed, just return */ - if (kprobes_all_disarmed) - goto already_disabled; + if (kprobes_all_disarmed) { + mutex_unlock(&kprobe_mutex); + return; + } kprobes_all_disarmed = true; printk(KERN_INFO "Kprobes globally disabled\n"); - /* - * Here we call get_online_cpus() for avoiding text_mutex deadlock, - * because disarming may also unoptimize kprobes. - */ - get_online_cpus(); mutex_lock(&text_mutex); for (i = 0; i < KPROBE_TABLE_SIZE; i++) { head = &kprobe_table[i]; hlist_for_each_entry_rcu(p, node, head, hlist) { if (!arch_trampoline_kprobe(p) && !kprobe_disabled(p)) - __disarm_kprobe(p); + __disarm_kprobe(p, false); } } - mutex_unlock(&text_mutex); - put_online_cpus(); mutex_unlock(&kprobe_mutex); - /* Allow all currently running kprobes to complete */ - synchronize_sched(); - return; -already_disabled: - mutex_unlock(&kprobe_mutex); - return; + /* Wait for disarming all kprobes by optimizer */ + wait_for_kprobe_optimizer(); } /* diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 2870feee81dd..11847bf1e8cc 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -13,6 +13,7 @@ #include <linux/mm.h> #include <linux/cpu.h> #include <linux/smp.h> +#include <linux/idr.h> #include <linux/file.h> #include <linux/poll.h> #include <linux/slab.h> @@ -21,7 +22,9 @@ #include <linux/dcache.h> #include <linux/percpu.h> #include <linux/ptrace.h> +#include <linux/reboot.h> #include <linux/vmstat.h> +#include <linux/device.h> #include <linux/vmalloc.h> #include <linux/hardirq.h> #include <linux/rculist.h> @@ -133,6 +136,28 @@ static void unclone_ctx(struct perf_event_context *ctx) } } +static u32 perf_event_pid(struct perf_event *event, struct task_struct *p) +{ + /* + * only top level events have the pid namespace they were created in + */ + if (event->parent) + event = event->parent; + + return task_tgid_nr_ns(p, event->ns); +} + +static u32 perf_event_tid(struct perf_event *event, struct task_struct *p) +{ + /* + * only top level events have the pid namespace they were created in + */ + if (event->parent) + event = event->parent; + + return task_pid_nr_ns(p, event->ns); +} + /* * If we inherit events we want to return the parent event id * to userspace. @@ -312,9 +337,84 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx) ctx->nr_stat++; } +/* + * Called at perf_event creation and when events are attached/detached from a + * group. + */ +static void perf_event__read_size(struct perf_event *event) +{ + int entry = sizeof(u64); /* value */ + int size = 0; + int nr = 1; + + if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + size += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + size += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_ID) + entry += sizeof(u64); + + if (event->attr.read_format & PERF_FORMAT_GROUP) { + nr += event->group_leader->nr_siblings; + size += sizeof(u64); + } + + size += entry * nr; + event->read_size = size; +} + +static void perf_event__header_size(struct perf_event *event) +{ + struct perf_sample_data *data; + u64 sample_type = event->attr.sample_type; + u16 size = 0; + + perf_event__read_size(event); + + if (sample_type & PERF_SAMPLE_IP) + size += sizeof(data->ip); + + if (sample_type & PERF_SAMPLE_ADDR) + size += sizeof(data->addr); + + if (sample_type & PERF_SAMPLE_PERIOD) + size += sizeof(data->period); + + if (sample_type & PERF_SAMPLE_READ) + size += event->read_size; + + event->header_size = size; +} + +static void perf_event__id_header_size(struct perf_event *event) +{ + struct perf_sample_data *data; + u64 sample_type = event->attr.sample_type; + u16 size = 0; + + if (sample_type & PERF_SAMPLE_TID) + size += sizeof(data->tid_entry); + + if (sample_type & PERF_SAMPLE_TIME) + size += sizeof(data->time); + + if (sample_type & PERF_SAMPLE_ID) + size += sizeof(data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + size += sizeof(data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + size += sizeof(data->cpu_entry); + + event->id_header_size = size; +} + static void perf_group_attach(struct perf_event *event) { - struct perf_event *group_leader = event->group_leader; + struct perf_event *group_leader = event->group_leader, *pos; /* * We can have double attach due to group movement in perf_event_open. @@ -333,6 +433,11 @@ static void perf_group_attach(struct perf_event *event) list_add_tail(&event->group_entry, &group_leader->sibling_list); group_leader->nr_siblings++; + + perf_event__header_size(group_leader); + + list_for_each_entry(pos, &group_leader->sibling_list, group_entry) + perf_event__header_size(pos); } /* @@ -391,7 +496,7 @@ static void perf_group_detach(struct perf_event *event) if (event->group_leader != event) { list_del_init(&event->group_entry); event->group_leader->nr_siblings--; - return; + goto out; } if (!list_empty(&event->group_entry)) @@ -410,6 +515,12 @@ static void perf_group_detach(struct perf_event *event) /* Inherit group flags from the previous leader */ sibling->group_flags = event->group_flags; } + +out: + perf_event__header_size(event->group_leader); + + list_for_each_entry(tmp, &event->group_leader->sibling_list, group_entry) + perf_event__header_size(tmp); } static inline int @@ -1073,7 +1184,7 @@ static int perf_event_refresh(struct perf_event *event, int refresh) /* * not supported on inherited events */ - if (event->attr.inherit) + if (event->attr.inherit || !is_sampling_event(event)) return -EINVAL; atomic_add(refresh, &event->event_limit); @@ -2289,31 +2400,6 @@ static int perf_release(struct inode *inode, struct file *file) return perf_event_release_kernel(event); } -static int perf_event_read_size(struct perf_event *event) -{ - int entry = sizeof(u64); /* value */ - int size = 0; - int nr = 1; - - if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) - size += sizeof(u64); - - if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) - size += sizeof(u64); - - if (event->attr.read_format & PERF_FORMAT_ID) - entry += sizeof(u64); - - if (event->attr.read_format & PERF_FORMAT_GROUP) { - nr += event->group_leader->nr_siblings; - size += sizeof(u64); - } - - size += entry * nr; - - return size; -} - u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running) { struct perf_event *child; @@ -2428,7 +2514,7 @@ perf_read_hw(struct perf_event *event, char __user *buf, size_t count) if (event->state == PERF_EVENT_STATE_ERROR) return 0; - if (count < perf_event_read_size(event)) + if (count < event->read_size) return -ENOSPC; WARN_ON_ONCE(event->ctx->parent_ctx); @@ -2514,7 +2600,7 @@ static int perf_event_period(struct perf_event *event, u64 __user *arg) int ret = 0; u64 value; - if (!event->attr.sample_period) + if (!is_sampling_event(event)) return -EINVAL; if (copy_from_user(&value, arg, sizeof(value))) @@ -3305,6 +3391,73 @@ __always_inline void perf_output_copy(struct perf_output_handle *handle, } while (len); } +static void __perf_event_header__init_id(struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event) +{ + u64 sample_type = event->attr.sample_type; + + data->type = sample_type; + header->size += event->id_header_size; + + if (sample_type & PERF_SAMPLE_TID) { + /* namespace issues */ + data->tid_entry.pid = perf_event_pid(event, current); + data->tid_entry.tid = perf_event_tid(event, current); + } + + if (sample_type & PERF_SAMPLE_TIME) + data->time = perf_clock(); + + if (sample_type & PERF_SAMPLE_ID) + data->id = primary_event_id(event); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + data->stream_id = event->id; + + if (sample_type & PERF_SAMPLE_CPU) { + data->cpu_entry.cpu = raw_smp_processor_id(); + data->cpu_entry.reserved = 0; + } +} + +static void perf_event_header__init_id(struct perf_event_header *header, + struct perf_sample_data *data, + struct perf_event *event) +{ + if (event->attr.sample_id_all) + __perf_event_header__init_id(header, data, event); +} + +static void __perf_event__output_id_sample(struct perf_output_handle *handle, + struct perf_sample_data *data) +{ + u64 sample_type = data->type; + + if (sample_type & PERF_SAMPLE_TID) + perf_output_put(handle, data->tid_entry); + + if (sample_type & PERF_SAMPLE_TIME) + perf_output_put(handle, data->time); + + if (sample_type & PERF_SAMPLE_ID) + perf_output_put(handle, data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + perf_output_put(handle, data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + perf_output_put(handle, data->cpu_entry); +} + +static void perf_event__output_id_sample(struct perf_event *event, + struct perf_output_handle *handle, + struct perf_sample_data *sample) +{ + if (event->attr.sample_id_all) + __perf_event__output_id_sample(handle, sample); +} + int perf_output_begin(struct perf_output_handle *handle, struct perf_event *event, unsigned int size, int nmi, int sample) @@ -3312,6 +3465,7 @@ int perf_output_begin(struct perf_output_handle *handle, struct perf_buffer *buffer; unsigned long tail, offset, head; int have_lost; + struct perf_sample_data sample_data; struct { struct perf_event_header header; u64 id; @@ -3338,8 +3492,12 @@ int perf_output_begin(struct perf_output_handle *handle, goto out; have_lost = local_read(&buffer->lost); - if (have_lost) - size += sizeof(lost_event); + if (have_lost) { + lost_event.header.size = sizeof(lost_event); + perf_event_header__init_id(&lost_event.header, &sample_data, + event); + size += lost_event.header.size; + } perf_output_get_handle(handle); @@ -3370,11 +3528,11 @@ int perf_output_begin(struct perf_output_handle *handle, if (have_lost) { lost_event.header.type = PERF_RECORD_LOST; lost_event.header.misc = 0; - lost_event.header.size = sizeof(lost_event); lost_event.id = event->id; lost_event.lost = local_xchg(&buffer->lost, 0); perf_output_put(handle, lost_event); + perf_event__output_id_sample(event, handle, &sample_data); } return 0; @@ -3407,28 +3565,6 @@ void perf_output_end(struct perf_output_handle *handle) rcu_read_unlock(); } -static u32 perf_event_pid(struct perf_event *event, struct task_struct *p) -{ - /* - * only top level events have the pid namespace they were created in - */ - if (event->parent) - event = event->parent; - - return task_tgid_nr_ns(p, event->ns); -} - -static u32 perf_event_tid(struct perf_event *event, struct task_struct *p) -{ - /* - * only top level events have the pid namespace they were created in - */ - if (event->parent) - event = event->parent; - - return task_pid_nr_ns(p, event->ns); -} - static void perf_output_read_one(struct perf_output_handle *handle, struct perf_event *event, u64 enabled, u64 running) @@ -3603,61 +3739,16 @@ void perf_prepare_sample(struct perf_event_header *header, { u64 sample_type = event->attr.sample_type; - data->type = sample_type; - header->type = PERF_RECORD_SAMPLE; - header->size = sizeof(*header); + header->size = sizeof(*header) + event->header_size; header->misc = 0; header->misc |= perf_misc_flags(regs); - if (sample_type & PERF_SAMPLE_IP) { - data->ip = perf_instruction_pointer(regs); - - header->size += sizeof(data->ip); - } - - if (sample_type & PERF_SAMPLE_TID) { - /* namespace issues */ - data->tid_entry.pid = perf_event_pid(event, current); - data->tid_entry.tid = perf_event_tid(event, current); - - header->size += sizeof(data->tid_entry); - } - - if (sample_type & PERF_SAMPLE_TIME) { - data->time = perf_clock(); - - header->size += sizeof(data->time); - } - - if (sample_type & PERF_SAMPLE_ADDR) - header->size += sizeof(data->addr); - - if (sample_type & PERF_SAMPLE_ID) { - data->id = primary_event_id(event); - - header->size += sizeof(data->id); - } - - if (sample_type & PERF_SAMPLE_STREAM_ID) { - data->stream_id = event->id; - - header->size += sizeof(data->stream_id); - } - - if (sample_type & PERF_SAMPLE_CPU) { - data->cpu_entry.cpu = raw_smp_processor_id(); - data->cpu_entry.reserved = 0; - - header->size += sizeof(data->cpu_entry); - } - - if (sample_type & PERF_SAMPLE_PERIOD) - header->size += sizeof(data->period); + __perf_event_header__init_id(header, data, event); - if (sample_type & PERF_SAMPLE_READ) - header->size += perf_event_read_size(event); + if (sample_type & PERF_SAMPLE_IP) + data->ip = perf_instruction_pointer(regs); if (sample_type & PERF_SAMPLE_CALLCHAIN) { int size = 1; @@ -3722,23 +3813,26 @@ perf_event_read_event(struct perf_event *event, struct task_struct *task) { struct perf_output_handle handle; + struct perf_sample_data sample; struct perf_read_event read_event = { .header = { .type = PERF_RECORD_READ, .misc = 0, - .size = sizeof(read_event) + perf_event_read_size(event), + .size = sizeof(read_event) + event->read_size, }, .pid = perf_event_pid(event, task), .tid = perf_event_tid(event, task), }; int ret; + perf_event_header__init_id(&read_event.header, &sample, event); ret = perf_output_begin(&handle, event, read_event.header.size, 0, 0); if (ret) return; perf_output_put(&handle, read_event); perf_output_read(&handle, event); + perf_event__output_id_sample(event, &handle, &sample); perf_output_end(&handle); } @@ -3768,14 +3862,16 @@ static void perf_event_task_output(struct perf_event *event, struct perf_task_event *task_event) { struct perf_output_handle handle; + struct perf_sample_data sample; struct task_struct *task = task_event->task; - int size, ret; + int ret, size = task_event->event_id.header.size; - size = task_event->event_id.header.size; - ret = perf_output_begin(&handle, event, size, 0, 0); + perf_event_header__init_id(&task_event->event_id.header, &sample, event); + ret = perf_output_begin(&handle, event, + task_event->event_id.header.size, 0, 0); if (ret) - return; + goto out; task_event->event_id.pid = perf_event_pid(event, task); task_event->event_id.ppid = perf_event_pid(event, current); @@ -3785,7 +3881,11 @@ static void perf_event_task_output(struct perf_event *event, perf_output_put(&handle, task_event->event_id); + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +out: + task_event->event_id.header.size = size; } static int perf_event_task_match(struct perf_event *event) @@ -3900,11 +4000,16 @@ static void perf_event_comm_output(struct perf_event *event, struct perf_comm_event *comm_event) { struct perf_output_handle handle; + struct perf_sample_data sample; int size = comm_event->event_id.header.size; - int ret = perf_output_begin(&handle, event, size, 0, 0); + int ret; + + perf_event_header__init_id(&comm_event->event_id.header, &sample, event); + ret = perf_output_begin(&handle, event, + comm_event->event_id.header.size, 0, 0); if (ret) - return; + goto out; comm_event->event_id.pid = perf_event_pid(event, comm_event->task); comm_event->event_id.tid = perf_event_tid(event, comm_event->task); @@ -3912,7 +4017,12 @@ static void perf_event_comm_output(struct perf_event *event, perf_output_put(&handle, comm_event->event_id); perf_output_copy(&handle, comm_event->comm, comm_event->comm_size); + + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +out: + comm_event->event_id.header.size = size; } static int perf_event_comm_match(struct perf_event *event) @@ -3957,7 +4067,6 @@ static void perf_event_comm_event(struct perf_comm_event *comm_event) comm_event->comm_size = size; comm_event->event_id.header.size = sizeof(comm_event->event_id) + size; - rcu_read_lock(); list_for_each_entry_rcu(pmu, &pmus, entry) { cpuctx = get_cpu_ptr(pmu->pmu_cpu_context); @@ -4038,11 +4147,15 @@ static void perf_event_mmap_output(struct perf_event *event, struct perf_mmap_event *mmap_event) { struct perf_output_handle handle; + struct perf_sample_data sample; int size = mmap_event->event_id.header.size; - int ret = perf_output_begin(&handle, event, size, 0, 0); + int ret; + perf_event_header__init_id(&mmap_event->event_id.header, &sample, event); + ret = perf_output_begin(&handle, event, + mmap_event->event_id.header.size, 0, 0); if (ret) - return; + goto out; mmap_event->event_id.pid = perf_event_pid(event, current); mmap_event->event_id.tid = perf_event_tid(event, current); @@ -4050,7 +4163,12 @@ static void perf_event_mmap_output(struct perf_event *event, perf_output_put(&handle, mmap_event->event_id); perf_output_copy(&handle, mmap_event->file_name, mmap_event->file_size); + + perf_event__output_id_sample(event, &handle, &sample); + perf_output_end(&handle); +out: + mmap_event->event_id.header.size = size; } static int perf_event_mmap_match(struct perf_event *event, @@ -4205,6 +4323,7 @@ void perf_event_mmap(struct vm_area_struct *vma) static void perf_log_throttle(struct perf_event *event, int enable) { struct perf_output_handle handle; + struct perf_sample_data sample; int ret; struct { @@ -4226,11 +4345,15 @@ static void perf_log_throttle(struct perf_event *event, int enable) if (enable) throttle_event.header.type = PERF_RECORD_UNTHROTTLE; - ret = perf_output_begin(&handle, event, sizeof(throttle_event), 1, 0); + perf_event_header__init_id(&throttle_event.header, &sample, event); + + ret = perf_output_begin(&handle, event, + throttle_event.header.size, 1, 0); if (ret) return; perf_output_put(&handle, throttle_event); + perf_event__output_id_sample(event, &handle, &sample); perf_output_end(&handle); } @@ -4246,6 +4369,13 @@ static int __perf_event_overflow(struct perf_event *event, int nmi, struct hw_perf_event *hwc = &event->hw; int ret = 0; + /* + * Non-sampling counters might still use the PMI to fold short + * hardware counters, ignore those. + */ + if (unlikely(!is_sampling_event(event))) + return 0; + if (!throttle) { hwc->interrupts++; } else { @@ -4391,7 +4521,7 @@ static void perf_swevent_event(struct perf_event *event, u64 nr, if (!regs) return; - if (!hwc->sample_period) + if (!is_sampling_event(event)) return; if (nr == 1 && hwc->sample_period == 1 && !event->attr.freq) @@ -4554,7 +4684,7 @@ static int perf_swevent_add(struct perf_event *event, int flags) struct hw_perf_event *hwc = &event->hw; struct hlist_head *head; - if (hwc->sample_period) { + if (is_sampling_event(event)) { hwc->last_period = hwc->sample_period; perf_swevent_set_period(event); } @@ -4811,15 +4941,6 @@ static int perf_tp_event_init(struct perf_event *event) if (event->attr.type != PERF_TYPE_TRACEPOINT) return -ENOENT; - /* - * Raw tracepoint data is a severe data leak, only allow root to - * have these. - */ - if ((event->attr.sample_type & PERF_SAMPLE_RAW) && - perf_paranoid_tracepoint_raw() && - !capable(CAP_SYS_ADMIN)) - return -EPERM; - err = perf_trace_init(event); if (err) return err; @@ -4842,7 +4963,7 @@ static struct pmu perf_tracepoint = { static inline void perf_tp_register(void) { - perf_pmu_register(&perf_tracepoint); + perf_pmu_register(&perf_tracepoint, "tracepoint", PERF_TYPE_TRACEPOINT); } static int perf_event_set_filter(struct perf_event *event, void __user *arg) @@ -4932,31 +5053,33 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) static void perf_swevent_start_hrtimer(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; + s64 period; + + if (!is_sampling_event(event)) + return; hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hwc->hrtimer.function = perf_swevent_hrtimer; - if (hwc->sample_period) { - s64 period = local64_read(&hwc->period_left); - if (period) { - if (period < 0) - period = 10000; + period = local64_read(&hwc->period_left); + if (period) { + if (period < 0) + period = 10000; - local64_set(&hwc->period_left, 0); - } else { - period = max_t(u64, 10000, hwc->sample_period); - } - __hrtimer_start_range_ns(&hwc->hrtimer, + local64_set(&hwc->period_left, 0); + } else { + period = max_t(u64, 10000, hwc->sample_period); + } + __hrtimer_start_range_ns(&hwc->hrtimer, ns_to_ktime(period), 0, HRTIMER_MODE_REL_PINNED, 0); - } } static void perf_swevent_cancel_hrtimer(struct perf_event *event) { struct hw_perf_event *hwc = &event->hw; - if (hwc->sample_period) { + if (is_sampling_event(event)) { ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); local64_set(&hwc->period_left, ktime_to_ns(remaining)); @@ -5184,8 +5307,61 @@ static void free_pmu_context(struct pmu *pmu) out: mutex_unlock(&pmus_lock); } +static struct idr pmu_idr; + +static ssize_t +type_show(struct device *dev, struct device_attribute *attr, char *page) +{ + struct pmu *pmu = dev_get_drvdata(dev); + + return snprintf(page, PAGE_SIZE-1, "%d\n", pmu->type); +} + +static struct device_attribute pmu_dev_attrs[] = { + __ATTR_RO(type), + __ATTR_NULL, +}; + +static int pmu_bus_running; +static struct bus_type pmu_bus = { + .name = "event_source", + .dev_attrs = pmu_dev_attrs, +}; + +static void pmu_dev_release(struct device *dev) +{ + kfree(dev); +} + +static int pmu_dev_alloc(struct pmu *pmu) +{ + int ret = -ENOMEM; + + pmu->dev = kzalloc(sizeof(struct device), GFP_KERNEL); + if (!pmu->dev) + goto out; + + device_initialize(pmu->dev); + ret = dev_set_name(pmu->dev, "%s", pmu->name); + if (ret) + goto free_dev; + + dev_set_drvdata(pmu->dev, pmu); + pmu->dev->bus = &pmu_bus; + pmu->dev->release = pmu_dev_release; + ret = device_add(pmu->dev); + if (ret) + goto free_dev; + +out: + return ret; + +free_dev: + put_device(pmu->dev); + goto out; +} -int perf_pmu_register(struct pmu *pmu) +int perf_pmu_register(struct pmu *pmu, char *name, int type) { int cpu, ret; @@ -5195,13 +5371,38 @@ int perf_pmu_register(struct pmu *pmu) if (!pmu->pmu_disable_count) goto unlock; + pmu->type = -1; + if (!name) + goto skip_type; + pmu->name = name; + + if (type < 0) { + int err = idr_pre_get(&pmu_idr, GFP_KERNEL); + if (!err) + goto free_pdc; + + err = idr_get_new_above(&pmu_idr, pmu, PERF_TYPE_MAX, &type); + if (err) { + ret = err; + goto free_pdc; + } + } + pmu->type = type; + + if (pmu_bus_running) { + ret = pmu_dev_alloc(pmu); + if (ret) + goto free_idr; + } + +skip_type: pmu->pmu_cpu_context = find_pmu_context(pmu->task_ctx_nr); if (pmu->pmu_cpu_context) goto got_cpu_context; pmu->pmu_cpu_context = alloc_percpu(struct perf_cpu_context); if (!pmu->pmu_cpu_context) - goto free_pdc; + goto free_dev; for_each_possible_cpu(cpu) { struct perf_cpu_context *cpuctx; @@ -5245,6 +5446,14 @@ unlock: return ret; +free_dev: + device_del(pmu->dev); + put_device(pmu->dev); + +free_idr: + if (pmu->type >= PERF_TYPE_MAX) + idr_remove(&pmu_idr, pmu->type); + free_pdc: free_percpu(pmu->pmu_disable_count); goto unlock; @@ -5264,6 +5473,10 @@ void perf_pmu_unregister(struct pmu *pmu) synchronize_rcu(); free_percpu(pmu->pmu_disable_count); + if (pmu->type >= PERF_TYPE_MAX) + idr_remove(&pmu_idr, pmu->type); + device_del(pmu->dev); + put_device(pmu->dev); free_pmu_context(pmu); } @@ -5273,6 +5486,13 @@ struct pmu *perf_init_event(struct perf_event *event) int idx; idx = srcu_read_lock(&pmus_srcu); + + rcu_read_lock(); + pmu = idr_find(&pmu_idr, event->attr.type); + rcu_read_unlock(); + if (pmu) + goto unlock; + list_for_each_entry_rcu(pmu, &pmus, entry) { int ret = pmu->event_init(event); if (!ret) @@ -5738,6 +5958,12 @@ SYSCALL_DEFINE5(perf_event_open, mutex_unlock(¤t->perf_event_mutex); /* + * Precalculate sample_data sizes + */ + perf_event__header_size(event); + perf_event__id_header_size(event); + + /* * Drop the reference on the group_event after placing the * new event on the sibling_list. This ensures destruction * of the group leader will find the pointer to itself in @@ -6090,6 +6316,12 @@ inherit_event(struct perf_event *parent_event, child_event->overflow_handler = parent_event->overflow_handler; /* + * Precalculate sample_data sizes + */ + perf_event__header_size(child_event); + perf_event__id_header_size(child_event); + + /* * Link it up in the child's context: */ raw_spin_lock_irqsave(&child_ctx->lock, flags); @@ -6320,7 +6552,7 @@ static void __cpuinit perf_event_init_cpu(int cpu) mutex_unlock(&swhash->hlist_mutex); } -#ifdef CONFIG_HOTPLUG_CPU +#if defined CONFIG_HOTPLUG_CPU || defined CONFIG_KEXEC static void perf_pmu_rotate_stop(struct pmu *pmu) { struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context); @@ -6374,6 +6606,26 @@ static void perf_event_exit_cpu(int cpu) static inline void perf_event_exit_cpu(int cpu) { } #endif +static int +perf_reboot(struct notifier_block *notifier, unsigned long val, void *v) +{ + int cpu; + + for_each_online_cpu(cpu) + perf_event_exit_cpu(cpu); + + return NOTIFY_OK; +} + +/* + * Run the perf reboot notifier at the very last possible moment so that + * the generic watchdog code runs as long as possible. + */ +static struct notifier_block perf_reboot_notifier = { + .notifier_call = perf_reboot, + .priority = INT_MIN, +}; + static int __cpuinit perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { @@ -6402,14 +6654,45 @@ void __init perf_event_init(void) { int ret; + idr_init(&pmu_idr); + perf_event_init_all_cpus(); init_srcu_struct(&pmus_srcu); - perf_pmu_register(&perf_swevent); - perf_pmu_register(&perf_cpu_clock); - perf_pmu_register(&perf_task_clock); + perf_pmu_register(&perf_swevent, "software", PERF_TYPE_SOFTWARE); + perf_pmu_register(&perf_cpu_clock, NULL, -1); + perf_pmu_register(&perf_task_clock, NULL, -1); perf_tp_register(); perf_cpu_notifier(perf_cpu_notify); + register_reboot_notifier(&perf_reboot_notifier); ret = init_hw_breakpoint(); WARN(ret, "hw_breakpoint initialization failed with: %d", ret); } + +static int __init perf_event_sysfs_init(void) +{ + struct pmu *pmu; + int ret; + + mutex_lock(&pmus_lock); + + ret = bus_register(&pmu_bus); + if (ret) + goto unlock; + + list_for_each_entry(pmu, &pmus, entry) { + if (!pmu->name || pmu->type < 0) + continue; + + ret = pmu_dev_alloc(pmu); + WARN(ret, "Failed to register pmu: %s, reason %d\n", pmu->name, ret); + } + pmu_bus_running = 1; + ret = 0; + +unlock: + mutex_unlock(&pmus_lock); + + return ret; +} +device_initcall(perf_event_sysfs_init); diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index ecf770509d0d..031d5e3a6197 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -22,6 +22,7 @@ #include <linux/mm.h> #include <linux/slab.h> #include <linux/suspend.h> +#include <trace/events/power.h> #include "power.h" @@ -201,6 +202,7 @@ int suspend_devices_and_enter(suspend_state_t state) if (!suspend_ops) return -ENOSYS; + trace_machine_suspend(state); if (suspend_ops->begin) { error = suspend_ops->begin(state); if (error) @@ -229,6 +231,7 @@ int suspend_devices_and_enter(suspend_state_t state) Close: if (suspend_ops->end) suspend_ops->end(); + trace_machine_suspend(PWR_EVENT_EXIT); return error; Recover_platform: diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index d806735342ac..034493724749 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -36,31 +36,16 @@ #include <linux/time.h> #include <linux/cpu.h> -/* Global control variables for rcupdate callback mechanism. */ -struct rcu_ctrlblk { - struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ - struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ - struct rcu_head **curtail; /* ->next pointer of last CB. */ -}; - -/* Definition for rcupdate control block. */ -static struct rcu_ctrlblk rcu_sched_ctrlblk = { - .donetail = &rcu_sched_ctrlblk.rcucblist, - .curtail = &rcu_sched_ctrlblk.rcucblist, -}; - -static struct rcu_ctrlblk rcu_bh_ctrlblk = { - .donetail = &rcu_bh_ctrlblk.rcucblist, - .curtail = &rcu_bh_ctrlblk.rcucblist, -}; - -#ifdef CONFIG_DEBUG_LOCK_ALLOC -int rcu_scheduler_active __read_mostly; -EXPORT_SYMBOL_GPL(rcu_scheduler_active); -#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ +/* Controls for rcu_kthread() kthread, replacing RCU_SOFTIRQ used previously. */ +static struct task_struct *rcu_kthread_task; +static DECLARE_WAIT_QUEUE_HEAD(rcu_kthread_wq); +static unsigned long have_rcu_kthread_work; +static void invoke_rcu_kthread(void); /* Forward declarations for rcutiny_plugin.h. */ -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp); +struct rcu_ctrlblk; +static void rcu_process_callbacks(struct rcu_ctrlblk *rcp); +static int rcu_kthread(void *arg); static void __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), struct rcu_ctrlblk *rcp); @@ -123,7 +108,7 @@ void rcu_sched_qs(int cpu) { if (rcu_qsctr_help(&rcu_sched_ctrlblk) + rcu_qsctr_help(&rcu_bh_ctrlblk)) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); } /* @@ -132,7 +117,7 @@ void rcu_sched_qs(int cpu) void rcu_bh_qs(int cpu) { if (rcu_qsctr_help(&rcu_bh_ctrlblk)) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); } /* @@ -152,13 +137,14 @@ void rcu_check_callbacks(int cpu, int user) } /* - * Helper function for rcu_process_callbacks() that operates on the - * specified rcu_ctrlkblk structure. + * Invoke the RCU callbacks on the specified rcu_ctrlkblk structure + * whose grace period has elapsed. */ -static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) +static void rcu_process_callbacks(struct rcu_ctrlblk *rcp) { struct rcu_head *next, *list; unsigned long flags; + RCU_TRACE(int cb_count = 0); /* If no RCU callbacks ready to invoke, just return. */ if (&rcp->rcucblist == rcp->donetail) @@ -180,19 +166,58 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp) next = list->next; prefetch(next); debug_rcu_head_unqueue(list); + local_bh_disable(); list->func(list); + local_bh_enable(); list = next; + RCU_TRACE(cb_count++); } + RCU_TRACE(rcu_trace_sub_qlen(rcp, cb_count)); } /* - * Invoke any callbacks whose grace period has completed. + * This kthread invokes RCU callbacks whose grace periods have + * elapsed. It is awakened as needed, and takes the place of the + * RCU_SOFTIRQ that was used previously for this purpose. + * This is a kthread, but it is never stopped, at least not until + * the system goes down. */ -static void rcu_process_callbacks(struct softirq_action *unused) +static int rcu_kthread(void *arg) { - __rcu_process_callbacks(&rcu_sched_ctrlblk); - __rcu_process_callbacks(&rcu_bh_ctrlblk); - rcu_preempt_process_callbacks(); + unsigned long work; + unsigned long morework; + unsigned long flags; + + for (;;) { + wait_event(rcu_kthread_wq, have_rcu_kthread_work != 0); + morework = rcu_boost(); + local_irq_save(flags); + work = have_rcu_kthread_work; + have_rcu_kthread_work = morework; + local_irq_restore(flags); + if (work) { + rcu_process_callbacks(&rcu_sched_ctrlblk); + rcu_process_callbacks(&rcu_bh_ctrlblk); + rcu_preempt_process_callbacks(); + } + schedule_timeout_interruptible(1); /* Leave CPU for others. */ + } + + return 0; /* Not reached, but needed to shut gcc up. */ +} + +/* + * Wake up rcu_kthread() to process callbacks now eligible for invocation + * or to boost readers. + */ +static void invoke_rcu_kthread(void) +{ + unsigned long flags; + + local_irq_save(flags); + have_rcu_kthread_work = 1; + wake_up(&rcu_kthread_wq); + local_irq_restore(flags); } /* @@ -230,6 +255,7 @@ static void __call_rcu(struct rcu_head *head, local_irq_save(flags); *rcp->curtail = head; rcp->curtail = &head->next; + RCU_TRACE(rcp->qlen++); local_irq_restore(flags); } @@ -282,7 +308,16 @@ void rcu_barrier_sched(void) } EXPORT_SYMBOL_GPL(rcu_barrier_sched); -void __init rcu_init(void) +/* + * Spawn the kthread that invokes RCU callbacks. + */ +static int __init rcu_spawn_kthreads(void) { - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); + struct sched_param sp; + + rcu_kthread_task = kthread_run(rcu_kthread, NULL, "rcu_kthread"); + sp.sched_priority = RCU_BOOST_PRIO; + sched_setscheduler_nocheck(rcu_kthread_task, SCHED_FIFO, &sp); + return 0; } +early_initcall(rcu_spawn_kthreads); diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 6ceca4f745ff..015abaea962a 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -22,6 +22,40 @@ * Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com> */ +#include <linux/kthread.h> +#include <linux/debugfs.h> +#include <linux/seq_file.h> + +#ifdef CONFIG_RCU_TRACE +#define RCU_TRACE(stmt) stmt +#else /* #ifdef CONFIG_RCU_TRACE */ +#define RCU_TRACE(stmt) +#endif /* #else #ifdef CONFIG_RCU_TRACE */ + +/* Global control variables for rcupdate callback mechanism. */ +struct rcu_ctrlblk { + struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */ + struct rcu_head **donetail; /* ->next pointer of last "done" CB. */ + struct rcu_head **curtail; /* ->next pointer of last CB. */ + RCU_TRACE(long qlen); /* Number of pending CBs. */ +}; + +/* Definition for rcupdate control block. */ +static struct rcu_ctrlblk rcu_sched_ctrlblk = { + .donetail = &rcu_sched_ctrlblk.rcucblist, + .curtail = &rcu_sched_ctrlblk.rcucblist, +}; + +static struct rcu_ctrlblk rcu_bh_ctrlblk = { + .donetail = &rcu_bh_ctrlblk.rcucblist, + .curtail = &rcu_bh_ctrlblk.rcucblist, +}; + +#ifdef CONFIG_DEBUG_LOCK_ALLOC +int rcu_scheduler_active __read_mostly; +EXPORT_SYMBOL_GPL(rcu_scheduler_active); +#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + #ifdef CONFIG_TINY_PREEMPT_RCU #include <linux/delay.h> @@ -46,17 +80,45 @@ struct rcu_preempt_ctrlblk { struct list_head *gp_tasks; /* Pointer to the first task blocking the */ /* current grace period, or NULL if there */ - /* is not such task. */ + /* is no such task. */ struct list_head *exp_tasks; /* Pointer to first task blocking the */ /* current expedited grace period, or NULL */ /* if there is no such task. If there */ /* is no current expedited grace period, */ /* then there cannot be any such task. */ +#ifdef CONFIG_RCU_BOOST + struct list_head *boost_tasks; + /* Pointer to first task that needs to be */ + /* priority-boosted, or NULL if no priority */ + /* boosting is needed. If there is no */ + /* current or expedited grace period, there */ + /* can be no such task. */ +#endif /* #ifdef CONFIG_RCU_BOOST */ u8 gpnum; /* Current grace period. */ u8 gpcpu; /* Last grace period blocked by the CPU. */ u8 completed; /* Last grace period completed. */ /* If all three are equal, RCU is idle. */ +#ifdef CONFIG_RCU_BOOST + s8 boosted_this_gp; /* Has boosting already happened? */ + unsigned long boost_time; /* When to start boosting (jiffies) */ +#endif /* #ifdef CONFIG_RCU_BOOST */ +#ifdef CONFIG_RCU_TRACE + unsigned long n_grace_periods; +#ifdef CONFIG_RCU_BOOST + unsigned long n_tasks_boosted; + unsigned long n_exp_boosts; + unsigned long n_normal_boosts; + unsigned long n_normal_balk_blkd_tasks; + unsigned long n_normal_balk_gp_tasks; + unsigned long n_normal_balk_boost_tasks; + unsigned long n_normal_balk_boosted; + unsigned long n_normal_balk_notyet; + unsigned long n_normal_balk_nos; + unsigned long n_exp_balk_blkd_tasks; + unsigned long n_exp_balk_nos; +#endif /* #ifdef CONFIG_RCU_BOOST */ +#endif /* #ifdef CONFIG_RCU_TRACE */ }; static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = { @@ -122,6 +184,210 @@ static int rcu_preempt_gp_in_progress(void) } /* + * Advance a ->blkd_tasks-list pointer to the next entry, instead + * returning NULL if at the end of the list. + */ +static struct list_head *rcu_next_node_entry(struct task_struct *t) +{ + struct list_head *np; + + np = t->rcu_node_entry.next; + if (np == &rcu_preempt_ctrlblk.blkd_tasks) + np = NULL; + return np; +} + +#ifdef CONFIG_RCU_TRACE + +#ifdef CONFIG_RCU_BOOST +static void rcu_initiate_boost_trace(void); +static void rcu_initiate_exp_boost_trace(void); +#endif /* #ifdef CONFIG_RCU_BOOST */ + +/* + * Dump additional statistice for TINY_PREEMPT_RCU. + */ +static void show_tiny_preempt_stats(struct seq_file *m) +{ + seq_printf(m, "rcu_preempt: qlen=%ld gp=%lu g%u/p%u/c%u tasks=%c%c%c\n", + rcu_preempt_ctrlblk.rcb.qlen, + rcu_preempt_ctrlblk.n_grace_periods, + rcu_preempt_ctrlblk.gpnum, + rcu_preempt_ctrlblk.gpcpu, + rcu_preempt_ctrlblk.completed, + "T."[list_empty(&rcu_preempt_ctrlblk.blkd_tasks)], + "N."[!rcu_preempt_ctrlblk.gp_tasks], + "E."[!rcu_preempt_ctrlblk.exp_tasks]); +#ifdef CONFIG_RCU_BOOST + seq_printf(m, " ttb=%c btg=", + "B."[!rcu_preempt_ctrlblk.boost_tasks]); + switch (rcu_preempt_ctrlblk.boosted_this_gp) { + case -1: + seq_puts(m, "exp"); + break; + case 0: + seq_puts(m, "no"); + break; + case 1: + seq_puts(m, "begun"); + break; + case 2: + seq_puts(m, "done"); + break; + default: + seq_printf(m, "?%d?", rcu_preempt_ctrlblk.boosted_this_gp); + } + seq_printf(m, " ntb=%lu neb=%lu nnb=%lu j=%04x bt=%04x\n", + rcu_preempt_ctrlblk.n_tasks_boosted, + rcu_preempt_ctrlblk.n_exp_boosts, + rcu_preempt_ctrlblk.n_normal_boosts, + (int)(jiffies & 0xffff), + (int)(rcu_preempt_ctrlblk.boost_time & 0xffff)); + seq_printf(m, " %s: nt=%lu gt=%lu bt=%lu b=%lu ny=%lu nos=%lu\n", + "normal balk", + rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks, + rcu_preempt_ctrlblk.n_normal_balk_gp_tasks, + rcu_preempt_ctrlblk.n_normal_balk_boost_tasks, + rcu_preempt_ctrlblk.n_normal_balk_boosted, + rcu_preempt_ctrlblk.n_normal_balk_notyet, + rcu_preempt_ctrlblk.n_normal_balk_nos); + seq_printf(m, " exp balk: bt=%lu nos=%lu\n", + rcu_preempt_ctrlblk.n_exp_balk_blkd_tasks, + rcu_preempt_ctrlblk.n_exp_balk_nos); +#endif /* #ifdef CONFIG_RCU_BOOST */ +} + +#endif /* #ifdef CONFIG_RCU_TRACE */ + +#ifdef CONFIG_RCU_BOOST + +#include "rtmutex_common.h" + +/* + * Carry out RCU priority boosting on the task indicated by ->boost_tasks, + * and advance ->boost_tasks to the next task in the ->blkd_tasks list. + */ +static int rcu_boost(void) +{ + unsigned long flags; + struct rt_mutex mtx; + struct list_head *np; + struct task_struct *t; + + if (rcu_preempt_ctrlblk.boost_tasks == NULL) + return 0; /* Nothing to boost. */ + raw_local_irq_save(flags); + rcu_preempt_ctrlblk.boosted_this_gp++; + t = container_of(rcu_preempt_ctrlblk.boost_tasks, struct task_struct, + rcu_node_entry); + np = rcu_next_node_entry(t); + rt_mutex_init_proxy_locked(&mtx, t); + t->rcu_boost_mutex = &mtx; + t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BOOSTED; + raw_local_irq_restore(flags); + rt_mutex_lock(&mtx); + RCU_TRACE(rcu_preempt_ctrlblk.n_tasks_boosted++); + rcu_preempt_ctrlblk.boosted_this_gp++; + rt_mutex_unlock(&mtx); + return rcu_preempt_ctrlblk.boost_tasks != NULL; +} + +/* + * Check to see if it is now time to start boosting RCU readers blocking + * the current grace period, and, if so, tell the rcu_kthread_task to + * start boosting them. If there is an expedited boost in progress, + * we wait for it to complete. + * + * If there are no blocked readers blocking the current grace period, + * return 0 to let the caller know, otherwise return 1. Note that this + * return value is independent of whether or not boosting was done. + */ +static int rcu_initiate_boost(void) +{ + if (!rcu_preempt_blocked_readers_cgp()) { + RCU_TRACE(rcu_preempt_ctrlblk.n_normal_balk_blkd_tasks++); + return 0; + } + if (rcu_preempt_ctrlblk.gp_tasks != NULL && + rcu_preempt_ctrlblk.boost_tasks == NULL && + rcu_preempt_ctrlblk.boosted_this_gp == 0 && + ULONG_CMP_GE(jiffies, rcu_preempt_ctrlblk.boost_time)) { + rcu_preempt_ctrlblk.boost_tasks = rcu_preempt_ctrlblk.gp_tasks; + invoke_rcu_kthread(); + RCU_TRACE(rcu_preempt_ctrlblk.n_normal_boosts++); + } else + RCU_TRACE(rcu_initiate_boost_trace()); + return 1; +} + +/* + * Initiate boosting for an expedited grace period. + */ +static void rcu_initiate_expedited_boost(void) +{ + unsigned long flags; + + raw_local_irq_save(flags); + if (!list_empty(&rcu_preempt_ctrlblk.blkd_tasks)) { + rcu_preempt_ctrlblk.boost_tasks = + rcu_preempt_ctrlblk.blkd_tasks.next; + rcu_preempt_ctrlblk.boosted_this_gp = -1; + invoke_rcu_kthread(); + RCU_TRACE(rcu_preempt_ctrlblk.n_exp_boosts++); + } else + RCU_TRACE(rcu_initiate_exp_boost_trace()); + raw_local_irq_restore(flags); +} + +#define RCU_BOOST_DELAY_JIFFIES DIV_ROUND_UP(CONFIG_RCU_BOOST_DELAY * HZ, 1000); + +/* + * Do priority-boost accounting for the start of a new grace period. + */ +static void rcu_preempt_boost_start_gp(void) +{ + rcu_preempt_ctrlblk.boost_time = jiffies + RCU_BOOST_DELAY_JIFFIES; + if (rcu_preempt_ctrlblk.boosted_this_gp > 0) + rcu_preempt_ctrlblk.boosted_this_gp = 0; +} + +#else /* #ifdef CONFIG_RCU_BOOST */ + +/* + * If there is no RCU priority boosting, we don't boost. + */ +static int rcu_boost(void) +{ + return 0; +} + +/* + * If there is no RCU priority boosting, we don't initiate boosting, + * but we do indicate whether there are blocked readers blocking the + * current grace period. + */ +static int rcu_initiate_boost(void) +{ + return rcu_preempt_blocked_readers_cgp(); +} + +/* + * If there is no RCU priority boosting, we don't initiate expedited boosting. + */ +static void rcu_initiate_expedited_boost(void) +{ +} + +/* + * If there is no RCU priority boosting, nothing to do at grace-period start. + */ +static void rcu_preempt_boost_start_gp(void) +{ +} + +#endif /* else #ifdef CONFIG_RCU_BOOST */ + +/* * Record a preemptible-RCU quiescent state for the specified CPU. Note * that this just means that the task currently running on the CPU is * in a quiescent state. There might be any number of tasks blocked @@ -148,11 +414,14 @@ static void rcu_preempt_cpu_qs(void) rcu_preempt_ctrlblk.gpcpu = rcu_preempt_ctrlblk.gpnum; current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + /* If there is no GP then there is nothing more to do. */ + if (!rcu_preempt_gp_in_progress()) + return; /* - * If there is no GP, or if blocked readers are still blocking GP, - * then there is nothing more to do. + * Check up on boosting. If there are no readers blocking the + * current grace period, leave. */ - if (!rcu_preempt_gp_in_progress() || rcu_preempt_blocked_readers_cgp()) + if (rcu_initiate_boost()) return; /* Advance callbacks. */ @@ -164,9 +433,9 @@ static void rcu_preempt_cpu_qs(void) if (!rcu_preempt_blocked_readers_any()) rcu_preempt_ctrlblk.rcb.donetail = rcu_preempt_ctrlblk.nexttail; - /* If there are done callbacks, make RCU_SOFTIRQ process them. */ + /* If there are done callbacks, cause them to be invoked. */ if (*rcu_preempt_ctrlblk.rcb.donetail != NULL) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); } /* @@ -178,12 +447,16 @@ static void rcu_preempt_start_gp(void) /* Official start of GP. */ rcu_preempt_ctrlblk.gpnum++; + RCU_TRACE(rcu_preempt_ctrlblk.n_grace_periods++); /* Any blocked RCU readers block new GP. */ if (rcu_preempt_blocked_readers_any()) rcu_preempt_ctrlblk.gp_tasks = rcu_preempt_ctrlblk.blkd_tasks.next; + /* Set up for RCU priority boosting. */ + rcu_preempt_boost_start_gp(); + /* If there is no running reader, CPU is done with GP. */ if (!rcu_preempt_running_reader()) rcu_preempt_cpu_qs(); @@ -304,14 +577,16 @@ static void rcu_read_unlock_special(struct task_struct *t) */ empty = !rcu_preempt_blocked_readers_cgp(); empty_exp = rcu_preempt_ctrlblk.exp_tasks == NULL; - np = t->rcu_node_entry.next; - if (np == &rcu_preempt_ctrlblk.blkd_tasks) - np = NULL; + np = rcu_next_node_entry(t); list_del(&t->rcu_node_entry); if (&t->rcu_node_entry == rcu_preempt_ctrlblk.gp_tasks) rcu_preempt_ctrlblk.gp_tasks = np; if (&t->rcu_node_entry == rcu_preempt_ctrlblk.exp_tasks) rcu_preempt_ctrlblk.exp_tasks = np; +#ifdef CONFIG_RCU_BOOST + if (&t->rcu_node_entry == rcu_preempt_ctrlblk.boost_tasks) + rcu_preempt_ctrlblk.boost_tasks = np; +#endif /* #ifdef CONFIG_RCU_BOOST */ INIT_LIST_HEAD(&t->rcu_node_entry); /* @@ -331,6 +606,14 @@ static void rcu_read_unlock_special(struct task_struct *t) if (!empty_exp && rcu_preempt_ctrlblk.exp_tasks == NULL) rcu_report_exp_done(); } +#ifdef CONFIG_RCU_BOOST + /* Unboost self if was boosted. */ + if (special & RCU_READ_UNLOCK_BOOSTED) { + t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BOOSTED; + rt_mutex_unlock(t->rcu_boost_mutex); + t->rcu_boost_mutex = NULL; + } +#endif /* #ifdef CONFIG_RCU_BOOST */ local_irq_restore(flags); } @@ -374,7 +657,7 @@ static void rcu_preempt_check_callbacks(void) rcu_preempt_cpu_qs(); if (&rcu_preempt_ctrlblk.rcb.rcucblist != rcu_preempt_ctrlblk.rcb.donetail) - raise_softirq(RCU_SOFTIRQ); + invoke_rcu_kthread(); if (rcu_preempt_gp_in_progress() && rcu_cpu_blocking_cur_gp() && rcu_preempt_running_reader()) @@ -383,7 +666,7 @@ static void rcu_preempt_check_callbacks(void) /* * TINY_PREEMPT_RCU has an extra callback-list tail pointer to - * update, so this is invoked from __rcu_process_callbacks() to + * update, so this is invoked from rcu_process_callbacks() to * handle that case. Of course, it is invoked for all flavors of * RCU, but RCU callbacks can appear only on one of the lists, and * neither ->nexttail nor ->donetail can possibly be NULL, so there @@ -400,7 +683,7 @@ static void rcu_preempt_remove_callbacks(struct rcu_ctrlblk *rcp) */ static void rcu_preempt_process_callbacks(void) { - __rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb); + rcu_process_callbacks(&rcu_preempt_ctrlblk.rcb); } /* @@ -417,6 +700,7 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) local_irq_save(flags); *rcu_preempt_ctrlblk.nexttail = head; rcu_preempt_ctrlblk.nexttail = &head->next; + RCU_TRACE(rcu_preempt_ctrlblk.rcb.qlen++); rcu_preempt_start_gp(); /* checks to see if GP needed. */ local_irq_restore(flags); } @@ -532,6 +816,7 @@ void synchronize_rcu_expedited(void) /* Wait for tail of ->blkd_tasks list to drain. */ if (rcu_preempted_readers_exp()) + rcu_initiate_expedited_boost(); wait_event(sync_rcu_preempt_exp_wq, !rcu_preempted_readers_exp()); @@ -572,6 +857,27 @@ void exit_rcu(void) #else /* #ifdef CONFIG_TINY_PREEMPT_RCU */ +#ifdef CONFIG_RCU_TRACE + +/* + * Because preemptible RCU does not exist, it is not necessary to + * dump out its statistics. + */ +static void show_tiny_preempt_stats(struct seq_file *m) +{ +} + +#endif /* #ifdef CONFIG_RCU_TRACE */ + +/* + * Because preemptible RCU does not exist, it is never necessary to + * boost preempted RCU readers. + */ +static int rcu_boost(void) +{ + return 0; +} + /* * Because preemptible RCU does not exist, it never has any callbacks * to check. @@ -599,17 +905,116 @@ static void rcu_preempt_process_callbacks(void) #endif /* #else #ifdef CONFIG_TINY_PREEMPT_RCU */ #ifdef CONFIG_DEBUG_LOCK_ALLOC - #include <linux/kernel_stat.h> /* * During boot, we forgive RCU lockdep issues. After this function is * invoked, we start taking RCU lockdep issues seriously. */ -void rcu_scheduler_starting(void) +void __init rcu_scheduler_starting(void) { WARN_ON(nr_context_switches() > 0); rcu_scheduler_active = 1; } #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ + +#ifdef CONFIG_RCU_BOOST +#define RCU_BOOST_PRIO CONFIG_RCU_BOOST_PRIO +#else /* #ifdef CONFIG_RCU_BOOST */ +#define RCU_BOOST_PRIO 1 +#endif /* #else #ifdef CONFIG_RCU_BOOST */ + +#ifdef CONFIG_RCU_TRACE + +#ifdef CONFIG_RCU_BOOST + +static void rcu_initiate_boost_trace(void) +{ + if (rcu_preempt_ctrlblk.gp_tasks == NULL) + rcu_preempt_ctrlblk.n_normal_balk_gp_tasks++; + else if (rcu_preempt_ctrlblk.boost_tasks != NULL) + rcu_preempt_ctrlblk.n_normal_balk_boost_tasks++; + else if (rcu_preempt_ctrlblk.boosted_this_gp != 0) + rcu_preempt_ctrlblk.n_normal_balk_boosted++; + else if (!ULONG_CMP_GE(jiffies, rcu_preempt_ctrlblk.boost_time)) + rcu_preempt_ctrlblk.n_normal_balk_notyet++; + else + rcu_preempt_ctrlblk.n_normal_balk_nos++; +} + +static void rcu_initiate_exp_boost_trace(void) +{ + if (list_empty(&rcu_preempt_ctrlblk.blkd_tasks)) + rcu_preempt_ctrlblk.n_exp_balk_blkd_tasks++; + else + rcu_preempt_ctrlblk.n_exp_balk_nos++; +} + +#endif /* #ifdef CONFIG_RCU_BOOST */ + +static void rcu_trace_sub_qlen(struct rcu_ctrlblk *rcp, int n) +{ + unsigned long flags; + + raw_local_irq_save(flags); + rcp->qlen -= n; + raw_local_irq_restore(flags); +} + +/* + * Dump statistics for TINY_RCU, such as they are. + */ +static int show_tiny_stats(struct seq_file *m, void *unused) +{ + show_tiny_preempt_stats(m); + seq_printf(m, "rcu_sched: qlen: %ld\n", rcu_sched_ctrlblk.qlen); + seq_printf(m, "rcu_bh: qlen: %ld\n", rcu_bh_ctrlblk.qlen); + return 0; +} + +static int show_tiny_stats_open(struct inode *inode, struct file *file) +{ + return single_open(file, show_tiny_stats, NULL); +} + +static const struct file_operations show_tiny_stats_fops = { + .owner = THIS_MODULE, + .open = show_tiny_stats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static struct dentry *rcudir; + +static int __init rcutiny_trace_init(void) +{ + struct dentry *retval; + + rcudir = debugfs_create_dir("rcu", NULL); + if (!rcudir) + goto free_out; + retval = debugfs_create_file("rcudata", 0444, rcudir, + NULL, &show_tiny_stats_fops); + if (!retval) + goto free_out; + return 0; +free_out: + debugfs_remove_recursive(rcudir); + return 1; +} + +static void __exit rcutiny_trace_cleanup(void) +{ + debugfs_remove_recursive(rcudir); +} + +module_init(rcutiny_trace_init); +module_exit(rcutiny_trace_cleanup); + +MODULE_AUTHOR("Paul E. McKenney"); +MODULE_DESCRIPTION("Read-Copy Update tracing for tiny implementation"); +MODULE_LICENSE("GPL"); + +#endif /* #ifdef CONFIG_RCU_TRACE */ diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 9d8e8fb2515f..89613f97ff26 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -47,6 +47,7 @@ #include <linux/srcu.h> #include <linux/slab.h> #include <asm/byteorder.h> +#include <linux/sched.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <paulmck@us.ibm.com> and " @@ -64,6 +65,9 @@ static int irqreader = 1; /* RCU readers from irq (timers). */ static int fqs_duration = 0; /* Duration of bursts (us), 0 to disable. */ static int fqs_holdoff = 0; /* Hold time within burst (us). */ static int fqs_stutter = 3; /* Wait time between bursts (s). */ +static int test_boost = 1; /* Test RCU prio boost: 0=no, 1=maybe, 2=yes. */ +static int test_boost_interval = 7; /* Interval between boost tests, seconds. */ +static int test_boost_duration = 4; /* Duration of each boost test, seconds. */ static char *torture_type = "rcu"; /* What RCU implementation to torture. */ module_param(nreaders, int, 0444); @@ -88,6 +92,12 @@ module_param(fqs_holdoff, int, 0444); MODULE_PARM_DESC(fqs_holdoff, "Holdoff time within fqs bursts (us)"); module_param(fqs_stutter, int, 0444); MODULE_PARM_DESC(fqs_stutter, "Wait time between fqs bursts (s)"); +module_param(test_boost, int, 0444); +MODULE_PARM_DESC(test_boost, "Test RCU prio boost: 0=no, 1=maybe, 2=yes."); +module_param(test_boost_interval, int, 0444); +MODULE_PARM_DESC(test_boost_interval, "Interval between boost tests, seconds."); +module_param(test_boost_duration, int, 0444); +MODULE_PARM_DESC(test_boost_duration, "Duration of each boost test, seconds."); module_param(torture_type, charp, 0444); MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, rcu_bh, srcu)"); @@ -109,6 +119,7 @@ static struct task_struct *stats_task; static struct task_struct *shuffler_task; static struct task_struct *stutter_task; static struct task_struct *fqs_task; +static struct task_struct *boost_tasks[NR_CPUS]; #define RCU_TORTURE_PIPE_LEN 10 @@ -134,6 +145,12 @@ static atomic_t n_rcu_torture_alloc_fail; static atomic_t n_rcu_torture_free; static atomic_t n_rcu_torture_mberror; static atomic_t n_rcu_torture_error; +static long n_rcu_torture_boost_ktrerror; +static long n_rcu_torture_boost_rterror; +static long n_rcu_torture_boost_allocerror; +static long n_rcu_torture_boost_afferror; +static long n_rcu_torture_boost_failure; +static long n_rcu_torture_boosts; static long n_rcu_torture_timers; static struct list_head rcu_torture_removed; static cpumask_var_t shuffle_tmp_mask; @@ -147,6 +164,16 @@ static int stutter_pause_test; #endif int rcutorture_runnable = RCUTORTURE_RUNNABLE_INIT; +#ifdef CONFIG_RCU_BOOST +#define rcu_can_boost() 1 +#else /* #ifdef CONFIG_RCU_BOOST */ +#define rcu_can_boost() 0 +#endif /* #else #ifdef CONFIG_RCU_BOOST */ + +static unsigned long boost_starttime; /* jiffies of next boost test start. */ +DEFINE_MUTEX(boost_mutex); /* protect setting boost_starttime */ + /* and boost task create/destroy. */ + /* Mediate rmmod and system shutdown. Concurrent rmmod & shutdown illegal! */ #define FULLSTOP_DONTSTOP 0 /* Normal operation. */ @@ -277,6 +304,7 @@ struct rcu_torture_ops { void (*fqs)(void); int (*stats)(char *page); int irq_capable; + int can_boost; char *name; }; @@ -366,6 +394,7 @@ static struct rcu_torture_ops rcu_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu" }; @@ -408,6 +437,7 @@ static struct rcu_torture_ops rcu_sync_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu_sync" }; @@ -424,6 +454,7 @@ static struct rcu_torture_ops rcu_expedited_ops = { .fqs = rcu_force_quiescent_state, .stats = NULL, .irq_capable = 1, + .can_boost = rcu_can_boost(), .name = "rcu_expedited" }; @@ -684,6 +715,110 @@ static struct rcu_torture_ops sched_expedited_ops = { }; /* + * RCU torture priority-boost testing. Runs one real-time thread per + * CPU for moderate bursts, repeatedly registering RCU callbacks and + * spinning waiting for them to be invoked. If a given callback takes + * too long to be invoked, we assume that priority inversion has occurred. + */ + +struct rcu_boost_inflight { + struct rcu_head rcu; + int inflight; +}; + +static void rcu_torture_boost_cb(struct rcu_head *head) +{ + struct rcu_boost_inflight *rbip = + container_of(head, struct rcu_boost_inflight, rcu); + + smp_mb(); /* Ensure RCU-core accesses precede clearing ->inflight */ + rbip->inflight = 0; +} + +static int rcu_torture_boost(void *arg) +{ + unsigned long call_rcu_time; + unsigned long endtime; + unsigned long oldstarttime; + struct rcu_boost_inflight rbi = { .inflight = 0 }; + struct sched_param sp; + + VERBOSE_PRINTK_STRING("rcu_torture_boost started"); + + /* Set real-time priority. */ + sp.sched_priority = 1; + if (sched_setscheduler(current, SCHED_FIFO, &sp) < 0) { + VERBOSE_PRINTK_STRING("rcu_torture_boost RT prio failed!"); + n_rcu_torture_boost_rterror++; + } + + /* Each pass through the following loop does one boost-test cycle. */ + do { + /* Wait for the next test interval. */ + oldstarttime = boost_starttime; + while (jiffies - oldstarttime > ULONG_MAX / 2) { + schedule_timeout_uninterruptible(1); + rcu_stutter_wait("rcu_torture_boost"); + if (kthread_should_stop() || + fullstop != FULLSTOP_DONTSTOP) + goto checkwait; + } + + /* Do one boost-test interval. */ + endtime = oldstarttime + test_boost_duration * HZ; + call_rcu_time = jiffies; + while (jiffies - endtime > ULONG_MAX / 2) { + /* If we don't have a callback in flight, post one. */ + if (!rbi.inflight) { + smp_mb(); /* RCU core before ->inflight = 1. */ + rbi.inflight = 1; + call_rcu(&rbi.rcu, rcu_torture_boost_cb); + if (jiffies - call_rcu_time > + test_boost_duration * HZ - HZ / 2) { + VERBOSE_PRINTK_STRING("rcu_torture_boost boosting failed"); + n_rcu_torture_boost_failure++; + } + call_rcu_time = jiffies; + } + cond_resched(); + rcu_stutter_wait("rcu_torture_boost"); + if (kthread_should_stop() || + fullstop != FULLSTOP_DONTSTOP) + goto checkwait; + } + + /* + * Set the start time of the next test interval. + * Yes, this is vulnerable to long delays, but such + * delays simply cause a false negative for the next + * interval. Besides, we are running at RT priority, + * so delays should be relatively rare. + */ + while (oldstarttime == boost_starttime) { + if (mutex_trylock(&boost_mutex)) { + boost_starttime = jiffies + + test_boost_interval * HZ; + n_rcu_torture_boosts++; + mutex_unlock(&boost_mutex); + break; + } + schedule_timeout_uninterruptible(1); + } + + /* Go do the stutter. */ +checkwait: rcu_stutter_wait("rcu_torture_boost"); + } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); + + /* Clean up and exit. */ + VERBOSE_PRINTK_STRING("rcu_torture_boost task stopping"); + rcutorture_shutdown_absorb("rcu_torture_boost"); + while (!kthread_should_stop() || rbi.inflight) + schedule_timeout_uninterruptible(1); + smp_mb(); /* order accesses to ->inflight before stack-frame death. */ + return 0; +} + +/* * RCU torture force-quiescent-state kthread. Repeatedly induces * bursts of calls to force_quiescent_state(), increasing the probability * of occurrence of some important types of race conditions. @@ -933,7 +1068,8 @@ rcu_torture_printk(char *page) cnt += sprintf(&page[cnt], "%s%s ", torture_type, TORTURE_FLAG); cnt += sprintf(&page[cnt], "rtc: %p ver: %ld tfle: %d rta: %d rtaf: %d rtf: %d " - "rtmbe: %d nt: %ld", + "rtmbe: %d rtbke: %ld rtbre: %ld rtbae: %ld rtbafe: %ld " + "rtbf: %ld rtb: %ld nt: %ld", rcu_torture_current, rcu_torture_current_version, list_empty(&rcu_torture_freelist), @@ -941,8 +1077,19 @@ rcu_torture_printk(char *page) atomic_read(&n_rcu_torture_alloc_fail), atomic_read(&n_rcu_torture_free), atomic_read(&n_rcu_torture_mberror), + n_rcu_torture_boost_ktrerror, + n_rcu_torture_boost_rterror, + n_rcu_torture_boost_allocerror, + n_rcu_torture_boost_afferror, + n_rcu_torture_boost_failure, + n_rcu_torture_boosts, n_rcu_torture_timers); - if (atomic_read(&n_rcu_torture_mberror) != 0) + if (atomic_read(&n_rcu_torture_mberror) != 0 || + n_rcu_torture_boost_ktrerror != 0 || + n_rcu_torture_boost_rterror != 0 || + n_rcu_torture_boost_allocerror != 0 || + n_rcu_torture_boost_afferror != 0 || + n_rcu_torture_boost_failure != 0) cnt += sprintf(&page[cnt], " !!!"); cnt += sprintf(&page[cnt], "\n%s%s ", torture_type, TORTURE_FLAG); if (i > 1) { @@ -1094,22 +1241,91 @@ rcu_torture_stutter(void *arg) } static inline void -rcu_torture_print_module_parms(char *tag) +rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, char *tag) { printk(KERN_ALERT "%s" TORTURE_FLAG "--- %s: nreaders=%d nfakewriters=%d " "stat_interval=%d verbose=%d test_no_idle_hz=%d " "shuffle_interval=%d stutter=%d irqreader=%d " - "fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d\n", + "fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d " + "test_boost=%d/%d test_boost_interval=%d " + "test_boost_duration=%d\n", torture_type, tag, nrealreaders, nfakewriters, stat_interval, verbose, test_no_idle_hz, shuffle_interval, - stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter); + stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter, + test_boost, cur_ops->can_boost, + test_boost_interval, test_boost_duration); } -static struct notifier_block rcutorture_nb = { +static struct notifier_block rcutorture_shutdown_nb = { .notifier_call = rcutorture_shutdown_notify, }; +static void rcutorture_booster_cleanup(int cpu) +{ + struct task_struct *t; + + if (boost_tasks[cpu] == NULL) + return; + mutex_lock(&boost_mutex); + VERBOSE_PRINTK_STRING("Stopping rcu_torture_boost task"); + t = boost_tasks[cpu]; + boost_tasks[cpu] = NULL; + mutex_unlock(&boost_mutex); + + /* This must be outside of the mutex, otherwise deadlock! */ + kthread_stop(t); +} + +static int rcutorture_booster_init(int cpu) +{ + int retval; + + if (boost_tasks[cpu] != NULL) + return 0; /* Already created, nothing more to do. */ + + /* Don't allow time recalculation while creating a new task. */ + mutex_lock(&boost_mutex); + VERBOSE_PRINTK_STRING("Creating rcu_torture_boost task"); + boost_tasks[cpu] = kthread_create(rcu_torture_boost, NULL, + "rcu_torture_boost"); + if (IS_ERR(boost_tasks[cpu])) { + retval = PTR_ERR(boost_tasks[cpu]); + VERBOSE_PRINTK_STRING("rcu_torture_boost task create failed"); + n_rcu_torture_boost_ktrerror++; + boost_tasks[cpu] = NULL; + mutex_unlock(&boost_mutex); + return retval; + } + kthread_bind(boost_tasks[cpu], cpu); + wake_up_process(boost_tasks[cpu]); + mutex_unlock(&boost_mutex); + return 0; +} + +static int rcutorture_cpu_notify(struct notifier_block *self, + unsigned long action, void *hcpu) +{ + long cpu = (long)hcpu; + + switch (action) { + case CPU_ONLINE: + case CPU_DOWN_FAILED: + (void)rcutorture_booster_init(cpu); + break; + case CPU_DOWN_PREPARE: + rcutorture_booster_cleanup(cpu); + break; + default: + break; + } + return NOTIFY_OK; +} + +static struct notifier_block rcutorture_cpu_nb = { + .notifier_call = rcutorture_cpu_notify, +}; + static void rcu_torture_cleanup(void) { @@ -1127,7 +1343,7 @@ rcu_torture_cleanup(void) } fullstop = FULLSTOP_RMMOD; mutex_unlock(&fullstop_mutex); - unregister_reboot_notifier(&rcutorture_nb); + unregister_reboot_notifier(&rcutorture_shutdown_nb); if (stutter_task) { VERBOSE_PRINTK_STRING("Stopping rcu_torture_stutter task"); kthread_stop(stutter_task); @@ -1184,6 +1400,12 @@ rcu_torture_cleanup(void) kthread_stop(fqs_task); } fqs_task = NULL; + if ((test_boost == 1 && cur_ops->can_boost) || + test_boost == 2) { + unregister_cpu_notifier(&rcutorture_cpu_nb); + for_each_possible_cpu(i) + rcutorture_booster_cleanup(i); + } /* Wait for all RCU callbacks to fire. */ @@ -1195,9 +1417,9 @@ rcu_torture_cleanup(void) if (cur_ops->cleanup) cur_ops->cleanup(); if (atomic_read(&n_rcu_torture_error)) - rcu_torture_print_module_parms("End of test: FAILURE"); + rcu_torture_print_module_parms(cur_ops, "End of test: FAILURE"); else - rcu_torture_print_module_parms("End of test: SUCCESS"); + rcu_torture_print_module_parms(cur_ops, "End of test: SUCCESS"); } static int __init @@ -1242,7 +1464,7 @@ rcu_torture_init(void) nrealreaders = nreaders; else nrealreaders = 2 * num_online_cpus(); - rcu_torture_print_module_parms("Start of test"); + rcu_torture_print_module_parms(cur_ops, "Start of test"); fullstop = FULLSTOP_DONTSTOP; /* Set up the freelist. */ @@ -1263,6 +1485,12 @@ rcu_torture_init(void) atomic_set(&n_rcu_torture_free, 0); atomic_set(&n_rcu_torture_mberror, 0); atomic_set(&n_rcu_torture_error, 0); + n_rcu_torture_boost_ktrerror = 0; + n_rcu_torture_boost_rterror = 0; + n_rcu_torture_boost_allocerror = 0; + n_rcu_torture_boost_afferror = 0; + n_rcu_torture_boost_failure = 0; + n_rcu_torture_boosts = 0; for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) atomic_set(&rcu_torture_wcount[i], 0); for_each_possible_cpu(cpu) { @@ -1376,7 +1604,27 @@ rcu_torture_init(void) goto unwind; } } - register_reboot_notifier(&rcutorture_nb); + if (test_boost_interval < 1) + test_boost_interval = 1; + if (test_boost_duration < 2) + test_boost_duration = 2; + if ((test_boost == 1 && cur_ops->can_boost) || + test_boost == 2) { + int retval; + + boost_starttime = jiffies + test_boost_interval * HZ; + register_cpu_notifier(&rcutorture_cpu_nb); + for_each_possible_cpu(i) { + if (cpu_is_offline(i)) + continue; /* Heuristic: CPU can go offline. */ + retval = rcutorture_booster_init(i); + if (retval < 0) { + firsterr = retval; + goto unwind; + } + } + } + register_reboot_notifier(&rcutorture_shutdown_nb); mutex_unlock(&fullstop_mutex); return 0; diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ccdc04c47981..d0ddfea6579d 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -67,9 +67,6 @@ static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; .gpnum = -300, \ .completed = -300, \ .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&structname.onofflock), \ - .orphan_cbs_list = NULL, \ - .orphan_cbs_tail = &structname.orphan_cbs_list, \ - .orphan_qlen = 0, \ .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&structname.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ @@ -620,9 +617,17 @@ static void __init check_cpu_stall_init(void) static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { if (rdp->gpnum != rnp->gpnum) { - rdp->qs_pending = 1; - rdp->passed_quiesc = 0; + /* + * If the current grace period is waiting for this CPU, + * set up to detect a quiescent state, otherwise don't + * go looking for one. + */ rdp->gpnum = rnp->gpnum; + if (rnp->qsmask & rdp->grpmask) { + rdp->qs_pending = 1; + rdp->passed_quiesc = 0; + } else + rdp->qs_pending = 0; } } @@ -681,6 +686,24 @@ __rcu_process_gp_end(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_dat /* Remember that we saw this grace-period completion. */ rdp->completed = rnp->completed; + + /* + * If we were in an extended quiescent state, we may have + * missed some grace periods that others CPUs handled on + * our behalf. Catch up with this state to avoid noting + * spurious new grace periods. If another grace period + * has started, then rnp->gpnum will have advanced, so + * we will detect this later on. + */ + if (ULONG_CMP_LT(rdp->gpnum, rdp->completed)) + rdp->gpnum = rdp->completed; + + /* + * If RCU does not need a quiescent state from this CPU, + * then make sure that this CPU doesn't go looking for one. + */ + if ((rnp->qsmask & rdp->grpmask) == 0) + rdp->qs_pending = 0; } } @@ -984,53 +1007,31 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) #ifdef CONFIG_HOTPLUG_CPU /* - * Move a dying CPU's RCU callbacks to the ->orphan_cbs_list for the - * specified flavor of RCU. The callbacks will be adopted by the next - * _rcu_barrier() invocation or by the CPU_DEAD notifier, whichever - * comes first. Because this is invoked from the CPU_DYING notifier, - * irqs are already disabled. + * Move a dying CPU's RCU callbacks to online CPU's callback list. + * Synchronization is not required because this function executes + * in stop_machine() context. */ -static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +static void rcu_send_cbs_to_online(struct rcu_state *rsp) { int i; + /* current DYING CPU is cleared in the cpu_online_mask */ + int receive_cpu = cpumask_any(cpu_online_mask); struct rcu_data *rdp = this_cpu_ptr(rsp->rda); + struct rcu_data *receive_rdp = per_cpu_ptr(rsp->rda, receive_cpu); if (rdp->nxtlist == NULL) return; /* irqs disabled, so comparison is stable. */ - raw_spin_lock(&rsp->onofflock); /* irqs already disabled. */ - *rsp->orphan_cbs_tail = rdp->nxtlist; - rsp->orphan_cbs_tail = rdp->nxttail[RCU_NEXT_TAIL]; + + *receive_rdp->nxttail[RCU_NEXT_TAIL] = rdp->nxtlist; + receive_rdp->nxttail[RCU_NEXT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; + receive_rdp->qlen += rdp->qlen; + receive_rdp->n_cbs_adopted += rdp->qlen; + rdp->n_cbs_orphaned += rdp->qlen; + rdp->nxtlist = NULL; for (i = 0; i < RCU_NEXT_SIZE; i++) rdp->nxttail[i] = &rdp->nxtlist; - rsp->orphan_qlen += rdp->qlen; - rdp->n_cbs_orphaned += rdp->qlen; rdp->qlen = 0; - raw_spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ -} - -/* - * Adopt previously orphaned RCU callbacks. - */ -static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) -{ - unsigned long flags; - struct rcu_data *rdp; - - raw_spin_lock_irqsave(&rsp->onofflock, flags); - rdp = this_cpu_ptr(rsp->rda); - if (rsp->orphan_cbs_list == NULL) { - raw_spin_unlock_irqrestore(&rsp->onofflock, flags); - return; - } - *rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_list; - rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_tail; - rdp->qlen += rsp->orphan_qlen; - rdp->n_cbs_adopted += rsp->orphan_qlen; - rsp->orphan_cbs_list = NULL; - rsp->orphan_cbs_tail = &rsp->orphan_cbs_list; - rsp->orphan_qlen = 0; - raw_spin_unlock_irqrestore(&rsp->onofflock, flags); } /* @@ -1081,8 +1082,6 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) raw_spin_unlock_irqrestore(&rnp->lock, flags); if (need_report & RCU_OFL_TASKS_EXP_GP) rcu_report_exp_rnp(rsp, rnp); - - rcu_adopt_orphan_cbs(rsp); } /* @@ -1100,11 +1099,7 @@ static void rcu_offline_cpu(int cpu) #else /* #ifdef CONFIG_HOTPLUG_CPU */ -static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) -{ -} - -static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) +static void rcu_send_cbs_to_online(struct rcu_state *rsp) { } @@ -1440,22 +1435,11 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), */ local_irq_save(flags); rdp = this_cpu_ptr(rsp->rda); - rcu_process_gp_end(rsp, rdp); - check_for_new_grace_period(rsp, rdp); /* Add the callback to our list. */ *rdp->nxttail[RCU_NEXT_TAIL] = head; rdp->nxttail[RCU_NEXT_TAIL] = &head->next; - /* Start a new grace period if one not already started. */ - if (!rcu_gp_in_progress(rsp)) { - unsigned long nestflag; - struct rcu_node *rnp_root = rcu_get_root(rsp); - - raw_spin_lock_irqsave(&rnp_root->lock, nestflag); - rcu_start_gp(rsp, nestflag); /* releases rnp_root->lock. */ - } - /* * Force the grace period if too many callbacks or too long waiting. * Enforce hysteresis, and don't invoke force_quiescent_state() @@ -1464,12 +1448,27 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), * is the only one waiting for a grace period to complete. */ if (unlikely(++rdp->qlen > rdp->qlen_last_fqs_check + qhimark)) { - rdp->blimit = LONG_MAX; - if (rsp->n_force_qs == rdp->n_force_qs_snap && - *rdp->nxttail[RCU_DONE_TAIL] != head) - force_quiescent_state(rsp, 0); - rdp->n_force_qs_snap = rsp->n_force_qs; - rdp->qlen_last_fqs_check = rdp->qlen; + + /* Are we ignoring a completed grace period? */ + rcu_process_gp_end(rsp, rdp); + check_for_new_grace_period(rsp, rdp); + + /* Start a new grace period if one not already started. */ + if (!rcu_gp_in_progress(rsp)) { + unsigned long nestflag; + struct rcu_node *rnp_root = rcu_get_root(rsp); + + raw_spin_lock_irqsave(&rnp_root->lock, nestflag); + rcu_start_gp(rsp, nestflag); /* rlses rnp_root->lock */ + } else { + /* Give the grace period a kick. */ + rdp->blimit = LONG_MAX; + if (rsp->n_force_qs == rdp->n_force_qs_snap && + *rdp->nxttail[RCU_DONE_TAIL] != head) + force_quiescent_state(rsp, 0); + rdp->n_force_qs_snap = rsp->n_force_qs; + rdp->qlen_last_fqs_check = rdp->qlen; + } } else if (ULONG_CMP_LT(ACCESS_ONCE(rsp->jiffies_force_qs), jiffies)) force_quiescent_state(rsp, 1); local_irq_restore(flags); @@ -1699,13 +1698,12 @@ static void _rcu_barrier(struct rcu_state *rsp, * decrement rcu_barrier_cpu_count -- otherwise the first CPU * might complete its grace period before all of the other CPUs * did their increment, causing this function to return too - * early. + * early. Note that on_each_cpu() disables irqs, which prevents + * any CPUs from coming online or going offline until each online + * CPU has queued its RCU-barrier callback. */ atomic_set(&rcu_barrier_cpu_count, 1); - preempt_disable(); /* stop CPU_DYING from filling orphan_cbs_list */ - rcu_adopt_orphan_cbs(rsp); on_each_cpu(rcu_barrier_func, (void *)call_rcu_func, 1); - preempt_enable(); /* CPU_DYING can again fill orphan_cbs_list */ if (atomic_dec_and_test(&rcu_barrier_cpu_count)) complete(&rcu_barrier_completion); wait_for_completion(&rcu_barrier_completion); @@ -1831,18 +1829,13 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, case CPU_DYING: case CPU_DYING_FROZEN: /* - * preempt_disable() in _rcu_barrier() prevents stop_machine(), - * so when "on_each_cpu(rcu_barrier_func, (void *)type, 1);" - * returns, all online cpus have queued rcu_barrier_func(). - * The dying CPU clears its cpu_online_mask bit and - * moves all of its RCU callbacks to ->orphan_cbs_list - * in the context of stop_machine(), so subsequent calls - * to _rcu_barrier() will adopt these callbacks and only - * then queue rcu_barrier_func() on all remaining CPUs. + * The whole machine is "stopped" except this CPU, so we can + * touch any data without introducing corruption. We send the + * dying CPU's callbacks to an arbitrarily chosen online CPU. */ - rcu_send_cbs_to_orphanage(&rcu_bh_state); - rcu_send_cbs_to_orphanage(&rcu_sched_state); - rcu_preempt_send_cbs_to_orphanage(); + rcu_send_cbs_to_online(&rcu_bh_state); + rcu_send_cbs_to_online(&rcu_sched_state); + rcu_preempt_send_cbs_to_online(); break; case CPU_DEAD: case CPU_DEAD_FROZEN: @@ -1880,8 +1873,9 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp) { int i; - for (i = NUM_RCU_LVLS - 1; i >= 0; i--) + for (i = NUM_RCU_LVLS - 1; i > 0; i--) rsp->levelspread[i] = CONFIG_RCU_FANOUT; + rsp->levelspread[0] = RCU_FANOUT_LEAF; } #else /* #ifdef CONFIG_RCU_FANOUT_EXACT */ static void __init rcu_init_levelspread(struct rcu_state *rsp) diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 91d4170c5c13..e8f057e44e3e 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -31,46 +31,51 @@ /* * Define shape of hierarchy based on NR_CPUS and CONFIG_RCU_FANOUT. * In theory, it should be possible to add more levels straightforwardly. - * In practice, this has not been tested, so there is probably some - * bug somewhere. + * In practice, this did work well going from three levels to four. + * Of course, your mileage may vary. */ #define MAX_RCU_LVLS 4 -#define RCU_FANOUT (CONFIG_RCU_FANOUT) -#define RCU_FANOUT_SQ (RCU_FANOUT * RCU_FANOUT) -#define RCU_FANOUT_CUBE (RCU_FANOUT_SQ * RCU_FANOUT) -#define RCU_FANOUT_FOURTH (RCU_FANOUT_CUBE * RCU_FANOUT) - -#if NR_CPUS <= RCU_FANOUT +#if CONFIG_RCU_FANOUT > 16 +#define RCU_FANOUT_LEAF 16 +#else /* #if CONFIG_RCU_FANOUT > 16 */ +#define RCU_FANOUT_LEAF (CONFIG_RCU_FANOUT) +#endif /* #else #if CONFIG_RCU_FANOUT > 16 */ +#define RCU_FANOUT_1 (RCU_FANOUT_LEAF) +#define RCU_FANOUT_2 (RCU_FANOUT_1 * CONFIG_RCU_FANOUT) +#define RCU_FANOUT_3 (RCU_FANOUT_2 * CONFIG_RCU_FANOUT) +#define RCU_FANOUT_4 (RCU_FANOUT_3 * CONFIG_RCU_FANOUT) + +#if NR_CPUS <= RCU_FANOUT_1 # define NUM_RCU_LVLS 1 # define NUM_RCU_LVL_0 1 # define NUM_RCU_LVL_1 (NR_CPUS) # define NUM_RCU_LVL_2 0 # define NUM_RCU_LVL_3 0 # define NUM_RCU_LVL_4 0 -#elif NR_CPUS <= RCU_FANOUT_SQ +#elif NR_CPUS <= RCU_FANOUT_2 # define NUM_RCU_LVLS 2 # define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) # define NUM_RCU_LVL_2 (NR_CPUS) # define NUM_RCU_LVL_3 0 # define NUM_RCU_LVL_4 0 -#elif NR_CPUS <= RCU_FANOUT_CUBE +#elif NR_CPUS <= RCU_FANOUT_3 # define NUM_RCU_LVLS 3 # define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_SQ) -# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) -# define NUM_RCU_LVL_3 NR_CPUS +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) +# define NUM_RCU_LVL_3 (NR_CPUS) # define NUM_RCU_LVL_4 0 -#elif NR_CPUS <= RCU_FANOUT_FOURTH +#elif NR_CPUS <= RCU_FANOUT_4 # define NUM_RCU_LVLS 4 # define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_CUBE) -# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_SQ) -# define NUM_RCU_LVL_3 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) -# define NUM_RCU_LVL_4 NR_CPUS +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) +# define NUM_RCU_LVL_3 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) +# define NUM_RCU_LVL_4 (NR_CPUS) #else # error "CONFIG_RCU_FANOUT insufficient for NR_CPUS" -#endif /* #if (NR_CPUS) <= RCU_FANOUT */ +#endif /* #if (NR_CPUS) <= RCU_FANOUT_1 */ #define RCU_SUM (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3 + NUM_RCU_LVL_4) #define NUM_RCU_NODES (RCU_SUM - NR_CPUS) @@ -203,8 +208,8 @@ struct rcu_data { long qlen_last_fqs_check; /* qlen at last check for QS forcing */ unsigned long n_cbs_invoked; /* count of RCU cbs invoked. */ - unsigned long n_cbs_orphaned; /* RCU cbs sent to orphanage. */ - unsigned long n_cbs_adopted; /* RCU cbs adopted from orphanage. */ + unsigned long n_cbs_orphaned; /* RCU cbs orphaned by dying CPU */ + unsigned long n_cbs_adopted; /* RCU cbs adopted from dying CPU */ unsigned long n_force_qs_snap; /* did other CPU force QS recently? */ long blimit; /* Upper limit on a processed batch */ @@ -309,15 +314,7 @@ struct rcu_state { /* End of fields guarded by root rcu_node's lock. */ raw_spinlock_t onofflock; /* exclude on/offline and */ - /* starting new GP. Also */ - /* protects the following */ - /* orphan_cbs fields. */ - struct rcu_head *orphan_cbs_list; /* list of rcu_head structs */ - /* orphaned by all CPUs in */ - /* a given leaf rcu_node */ - /* going offline. */ - struct rcu_head **orphan_cbs_tail; /* And tail pointer. */ - long orphan_qlen; /* Number of orphaned cbs. */ + /* starting new GP. */ raw_spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ @@ -390,7 +387,7 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp); static int rcu_preempt_pending(int cpu); static int rcu_preempt_needs_cpu(int cpu); static void __cpuinit rcu_preempt_init_percpu_data(int cpu); -static void rcu_preempt_send_cbs_to_orphanage(void); +static void rcu_preempt_send_cbs_to_online(void); static void __init __rcu_init_preempt(void); static void rcu_needs_cpu_flush(void); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 71a4147473f9..a3638710dc67 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -25,6 +25,7 @@ */ #include <linux/delay.h> +#include <linux/stop_machine.h> /* * Check the RCU kernel configuration parameters and print informative @@ -773,11 +774,11 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) } /* - * Move preemptable RCU's callbacks to ->orphan_cbs_list. + * Move preemptable RCU's callbacks from dying CPU to other online CPU. */ -static void rcu_preempt_send_cbs_to_orphanage(void) +static void rcu_preempt_send_cbs_to_online(void) { - rcu_send_cbs_to_orphanage(&rcu_preempt_state); + rcu_send_cbs_to_online(&rcu_preempt_state); } /* @@ -1001,7 +1002,7 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) /* * Because there is no preemptable RCU, there are no callbacks to move. */ -static void rcu_preempt_send_cbs_to_orphanage(void) +static void rcu_preempt_send_cbs_to_online(void) { } @@ -1014,6 +1015,132 @@ static void __init __rcu_init_preempt(void) #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifndef CONFIG_SMP + +void synchronize_sched_expedited(void) +{ + cond_resched(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#else /* #ifndef CONFIG_SMP */ + +static atomic_t sync_sched_expedited_started = ATOMIC_INIT(0); +static atomic_t sync_sched_expedited_done = ATOMIC_INIT(0); + +static int synchronize_sched_expedited_cpu_stop(void *data) +{ + /* + * There must be a full memory barrier on each affected CPU + * between the time that try_stop_cpus() is called and the + * time that it returns. + * + * In the current initial implementation of cpu_stop, the + * above condition is already met when the control reaches + * this point and the following smp_mb() is not strictly + * necessary. Do smp_mb() anyway for documentation and + * robustness against future implementation changes. + */ + smp_mb(); /* See above comment block. */ + return 0; +} + +/* + * Wait for an rcu-sched grace period to elapse, but use "big hammer" + * approach to force grace period to end quickly. This consumes + * significant time on all CPUs, and is thus not recommended for + * any sort of common-case code. + * + * Note that it is illegal to call this function while holding any + * lock that is acquired by a CPU-hotplug notifier. Failing to + * observe this restriction will result in deadlock. + * + * This implementation can be thought of as an application of ticket + * locking to RCU, with sync_sched_expedited_started and + * sync_sched_expedited_done taking on the roles of the halves + * of the ticket-lock word. Each task atomically increments + * sync_sched_expedited_started upon entry, snapshotting the old value, + * then attempts to stop all the CPUs. If this succeeds, then each + * CPU will have executed a context switch, resulting in an RCU-sched + * grace period. We are then done, so we use atomic_cmpxchg() to + * update sync_sched_expedited_done to match our snapshot -- but + * only if someone else has not already advanced past our snapshot. + * + * On the other hand, if try_stop_cpus() fails, we check the value + * of sync_sched_expedited_done. If it has advanced past our + * initial snapshot, then someone else must have forced a grace period + * some time after we took our snapshot. In this case, our work is + * done for us, and we can simply return. Otherwise, we try again, + * but keep our initial snapshot for purposes of checking for someone + * doing our work for us. + * + * If we fail too many times in a row, we fall back to synchronize_sched(). + */ +void synchronize_sched_expedited(void) +{ + int firstsnap, s, snap, trycount = 0; + + /* Note that atomic_inc_return() implies full memory barrier. */ + firstsnap = snap = atomic_inc_return(&sync_sched_expedited_started); + get_online_cpus(); + + /* + * Each pass through the following loop attempts to force a + * context switch on each CPU. + */ + while (try_stop_cpus(cpu_online_mask, + synchronize_sched_expedited_cpu_stop, + NULL) == -EAGAIN) { + put_online_cpus(); + + /* No joy, try again later. Or just synchronize_sched(). */ + if (trycount++ < 10) + udelay(trycount * num_online_cpus()); + else { + synchronize_sched(); + return; + } + + /* Check to see if someone else did our work for us. */ + s = atomic_read(&sync_sched_expedited_done); + if (UINT_CMP_GE((unsigned)s, (unsigned)firstsnap)) { + smp_mb(); /* ensure test happens before caller kfree */ + return; + } + + /* + * Refetching sync_sched_expedited_started allows later + * callers to piggyback on our grace period. We subtract + * 1 to get the same token that the last incrementer got. + * We retry after they started, so our grace period works + * for them, and they started after our first try, so their + * grace period works for us. + */ + get_online_cpus(); + snap = atomic_read(&sync_sched_expedited_started) - 1; + smp_mb(); /* ensure read is before try_stop_cpus(). */ + } + + /* + * Everyone up to our most recent fetch is covered by our grace + * period. Update the counter, but only if our work is still + * relevant -- which it won't be if someone who started later + * than we did beat us to the punch. + */ + do { + s = atomic_read(&sync_sched_expedited_done); + if (UINT_CMP_GE((unsigned)s, (unsigned)snap)) { + smp_mb(); /* ensure test happens before caller kfree */ + break; + } + } while (atomic_cmpxchg(&sync_sched_expedited_done, s, snap) != s); + + put_online_cpus(); +} +EXPORT_SYMBOL_GPL(synchronize_sched_expedited); + +#endif /* #else #ifndef CONFIG_SMP */ + #if !defined(CONFIG_RCU_FAST_NO_HZ) /* diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index d15430b9d122..c8e97853b970 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -166,13 +166,13 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) gpnum = rsp->gpnum; seq_printf(m, "c=%lu g=%lu s=%d jfq=%ld j=%x " - "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld\n", + "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu\n", rsp->completed, gpnum, rsp->signaled, (long)(rsp->jiffies_force_qs - jiffies), (int)(jiffies & 0xffff), rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, - rsp->n_force_qs_lh, rsp->orphan_qlen); + rsp->n_force_qs_lh); for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < NUM_RCU_NODES; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); @@ -300,7 +300,7 @@ static const struct file_operations rcu_pending_fops = { static struct dentry *rcudir; -static int __init rcuclassic_trace_init(void) +static int __init rcutree_trace_init(void) { struct dentry *retval; @@ -337,14 +337,14 @@ free_out: return 1; } -static void __exit rcuclassic_trace_cleanup(void) +static void __exit rcutree_trace_cleanup(void) { debugfs_remove_recursive(rcudir); } -module_init(rcuclassic_trace_init); -module_exit(rcuclassic_trace_cleanup); +module_init(rcutree_trace_init); +module_exit(rcutree_trace_cleanup); MODULE_AUTHOR("Paul E. McKenney"); MODULE_DESCRIPTION("Read-Copy Update tracing for hierarchical implementation"); diff --git a/kernel/sched.c b/kernel/sched.c index 114a0deb2b01..04949089e760 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -8067,8 +8067,6 @@ void __init sched_init(void) zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); #endif /* SMP */ - perf_event_init(); - scheduler_running = 1; } @@ -9241,72 +9239,3 @@ struct cgroup_subsys cpuacct_subsys = { }; #endif /* CONFIG_CGROUP_CPUACCT */ -#ifndef CONFIG_SMP - -void synchronize_sched_expedited(void) -{ - barrier(); -} -EXPORT_SYMBOL_GPL(synchronize_sched_expedited); - -#else /* #ifndef CONFIG_SMP */ - -static atomic_t synchronize_sched_expedited_count = ATOMIC_INIT(0); - -static int synchronize_sched_expedited_cpu_stop(void *data) -{ - /* - * There must be a full memory barrier on each affected CPU - * between the time that try_stop_cpus() is called and the - * time that it returns. - * - * In the current initial implementation of cpu_stop, the - * above condition is already met when the control reaches - * this point and the following smp_mb() is not strictly - * necessary. Do smp_mb() anyway for documentation and - * robustness against future implementation changes. - */ - smp_mb(); /* See above comment block. */ - return 0; -} - -/* - * Wait for an rcu-sched grace period to elapse, but use "big hammer" - * approach to force grace period to end quickly. This consumes - * significant time on all CPUs, and is thus not recommended for - * any sort of common-case code. - * - * Note that it is illegal to call this function while holding any - * lock that is acquired by a CPU-hotplug notifier. Failing to - * observe this restriction will result in deadlock. - */ -void synchronize_sched_expedited(void) -{ - int snap, trycount = 0; - - smp_mb(); /* ensure prior mod happens before capturing snap. */ - snap = atomic_read(&synchronize_sched_expedited_count) + 1; - get_online_cpus(); - while (try_stop_cpus(cpu_online_mask, - synchronize_sched_expedited_cpu_stop, - NULL) == -EAGAIN) { - put_online_cpus(); - if (trycount++ < 10) - udelay(trycount * num_online_cpus()); - else { - synchronize_sched(); - return; - } - if (atomic_read(&synchronize_sched_expedited_count) - snap > 0) { - smp_mb(); /* ensure test happens before caller kfree */ - return; - } - get_online_cpus(); - } - atomic_inc(&synchronize_sched_expedited_count); - smp_mb__after_atomic_inc(); /* ensure post-GP actions seen after GP. */ - put_online_cpus(); -} -EXPORT_SYMBOL_GPL(synchronize_sched_expedited); - -#endif /* #else #ifndef CONFIG_SMP */ diff --git a/kernel/srcu.c b/kernel/srcu.c index c71e07500536..98d8c1e80edb 100644 --- a/kernel/srcu.c +++ b/kernel/srcu.c @@ -31,6 +31,7 @@ #include <linux/rcupdate.h> #include <linux/sched.h> #include <linux/smp.h> +#include <linux/delay.h> #include <linux/srcu.h> static int init_srcu_struct_fields(struct srcu_struct *sp) @@ -203,9 +204,14 @@ static void __synchronize_srcu(struct srcu_struct *sp, void (*sync_func)(void)) * all srcu_read_lock() calls using the old counters have completed. * Their corresponding critical sections might well be still * executing, but the srcu_read_lock() primitives themselves - * will have finished executing. + * will have finished executing. We initially give readers + * an arbitrarily chosen 10 microseconds to get out of their + * SRCU read-side critical sections, then loop waiting 1/HZ + * seconds per iteration. */ + if (srcu_readers_active_idx(sp, idx)) + udelay(CONFIG_SRCU_SYNCHRONIZE_DELAY); while (srcu_readers_active_idx(sp, idx)) schedule_timeout_interruptible(1); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 121e4fff03d1..ae5cbb1e3ced 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -744,21 +744,21 @@ static struct ctl_table kern_table[] = { .extra1 = &zero, .extra2 = &one, }, -#endif -#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) && !defined(CONFIG_LOCKUP_DETECTOR) { - .procname = "unknown_nmi_panic", - .data = &unknown_nmi_panic, + .procname = "nmi_watchdog", + .data = &watchdog_enabled, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dowatchdog_enabled, }, +#endif +#if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) { - .procname = "nmi_watchdog", - .data = &nmi_watchdog_enabled, + .procname = "unknown_nmi_panic", + .data = &unknown_nmi_panic, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = proc_nmi_enabled, + .proc_handler = proc_dointvec, }, #endif #if defined(CONFIG_X86) diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 1357c5786064..4b2545a136ff 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -136,7 +136,6 @@ static const struct bin_table bin_kern_table[] = { { CTL_INT, KERN_IA64_UNALIGNED, "ignore-unaligned-usertrap" }, { CTL_INT, KERN_COMPAT_LOG, "compat-log" }, { CTL_INT, KERN_MAX_LOCK_DEPTH, "max_lock_depth" }, - { CTL_INT, KERN_NMI_WATCHDOG, "nmi_watchdog" }, { CTL_INT, KERN_PANIC_ON_NMI, "panic_on_unrecovered_nmi" }, {} }; diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index ea37e2ff4164..14674dce77a6 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -69,6 +69,21 @@ config EVENT_TRACING select CONTEXT_SWITCH_TRACER bool +config EVENT_POWER_TRACING_DEPRECATED + depends on EVENT_TRACING + bool "Deprecated power event trace API, to be removed" + default y + help + Provides old power event types: + C-state/idle accounting events: + power:power_start + power:power_end + and old cpufreq accounting event: + power:power_frequency + This is for userspace compatibility + and will vanish after 5 kernel iterations, + namely 2.6.41. + config CONTEXT_SWITCH_TRACER bool diff --git a/kernel/trace/power-traces.c b/kernel/trace/power-traces.c index a22582a06161..f55fcf61b223 100644 --- a/kernel/trace/power-traces.c +++ b/kernel/trace/power-traces.c @@ -13,5 +13,8 @@ #define CREATE_TRACE_POINTS #include <trace/events/power.h> -EXPORT_TRACEPOINT_SYMBOL_GPL(power_frequency); +#ifdef EVENT_POWER_TRACING_DEPRECATED +EXPORT_TRACEPOINT_SYMBOL_GPL(power_start); +#endif +EXPORT_TRACEPOINT_SYMBOL_GPL(cpu_idle); diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 39c059ca670e..19a359d5e6d5 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -21,17 +21,46 @@ typedef typeof(unsigned long [PERF_MAX_TRACE_SIZE / sizeof(unsigned long)]) /* Count the events in use (per event id, not per instance) */ static int total_ref_count; +static int perf_trace_event_perm(struct ftrace_event_call *tp_event, + struct perf_event *p_event) +{ + /* No tracing, just counting, so no obvious leak */ + if (!(p_event->attr.sample_type & PERF_SAMPLE_RAW)) + return 0; + + /* Some events are ok to be traced by non-root users... */ + if (p_event->attach_state == PERF_ATTACH_TASK) { + if (tp_event->flags & TRACE_EVENT_FL_CAP_ANY) + return 0; + } + + /* + * ...otherwise raw tracepoint data can be a severe data leak, + * only allow root to have these. + */ + if (perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN)) + return -EPERM; + + return 0; +} + static int perf_trace_event_init(struct ftrace_event_call *tp_event, struct perf_event *p_event) { struct hlist_head __percpu *list; - int ret = -ENOMEM; + int ret; int cpu; + ret = perf_trace_event_perm(tp_event, p_event); + if (ret) + return ret; + p_event->tp_event = tp_event; if (tp_event->perf_refcount++ > 0) return 0; + ret = -ENOMEM; + list = alloc_percpu(struct hlist_head); if (!list) goto fail; diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 0725eeab1937..35fde09b81de 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -27,6 +27,12 @@ DEFINE_MUTEX(event_mutex); +DEFINE_MUTEX(event_storage_mutex); +EXPORT_SYMBOL_GPL(event_storage_mutex); + +char event_storage[EVENT_STORAGE_SIZE]; +EXPORT_SYMBOL_GPL(event_storage); + LIST_HEAD(ftrace_events); LIST_HEAD(ftrace_common_fields); diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index 4ba44deaac25..4b74d71705c0 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -83,13 +83,19 @@ static void __always_unused ____ftrace_check_##name(void) \ #undef __array #define __array(type, item, len) \ - BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ - ret = trace_define_field(event_call, #type "[" #len "]", #item, \ + do { \ + BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ + mutex_lock(&event_storage_mutex); \ + snprintf(event_storage, sizeof(event_storage), \ + "%s[%d]", #type, len); \ + ret = trace_define_field(event_call, event_storage, #item, \ offsetof(typeof(field), item), \ sizeof(field.item), \ is_signed_type(type), FILTER_OTHER); \ - if (ret) \ - return ret; + mutex_unlock(&event_storage_mutex); \ + if (ret) \ + return ret; \ + } while (0); #undef __array_desc #define __array_desc(type, container, item, len) \ diff --git a/kernel/watchdog.c b/kernel/watchdog.c index c812c4927cab..6e7b575ac33c 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -57,6 +57,8 @@ static int __init hardlockup_panic_setup(char *str) { if (!strncmp(str, "panic", 5)) hardlockup_panic = 1; + else if (!strncmp(str, "0", 1)) + no_watchdog = 1; return 1; } __setup("nmi_watchdog=", hardlockup_panic_setup); @@ -548,13 +550,13 @@ static struct notifier_block __cpuinitdata cpu_nfb = { .notifier_call = cpu_callback }; -static int __init spawn_watchdog_task(void) +void __init lockup_detector_init(void) { void *cpu = (void *)(long)smp_processor_id(); int err; if (no_watchdog) - return 0; + return; err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu); WARN_ON(notifier_to_errno(err)); @@ -562,6 +564,5 @@ static int __init spawn_watchdog_task(void) cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); register_cpu_notifier(&cpu_nfb); - return 0; + return; } -early_initcall(spawn_watchdog_task); diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 28b42b9274d0..2d05adb98401 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -173,7 +173,8 @@ config LOCKUP_DETECTOR An NMI is generated every 60 seconds or so to check for hardlockups. config HARDLOCKUP_DETECTOR - def_bool LOCKUP_DETECTOR && PERF_EVENTS && HAVE_PERF_EVENTS_NMI + def_bool LOCKUP_DETECTOR && PERF_EVENTS && HAVE_PERF_EVENTS_NMI && \ + !ARCH_HAS_NMI_WATCHDOG config BOOTPARAM_SOFTLOCKUP_PANIC bool "Panic (Reboot) On Soft Lockups" diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 5ad25e17b6cb..4eb99ab34053 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -214,17 +214,22 @@ ifdef BUILD_C_RECORDMCOUNT # The empty.o file is created in the make process in order to determine # the target endianness and word size. It is made before all other C # files, including recordmcount. -cmd_record_mcount = if [ $(@) != "scripts/mod/empty.o" ]; then \ - $(objtree)/scripts/recordmcount "$(@)"; \ - fi; +sub_cmd_record_mcount = \ + if [ $(@) != "scripts/mod/empty.o" ]; then \ + $(objtree)/scripts/recordmcount "$(@)"; \ + fi; else -cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \ +sub_cmd_record_mcount = set -e ; perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \ "$(if $(CONFIG_CPU_BIG_ENDIAN),big,little)" \ "$(if $(CONFIG_64BIT),64,32)" \ "$(OBJDUMP)" "$(OBJCOPY)" "$(CC) $(KBUILD_CFLAGS)" \ "$(LD)" "$(NM)" "$(RM)" "$(MV)" \ "$(if $(part-of-module),1,0)" "$(@)"; endif +cmd_record_mcount = \ + if [ "$(findstring -pg,$(_c_flags))" = "-pg" ]; then \ + $(sub_cmd_record_mcount) \ + fi; endif define rule_cc_o_c diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 39580a5dc5df..9f85012acf0d 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -155,6 +155,8 @@ use strict; # '@parameter' - name of a parameter # '%CONST' - name of a constant. +## init lots of data + my $errors = 0; my $warnings = 0; my $anon_struct_union = 0; @@ -218,21 +220,14 @@ my %highlights_list = ( $type_constant, "\$1", $type_param, "\$1" ); my $blankline_list = ""; -sub usage { - print "Usage: $0 [ -v ] [ -docbook | -html | -text | -man | -list ]\n"; - print " [ -no-doc-sections ]\n"; - print " [ -function funcname [ -function funcname ...] ]\n"; - print " [ -nofunction funcname [ -nofunction funcname ...] ]\n"; - print " c source file(s) > outputfile\n"; - print " -v : verbose output, more warnings & other info listed\n"; - exit 1; -} - # read arguments if ($#ARGV == -1) { usage(); } +my $kernelversion; +my $dohighlight = ""; + my $verbose = 0; my $output_mode = "man"; my $no_doc_sections = 0; @@ -245,7 +240,7 @@ my $man_date = ('January', 'February', 'March', 'April', 'May', 'June', 'November', 'December')[(localtime)[4]] . " " . ((localtime)[5]+1900); -# Essentially these are globals +# Essentially these are globals. # They probably want to be tidied up, made more localised or something. # CAVEAT EMPTOR! Some of the others I localised may not want to be, which # could cause "use of undefined value" or other bugs. @@ -353,6 +348,18 @@ while ($ARGV[0] =~ m/^-(.*)/) { } } +# continue execution near EOF; + +sub usage { + print "Usage: $0 [ -v ] [ -docbook | -html | -text | -man | -list ]\n"; + print " [ -no-doc-sections ]\n"; + print " [ -function funcname [ -function funcname ...] ]\n"; + print " [ -nofunction funcname [ -nofunction funcname ...] ]\n"; + print " c source file(s) > outputfile\n"; + print " -v : verbose output, more warnings & other info listed\n"; + exit 1; +} + # get kernel version from env sub get_kernel_version() { my $version = 'unknown kernel version'; @@ -362,15 +369,6 @@ sub get_kernel_version() { } return $version; } -my $kernelversion = get_kernel_version(); - -# generate a sequence of code that will splice in highlighting information -# using the s// operator. -my $dohighlight = ""; -foreach my $pattern (keys %highlights) { -# print STDERR "scanning pattern:$pattern, highlight:($highlights{$pattern})\n"; - $dohighlight .= "\$contents =~ s:$pattern:$highlights{$pattern}:gs;\n"; -} ## # dumps section contents to arrays/hashes intended for that purpose. @@ -1851,34 +1849,6 @@ sub dump_function($$) { }); } -sub process_file($); - -# Read the file that maps relative names to absolute names for -# separate source and object directories and for shadow trees. -if (open(SOURCE_MAP, "<.tmp_filelist.txt")) { - my ($relname, $absname); - while(<SOURCE_MAP>) { - chop(); - ($relname, $absname) = (split())[0..1]; - $relname =~ s:^/+::; - $source_map{$relname} = $absname; - } - close(SOURCE_MAP); -} - -foreach (@ARGV) { - chomp; - process_file($_); -} -if ($verbose && $errors) { - print STDERR "$errors errors\n"; -} -if ($verbose && $warnings) { - print STDERR "$warnings warnings\n"; -} - -exit($errors); - sub reset_state { $function = ""; %constants = (); @@ -2285,3 +2255,39 @@ sub process_file($) { } } } + + +$kernelversion = get_kernel_version(); + +# generate a sequence of code that will splice in highlighting information +# using the s// operator. +foreach my $pattern (keys %highlights) { +# print STDERR "scanning pattern:$pattern, highlight:($highlights{$pattern})\n"; + $dohighlight .= "\$contents =~ s:$pattern:$highlights{$pattern}:gs;\n"; +} + +# Read the file that maps relative names to absolute names for +# separate source and object directories and for shadow trees. +if (open(SOURCE_MAP, "<.tmp_filelist.txt")) { + my ($relname, $absname); + while(<SOURCE_MAP>) { + chop(); + ($relname, $absname) = (split())[0..1]; + $relname =~ s:^/+::; + $source_map{$relname} = $absname; + } + close(SOURCE_MAP); +} + +foreach (@ARGV) { + chomp; + process_file($_); +} +if ($verbose && $errors) { + print STDERR "$errors errors\n"; +} +if ($verbose && $warnings) { + print STDERR "$warnings warnings\n"; +} + +exit($errors); diff --git a/tools/perf/Documentation/perf-annotate.txt b/tools/perf/Documentation/perf-annotate.txt index b2c63309a651..6f5a498608b2 100644 --- a/tools/perf/Documentation/perf-annotate.txt +++ b/tools/perf/Documentation/perf-annotate.txt @@ -24,12 +24,47 @@ OPTIONS --input=:: Input file name. (default: perf.data) +-d:: +--dsos=<dso[,dso...]>:: + Only consider symbols in these dsos. +-s:: +--symbol=<symbol>:: + Symbol to annotate. + +-f:: +--force:: + Don't complain, do it. + +-v:: +--verbose:: + Be more verbose. (Show symbol address, etc) + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-k:: +--vmlinux=<file>:: + vmlinux pathname. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel. + +-l:: +--print-line:: + Print matching source lines (may be slow). + +-P:: +--full-paths:: + Don't shorten the displayed pathnames. + --stdio:: Use the stdio interface. --tui:: Use the TUI interface Use of --tui requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. This interfaces starts by centering on the line with more - samples, TAB/UNTAB cycles thru the lines with more samples. + samples, TAB/UNTAB cycles through the lines with more samples. SEE ALSO -------- diff --git a/tools/perf/Documentation/perf-buildid-list.txt b/tools/perf/Documentation/perf-buildid-list.txt index 01b642c0bf8f..5eaac6f26d51 100644 --- a/tools/perf/Documentation/perf-buildid-list.txt +++ b/tools/perf/Documentation/perf-buildid-list.txt @@ -18,6 +18,9 @@ perf report. OPTIONS ------- +-H:: +--with-hits:: + Show only DSOs with hits. -i:: --input=:: Input file name. (default: perf.data) diff --git a/tools/perf/Documentation/perf-diff.txt b/tools/perf/Documentation/perf-diff.txt index 20d97d84ea1c..74d7481ed7a6 100644 --- a/tools/perf/Documentation/perf-diff.txt +++ b/tools/perf/Documentation/perf-diff.txt @@ -19,6 +19,18 @@ If no parameters are passed it will assume perf.data.old and perf.data. OPTIONS ------- +-M:: +--displacement:: + Show position displacement relative to baseline. + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + +-m:: +--modules:: + Load module symbols. WARNING: use only with -k and LIVE kernel + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands @@ -42,7 +54,7 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. -v:: @@ -50,6 +62,13 @@ OPTIONS Be verbose, for instance, show the raw counts in addition to the diff. +-f:: +--force:: + Don't complain, do it. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + SEE ALSO -------- linkperf:perf-record[1] diff --git a/tools/perf/Documentation/perf-kvm.txt b/tools/perf/Documentation/perf-kvm.txt index d004e19fe6d6..dd84cb2f0a88 100644 --- a/tools/perf/Documentation/perf-kvm.txt +++ b/tools/perf/Documentation/perf-kvm.txt @@ -22,7 +22,7 @@ There are a couple of variants of perf kvm: a performance counter profile of guest os in realtime of an arbitrary workload. - 'perf kvm record <command>' to record the performance couinter profile + 'perf kvm record <command>' to record the performance counter profile of an arbitrary workload and save it into a perf data file. If both --host and --guest are input, the perf data file name is perf.data.kvm. If there is no --host but --guest, the file name is perf.data.guest. @@ -40,6 +40,12 @@ There are a couple of variants of perf kvm: OPTIONS ------- +-i:: +--input=:: + Input file name. +-o:: +--output:: + Output file name. --host=:: Collect host side performance profile. --guest=:: diff --git a/tools/perf/Documentation/perf-lock.txt b/tools/perf/Documentation/perf-lock.txt index b317102138c8..921de259ea10 100644 --- a/tools/perf/Documentation/perf-lock.txt +++ b/tools/perf/Documentation/perf-lock.txt @@ -24,6 +24,21 @@ and statistics with this 'perf lock' command. 'perf lock report' reports statistical data. +OPTIONS +------- + +-i:: +--input=<file>:: + Input file name. + +-v:: +--verbose:: + Be more verbose (show symbol address, etc). + +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + SEE ALSO -------- linkperf:perf[1] diff --git a/tools/perf/Documentation/perf-probe.txt b/tools/perf/Documentation/perf-probe.txt index 62de1b7f4e76..86b797a35aa6 100644 --- a/tools/perf/Documentation/perf-probe.txt +++ b/tools/perf/Documentation/perf-probe.txt @@ -115,9 +115,9 @@ Each probe argument follows below syntax. LINE SYNTAX ----------- -Line range is descripted by following syntax. +Line range is described by following syntax. - "FUNC[:RLN[+NUM|-RLN2]]|SRC:ALN[+NUM|-ALN2]" + "FUNC[:RLN[+NUM|-RLN2]]|SRC[:ALN[+NUM|-ALN2]]" FUNC specifies the function name of showing lines. 'RLN' is the start line number from function entry line, and 'RLN2' is the end line number. As same as diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt index a91f9f9e6e5c..52462ae26455 100644 --- a/tools/perf/Documentation/perf-record.txt +++ b/tools/perf/Documentation/perf-record.txt @@ -39,15 +39,24 @@ OPTIONS be passed as follows: '\mem:addr[:[r][w][x]]'. If you want to profile read-write accesses in 0x1000, just set 'mem:0x1000:rw'. + +--filter=<filter>:: + Event filter. + -a:: - System-wide collection. +--all-cpus:: + System-wide collection from all CPUs. -l:: Scale counter values. -p:: --pid=:: - Record events on existing pid. + Record events on existing process ID. + +-t:: +--tid=:: + Record events on existing thread ID. -r:: --realtime=:: @@ -99,6 +108,11 @@ OPTIONS --data:: Sample addresses. +-T:: +--timestamp:: + Sample timestamps. Use it with 'perf report -D' to see the timestamps, + for instance. + -n:: --no-samples:: Don't sample. @@ -109,8 +123,8 @@ Collect raw sample records from all opened counters (default for tracepoint coun -C:: --cpu:: -Collect samples only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Collect samples only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode with inheritance mode on (default), samples are captured only when the thread executes on the designated CPUs. Default is to monitor all CPUs. diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index 12052c9ed0ba..8ba03d6e5398 100644 --- a/tools/perf/Documentation/perf-report.txt +++ b/tools/perf/Documentation/perf-report.txt @@ -20,6 +20,11 @@ OPTIONS -i:: --input=:: Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -d:: --dsos=:: Only consider symbols in these dsos. CSV that understands @@ -27,6 +32,10 @@ OPTIONS -n:: --show-nr-samples:: Show the number of samples for each symbol + +--showcpuutilization:: + Show sample percentage for different cpu modes. + -T:: --threads:: Show per-thread event counters @@ -39,12 +48,24 @@ OPTIONS Only consider these symbols. CSV that understands file://filename entries. +-U:: +--hide-unresolved:: + Only display entries resolved to a symbol. + -s:: --sort=:: Sort by key(s): pid, comm, dso, symbol, parent. +-p:: +--parent=<regex>:: + regex filter to identify parent, see: '--sort parent' + +-x:: +--exclude-other:: + Only display entries with parent-match. + -w:: ---field-width=:: +--column-widths=<width[,width...]>:: Force each column width to the provided list, for large terminal readability. @@ -52,19 +73,26 @@ OPTIONS --field-separator=:: Use a special separator character and don't pad with spaces, replacing - all occurances of this separator in symbol names (and other output) + all occurrences of this separator in symbol names (and other output) with a '.' character, that thus it's the only non valid separator. +-D:: +--dump-raw-trace:: + Dump raw trace in ASCII. + -g [type,min]:: --call-graph:: - Display callchains using type and min percent threshold. + Display call chains using type and min percent threshold. type can be either: - - flat: single column, linear exposure of callchains. + - flat: single column, linear exposure of call chains. - graph: use a graph tree, displaying absolute overhead rates. - fractal: like graph, but displays relative rates. Each branch of the tree is considered as a new profiled object. + Default: fractal,0.5. +--pretty=<key>:: + Pretty printing style. key: normal, raw + --stdio:: Use the stdio interface. --tui:: Use the TUI interface, that is integrated with annotate and allows @@ -72,6 +100,25 @@ OPTIONS requires a tty, if one is not present, as when piping to other commands, the stdio interface is used. +-k:: +--vmlinux=<file>:: + vmlinux pathname + +--kallsyms=<file>:: + kallsyms pathname + +-m:: +--modules:: + Load module symbols. WARNING: This should only be used with -k and + a LIVE kernel. + +-f:: +--force:: + Don't complain, do it. + +--symfs=<directory>:: + Look for files with symbols relative to this directory. + SEE ALSO -------- linkperf:perf-stat[1] diff --git a/tools/perf/Documentation/perf-sched.txt b/tools/perf/Documentation/perf-sched.txt index 8417644a6166..46822d5fde1c 100644 --- a/tools/perf/Documentation/perf-sched.txt +++ b/tools/perf/Documentation/perf-sched.txt @@ -8,11 +8,11 @@ perf-sched - Tool to trace/measure scheduler properties (latencies) SYNOPSIS -------- [verse] -'perf sched' {record|latency|replay|trace} +'perf sched' {record|latency|map|replay|trace} DESCRIPTION ----------- -There are four variants of perf sched: +There are five variants of perf sched: 'perf sched record <command>' to record the scheduling events of an arbitrary workload. @@ -30,8 +30,22 @@ There are four variants of perf sched: of the workload as it occurred when it was recorded - and can repeat it a number of times, measuring its performance.) + 'perf sched map' to print a textual context-switching outline of + workload captured via perf sched record. Columns stand for + individual CPUs, and the two-letter shortcuts stand for tasks that + are running on a CPU. A '*' denotes the CPU that had the event, and + a dot signals an idle CPU. + OPTIONS ------- +-i:: +--input=<file>:: + Input file name. (default: perf.data) + +-v:: +--verbose:: + Be more verbose. (show symbol address, etc) + -D:: --dump-raw-trace=:: Display verbose dump of the sched data. diff --git a/tools/perf/Documentation/perf-trace-perl.txt b/tools/perf/Documentation/perf-script-perl.txt index ee6525ee6d69..5bb41e55a3ac 100644 --- a/tools/perf/Documentation/perf-trace-perl.txt +++ b/tools/perf/Documentation/perf-script-perl.txt @@ -1,19 +1,19 @@ -perf-trace-perl(1) +perf-script-perl(1) ================== NAME ---- -perf-trace-perl - Process trace data with a Perl script +perf-script-perl - Process trace data with a Perl script SYNOPSIS -------- [verse] -'perf trace' [-s [Perl]:script[.pl] ] +'perf script' [-s [Perl]:script[.pl] ] DESCRIPTION ----------- -This perf trace option is used to process perf trace data using perf's +This perf script option is used to process perf script data using perf's built-in Perl interpreter. It reads and processes the input file and displays the results of the trace analysis implemented in the given Perl script, if any. @@ -21,7 +21,7 @@ Perl script, if any. STARTER SCRIPTS --------------- -You can avoid reading the rest of this document by running 'perf trace +You can avoid reading the rest of this document by running 'perf script -g perl' in the same directory as an existing perf.data trace file. That will generate a starter script containing a handler for each of the event types in the trace file; it simply prints every available @@ -30,13 +30,13 @@ field for each event in the trace file. You can also look at the existing scripts in ~/libexec/perf-core/scripts/perl for typical examples showing how to do basic things like aggregate event data, print results, etc. Also, -the check-perf-trace.pl script, while not interesting for its results, +the check-perf-script.pl script, while not interesting for its results, attempts to exercise all of the main scripting features. EVENT HANDLERS -------------- -When perf trace is invoked using a trace script, a user-defined +When perf script is invoked using a trace script, a user-defined 'handler function' is called for each event in the trace. If there's no handler function defined for a given event type, the event is ignored (or passed to a 'trace_handled' function, see below) and the @@ -112,13 +112,13 @@ write a useful trace script. The sections below cover the rest. SCRIPT LAYOUT ------------- -Every perf trace Perl script should start by setting up a Perl module +Every perf script Perl script should start by setting up a Perl module search path and 'use'ing a few support modules (see module descriptions below): ---- - use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/Perf-Trace-Util/lib"; - use lib "./Perf-Trace-Util/lib"; + use lib "$ENV{'PERF_EXEC_PATH'}/scripts/perl/perf-script-Util/lib"; + use lib "./perf-script-Util/lib"; use Perf::Trace::Core; use Perf::Trace::Context; use Perf::Trace::Util; @@ -162,7 +162,7 @@ sub trace_unhandled ---- The remaining sections provide descriptions of each of the available -built-in perf trace Perl modules and their associated functions. +built-in perf script Perl modules and their associated functions. AVAILABLE MODULES AND FUNCTIONS ------------------------------- @@ -170,7 +170,7 @@ AVAILABLE MODULES AND FUNCTIONS The following sections describe the functions and variables available via the various Perf::Trace::* Perl modules. To use the functions and variables from the given module, add the corresponding 'use -Perf::Trace::XXX' line to your perf trace script. +Perf::Trace::XXX' line to your perf script script. Perf::Trace::Core Module ~~~~~~~~~~~~~~~~~~~~~~~~ @@ -204,7 +204,7 @@ argument. Perf::Trace::Util Module ~~~~~~~~~~~~~~~~~~~~~~~~ -Various utility functions for use with perf trace: +Various utility functions for use with perf script: nsecs($secs, $nsecs) - returns total nsecs given secs/nsecs pair nsecs_secs($nsecs) - returns whole secs portion given nsecs @@ -214,4 +214,4 @@ Various utility functions for use with perf trace: SEE ALSO -------- -linkperf:perf-trace[1] +linkperf:perf-script[1] diff --git a/tools/perf/Documentation/perf-trace-python.txt b/tools/perf/Documentation/perf-script-python.txt index 693be804dd3d..36b38277422c 100644 --- a/tools/perf/Documentation/perf-trace-python.txt +++ b/tools/perf/Documentation/perf-script-python.txt @@ -1,19 +1,19 @@ -perf-trace-python(1) +perf-script-python(1) ==================== NAME ---- -perf-trace-python - Process trace data with a Python script +perf-script-python - Process trace data with a Python script SYNOPSIS -------- [verse] -'perf trace' [-s [Python]:script[.py] ] +'perf script' [-s [Python]:script[.py] ] DESCRIPTION ----------- -This perf trace option is used to process perf trace data using perf's +This perf script option is used to process perf script data using perf's built-in Python interpreter. It reads and processes the input file and displays the results of the trace analysis implemented in the given Python script, if any. @@ -23,15 +23,15 @@ A QUICK EXAMPLE This section shows the process, start to finish, of creating a working Python script that aggregates and extracts useful information from a -raw perf trace stream. You can avoid reading the rest of this +raw perf script stream. You can avoid reading the rest of this document if an example is enough for you; the rest of the document provides more details on each step and lists the library functions available to script writers. This example actually details the steps that were used to create the -'syscall-counts' script you see when you list the available perf trace -scripts via 'perf trace -l'. As such, this script also shows how to -integrate your script into the list of general-purpose 'perf trace' +'syscall-counts' script you see when you list the available perf script +scripts via 'perf script -l'. As such, this script also shows how to +integrate your script into the list of general-purpose 'perf script' scripts listed by that command. The syscall-counts script is a simple script, but demonstrates all the @@ -105,31 +105,31 @@ That single stream will be recorded in a file in the current directory called perf.data. Once we have a perf.data file containing our data, we can use the -g -'perf trace' option to generate a Python script that will contain a +'perf script' option to generate a Python script that will contain a callback handler for each event type found in the perf.data trace stream (for more details, see the STARTER SCRIPTS section). ---- -# perf trace -g python -generated Python script: perf-trace.py +# perf script -g python +generated Python script: perf-script.py The output file created also in the current directory is named -perf-trace.py. Here's the file in its entirety: +perf-script.py. Here's the file in its entirety: -# perf trace event handlers, generated by perf trace -g python +# perf script event handlers, generated by perf script -g python # Licensed under the terms of the GNU GPL License version 2 # The common_* event handler fields are the most useful fields common to # all events. They don't necessarily correspond to the 'common_*' fields # in the format files. Those fields not available as handler params can # be retrieved using Python functions of the form common_*(context). -# See the perf-trace-python Documentation for the list of available functions. +# See the perf-script-python Documentation for the list of available functions. import os import sys sys.path.append(os.environ['PERF_EXEC_PATH'] + \ - '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + '/scripts/python/perf-script-Util/lib/Perf/Trace') from perf_trace_context import * from Core import * @@ -160,7 +160,7 @@ def print_header(event_name, cpu, secs, nsecs, pid, comm): ---- At the top is a comment block followed by some import statements and a -path append which every perf trace script should include. +path append which every perf script script should include. Following that are a couple generated functions, trace_begin() and trace_end(), which are called at the beginning and the end of the @@ -189,8 +189,8 @@ simply a utility function used for that purpose. Let's rename the script and run it to see the default output: ---- -# mv perf-trace.py syscall-counts.py -# perf trace -s syscall-counts.py +# mv perf-script.py syscall-counts.py +# perf script -s syscall-counts.py raw_syscalls__sys_enter 1 00840.847582083 7506 perf id=1, args= raw_syscalls__sys_enter 1 00840.847595764 7506 perf id=1, args= @@ -216,7 +216,7 @@ import os import sys sys.path.append(os.environ['PERF_EXEC_PATH'] + \ - '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + '/scripts/python/perf-script-Util/lib/Perf/Trace') from perf_trace_context import * from Core import * @@ -279,7 +279,7 @@ import os import sys sys.path.append(os.environ['PERF_EXEC_PATH'] + \ - '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + '/scripts/python/perf-script-Util/lib/Perf/Trace') from perf_trace_context import * from Core import * @@ -315,7 +315,7 @@ def print_syscall_totals(): The script can be run just as before: - # perf trace -s syscall-counts.py + # perf script -s syscall-counts.py So those are the essential steps in writing and running a script. The process can be generalized to any tracepoint or set of tracepoints @@ -324,17 +324,17 @@ interested in by looking at the list of available events shown by 'perf list' and/or look in /sys/kernel/debug/tracing events for detailed event and field info, record the corresponding trace data using 'perf record', passing it the list of interesting events, -generate a skeleton script using 'perf trace -g python' and modify the +generate a skeleton script using 'perf script -g python' and modify the code to aggregate and display it for your particular needs. After you've done that you may end up with a general-purpose script that you want to keep around and have available for future use. By writing a couple of very simple shell scripts and putting them in the right place, you can have your script listed alongside the other -scripts listed by the 'perf trace -l' command e.g.: +scripts listed by the 'perf script -l' command e.g.: ---- -root@tropicana:~# perf trace -l +root@tropicana:~# perf script -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency @@ -365,14 +365,14 @@ perf record -a -e raw_syscalls:sys_enter The 'report' script is also a shell script with the same base name as your script, but with -report appended. It should also be located in the perf/scripts/python/bin directory. In that script, you write the -'perf trace -s' command-line needed for running your script: +'perf script -s' command-line needed for running your script: ---- # cat kernel-source/tools/perf/scripts/python/bin/syscall-counts-report #!/bin/bash # description: system-wide syscall counts -perf trace -s ~/libexec/perf-core/scripts/python/syscall-counts.py +perf script -s ~/libexec/perf-core/scripts/python/syscall-counts.py ---- Note that the location of the Python script given in the shell script @@ -390,17 +390,17 @@ total 32 drwxr-xr-x 4 trz trz 4096 2010-01-26 22:30 . drwxr-xr-x 4 trz trz 4096 2010-01-26 22:29 .. drwxr-xr-x 2 trz trz 4096 2010-01-26 22:29 bin --rw-r--r-- 1 trz trz 2548 2010-01-26 22:29 check-perf-trace.py -drwxr-xr-x 3 trz trz 4096 2010-01-26 22:49 Perf-Trace-Util +-rw-r--r-- 1 trz trz 2548 2010-01-26 22:29 check-perf-script.py +drwxr-xr-x 3 trz trz 4096 2010-01-26 22:49 perf-script-Util -rw-r--r-- 1 trz trz 1462 2010-01-26 22:30 syscall-counts.py ---- Once you've done that (don't forget to do a new 'make install', -otherwise your script won't show up at run-time), 'perf trace -l' +otherwise your script won't show up at run-time), 'perf script -l' should show a new entry for your script: ---- -root@tropicana:~# perf trace -l +root@tropicana:~# perf script -l List of available trace scripts: workqueue-stats workqueue stats (ins/exe/create/destroy) wakeup-latency system-wide min/max/avg wakeup latency @@ -409,19 +409,19 @@ List of available trace scripts: syscall-counts system-wide syscall counts ---- -You can now perform the record step via 'perf trace record': +You can now perform the record step via 'perf script record': - # perf trace record syscall-counts + # perf script record syscall-counts -and display the output using 'perf trace report': +and display the output using 'perf script report': - # perf trace report syscall-counts + # perf script report syscall-counts STARTER SCRIPTS --------------- You can quickly get started writing a script for a particular set of -trace data by generating a skeleton script using 'perf trace -g +trace data by generating a skeleton script using 'perf script -g python' in the same directory as an existing perf.data trace file. That will generate a starter script containing a handler for each of the event types in the trace file; it simply prints every available @@ -430,13 +430,13 @@ field for each event in the trace file. You can also look at the existing scripts in ~/libexec/perf-core/scripts/python for typical examples showing how to do basic things like aggregate event data, print results, etc. Also, -the check-perf-trace.py script, while not interesting for its results, +the check-perf-script.py script, while not interesting for its results, attempts to exercise all of the main scripting features. EVENT HANDLERS -------------- -When perf trace is invoked using a trace script, a user-defined +When perf script is invoked using a trace script, a user-defined 'handler function' is called for each event in the trace. If there's no handler function defined for a given event type, the event is ignored (or passed to a 'trace_handled' function, see below) and the @@ -510,7 +510,7 @@ write a useful trace script. The sections below cover the rest. SCRIPT LAYOUT ------------- -Every perf trace Python script should start by setting up a Python +Every perf script Python script should start by setting up a Python module search path and 'import'ing a few support modules (see module descriptions below): @@ -519,7 +519,7 @@ descriptions below): import sys sys.path.append(os.environ['PERF_EXEC_PATH'] + \ - '/scripts/python/Perf-Trace-Util/lib/Perf/Trace') + '/scripts/python/perf-script-Util/lib/Perf/Trace') from perf_trace_context import * from Core import * @@ -559,15 +559,15 @@ def trace_unhandled(event_name, context, common_cpu, common_secs, ---- The remaining sections provide descriptions of each of the available -built-in perf trace Python modules and their associated functions. +built-in perf script Python modules and their associated functions. AVAILABLE MODULES AND FUNCTIONS ------------------------------- The following sections describe the functions and variables available -via the various perf trace Python modules. To use the functions and +via the various perf script Python modules. To use the functions and variables from the given module, add the corresponding 'from XXXX -import' line to your perf trace script. +import' line to your perf script script. Core.py Module ~~~~~~~~~~~~~~ @@ -610,7 +610,7 @@ argument. Util.py Module ~~~~~~~~~~~~~~ -Various utility functions for use with perf trace: +Various utility functions for use with perf script: nsecs(secs, nsecs) - returns total nsecs given secs/nsecs pair nsecs_secs(nsecs) - returns whole secs portion given nsecs @@ -620,4 +620,4 @@ Various utility functions for use with perf trace: SEE ALSO -------- -linkperf:perf-trace[1] +linkperf:perf-script[1] diff --git a/tools/perf/Documentation/perf-trace.txt b/tools/perf/Documentation/perf-script.txt index 26aff6bf9e50..29ad94293cd2 100644 --- a/tools/perf/Documentation/perf-trace.txt +++ b/tools/perf/Documentation/perf-script.txt @@ -1,71 +1,71 @@ -perf-trace(1) +perf-script(1) ============= NAME ---- -perf-trace - Read perf.data (created by perf record) and display trace output +perf-script - Read perf.data (created by perf record) and display trace output SYNOPSIS -------- [verse] -'perf trace' [<options>] -'perf trace' [<options>] record <script> [<record-options>] <command> -'perf trace' [<options>] report <script> [script-args] -'perf trace' [<options>] <script> <required-script-args> [<record-options>] <command> -'perf trace' [<options>] <top-script> [script-args] +'perf script' [<options>] +'perf script' [<options>] record <script> [<record-options>] <command> +'perf script' [<options>] report <script> [script-args] +'perf script' [<options>] <script> <required-script-args> [<record-options>] <command> +'perf script' [<options>] <top-script> [script-args] DESCRIPTION ----------- This command reads the input file and displays the trace recorded. -There are several variants of perf trace: +There are several variants of perf script: - 'perf trace' to see a detailed trace of the workload that was + 'perf script' to see a detailed trace of the workload that was recorded. You can also run a set of pre-canned scripts that aggregate and summarize the raw trace data in various ways (the list of scripts is - available via 'perf trace -l'). The following variants allow you to + available via 'perf script -l'). The following variants allow you to record and run those scripts: - 'perf trace record <script> <command>' to record the events required - for 'perf trace report'. <script> is the name displayed in the - output of 'perf trace --list' i.e. the actual script name minus any + 'perf script record <script> <command>' to record the events required + for 'perf script report'. <script> is the name displayed in the + output of 'perf script --list' i.e. the actual script name minus any language extension. If <command> is not specified, the events are recorded using the -a (system-wide) 'perf record' option. - 'perf trace report <script> [args]' to run and display the results + 'perf script report <script> [args]' to run and display the results of <script>. <script> is the name displayed in the output of 'perf trace --list' i.e. the actual script name minus any language - extension. The perf.data output from a previous run of 'perf trace + extension. The perf.data output from a previous run of 'perf script record <script>' is used and should be present for this command to succeed. [args] refers to the (mainly optional) args expected by the script. - 'perf trace <script> <required-script-args> <command>' to both + 'perf script <script> <required-script-args> <command>' to both record the events required for <script> and to run the <script> using 'live-mode' i.e. without writing anything to disk. <script> - is the name displayed in the output of 'perf trace --list' i.e. the + is the name displayed in the output of 'perf script --list' i.e. the actual script name minus any language extension. If <command> is not specified, the events are recorded using the -a (system-wide) 'perf record' option. If <script> has any required args, they should be specified before <command>. This mode doesn't allow for optional script args to be specified; if optional script args are - desired, they can be specified using separate 'perf trace record' - and 'perf trace report' commands, with the stdout of the record step + desired, they can be specified using separate 'perf script record' + and 'perf script report' commands, with the stdout of the record step piped to the stdin of the report script, using the '-o -' and '-i -' options of the corresponding commands. - 'perf trace <top-script>' to both record the events required for + 'perf script <top-script>' to both record the events required for <top-script> and to run the <top-script> using 'live-mode' i.e. without writing anything to disk. <top-script> is the name - displayed in the output of 'perf trace --list' i.e. the actual + displayed in the output of 'perf script --list' i.e. the actual script name minus any language extension; a <top-script> is defined as any script name ending with the string 'top'. - [<record-options>] can be passed to the record steps of 'perf trace + [<record-options>] can be passed to the record steps of 'perf script record' and 'live-mode' variants; this isn't possible however for - <top-script> 'live-mode' or 'perf trace report' variants. + <top-script> 'live-mode' or 'perf script report' variants. See the 'SEE ALSO' section for links to language-specific information on how to write and run your own trace scripts. @@ -76,7 +76,7 @@ OPTIONS Any command you can specify in a shell. -D:: ---dump-raw-trace=:: +--dump-raw-script=:: Display verbose dump of the trace data. -L:: @@ -95,7 +95,7 @@ OPTIONS -g:: --gen-script=:: - Generate perf-trace.[ext] starter script for given language, + Generate perf-script.[ext] starter script for given language, using current perf.data. -a:: @@ -104,8 +104,15 @@ OPTIONS normally don't - this option allows the latter to be run in system-wide mode. +-i:: +--input=:: + Input file name. + +-d:: +--debug-mode:: + Do various checks like samples ordering and lost events. SEE ALSO -------- -linkperf:perf-record[1], linkperf:perf-trace-perl[1], -linkperf:perf-trace-python[1] +linkperf:perf-record[1], linkperf:perf-script-perl[1], +linkperf:perf-script-python[1] diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt index 4b3a2d46b437..b6da7affbbee 100644 --- a/tools/perf/Documentation/perf-stat.txt +++ b/tools/perf/Documentation/perf-stat.txt @@ -8,8 +8,8 @@ perf-stat - Run a command and gather performance counter statistics SYNOPSIS -------- [verse] -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] <command> -'perf stat' [-e <EVENT> | --event=EVENT] [-S] [-a] -- <command> [<options>] +'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command> +'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>] DESCRIPTION ----------- @@ -35,24 +35,54 @@ OPTIONS child tasks do not inherit counters -p:: --pid=<pid>:: - stat events on existing pid + stat events on existing process id + +-t:: +--tid=<tid>:: + stat events on existing thread id + -a:: - system-wide collection +--all-cpus:: + system-wide collection from all CPUs -c:: - scale counter values +--scale:: + scale/normalize counter values + +-r:: +--repeat=<n>:: + repeat command and print average + stddev (max: 100) -B:: +--big-num:: print large numbers with thousands' separators according to locale -C:: --cpu=:: -Count only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Count only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. In per-thread mode, this option is ignored. The -a option is still necessary to activate system-wide monitoring. Default is to count on all CPUs. +-A:: +--no-aggr:: +Do not aggregate counts across all monitored CPUs in system-wide mode (-a). +This option is only valid in system-wide mode. + +-n:: +--null:: + null run - don't start any counters + +-v:: +--verbose:: + be more verbose (show counter open errors, etc) + +-x SEP:: +--field-separator SEP:: +print counts using a CSV-style output to make it easy to import directly into +spreadsheets. Columns are separated by the string specified in SEP. + EXAMPLES -------- diff --git a/tools/perf/Documentation/perf-test.txt b/tools/perf/Documentation/perf-test.txt index 1c4b5f5b7f71..2c3b462f64b0 100644 --- a/tools/perf/Documentation/perf-test.txt +++ b/tools/perf/Documentation/perf-test.txt @@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command does assorted sanity tests, initially thru linked routines but +This command does assorted sanity tests, initially through linked routines but also will look for a directory with more tests in the form of scripts. OPTIONS diff --git a/tools/perf/Documentation/perf-timechart.txt b/tools/perf/Documentation/perf-timechart.txt index 4b1788355eca..d7b79e2ba2ad 100644 --- a/tools/perf/Documentation/perf-timechart.txt +++ b/tools/perf/Documentation/perf-timechart.txt @@ -38,6 +38,8 @@ OPTIONS --process:: Select the processes to display, by name or PID +--symfs=<directory>:: + Look for files with symbols relative to this directory. SEE ALSO -------- diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt index 1f9687663f2a..f6eb1cdafb77 100644 --- a/tools/perf/Documentation/perf-top.txt +++ b/tools/perf/Documentation/perf-top.txt @@ -12,7 +12,7 @@ SYNOPSIS DESCRIPTION ----------- -This command generates and displays a performance counter profile in realtime. +This command generates and displays a performance counter profile in real time. OPTIONS @@ -27,8 +27,8 @@ OPTIONS -C <cpu-list>:: --cpu=<cpu>:: -Monitor only on the list of cpus provided. Multiple CPUs can be provided as a -comma-sperated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. +Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a +comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. Default is to monitor all CPUS. -d <seconds>:: @@ -50,6 +50,10 @@ Default is to monitor all CPUS. --count-filter=<count>:: Only display functions with more events than this. +-g:: +--group:: + Put the counters into a counter group. + -F <freq>:: --freq=<freq>:: Profile at this frequency. @@ -68,7 +72,11 @@ Default is to monitor all CPUS. -p <pid>:: --pid=<pid>:: - Profile events on existing pid. + Profile events on existing Process ID. + +-t <tid>:: +--tid=<tid>:: + Profile events on existing thread ID. -r <priority>:: --realtime=<priority>:: @@ -78,6 +86,18 @@ Default is to monitor all CPUS. --sym-annotate=<symbol>:: Annotate this symbol. +-K:: +--hide_kernel_symbols:: + Hide kernel symbols. + +-U:: +--hide_user_symbols:: + Hide user symbols. + +-D:: +--dump-symtab:: + Dump the symbol table used for profiling. + -v:: --verbose:: Be more verbose (show counter open errors, etc). diff --git a/tools/perf/MANIFEST b/tools/perf/MANIFEST index 8c7fc0c8f0b8..c12659d8cb26 100644 --- a/tools/perf/MANIFEST +++ b/tools/perf/MANIFEST @@ -7,6 +7,7 @@ include/linux/stringify.h lib/rbtree.c include/linux/swab.h arch/*/include/asm/unistd*.h +arch/*/lib/memcpy*.S include/linux/poison.h include/linux/magic.h include/linux/hw_breakpoint.h diff --git a/tools/perf/Makefile b/tools/perf/Makefile index d1db0f676a4b..1b9b13ee2a72 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -185,7 +185,10 @@ ifeq ($(ARCH),i386) ARCH := x86 endif ifeq ($(ARCH),x86_64) + RAW_ARCH := x86_64 ARCH := x86 + ARCH_CFLAGS := -DARCH_X86_64 + ARCH_INCLUDE = ../../arch/x86/lib/memcpy_64.S endif # CFLAGS and LDFLAGS are for the users to override from the command line. @@ -375,6 +378,7 @@ LIB_H += util/include/linux/prefetch.h LIB_H += util/include/linux/rbtree.h LIB_H += util/include/linux/string.h LIB_H += util/include/linux/types.h +LIB_H += util/include/linux/linkage.h LIB_H += util/include/asm/asm-offsets.h LIB_H += util/include/asm/bug.h LIB_H += util/include/asm/byteorder.h @@ -383,6 +387,8 @@ LIB_H += util/include/asm/swab.h LIB_H += util/include/asm/system.h LIB_H += util/include/asm/uaccess.h LIB_H += util/include/dwarf-regs.h +LIB_H += util/include/asm/dwarf2.h +LIB_H += util/include/asm/cpufeature.h LIB_H += perf.h LIB_H += util/cache.h LIB_H += util/callchain.h @@ -390,6 +396,7 @@ LIB_H += util/build-id.h LIB_H += util/debug.h LIB_H += util/debugfs.h LIB_H += util/event.h +LIB_H += util/evsel.h LIB_H += util/exec_cmd.h LIB_H += util/types.h LIB_H += util/levenshtein.h @@ -398,6 +405,7 @@ LIB_H += util/parse-options.h LIB_H += util/parse-events.h LIB_H += util/quote.h LIB_H += util/util.h +LIB_H += util/xyarray.h LIB_H += util/header.h LIB_H += util/help.h LIB_H += util/session.h @@ -417,6 +425,7 @@ LIB_H += util/probe-finder.h LIB_H += util/probe-event.h LIB_H += util/pstack.h LIB_H += util/cpumap.h +LIB_H += $(ARCH_INCLUDE) LIB_OBJS += $(OUTPUT)util/abspath.o LIB_OBJS += $(OUTPUT)util/alias.o @@ -426,6 +435,7 @@ LIB_OBJS += $(OUTPUT)util/ctype.o LIB_OBJS += $(OUTPUT)util/debugfs.o LIB_OBJS += $(OUTPUT)util/environment.o LIB_OBJS += $(OUTPUT)util/event.o +LIB_OBJS += $(OUTPUT)util/evsel.o LIB_OBJS += $(OUTPUT)util/exec_cmd.o LIB_OBJS += $(OUTPUT)util/help.o LIB_OBJS += $(OUTPUT)util/levenshtein.o @@ -463,6 +473,7 @@ LIB_OBJS += $(OUTPUT)util/sort.o LIB_OBJS += $(OUTPUT)util/hist.o LIB_OBJS += $(OUTPUT)util/probe-event.o LIB_OBJS += $(OUTPUT)util/util.o +LIB_OBJS += $(OUTPUT)util/xyarray.o LIB_OBJS += $(OUTPUT)util/cpumap.o BUILTIN_OBJS += $(OUTPUT)builtin-annotate.o @@ -472,6 +483,9 @@ BUILTIN_OBJS += $(OUTPUT)builtin-bench.o # Benchmark modules BUILTIN_OBJS += $(OUTPUT)bench/sched-messaging.o BUILTIN_OBJS += $(OUTPUT)bench/sched-pipe.o +ifeq ($(RAW_ARCH),x86_64) +BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy-x86-64-asm.o +endif BUILTIN_OBJS += $(OUTPUT)bench/mem-memcpy.o BUILTIN_OBJS += $(OUTPUT)builtin-diff.o @@ -485,7 +499,7 @@ BUILTIN_OBJS += $(OUTPUT)builtin-report.o BUILTIN_OBJS += $(OUTPUT)builtin-stat.o BUILTIN_OBJS += $(OUTPUT)builtin-timechart.o BUILTIN_OBJS += $(OUTPUT)builtin-top.o -BUILTIN_OBJS += $(OUTPUT)builtin-trace.o +BUILTIN_OBJS += $(OUTPUT)builtin-script.o BUILTIN_OBJS += $(OUTPUT)builtin-probe.o BUILTIN_OBJS += $(OUTPUT)builtin-kmem.o BUILTIN_OBJS += $(OUTPUT)builtin-lock.o @@ -507,7 +521,7 @@ PERFLIBS = $(LIB_FILE) -include config.mak ifndef NO_DWARF -FLAGS_DWARF=$(ALL_CFLAGS) -I/usr/include/elfutils -ldw -lelf $(ALL_LDFLAGS) $(EXTLIBS) +FLAGS_DWARF=$(ALL_CFLAGS) -ldw -lelf $(ALL_LDFLAGS) $(EXTLIBS) ifneq ($(call try-cc,$(SOURCE_DWARF),$(FLAGS_DWARF)),y) msg := $(warning No libdw.h found or old libdw.h found or elfutils is older than 0.138, disables dwarf support. Please install new elfutils-devel/libdw-dev); NO_DWARF := 1 @@ -554,7 +568,7 @@ ifndef NO_DWARF ifeq ($(origin PERF_HAVE_DWARF_REGS), undefined) msg := $(warning DWARF register mappings have not been defined for architecture $(ARCH), DWARF support disabled); else - BASIC_CFLAGS += -I/usr/include/elfutils -DDWARF_SUPPORT + BASIC_CFLAGS += -DDWARF_SUPPORT EXTLIBS += -lelf -ldw LIB_OBJS += $(OUTPUT)util/probe-finder.o endif # PERF_HAVE_DWARF_REGS @@ -891,13 +905,14 @@ prefix_SQ = $(subst ','\'',$(prefix)) SHELL_PATH_SQ = $(subst ','\'',$(SHELL_PATH)) PERL_PATH_SQ = $(subst ','\'',$(PERL_PATH)) -LIBS = $(PERFLIBS) $(EXTLIBS) +LIBS = -Wl,--whole-archive $(PERFLIBS) -Wl,--no-whole-archive $(EXTLIBS) BASIC_CFLAGS += -DSHA1_HEADER='$(SHA1_HEADER_SQ)' \ $(COMPAT_CFLAGS) LIB_OBJS += $(COMPAT_OBJS) ALL_CFLAGS += $(BASIC_CFLAGS) +ALL_CFLAGS += $(ARCH_CFLAGS) ALL_LDFLAGS += $(BASIC_LDFLAGS) export TAR INSTALL DESTDIR SHELL_PATH diff --git a/tools/perf/bench/mem-memcpy-arch.h b/tools/perf/bench/mem-memcpy-arch.h new file mode 100644 index 000000000000..a72e36cb5394 --- /dev/null +++ b/tools/perf/bench/mem-memcpy-arch.h @@ -0,0 +1,12 @@ + +#ifdef ARCH_X86_64 + +#define MEMCPY_FN(fn, name, desc) \ + extern void *fn(void *, const void *, size_t); + +#include "mem-memcpy-x86-64-asm-def.h" + +#undef MEMCPY_FN + +#endif + diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm-def.h b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h new file mode 100644 index 000000000000..d588b87696fc --- /dev/null +++ b/tools/perf/bench/mem-memcpy-x86-64-asm-def.h @@ -0,0 +1,4 @@ + +MEMCPY_FN(__memcpy, + "x86-64-unrolled", + "unrolled memcpy() in arch/x86/lib/memcpy_64.S") diff --git a/tools/perf/bench/mem-memcpy-x86-64-asm.S b/tools/perf/bench/mem-memcpy-x86-64-asm.S new file mode 100644 index 000000000000..a57b66e853c2 --- /dev/null +++ b/tools/perf/bench/mem-memcpy-x86-64-asm.S @@ -0,0 +1,2 @@ + +#include "../../../arch/x86/lib/memcpy_64.S" diff --git a/tools/perf/bench/mem-memcpy.c b/tools/perf/bench/mem-memcpy.c index 38dae7465142..db82021f4b91 100644 --- a/tools/perf/bench/mem-memcpy.c +++ b/tools/perf/bench/mem-memcpy.c @@ -12,6 +12,7 @@ #include "../util/parse-options.h" #include "../util/header.h" #include "bench.h" +#include "mem-memcpy-arch.h" #include <stdio.h> #include <stdlib.h> @@ -23,8 +24,10 @@ static const char *length_str = "1MB"; static const char *routine = "default"; -static bool use_clock = false; +static bool use_clock; static int clock_fd; +static bool only_prefault; +static bool no_prefault; static const struct option options[] = { OPT_STRING('l', "length", &length_str, "1MB", @@ -34,19 +37,33 @@ static const struct option options[] = { "Specify routine to copy"), OPT_BOOLEAN('c', "clock", &use_clock, "Use CPU clock for measuring"), + OPT_BOOLEAN('o', "only-prefault", &only_prefault, + "Show only the result with page faults before memcpy()"), + OPT_BOOLEAN('n', "no-prefault", &no_prefault, + "Show only the result without page faults before memcpy()"), OPT_END() }; +typedef void *(*memcpy_t)(void *, const void *, size_t); + struct routine { const char *name; const char *desc; - void * (*fn)(void *dst, const void *src, size_t len); + memcpy_t fn; }; struct routine routines[] = { { "default", "Default memcpy() provided by glibc", memcpy }, +#ifdef ARCH_X86_64 + +#define MEMCPY_FN(fn, name, desc) { name, desc, fn }, +#include "mem-memcpy-x86-64-asm-def.h" +#undef MEMCPY_FN + +#endif + { NULL, NULL, NULL } @@ -89,29 +106,98 @@ static double timeval2double(struct timeval *ts) (double)ts->tv_usec / (double)1000000; } +static void alloc_mem(void **dst, void **src, size_t length) +{ + *dst = zalloc(length); + if (!dst) + die("memory allocation failed - maybe length is too large?\n"); + + *src = zalloc(length); + if (!src) + die("memory allocation failed - maybe length is too large?\n"); +} + +static u64 do_memcpy_clock(memcpy_t fn, size_t len, bool prefault) +{ + u64 clock_start = 0ULL, clock_end = 0ULL; + void *src = NULL, *dst = NULL; + + alloc_mem(&src, &dst, len); + + if (prefault) + fn(dst, src, len); + + clock_start = get_clock(); + fn(dst, src, len); + clock_end = get_clock(); + + free(src); + free(dst); + return clock_end - clock_start; +} + +static double do_memcpy_gettimeofday(memcpy_t fn, size_t len, bool prefault) +{ + struct timeval tv_start, tv_end, tv_diff; + void *src = NULL, *dst = NULL; + + alloc_mem(&src, &dst, len); + + if (prefault) + fn(dst, src, len); + + BUG_ON(gettimeofday(&tv_start, NULL)); + fn(dst, src, len); + BUG_ON(gettimeofday(&tv_end, NULL)); + + timersub(&tv_end, &tv_start, &tv_diff); + + free(src); + free(dst); + return (double)((double)len / timeval2double(&tv_diff)); +} + +#define pf (no_prefault ? 0 : 1) + +#define print_bps(x) do { \ + if (x < K) \ + printf(" %14lf B/Sec", x); \ + else if (x < K * K) \ + printf(" %14lfd KB/Sec", x / K); \ + else if (x < K * K * K) \ + printf(" %14lf MB/Sec", x / K / K); \ + else \ + printf(" %14lf GB/Sec", x / K / K / K); \ + } while (0) + int bench_mem_memcpy(int argc, const char **argv, const char *prefix __used) { int i; - void *dst, *src; - size_t length; - double bps = 0.0; - struct timeval tv_start, tv_end, tv_diff; - u64 clock_start, clock_end, clock_diff; + size_t len; + double result_bps[2]; + u64 result_clock[2]; - clock_start = clock_end = clock_diff = 0ULL; argc = parse_options(argc, argv, options, bench_mem_memcpy_usage, 0); - tv_diff.tv_sec = 0; - tv_diff.tv_usec = 0; - length = (size_t)perf_atoll((char *)length_str); + if (use_clock) + init_clock(); + + len = (size_t)perf_atoll((char *)length_str); - if ((s64)length <= 0) { + result_clock[0] = result_clock[1] = 0ULL; + result_bps[0] = result_bps[1] = 0.0; + + if ((s64)len <= 0) { fprintf(stderr, "Invalid length:%s\n", length_str); return 1; } + /* same to without specifying either of prefault and no-prefault */ + if (only_prefault && no_prefault) + only_prefault = no_prefault = false; + for (i = 0; routines[i].name; i++) { if (!strcmp(routines[i].name, routine)) break; @@ -126,61 +212,80 @@ int bench_mem_memcpy(int argc, const char **argv, return 1; } - dst = zalloc(length); - if (!dst) - die("memory allocation failed - maybe length is too large?\n"); - - src = zalloc(length); - if (!src) - die("memory allocation failed - maybe length is too large?\n"); - - if (bench_format == BENCH_FORMAT_DEFAULT) { - printf("# Copying %s Bytes from %p to %p ...\n\n", - length_str, src, dst); - } - - if (use_clock) { - init_clock(); - clock_start = get_clock(); - } else { - BUG_ON(gettimeofday(&tv_start, NULL)); - } - - routines[i].fn(dst, src, length); + if (bench_format == BENCH_FORMAT_DEFAULT) + printf("# Copying %s Bytes ...\n\n", length_str); - if (use_clock) { - clock_end = get_clock(); - clock_diff = clock_end - clock_start; + if (!only_prefault && !no_prefault) { + /* show both of results */ + if (use_clock) { + result_clock[0] = + do_memcpy_clock(routines[i].fn, len, false); + result_clock[1] = + do_memcpy_clock(routines[i].fn, len, true); + } else { + result_bps[0] = + do_memcpy_gettimeofday(routines[i].fn, + len, false); + result_bps[1] = + do_memcpy_gettimeofday(routines[i].fn, + len, true); + } } else { - BUG_ON(gettimeofday(&tv_end, NULL)); - timersub(&tv_end, &tv_start, &tv_diff); - bps = (double)((double)length / timeval2double(&tv_diff)); + if (use_clock) { + result_clock[pf] = + do_memcpy_clock(routines[i].fn, + len, only_prefault); + } else { + result_bps[pf] = + do_memcpy_gettimeofday(routines[i].fn, + len, only_prefault); + } } switch (bench_format) { case BENCH_FORMAT_DEFAULT: - if (use_clock) { - printf(" %14lf Clock/Byte\n", - (double)clock_diff / (double)length); - } else { - if (bps < K) - printf(" %14lf B/Sec\n", bps); - else if (bps < K * K) - printf(" %14lfd KB/Sec\n", bps / 1024); - else if (bps < K * K * K) - printf(" %14lf MB/Sec\n", bps / 1024 / 1024); - else { - printf(" %14lf GB/Sec\n", - bps / 1024 / 1024 / 1024); + if (!only_prefault && !no_prefault) { + if (use_clock) { + printf(" %14lf Clock/Byte\n", + (double)result_clock[0] + / (double)len); + printf(" %14lf Clock/Byte (with prefault)\n", + (double)result_clock[1] + / (double)len); + } else { + print_bps(result_bps[0]); + printf("\n"); + print_bps(result_bps[1]); + printf(" (with prefault)\n"); } + } else { + if (use_clock) { + printf(" %14lf Clock/Byte", + (double)result_clock[pf] + / (double)len); + } else + print_bps(result_bps[pf]); + + printf("%s\n", only_prefault ? " (with prefault)" : ""); } break; case BENCH_FORMAT_SIMPLE: - if (use_clock) { - printf("%14lf\n", - (double)clock_diff / (double)length); - } else - printf("%lf\n", bps); + if (!only_prefault && !no_prefault) { + if (use_clock) { + printf("%lf %lf\n", + (double)result_clock[0] / (double)len, + (double)result_clock[1] / (double)len); + } else { + printf("%lf %lf\n", + result_bps[0], result_bps[1]); + } + } else { + if (use_clock) { + printf("%lf\n", (double)result_clock[pf] + / (double)len); + } else + printf("%lf\n", result_bps[pf]); + } break; default: /* reaching this means there's some disaster: */ diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 6d5604d8df95..c056cdc06912 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -58,12 +58,12 @@ static int hists__add_entry(struct hists *self, struct addr_location *al) return hist_entry__inc_addr_samples(he, al->addr); } -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event, struct sample_data *sample, + struct perf_session *session) { struct addr_location al; - struct sample_data data; - if (event__preprocess_sample(event, session, &al, &data, NULL) < 0) { + if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) { pr_warning("problem processing %d event, skipping it.\n", event->header.type); return -1; @@ -375,6 +375,8 @@ static struct perf_event_ops event_ops = { .mmap = event__process_mmap, .comm = event__process_comm, .fork = event__process_task, + .ordered_samples = true, + .ordering_requires_timestamps = true, }; static int __cmd_annotate(void) @@ -382,7 +384,7 @@ static int __cmd_annotate(void) int ret; struct perf_session *session; - session = perf_session__new(input_name, O_RDONLY, force, false); + session = perf_session__new(input_name, O_RDONLY, force, false, &event_ops); if (session == NULL) return -ENOMEM; diff --git a/tools/perf/builtin-buildid-list.c b/tools/perf/builtin-buildid-list.c index c49837de7d3f..5af32ae9031e 100644 --- a/tools/perf/builtin-buildid-list.c +++ b/tools/perf/builtin-buildid-list.c @@ -38,7 +38,8 @@ static int __cmd_buildid_list(void) { struct perf_session *session; - session = perf_session__new(input_name, O_RDONLY, force, false); + session = perf_session__new(input_name, O_RDONLY, force, false, + &build_id__mark_dso_hit_ops); if (session == NULL) return -1; diff --git a/tools/perf/builtin-diff.c b/tools/perf/builtin-diff.c index fca1d4402910..3153e492dbcc 100644 --- a/tools/perf/builtin-diff.c +++ b/tools/perf/builtin-diff.c @@ -30,12 +30,13 @@ static int hists__add_entry(struct hists *self, return -ENOMEM; } -static int diff__process_sample_event(event_t *event, struct perf_session *session) +static int diff__process_sample_event(event_t *event, + struct sample_data *sample, + struct perf_session *session) { struct addr_location al; - struct sample_data data = { .period = 1, }; - if (event__preprocess_sample(event, session, &al, &data, NULL) < 0) { + if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) { pr_warning("problem processing %d event, skipping it.\n", event->header.type); return -1; @@ -44,12 +45,12 @@ static int diff__process_sample_event(event_t *event, struct perf_session *sessi if (al.filtered || al.sym == NULL) return 0; - if (hists__add_entry(&session->hists, &al, data.period)) { + if (hists__add_entry(&session->hists, &al, sample->period)) { pr_warning("problem incrementing symbol period, skipping event\n"); return -1; } - session->hists.stats.total_period += data.period; + session->hists.stats.total_period += sample->period; return 0; } @@ -60,6 +61,8 @@ static struct perf_event_ops event_ops = { .exit = event__process_task, .fork = event__process_task, .lost = event__process_lost, + .ordered_samples = true, + .ordering_requires_timestamps = true, }; static void perf_session__insert_hist_entry_by_name(struct rb_root *root, @@ -141,8 +144,8 @@ static int __cmd_diff(void) int ret, i; struct perf_session *session[2]; - session[0] = perf_session__new(input_old, O_RDONLY, force, false); - session[1] = perf_session__new(input_new, O_RDONLY, force, false); + session[0] = perf_session__new(input_old, O_RDONLY, force, false, &event_ops); + session[1] = perf_session__new(input_new, O_RDONLY, force, false, &event_ops); if (session[0] == NULL || session[1] == NULL) return -ENOMEM; @@ -173,7 +176,7 @@ static const char * const diff_usage[] = { static const struct option options[] = { OPT_INCR('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), - OPT_BOOLEAN('m', "displacement", &show_displacement, + OPT_BOOLEAN('M', "displacement", &show_displacement, "Show position displacement relative to baseline"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), @@ -191,6 +194,8 @@ static const struct option options[] = { OPT_STRING('t', "field-separator", &symbol_conf.field_sep, "separator", "separator for columns, no spaces will be added between " "columns '.' is reserved."), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), OPT_END() }; diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c index 8e3e47b064ce..0c78ffa7bf67 100644 --- a/tools/perf/builtin-inject.c +++ b/tools/perf/builtin-inject.c @@ -16,8 +16,8 @@ static char const *input_name = "-"; static bool inject_build_ids; -static int event__repipe(event_t *event __used, - struct perf_session *session __used) +static int event__repipe_synth(event_t *event, + struct perf_session *session __used) { uint32_t size; void *buf = event; @@ -36,22 +36,30 @@ static int event__repipe(event_t *event __used, return 0; } -static int event__repipe_mmap(event_t *self, struct perf_session *session) +static int event__repipe(event_t *event, struct sample_data *sample __used, + struct perf_session *session) +{ + return event__repipe_synth(event, session); +} + +static int event__repipe_mmap(event_t *self, struct sample_data *sample, + struct perf_session *session) { int err; - err = event__process_mmap(self, session); - event__repipe(self, session); + err = event__process_mmap(self, sample, session); + event__repipe(self, sample, session); return err; } -static int event__repipe_task(event_t *self, struct perf_session *session) +static int event__repipe_task(event_t *self, struct sample_data *sample, + struct perf_session *session) { int err; - err = event__process_task(self, session); - event__repipe(self, session); + err = event__process_task(self, sample, session); + event__repipe(self, sample, session); return err; } @@ -61,7 +69,7 @@ static int event__repipe_tracing_data(event_t *self, { int err; - event__repipe(self, session); + event__repipe_synth(self, session); err = event__process_tracing_data(self, session); return err; @@ -111,7 +119,8 @@ static int dso__inject_build_id(struct dso *self, struct perf_session *session) return 0; } -static int event__inject_buildid(event_t *event, struct perf_session *session) +static int event__inject_buildid(event_t *event, struct sample_data *sample, + struct perf_session *session) { struct addr_location al; struct thread *thread; @@ -146,7 +155,7 @@ static int event__inject_buildid(event_t *event, struct perf_session *session) } repipe: - event__repipe(event, session); + event__repipe(event, sample, session); return 0; } @@ -160,10 +169,10 @@ struct perf_event_ops inject_ops = { .read = event__repipe, .throttle = event__repipe, .unthrottle = event__repipe, - .attr = event__repipe, - .event_type = event__repipe, - .tracing_data = event__repipe, - .build_id = event__repipe, + .attr = event__repipe_synth, + .event_type = event__repipe_synth, + .tracing_data = event__repipe_synth, + .build_id = event__repipe_synth, }; extern volatile int session_done; @@ -187,7 +196,7 @@ static int __cmd_inject(void) inject_ops.tracing_data = event__repipe_tracing_data; } - session = perf_session__new(input_name, O_RDONLY, false, true); + session = perf_session__new(input_name, O_RDONLY, false, true, &inject_ops); if (session == NULL) return -ENOMEM; diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index 31f60a2535e0..def7ddc2fd4f 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -304,22 +304,11 @@ process_raw_event(event_t *raw_event __used, void *data, } } -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event, struct sample_data *sample, + struct perf_session *session) { - struct sample_data data; - struct thread *thread; + struct thread *thread = perf_session__findnew(session, event->ip.pid); - memset(&data, 0, sizeof(data)); - data.time = -1; - data.cpu = -1; - data.period = 1; - - event__parse_sample(event, session->sample_type, &data); - - dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc, - data.pid, data.tid, data.ip, data.period); - - thread = perf_session__findnew(session, event->ip.pid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", event->header.type); @@ -328,8 +317,8 @@ static int process_sample_event(event_t *event, struct perf_session *session) dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); - process_raw_event(event, data.raw_data, data.cpu, - data.time, thread); + process_raw_event(event, sample->raw_data, sample->cpu, + sample->time, thread); return 0; } @@ -492,7 +481,8 @@ static void sort_result(void) static int __cmd_kmem(void) { int err = -EINVAL; - struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0, false); + struct perf_session *session = perf_session__new(input_name, O_RDONLY, + 0, false, &event_ops); if (session == NULL) return -ENOMEM; @@ -747,6 +737,9 @@ static int __cmd_record(int argc, const char **argv) rec_argc = ARRAY_SIZE(record_args) + argc - 1; rec_argv = calloc(rec_argc + 1, sizeof(char *)); + if (rec_argv == NULL) + return -ENOMEM; + for (i = 0; i < ARRAY_SIZE(record_args); i++) rec_argv[i] = strdup(record_args[i]); diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c index 821c1586a22b..b9c6e5432971 100644 --- a/tools/perf/builtin-lock.c +++ b/tools/perf/builtin-lock.c @@ -834,22 +834,18 @@ static void dump_info(void) die("Unknown type of information\n"); } -static int process_sample_event(event_t *self, struct perf_session *s) +static int process_sample_event(event_t *self, struct sample_data *sample, + struct perf_session *s) { - struct sample_data data; - struct thread *thread; + struct thread *thread = perf_session__findnew(s, sample->tid); - bzero(&data, sizeof(data)); - event__parse_sample(self, s->sample_type, &data); - - thread = perf_session__findnew(s, data.tid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", self->header.type); return -1; } - process_raw_event(data.raw_data, data.cpu, data.time, thread); + process_raw_event(sample->raw_data, sample->cpu, sample->time, thread); return 0; } @@ -862,7 +858,7 @@ static struct perf_event_ops eops = { static int read_events(void) { - session = perf_session__new(input_name, O_RDONLY, 0, false); + session = perf_session__new(input_name, O_RDONLY, 0, false, &eops); if (!session) die("Initializing perf session failed\n"); @@ -947,6 +943,9 @@ static int __cmd_record(int argc, const char **argv) rec_argc = ARRAY_SIZE(record_args) + argc - 1; rec_argv = calloc(rec_argc + 1, sizeof(char *)); + if (rec_argv == NULL) + return -ENOMEM; + for (i = 0; i < ARRAY_SIZE(record_args); i++) rec_argv[i] = strdup(record_args[i]); @@ -982,9 +981,9 @@ int cmd_lock(int argc, const char **argv, const char *prefix __used) usage_with_options(report_usage, report_options); } __cmd_report(); - } else if (!strcmp(argv[0], "trace")) { - /* Aliased to 'perf trace' */ - return cmd_trace(argc, argv, prefix); + } else if (!strcmp(argv[0], "script")) { + /* Aliased to 'perf script' */ + return cmd_script(argc, argv, prefix); } else if (!strcmp(argv[0], "info")) { if (argc) { argc = parse_options(argc, argv, diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 564491fa18b2..7bc049035484 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -18,6 +18,7 @@ #include "util/header.h" #include "util/event.h" +#include "util/evsel.h" #include "util/debug.h" #include "util/session.h" #include "util/symbol.h" @@ -27,17 +28,18 @@ #include <sched.h> #include <sys/mman.h> +#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y)) + enum write_mode_t { WRITE_FORCE, WRITE_APPEND }; -static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; - static u64 user_interval = ULLONG_MAX; static u64 default_interval = 0; +static u64 sample_type; -static int nr_cpus = 0; +static struct cpu_map *cpus; static unsigned int page_size; static unsigned int mmap_pages = 128; static unsigned int user_freq = UINT_MAX; @@ -48,11 +50,11 @@ static const char *output_name = "perf.data"; static int group = 0; static int realtime_prio = 0; static bool raw_samples = false; +static bool sample_id_all_avail = true; static bool system_wide = false; static pid_t target_pid = -1; static pid_t target_tid = -1; -static pid_t *all_tids = NULL; -static int thread_num = 0; +static struct thread_map *threads; static pid_t child_pid = -1; static bool no_inherit = false; static enum write_mode_t write_mode = WRITE_FORCE; @@ -60,7 +62,9 @@ static bool call_graph = false; static bool inherit_stat = false; static bool no_samples = false; static bool sample_address = false; +static bool sample_time = false; static bool no_buildid = false; +static bool no_buildid_cache = false; static long samples = 0; static u64 bytes_written = 0; @@ -77,7 +81,6 @@ static struct perf_session *session; static const char *cpu_list; struct mmap_data { - int counter; void *base; unsigned int mask; unsigned int prev; @@ -128,6 +131,7 @@ static void write_output(void *buf, size_t size) } static int process_synthesized_event(event_t *event, + struct sample_data *sample __used, struct perf_session *self __used) { write_output(event, event->header.size); @@ -224,12 +228,12 @@ static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int n return h_attr; } -static void create_counter(int counter, int cpu) +static void create_counter(struct perf_evsel *evsel, int cpu) { - char *filter = filters[counter]; - struct perf_event_attr *attr = attrs + counter; + char *filter = evsel->filter; + struct perf_event_attr *attr = &evsel->attr; struct perf_header_attr *h_attr; - int track = !counter; /* only the first counter needs these */ + int track = !evsel->idx; /* only the first counter needs these */ int thread_index; int ret; struct { @@ -238,6 +242,19 @@ static void create_counter(int counter, int cpu) u64 time_running; u64 id; } read_data; + /* + * Check if parse_single_tracepoint_event has already asked for + * PERF_SAMPLE_TIME. + * + * XXX this is kludgy but short term fix for problems introduced by + * eac23d1c that broke 'perf script' by having different sample_types + * when using multiple tracepoint events when we use a perf binary + * that tries to use sample_id_all on an older kernel. + * + * We need to move counter creation to perf_session, support + * different sample_types, etc. + */ + bool time_needed = attr->sample_type & PERF_SAMPLE_TIME; attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED | PERF_FORMAT_TOTAL_TIME_RUNNING | @@ -280,6 +297,10 @@ static void create_counter(int counter, int cpu) if (system_wide) attr->sample_type |= PERF_SAMPLE_CPU; + if (sample_id_all_avail && + (sample_time || system_wide || !no_inherit || cpu_list)) + attr->sample_type |= PERF_SAMPLE_TIME; + if (raw_samples) { attr->sample_type |= PERF_SAMPLE_TIME; attr->sample_type |= PERF_SAMPLE_RAW; @@ -293,13 +314,14 @@ static void create_counter(int counter, int cpu) attr->disabled = 1; attr->enable_on_exec = 1; } +retry_sample_id: + attr->sample_id_all = sample_id_all_avail ? 1 : 0; - for (thread_index = 0; thread_index < thread_num; thread_index++) { + for (thread_index = 0; thread_index < threads->nr; thread_index++) { try_again: - fd[nr_cpu][counter][thread_index] = sys_perf_event_open(attr, - all_tids[thread_index], cpu, group_fd, 0); + FD(evsel, nr_cpu, thread_index) = sys_perf_event_open(attr, threads->map[thread_index], cpu, group_fd, 0); - if (fd[nr_cpu][counter][thread_index] < 0) { + if (FD(evsel, nr_cpu, thread_index) < 0) { int err = errno; if (err == EPERM || err == EACCES) @@ -309,6 +331,15 @@ try_again: else if (err == ENODEV && cpu_list) { die("No such device - did you specify" " an out-of-range profile CPU?\n"); + } else if (err == EINVAL && sample_id_all_avail) { + /* + * Old kernel, no attr->sample_id_type_all field + */ + sample_id_all_avail = false; + if (!sample_time && !raw_samples && !time_needed) + attr->sample_type &= ~PERF_SAMPLE_TIME; + + goto retry_sample_id; } /* @@ -326,8 +357,8 @@ try_again: goto try_again; } printf("\n"); - error("perfcounter syscall returned with %d (%s)\n", - fd[nr_cpu][counter][thread_index], strerror(err)); + error("sys_perf_event_open() syscall returned with %d (%s). /bin/dmesg may provide additional information.\n", + FD(evsel, nr_cpu, thread_index), strerror(err)); #if defined(__i386__) || defined(__x86_64__) if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP) @@ -341,7 +372,7 @@ try_again: exit(-1); } - h_attr = get_header_attr(attr, counter); + h_attr = get_header_attr(attr, evsel->idx); if (h_attr == NULL) die("nomem\n"); @@ -352,7 +383,7 @@ try_again: } } - if (read(fd[nr_cpu][counter][thread_index], &read_data, sizeof(read_data)) == -1) { + if (read(FD(evsel, nr_cpu, thread_index), &read_data, sizeof(read_data)) == -1) { perror("Unable to read perf file descriptor"); exit(-1); } @@ -362,43 +393,44 @@ try_again: exit(-1); } - assert(fd[nr_cpu][counter][thread_index] >= 0); - fcntl(fd[nr_cpu][counter][thread_index], F_SETFL, O_NONBLOCK); + assert(FD(evsel, nr_cpu, thread_index) >= 0); + fcntl(FD(evsel, nr_cpu, thread_index), F_SETFL, O_NONBLOCK); /* * First counter acts as the group leader: */ if (group && group_fd == -1) - group_fd = fd[nr_cpu][counter][thread_index]; - - if (counter || thread_index) { - ret = ioctl(fd[nr_cpu][counter][thread_index], - PERF_EVENT_IOC_SET_OUTPUT, - fd[nr_cpu][0][0]); + group_fd = FD(evsel, nr_cpu, thread_index); + + if (evsel->idx || thread_index) { + struct perf_evsel *first; + first = list_entry(evsel_list.next, struct perf_evsel, node); + ret = ioctl(FD(evsel, nr_cpu, thread_index), + PERF_EVENT_IOC_SET_OUTPUT, + FD(first, nr_cpu, 0)); if (ret) { error("failed to set output: %d (%s)\n", errno, strerror(errno)); exit(-1); } } else { - mmap_array[nr_cpu].counter = counter; mmap_array[nr_cpu].prev = 0; mmap_array[nr_cpu].mask = mmap_pages*page_size - 1; mmap_array[nr_cpu].base = mmap(NULL, (mmap_pages+1)*page_size, - PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter][thread_index], 0); + PROT_READ | PROT_WRITE, MAP_SHARED, FD(evsel, nr_cpu, thread_index), 0); if (mmap_array[nr_cpu].base == MAP_FAILED) { error("failed to mmap with %d (%s)\n", errno, strerror(errno)); exit(-1); } - event_array[nr_poll].fd = fd[nr_cpu][counter][thread_index]; + event_array[nr_poll].fd = FD(evsel, nr_cpu, thread_index); event_array[nr_poll].events = POLLIN; nr_poll++; } if (filter != NULL) { - ret = ioctl(fd[nr_cpu][counter][thread_index], - PERF_EVENT_IOC_SET_FILTER, filter); + ret = ioctl(FD(evsel, nr_cpu, thread_index), + PERF_EVENT_IOC_SET_FILTER, filter); if (ret) { error("failed to set filter with %d (%s)\n", errno, strerror(errno)); @@ -406,15 +438,19 @@ try_again: } } } + + if (!sample_type) + sample_type = attr->sample_type; } static void open_counters(int cpu) { - int counter; + struct perf_evsel *pos; group_fd = -1; - for (counter = 0; counter < nr_counters; counter++) - create_counter(counter, cpu); + + list_for_each_entry(pos, &evsel_list, node) + create_counter(pos, cpu); nr_cpu++; } @@ -437,7 +473,8 @@ static void atexit_header(void) if (!pipe_output) { session->header.data_size += bytes_written; - process_buildids(); + if (!no_buildid) + process_buildids(); perf_header__write(&session->header, output, true); perf_session__delete(session); symbol__exit(); @@ -500,7 +537,7 @@ static void mmap_read_all(void) static int __cmd_record(int argc, const char **argv) { - int i, counter; + int i; struct stat st; int flags; int err; @@ -552,19 +589,22 @@ static int __cmd_record(int argc, const char **argv) } session = perf_session__new(output_name, O_WRONLY, - write_mode == WRITE_FORCE, false); + write_mode == WRITE_FORCE, false, NULL); if (session == NULL) { pr_err("Not enough memory for reading perf file header\n"); return -1; } + if (!no_buildid) + perf_header__set_feat(&session->header, HEADER_BUILD_ID); + if (!file_new) { err = perf_header__read(session, output); if (err < 0) goto out_delete_session; } - if (have_tracepoints(attrs, nr_counters)) + if (have_tracepoints(&evsel_list)) perf_header__set_feat(&session->header, HEADER_TRACE_INFO); /* @@ -612,7 +652,7 @@ static int __cmd_record(int argc, const char **argv) } if (!system_wide && target_tid == -1 && target_pid == -1) - all_tids[0] = child_pid; + threads->map[0] = child_pid; close(child_ready_pipe[1]); close(go_pipe[0]); @@ -626,19 +666,15 @@ static int __cmd_record(int argc, const char **argv) close(child_ready_pipe[0]); } - nr_cpus = read_cpu_map(cpu_list); - if (nr_cpus < 1) { - perror("failed to collect number of CPUs"); - return -1; - } - if (!system_wide && no_inherit && !cpu_list) { open_counters(-1); } else { - for (i = 0; i < nr_cpus; i++) - open_counters(cpumap[i]); + for (i = 0; i < cpus->nr; i++) + open_counters(cpus->map[i]); } + perf_session__set_sample_type(session, sample_type); + if (pipe_output) { err = perf_header__write_pipe(output); if (err < 0) @@ -651,6 +687,8 @@ static int __cmd_record(int argc, const char **argv) post_processing_offset = lseek(output, 0, SEEK_CUR); + perf_session__set_sample_id_all(session, sample_id_all_avail); + if (pipe_output) { err = event__synthesize_attrs(&session->header, process_synthesized_event, @@ -667,7 +705,7 @@ static int __cmd_record(int argc, const char **argv) return err; } - if (have_tracepoints(attrs, nr_counters)) { + if (have_tracepoints(&evsel_list)) { /* * FIXME err <= 0 here actually means that * there were no tracepoints so its not really @@ -676,8 +714,7 @@ static int __cmd_record(int argc, const char **argv) * return this more properly and also * propagate errors that now are calling die() */ - err = event__synthesize_tracing_data(output, attrs, - nr_counters, + err = event__synthesize_tracing_data(output, &evsel_list, process_synthesized_event, session); if (err <= 0) { @@ -751,13 +788,13 @@ static int __cmd_record(int argc, const char **argv) if (done) { for (i = 0; i < nr_cpu; i++) { - for (counter = 0; - counter < nr_counters; - counter++) { + struct perf_evsel *pos; + + list_for_each_entry(pos, &evsel_list, node) { for (thread = 0; - thread < thread_num; + thread < threads->nr; thread++) - ioctl(fd[i][counter][thread], + ioctl(FD(pos, i, thread), PERF_EVENT_IOC_DISABLE); } } @@ -831,16 +868,20 @@ const struct option record_options[] = { "per thread counts"), OPT_BOOLEAN('d', "data", &sample_address, "Sample addresses"), + OPT_BOOLEAN('T', "timestamp", &sample_time, "Sample timestamps"), OPT_BOOLEAN('n', "no-samples", &no_samples, "don't sample"), - OPT_BOOLEAN('N', "no-buildid-cache", &no_buildid, + OPT_BOOLEAN('N', "no-buildid-cache", &no_buildid_cache, "do not update the buildid cache"), + OPT_BOOLEAN('B', "no-buildid", &no_buildid, + "do not collect buildids in perf.data"), OPT_END() }; int cmd_record(int argc, const char **argv, const char *prefix __used) { - int i, j, err = -ENOMEM; + int err = -ENOMEM; + struct perf_evsel *pos; argc = parse_options(argc, argv, record_options, record_usage, PARSE_OPT_STOP_AT_NON_OPTION); @@ -859,41 +900,36 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) } symbol__init(); - if (no_buildid) + + if (no_buildid_cache || no_buildid) disable_buildid_cache(); - if (!nr_counters) { - nr_counters = 1; - attrs[0].type = PERF_TYPE_HARDWARE; - attrs[0].config = PERF_COUNT_HW_CPU_CYCLES; + if (list_empty(&evsel_list) && perf_evsel_list__create_default() < 0) { + pr_err("Not enough memory for event selector list\n"); + goto out_symbol_exit; } - if (target_pid != -1) { + if (target_pid != -1) target_tid = target_pid; - thread_num = find_all_tid(target_pid, &all_tids); - if (thread_num <= 0) { - fprintf(stderr, "Can't find all threads of pid %d\n", - target_pid); - usage_with_options(record_usage, record_options); - } - } else { - all_tids=malloc(sizeof(pid_t)); - if (!all_tids) - goto out_symbol_exit; - all_tids[0] = target_tid; - thread_num = 1; + threads = thread_map__new(target_pid, target_tid); + if (threads == NULL) { + pr_err("Problems finding threads of monitor\n"); + usage_with_options(record_usage, record_options); } - for (i = 0; i < MAX_NR_CPUS; i++) { - for (j = 0; j < MAX_COUNTERS; j++) { - fd[i][j] = malloc(sizeof(int)*thread_num); - if (!fd[i][j]) - goto out_free_fd; - } + cpus = cpu_map__new(cpu_list); + if (cpus == NULL) { + perror("failed to parse CPUs map"); + return -1; } - event_array = malloc( - sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num); + + list_for_each_entry(pos, &evsel_list, node) { + if (perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0) + goto out_free_fd; + } + event_array = malloc((sizeof(struct pollfd) * MAX_NR_CPUS * + MAX_COUNTERS * threads->nr)); if (!event_array) goto out_free_fd; @@ -920,12 +956,8 @@ int cmd_record(int argc, const char **argv, const char *prefix __used) out_free_event_array: free(event_array); out_free_fd: - for (i = 0; i < MAX_NR_CPUS; i++) { - for (j = 0; j < MAX_COUNTERS; j++) - free(fd[i][j]); - } - free(all_tids); - all_tids = NULL; + thread_map__delete(threads); + threads = NULL; out_symbol_exit: symbol__exit(); return err; diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 5de405d45230..75183a4518e6 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -150,13 +150,13 @@ static int add_event_total(struct perf_session *session, return 0; } -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event, struct sample_data *sample, + struct perf_session *session) { - struct sample_data data = { .period = 1, }; struct addr_location al; struct perf_event_attr *attr; - if (event__preprocess_sample(event, session, &al, &data, NULL) < 0) { + if (event__preprocess_sample(event, session, &al, sample, NULL) < 0) { fprintf(stderr, "problem processing %d event, skipping it.\n", event->header.type); return -1; @@ -165,14 +165,14 @@ static int process_sample_event(event_t *event, struct perf_session *session) if (al.filtered || (hide_unresolved && al.sym == NULL)) return 0; - if (perf_session__add_hist_entry(session, &al, &data)) { + if (perf_session__add_hist_entry(session, &al, sample)) { pr_debug("problem incrementing symbol period, skipping event\n"); return -1; } - attr = perf_header__find_attr(data.id, &session->header); + attr = perf_header__find_attr(sample->id, &session->header); - if (add_event_total(session, &data, attr)) { + if (add_event_total(session, sample, attr)) { pr_debug("problem adding event period\n"); return -1; } @@ -180,7 +180,8 @@ static int process_sample_event(event_t *event, struct perf_session *session) return 0; } -static int process_read_event(event_t *event, struct perf_session *session __used) +static int process_read_event(event_t *event, struct sample_data *sample __used, + struct perf_session *session __used) { struct perf_event_attr *attr; @@ -243,6 +244,8 @@ static struct perf_event_ops event_ops = { .event_type = event__process_event_type, .tracing_data = event__process_tracing_data, .build_id = event__process_build_id, + .ordered_samples = true, + .ordering_requires_timestamps = true, }; extern volatile int session_done; @@ -307,7 +310,7 @@ static int __cmd_report(void) signal(SIGINT, sig_handler); - session = perf_session__new(input_name, O_RDONLY, force, false); + session = perf_session__new(input_name, O_RDONLY, force, false, &event_ops); if (session == NULL) return -ENOMEM; @@ -442,6 +445,8 @@ static const struct option options[] = { "dump raw trace in ASCII"), OPT_STRING('k', "vmlinux", &symbol_conf.vmlinux_name, "file", "vmlinux pathname"), + OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name, + "file", "kallsyms pathname"), OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), OPT_BOOLEAN('m', "modules", &symbol_conf.use_modules, "load module symbols - WARNING: use only with -k and LIVE kernel"), @@ -478,6 +483,8 @@ static const struct option options[] = { "columns '.' is reserved."), OPT_BOOLEAN('U', "hide-unresolved", &hide_unresolved, "Only display entries resolved to a symbol"), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), OPT_END() }; diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 55f3b5dcc731..7a4ebeb8b016 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -1606,25 +1606,15 @@ process_raw_event(event_t *raw_event __used, struct perf_session *session, process_sched_migrate_task_event(data, session, event, cpu, timestamp, thread); } -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event, struct sample_data *sample, + struct perf_session *session) { - struct sample_data data; struct thread *thread; if (!(session->sample_type & PERF_SAMPLE_RAW)) return 0; - memset(&data, 0, sizeof(data)); - data.time = -1; - data.cpu = -1; - data.period = -1; - - event__parse_sample(event, session->sample_type, &data); - - dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc, - data.pid, data.tid, data.ip, data.period); - - thread = perf_session__findnew(session, data.pid); + thread = perf_session__findnew(session, sample->pid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", event->header.type); @@ -1633,10 +1623,11 @@ static int process_sample_event(event_t *event, struct perf_session *session) dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); - if (profile_cpu != -1 && profile_cpu != (int)data.cpu) + if (profile_cpu != -1 && profile_cpu != (int)sample->cpu) return 0; - process_raw_event(event, session, data.raw_data, data.cpu, data.time, thread); + process_raw_event(event, session, sample->raw_data, sample->cpu, + sample->time, thread); return 0; } @@ -1652,7 +1643,8 @@ static struct perf_event_ops event_ops = { static int read_events(void) { int err = -EINVAL; - struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0, false); + struct perf_session *session = perf_session__new(input_name, O_RDONLY, + 0, false, &event_ops); if (session == NULL) return -ENOMEM; @@ -1869,6 +1861,9 @@ static int __cmd_record(int argc, const char **argv) rec_argc = ARRAY_SIZE(record_args) + argc - 1; rec_argv = calloc(rec_argc + 1, sizeof(char *)); + if (rec_argv) + return -ENOMEM; + for (i = 0; i < ARRAY_SIZE(record_args); i++) rec_argv[i] = strdup(record_args[i]); @@ -1888,10 +1883,10 @@ int cmd_sched(int argc, const char **argv, const char *prefix __used) usage_with_options(sched_usage, sched_options); /* - * Aliased to 'perf trace' for now: + * Aliased to 'perf script' for now: */ - if (!strcmp(argv[0], "trace")) - return cmd_trace(argc, argv, prefix); + if (!strcmp(argv[0], "script")) + return cmd_script(argc, argv, prefix); symbol__init(); if (!strncmp(argv[0], "rec", 3)) { diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-script.c index 86cfe3800e6b..150a606002eb 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-script.c @@ -56,29 +56,18 @@ static void setup_scripting(void) static int cleanup_scripting(void) { - pr_debug("\nperf trace script stopped\n"); + pr_debug("\nperf script stopped\n"); return scripting_ops->stop_script(); } static char const *input_name = "perf.data"; -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event, struct sample_data *sample, + struct perf_session *session) { - struct sample_data data; - struct thread *thread; + struct thread *thread = perf_session__findnew(session, event->ip.pid); - memset(&data, 0, sizeof(data)); - data.time = -1; - data.cpu = -1; - data.period = 1; - - event__parse_sample(event, session->sample_type, &data); - - dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc, - data.pid, data.tid, data.ip, data.period); - - thread = perf_session__findnew(session, event->ip.pid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", event->header.type); @@ -87,13 +76,13 @@ static int process_sample_event(event_t *event, struct perf_session *session) if (session->sample_type & PERF_SAMPLE_RAW) { if (debug_mode) { - if (data.time < last_timestamp) { + if (sample->time < last_timestamp) { pr_err("Samples misordered, previous: %llu " "this: %llu\n", last_timestamp, - data.time); + sample->time); nr_unordered++; } - last_timestamp = data.time; + last_timestamp = sample->time; return 0; } /* @@ -101,21 +90,12 @@ static int process_sample_event(event_t *event, struct perf_session *session) * field, although it should be the same than this perf * event pid */ - scripting_ops->process_event(data.cpu, data.raw_data, - data.raw_size, - data.time, thread->comm); + scripting_ops->process_event(sample->cpu, sample->raw_data, + sample->raw_size, + sample->time, thread->comm); } - session->hists.stats.total_period += data.period; - return 0; -} - -static u64 nr_lost; - -static int process_lost_event(event_t *event, struct perf_session *session __used) -{ - nr_lost += event->lost.lost; - + session->hists.stats.total_period += sample->period; return 0; } @@ -126,7 +106,7 @@ static struct perf_event_ops event_ops = { .event_type = event__process_event_type, .tracing_data = event__process_tracing_data, .build_id = event__process_build_id, - .lost = process_lost_event, + .ordering_requires_timestamps = true, .ordered_samples = true, }; @@ -137,7 +117,7 @@ static void sig_handler(int sig __unused) session_done = 1; } -static int __cmd_trace(struct perf_session *session) +static int __cmd_script(struct perf_session *session) { int ret; @@ -145,10 +125,8 @@ static int __cmd_trace(struct perf_session *session) ret = perf_session__process_events(session, &event_ops); - if (debug_mode) { + if (debug_mode) pr_err("Misordered timestamps: %llu\n", nr_unordered); - pr_err("Lost events: %llu\n", nr_lost); - } return ret; } @@ -159,7 +137,7 @@ struct script_spec { char spec[0]; }; -LIST_HEAD(script_specs); +static LIST_HEAD(script_specs); static struct script_spec *script_spec__new(const char *spec, struct scripting_ops *ops) @@ -247,7 +225,7 @@ static void list_available_languages(void) fprintf(stderr, "\n"); fprintf(stderr, "Scripting language extensions (used in " - "perf trace -s [spec:]script.[spec]):\n\n"); + "perf script -s [spec:]script.[spec]):\n\n"); list_for_each_entry(s, &script_specs, node) fprintf(stderr, " %-42s [%s]\n", s->spec, s->ops->name); @@ -301,17 +279,34 @@ static int parse_scriptname(const struct option *opt __used, return 0; } -#define for_each_lang(scripts_dir, lang_dirent, lang_next) \ +/* Helper function for filesystems that return a dent->d_type DT_UNKNOWN */ +static int is_directory(const char *base_path, const struct dirent *dent) +{ + char path[PATH_MAX]; + struct stat st; + + sprintf(path, "%s/%s", base_path, dent->d_name); + if (stat(path, &st)) + return 0; + + return S_ISDIR(st.st_mode); +} + +#define for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next)\ while (!readdir_r(scripts_dir, &lang_dirent, &lang_next) && \ lang_next) \ - if (lang_dirent.d_type == DT_DIR && \ + if ((lang_dirent.d_type == DT_DIR || \ + (lang_dirent.d_type == DT_UNKNOWN && \ + is_directory(scripts_path, &lang_dirent))) && \ (strcmp(lang_dirent.d_name, ".")) && \ (strcmp(lang_dirent.d_name, ".."))) -#define for_each_script(lang_dir, script_dirent, script_next) \ +#define for_each_script(lang_path, lang_dir, script_dirent, script_next)\ while (!readdir_r(lang_dir, &script_dirent, &script_next) && \ script_next) \ - if (script_dirent.d_type != DT_DIR) + if (script_dirent.d_type != DT_DIR && \ + (script_dirent.d_type != DT_UNKNOWN || \ + !is_directory(lang_path, &script_dirent))) #define RECORD_SUFFIX "-record" @@ -324,7 +319,7 @@ struct script_desc { char *args; }; -LIST_HEAD(script_descs); +static LIST_HEAD(script_descs); static struct script_desc *script_desc__new(const char *name) { @@ -380,10 +375,10 @@ out_delete_desc: return NULL; } -static char *ends_with(char *str, const char *suffix) +static const char *ends_with(const char *str, const char *suffix) { size_t suffix_len = strlen(suffix); - char *p = str; + const char *p = str; if (strlen(str) > suffix_len) { p = str + strlen(str) - suffix_len; @@ -466,16 +461,16 @@ static int list_available_scripts(const struct option *opt __used, if (!scripts_dir) return -1; - for_each_lang(scripts_dir, lang_dirent, lang_next) { + for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) { snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path, lang_dirent.d_name); lang_dir = opendir(lang_path); if (!lang_dir) continue; - for_each_script(lang_dir, script_dirent, script_next) { + for_each_script(lang_path, lang_dir, script_dirent, script_next) { script_root = strdup(script_dirent.d_name); - str = ends_with(script_root, REPORT_SUFFIX); + str = (char *)ends_with(script_root, REPORT_SUFFIX); if (str) { *str = '\0'; desc = script_desc__findnew(script_root); @@ -514,16 +509,16 @@ static char *get_script_path(const char *script_root, const char *suffix) if (!scripts_dir) return NULL; - for_each_lang(scripts_dir, lang_dirent, lang_next) { + for_each_lang(scripts_path, scripts_dir, lang_dirent, lang_next) { snprintf(lang_path, MAXPATHLEN, "%s/%s/bin", scripts_path, lang_dirent.d_name); lang_dir = opendir(lang_path); if (!lang_dir) continue; - for_each_script(lang_dir, script_dirent, script_next) { + for_each_script(lang_path, lang_dir, script_dirent, script_next) { __script_root = strdup(script_dirent.d_name); - str = ends_with(__script_root, suffix); + str = (char *)ends_with(__script_root, suffix); if (str) { *str = '\0'; if (strcmp(__script_root, script_root)) @@ -543,7 +538,7 @@ static char *get_script_path(const char *script_root, const char *suffix) static bool is_top_script(const char *script_path) { - return ends_with((char *)script_path, "top") == NULL ? false : true; + return ends_with(script_path, "top") == NULL ? false : true; } static int has_required_arg(char *script_path) @@ -569,12 +564,12 @@ out: return n_args; } -static const char * const trace_usage[] = { - "perf trace [<options>]", - "perf trace [<options>] record <script> [<record-options>] <command>", - "perf trace [<options>] report <script> [script-args]", - "perf trace [<options>] <script> [<record-options>] <command>", - "perf trace [<options>] <top-script> [script-args]", +static const char * const script_usage[] = { + "perf script [<options>]", + "perf script [<options>] record <script> [<record-options>] <command>", + "perf script [<options>] report <script> [script-args]", + "perf script [<options>] <script> [<record-options>] <command>", + "perf script [<options>] <top-script> [script-args]", NULL }; @@ -591,7 +586,7 @@ static const struct option options[] = { "script file name (lang:script name, script name, or *)", parse_scriptname), OPT_STRING('g', "gen-script", &generate_script_lang, "lang", - "generate perf-trace.xx script in specified language"), + "generate perf-script.xx script in specified language"), OPT_STRING('i', "input", &input_name, "file", "input file name"), OPT_BOOLEAN('d', "debug-mode", &debug_mode, @@ -614,7 +609,7 @@ static bool have_cmd(int argc, const char **argv) return argc != 0; } -int cmd_trace(int argc, const char **argv, const char *prefix __used) +int cmd_script(int argc, const char **argv, const char *prefix __used) { char *rec_script_path = NULL; char *rep_script_path = NULL; @@ -626,7 +621,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) setup_scripting(); - argc = parse_options(argc, argv, options, trace_usage, + argc = parse_options(argc, argv, options, script_usage, PARSE_OPT_STOP_AT_NON_OPTION); if (argc > 1 && !strncmp(argv[0], "rec", strlen("rec"))) { @@ -640,7 +635,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) if (!rep_script_path) { fprintf(stderr, "Please specify a valid report script" - "(see 'perf trace -l' for listing)\n"); + "(see 'perf script -l' for listing)\n"); return -1; } } @@ -658,8 +653,8 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) if (!rec_script_path && !rep_script_path) { fprintf(stderr, " Couldn't find script %s\n\n See perf" - " trace -l for available scripts.\n", argv[0]); - usage_with_options(trace_usage, options); + " script -l for available scripts.\n", argv[0]); + usage_with_options(script_usage, options); } if (is_top_script(argv[0])) { @@ -671,9 +666,9 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) rec_args = (argc - 1) - rep_args; if (rec_args < 0) { fprintf(stderr, " %s script requires options." - "\n\n See perf trace -l for available " + "\n\n See perf script -l for available " "scripts and options.\n", argv[0]); - usage_with_options(trace_usage, options); + usage_with_options(script_usage, options); } } @@ -772,7 +767,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) if (!script_name) setup_pager(); - session = perf_session__new(input_name, O_RDONLY, 0, false); + session = perf_session__new(input_name, O_RDONLY, 0, false, &event_ops); if (session == NULL) return -ENOMEM; @@ -806,7 +801,7 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) return -1; } - err = scripting_ops->generate_script("perf-trace"); + err = scripting_ops->generate_script("perf-script"); goto out; } @@ -814,10 +809,10 @@ int cmd_trace(int argc, const char **argv, const char *prefix __used) err = scripting_ops->start_script(script_name, argc, argv); if (err) goto out; - pr_debug("perf trace started with script %s\n\n", script_name); + pr_debug("perf script started with script %s\n\n", script_name); } - err = __cmd_trace(session); + err = __cmd_script(session); perf_session__delete(session); cleanup_scripting(); diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index a6b4d44f9502..02b2d8013a61 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -43,6 +43,7 @@ #include "util/parse-options.h" #include "util/parse-events.h" #include "util/event.h" +#include "util/evsel.h" #include "util/debug.h" #include "util/header.h" #include "util/cpumap.h" @@ -52,6 +53,8 @@ #include <math.h> #include <locale.h> +#define DEFAULT_SEPARATOR " " + static struct perf_event_attr default_attrs[] = { { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, @@ -69,25 +72,23 @@ static struct perf_event_attr default_attrs[] = { }; static bool system_wide = false; -static int nr_cpus = 0; +static struct cpu_map *cpus; static int run_idx = 0; static int run_count = 1; static bool no_inherit = false; static bool scale = true; +static bool no_aggr = false; static pid_t target_pid = -1; static pid_t target_tid = -1; -static pid_t *all_tids = NULL; -static int thread_num = 0; +static struct thread_map *threads; static pid_t child_pid = -1; static bool null_run = false; -static bool big_num = false; +static bool big_num = true; +static int big_num_opt = -1; static const char *cpu_list; - - -static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; - -static int event_scaled[MAX_COUNTERS]; +static const char *csv_sep = NULL; +static bool csv_output = false; static volatile int done = 0; @@ -96,6 +97,22 @@ struct stats double n, mean, M2; }; +struct perf_stat { + struct stats res_stats[3]; +}; + +static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel) +{ + evsel->priv = zalloc(sizeof(struct perf_stat)); + return evsel->priv == NULL ? -ENOMEM : 0; +} + +static void perf_evsel__free_stat_priv(struct perf_evsel *evsel) +{ + free(evsel->priv); + evsel->priv = NULL; +} + static void update_stats(struct stats *stats, u64 val) { double delta; @@ -135,69 +152,38 @@ static double stddev_stats(struct stats *stats) return sqrt(variance_mean); } -struct stats event_res_stats[MAX_COUNTERS][3]; -struct stats runtime_nsecs_stats; +struct stats runtime_nsecs_stats[MAX_NR_CPUS]; +struct stats runtime_cycles_stats[MAX_NR_CPUS]; +struct stats runtime_branches_stats[MAX_NR_CPUS]; struct stats walltime_nsecs_stats; -struct stats runtime_cycles_stats; -struct stats runtime_branches_stats; -#define MATCH_EVENT(t, c, counter) \ - (attrs[counter].type == PERF_TYPE_##t && \ - attrs[counter].config == PERF_COUNT_##c) - -#define ERR_PERF_OPEN \ -"Error: counter %d, sys_perf_event_open() syscall returned with %d (%s)\n" - -static int create_perf_stat_counter(int counter) +static int create_perf_stat_counter(struct perf_evsel *evsel) { - struct perf_event_attr *attr = attrs + counter; - int thread; - int ncreated = 0; + struct perf_event_attr *attr = &evsel->attr; if (scale) attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED | PERF_FORMAT_TOTAL_TIME_RUNNING; - if (system_wide) { - int cpu; - - for (cpu = 0; cpu < nr_cpus; cpu++) { - fd[cpu][counter][0] = sys_perf_event_open(attr, - -1, cpumap[cpu], -1, 0); - if (fd[cpu][counter][0] < 0) - pr_debug(ERR_PERF_OPEN, counter, - fd[cpu][counter][0], strerror(errno)); - else - ++ncreated; - } - } else { - attr->inherit = !no_inherit; - if (target_pid == -1 && target_tid == -1) { - attr->disabled = 1; - attr->enable_on_exec = 1; - } - for (thread = 0; thread < thread_num; thread++) { - fd[0][counter][thread] = sys_perf_event_open(attr, - all_tids[thread], -1, -1, 0); - if (fd[0][counter][thread] < 0) - pr_debug(ERR_PERF_OPEN, counter, - fd[0][counter][thread], - strerror(errno)); - else - ++ncreated; - } + if (system_wide) + return perf_evsel__open_per_cpu(evsel, cpus); + + attr->inherit = !no_inherit; + if (target_pid == -1 && target_tid == -1) { + attr->disabled = 1; + attr->enable_on_exec = 1; } - return ncreated; + return perf_evsel__open_per_thread(evsel, threads); } /* * Does the counter have nsecs as a unit? */ -static inline int nsec_counter(int counter) +static inline int nsec_counter(struct perf_evsel *evsel) { - if (MATCH_EVENT(SOFTWARE, SW_CPU_CLOCK, counter) || - MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter)) + if (perf_evsel__match(evsel, SOFTWARE, SW_CPU_CLOCK) || + perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK)) return 1; return 0; @@ -205,55 +191,19 @@ static inline int nsec_counter(int counter) /* * Read out the results of a single counter: + * aggregate counts across CPUs in system-wide mode */ -static void read_counter(int counter) +static int read_counter_aggr(struct perf_evsel *counter) { - u64 count[3], single_count[3]; - int cpu; - size_t res, nv; - int scaled; - int i, thread; - - count[0] = count[1] = count[2] = 0; - - nv = scale ? 3 : 1; - for (cpu = 0; cpu < nr_cpus; cpu++) { - for (thread = 0; thread < thread_num; thread++) { - if (fd[cpu][counter][thread] < 0) - continue; - - res = read(fd[cpu][counter][thread], - single_count, nv * sizeof(u64)); - assert(res == nv * sizeof(u64)); - - close(fd[cpu][counter][thread]); - fd[cpu][counter][thread] = -1; - - count[0] += single_count[0]; - if (scale) { - count[1] += single_count[1]; - count[2] += single_count[2]; - } - } - } - - scaled = 0; - if (scale) { - if (count[2] == 0) { - event_scaled[counter] = -1; - count[0] = 0; - return; - } + struct perf_stat *ps = counter->priv; + u64 *count = counter->counts->aggr.values; + int i; - if (count[2] < count[1]) { - event_scaled[counter] = 1; - count[0] = (unsigned long long) - ((double)count[0] * count[1] / count[2] + 0.5); - } - } + if (__perf_evsel__read(counter, cpus->nr, threads->nr, scale) < 0) + return -1; for (i = 0; i < 3; i++) - update_stats(&event_res_stats[counter][i], count[i]); + update_stats(&ps->res_stats[i], count[i]); if (verbose) { fprintf(stderr, "%s: %Ld %Ld %Ld\n", event_name(counter), @@ -263,26 +213,51 @@ static void read_counter(int counter) /* * Save the full runtime - to allow normalization during printout: */ - if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter)) - update_stats(&runtime_nsecs_stats, count[0]); - if (MATCH_EVENT(HARDWARE, HW_CPU_CYCLES, counter)) - update_stats(&runtime_cycles_stats, count[0]); - if (MATCH_EVENT(HARDWARE, HW_BRANCH_INSTRUCTIONS, counter)) - update_stats(&runtime_branches_stats, count[0]); + if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK)) + update_stats(&runtime_nsecs_stats[0], count[0]); + if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES)) + update_stats(&runtime_cycles_stats[0], count[0]); + if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS)) + update_stats(&runtime_branches_stats[0], count[0]); + + return 0; +} + +/* + * Read out the results of a single counter: + * do not aggregate counts across CPUs in system-wide mode + */ +static int read_counter(struct perf_evsel *counter) +{ + u64 *count; + int cpu; + + for (cpu = 0; cpu < cpus->nr; cpu++) { + if (__perf_evsel__read_on_cpu(counter, cpu, 0, scale) < 0) + return -1; + + count = counter->counts->cpu[cpu].values; + + if (perf_evsel__match(counter, SOFTWARE, SW_TASK_CLOCK)) + update_stats(&runtime_nsecs_stats[cpu], count[0]); + if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES)) + update_stats(&runtime_cycles_stats[cpu], count[0]); + if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS)) + update_stats(&runtime_branches_stats[cpu], count[0]); + } + + return 0; } static int run_perf_stat(int argc __used, const char **argv) { unsigned long long t0, t1; + struct perf_evsel *counter; int status = 0; - int counter, ncreated = 0; int child_ready_pipe[2], go_pipe[2]; const bool forks = (argc > 0); char buf; - if (!system_wide) - nr_cpus = 1; - if (forks && (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0)) { perror("failed to create pipes"); exit(1); @@ -322,7 +297,7 @@ static int run_perf_stat(int argc __used, const char **argv) } if (target_tid == -1 && target_pid == -1 && !system_wide) - all_tids[0] = child_pid; + threads->map[0] = child_pid; /* * Wait for the child to be ready to exec. @@ -334,16 +309,23 @@ static int run_perf_stat(int argc __used, const char **argv) close(child_ready_pipe[0]); } - for (counter = 0; counter < nr_counters; counter++) - ncreated += create_perf_stat_counter(counter); - - if (ncreated == 0) { - pr_err("No permission to collect %sstats.\n" - "Consider tweaking /proc/sys/kernel/perf_event_paranoid.\n", - system_wide ? "system-wide " : ""); - if (child_pid != -1) - kill(child_pid, SIGTERM); - return -1; + list_for_each_entry(counter, &evsel_list, node) { + if (create_perf_stat_counter(counter) < 0) { + if (errno == -EPERM || errno == -EACCES) { + error("You may not have permission to collect %sstats.\n" + "\t Consider tweaking" + " /proc/sys/kernel/perf_event_paranoid or running as root.", + system_wide ? "system-wide " : ""); + } else { + error("open_counter returned with %d (%s). " + "/bin/dmesg may provide additional information.\n", + errno, strerror(errno)); + } + if (child_pid != -1) + kill(child_pid, SIGTERM); + die("Not all events could be opened.\n"); + return -1; + } } /* @@ -362,60 +344,97 @@ static int run_perf_stat(int argc __used, const char **argv) update_stats(&walltime_nsecs_stats, t1 - t0); - for (counter = 0; counter < nr_counters; counter++) - read_counter(counter); + if (no_aggr) { + list_for_each_entry(counter, &evsel_list, node) { + read_counter(counter); + perf_evsel__close_fd(counter, cpus->nr, 1); + } + } else { + list_for_each_entry(counter, &evsel_list, node) { + read_counter_aggr(counter); + perf_evsel__close_fd(counter, cpus->nr, threads->nr); + } + } return WEXITSTATUS(status); } -static void print_noise(int counter, double avg) +static void print_noise(struct perf_evsel *evsel, double avg) { + struct perf_stat *ps; + if (run_count == 1) return; + ps = evsel->priv; fprintf(stderr, " ( +- %7.3f%% )", - 100 * stddev_stats(&event_res_stats[counter][0]) / avg); + 100 * stddev_stats(&ps->res_stats[0]) / avg); } -static void nsec_printout(int counter, double avg) +static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg) { double msecs = avg / 1e6; + char cpustr[16] = { '\0', }; + const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-24s"; - fprintf(stderr, " %18.6f %-24s", msecs, event_name(counter)); + if (no_aggr) + sprintf(cpustr, "CPU%*d%s", + csv_output ? 0 : -4, + cpus->map[cpu], csv_sep); + + fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel)); + + if (csv_output) + return; - if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter)) { + if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK)) fprintf(stderr, " # %10.3f CPUs ", avg / avg_stats(&walltime_nsecs_stats)); - } } -static void abs_printout(int counter, double avg) +static void abs_printout(int cpu, struct perf_evsel *evsel, double avg) { double total, ratio = 0.0; + char cpustr[16] = { '\0', }; + const char *fmt; + + if (csv_output) + fmt = "%s%.0f%s%s"; + else if (big_num) + fmt = "%s%'18.0f%s%-24s"; + else + fmt = "%s%18.0f%s%-24s"; - if (big_num) - fprintf(stderr, " %'18.0f %-24s", avg, event_name(counter)); + if (no_aggr) + sprintf(cpustr, "CPU%*d%s", + csv_output ? 0 : -4, + cpus->map[cpu], csv_sep); else - fprintf(stderr, " %18.0f %-24s", avg, event_name(counter)); + cpu = 0; + + fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel)); - if (MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter)) { - total = avg_stats(&runtime_cycles_stats); + if (csv_output) + return; + + if (perf_evsel__match(evsel, HARDWARE, HW_INSTRUCTIONS)) { + total = avg_stats(&runtime_cycles_stats[cpu]); if (total) ratio = avg / total; fprintf(stderr, " # %10.3f IPC ", ratio); - } else if (MATCH_EVENT(HARDWARE, HW_BRANCH_MISSES, counter) && - runtime_branches_stats.n != 0) { - total = avg_stats(&runtime_branches_stats); + } else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) && + runtime_branches_stats[cpu].n != 0) { + total = avg_stats(&runtime_branches_stats[cpu]); if (total) ratio = avg * 100 / total; fprintf(stderr, " # %10.3f %% ", ratio); - } else if (runtime_nsecs_stats.n != 0) { - total = avg_stats(&runtime_nsecs_stats); + } else if (runtime_nsecs_stats[cpu].n != 0) { + total = avg_stats(&runtime_nsecs_stats[cpu]); if (total) ratio = 1000.0 * avg / total; @@ -426,30 +445,38 @@ static void abs_printout(int counter, double avg) /* * Print out the results of a single counter: + * aggregated counts in system-wide mode */ -static void print_counter(int counter) +static void print_counter_aggr(struct perf_evsel *counter) { - double avg = avg_stats(&event_res_stats[counter][0]); - int scaled = event_scaled[counter]; + struct perf_stat *ps = counter->priv; + double avg = avg_stats(&ps->res_stats[0]); + int scaled = counter->counts->scaled; if (scaled == -1) { - fprintf(stderr, " %18s %-24s\n", - "<not counted>", event_name(counter)); + fprintf(stderr, "%*s%s%-24s\n", + csv_output ? 0 : 18, + "<not counted>", csv_sep, event_name(counter)); return; } if (nsec_counter(counter)) - nsec_printout(counter, avg); + nsec_printout(-1, counter, avg); else - abs_printout(counter, avg); + abs_printout(-1, counter, avg); + + if (csv_output) { + fputc('\n', stderr); + return; + } print_noise(counter, avg); if (scaled) { double avg_enabled, avg_running; - avg_enabled = avg_stats(&event_res_stats[counter][1]); - avg_running = avg_stats(&event_res_stats[counter][2]); + avg_enabled = avg_stats(&ps->res_stats[1]); + avg_running = avg_stats(&ps->res_stats[2]); fprintf(stderr, " (scaled from %.2f%%)", 100 * avg_running / avg_enabled); @@ -458,40 +485,92 @@ static void print_counter(int counter) fprintf(stderr, "\n"); } +/* + * Print out the results of a single counter: + * does not use aggregated count in system-wide + */ +static void print_counter(struct perf_evsel *counter) +{ + u64 ena, run, val; + int cpu; + + for (cpu = 0; cpu < cpus->nr; cpu++) { + val = counter->counts->cpu[cpu].val; + ena = counter->counts->cpu[cpu].ena; + run = counter->counts->cpu[cpu].run; + if (run == 0 || ena == 0) { + fprintf(stderr, "CPU%*d%s%*s%s%-24s", + csv_output ? 0 : -4, + cpus->map[cpu], csv_sep, + csv_output ? 0 : 18, + "<not counted>", csv_sep, + event_name(counter)); + + fprintf(stderr, "\n"); + continue; + } + + if (nsec_counter(counter)) + nsec_printout(cpu, counter, val); + else + abs_printout(cpu, counter, val); + + if (!csv_output) { + print_noise(counter, 1.0); + + if (run != ena) { + fprintf(stderr, " (scaled from %.2f%%)", + 100.0 * run / ena); + } + } + fprintf(stderr, "\n"); + } +} + static void print_stat(int argc, const char **argv) { - int i, counter; + struct perf_evsel *counter; + int i; fflush(stdout); - fprintf(stderr, "\n"); - fprintf(stderr, " Performance counter stats for "); - if(target_pid == -1 && target_tid == -1) { - fprintf(stderr, "\'%s", argv[0]); - for (i = 1; i < argc; i++) - fprintf(stderr, " %s", argv[i]); - } else if (target_pid != -1) - fprintf(stderr, "process id \'%d", target_pid); - else - fprintf(stderr, "thread id \'%d", target_tid); - - fprintf(stderr, "\'"); - if (run_count > 1) - fprintf(stderr, " (%d runs)", run_count); - fprintf(stderr, ":\n\n"); + if (!csv_output) { + fprintf(stderr, "\n"); + fprintf(stderr, " Performance counter stats for "); + if(target_pid == -1 && target_tid == -1) { + fprintf(stderr, "\'%s", argv[0]); + for (i = 1; i < argc; i++) + fprintf(stderr, " %s", argv[i]); + } else if (target_pid != -1) + fprintf(stderr, "process id \'%d", target_pid); + else + fprintf(stderr, "thread id \'%d", target_tid); + + fprintf(stderr, "\'"); + if (run_count > 1) + fprintf(stderr, " (%d runs)", run_count); + fprintf(stderr, ":\n\n"); + } - for (counter = 0; counter < nr_counters; counter++) - print_counter(counter); + if (no_aggr) { + list_for_each_entry(counter, &evsel_list, node) + print_counter(counter); + } else { + list_for_each_entry(counter, &evsel_list, node) + print_counter_aggr(counter); + } - fprintf(stderr, "\n"); - fprintf(stderr, " %18.9f seconds time elapsed", - avg_stats(&walltime_nsecs_stats)/1e9); - if (run_count > 1) { - fprintf(stderr, " ( +- %7.3f%% )", + if (!csv_output) { + fprintf(stderr, "\n"); + fprintf(stderr, " %18.9f seconds time elapsed", + avg_stats(&walltime_nsecs_stats)/1e9); + if (run_count > 1) { + fprintf(stderr, " ( +- %7.3f%% )", 100*stddev_stats(&walltime_nsecs_stats) / avg_stats(&walltime_nsecs_stats)); + } + fprintf(stderr, "\n\n"); } - fprintf(stderr, "\n\n"); } static volatile int signr = -1; @@ -521,6 +600,13 @@ static const char * const stat_usage[] = { NULL }; +static int stat__set_big_num(const struct option *opt __used, + const char *s __used, int unset) +{ + big_num_opt = unset ? 0 : 1; + return 0; +} + static const struct option options[] = { OPT_CALLBACK('e', "event", NULL, "event", "event selector. use 'perf list' to list available events", @@ -541,64 +627,96 @@ static const struct option options[] = { "repeat command and print average + stddev (max: 100)"), OPT_BOOLEAN('n', "null", &null_run, "null run - dont start any counters"), - OPT_BOOLEAN('B', "big-num", &big_num, - "print large numbers with thousands\' separators"), + OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL, + "print large numbers with thousands\' separators", + stat__set_big_num), OPT_STRING('C', "cpu", &cpu_list, "cpu", "list of cpus to monitor in system-wide"), + OPT_BOOLEAN('A', "no-aggr", &no_aggr, + "disable CPU count aggregation"), + OPT_STRING('x', "field-separator", &csv_sep, "separator", + "print counts with custom separator"), OPT_END() }; int cmd_stat(int argc, const char **argv, const char *prefix __used) { - int status; - int i,j; + struct perf_evsel *pos; + int status = -ENOMEM; setlocale(LC_ALL, ""); argc = parse_options(argc, argv, options, stat_usage, PARSE_OPT_STOP_AT_NON_OPTION); + + if (csv_sep) + csv_output = true; + else + csv_sep = DEFAULT_SEPARATOR; + + /* + * let the spreadsheet do the pretty-printing + */ + if (csv_output) { + /* User explicitely passed -B? */ + if (big_num_opt == 1) { + fprintf(stderr, "-B option not supported with -x\n"); + usage_with_options(stat_usage, options); + } else /* Nope, so disable big number formatting */ + big_num = false; + } else if (big_num_opt == 0) /* User passed --no-big-num */ + big_num = false; + if (!argc && target_pid == -1 && target_tid == -1) usage_with_options(stat_usage, options); if (run_count <= 0) usage_with_options(stat_usage, options); + /* no_aggr is for system-wide only */ + if (no_aggr && !system_wide) + usage_with_options(stat_usage, options); + /* Set attrs and nr_counters if no event is selected and !null_run */ if (!null_run && !nr_counters) { - memcpy(attrs, default_attrs, sizeof(default_attrs)); + size_t c; + nr_counters = ARRAY_SIZE(default_attrs); + + for (c = 0; c < ARRAY_SIZE(default_attrs); ++c) { + pos = perf_evsel__new(default_attrs[c].type, + default_attrs[c].config, + nr_counters); + if (pos == NULL) + goto out; + list_add(&pos->node, &evsel_list); + } } - if (system_wide) - nr_cpus = read_cpu_map(cpu_list); - else - nr_cpus = 1; + if (target_pid != -1) + target_tid = target_pid; - if (nr_cpus < 1) + threads = thread_map__new(target_pid, target_tid); + if (threads == NULL) { + pr_err("Problems finding threads of monitor\n"); usage_with_options(stat_usage, options); + } - if (target_pid != -1) { - target_tid = target_pid; - thread_num = find_all_tid(target_pid, &all_tids); - if (thread_num <= 0) { - fprintf(stderr, "Can't find all threads of pid %d\n", - target_pid); - usage_with_options(stat_usage, options); - } - } else { - all_tids=malloc(sizeof(pid_t)); - if (!all_tids) - return -ENOMEM; + if (system_wide) + cpus = cpu_map__new(cpu_list); + else + cpus = cpu_map__dummy_new(); - all_tids[0] = target_tid; - thread_num = 1; + if (cpus == NULL) { + perror("failed to parse CPUs map"); + usage_with_options(stat_usage, options); + return -1; } - for (i = 0; i < MAX_NR_CPUS; i++) { - for (j = 0; j < MAX_COUNTERS; j++) { - fd[i][j] = malloc(sizeof(int)*thread_num); - if (!fd[i][j]) - return -ENOMEM; - } + list_for_each_entry(pos, &evsel_list, node) { + if (perf_evsel__alloc_stat_priv(pos) < 0 || + perf_evsel__alloc_counts(pos, cpus->nr) < 0 || + perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0) + goto out_free_fd; } /* @@ -621,6 +739,11 @@ int cmd_stat(int argc, const char **argv, const char *prefix __used) if (status != -1) print_stat(argc, argv); - +out_free_fd: + list_for_each_entry(pos, &evsel_list, node) + perf_evsel__free_stat_priv(pos); +out: + thread_map__delete(threads); + threads = NULL; return status; } diff --git a/tools/perf/builtin-test.c b/tools/perf/builtin-test.c index 035b9fa063a9..1c984342a579 100644 --- a/tools/perf/builtin-test.c +++ b/tools/perf/builtin-test.c @@ -119,10 +119,16 @@ static int test__vmlinux_matches_kallsyms(void) * end addresses too. */ for (nd = rb_first(&vmlinux_map->dso->symbols[type]); nd; nd = rb_next(nd)) { - struct symbol *pair; + struct symbol *pair, *first_pair; + bool backwards = true; sym = rb_entry(nd, struct symbol, rb_node); - pair = machine__find_kernel_symbol(&kallsyms, type, sym->start, NULL, NULL); + + if (sym->start == sym->end) + continue; + + first_pair = machine__find_kernel_symbol(&kallsyms, type, sym->start, NULL, NULL); + pair = first_pair; if (pair && pair->start == sym->start) { next_pair: @@ -143,8 +149,10 @@ next_pair: pr_debug("%#Lx: diff end addr for %s v: %#Lx k: %#Lx\n", sym->start, sym->name, sym->end, pair->end); } else { - struct rb_node *nnd = rb_prev(&pair->rb_node); - + struct rb_node *nnd; +detour: + nnd = backwards ? rb_prev(&pair->rb_node) : + rb_next(&pair->rb_node); if (nnd) { struct symbol *next = rb_entry(nnd, struct symbol, rb_node); @@ -153,6 +161,13 @@ next_pair: goto next_pair; } } + + if (backwards) { + backwards = false; + pair = first_pair; + goto detour; + } + pr_debug("%#Lx: diff name v: %s k: %s\n", sym->start, sym->name, pair->name); } @@ -219,6 +234,89 @@ out: return err; } +#include "util/evsel.h" +#include <sys/types.h> + +static int trace_event__id(const char *event_name) +{ + char *filename; + int err = -1, fd; + + if (asprintf(&filename, + "/sys/kernel/debug/tracing/events/syscalls/%s/id", + event_name) < 0) + return -1; + + fd = open(filename, O_RDONLY); + if (fd >= 0) { + char id[16]; + if (read(fd, id, sizeof(id)) > 0) + err = atoi(id); + close(fd); + } + + free(filename); + return err; +} + +static int test__open_syscall_event(void) +{ + int err = -1, fd; + struct thread_map *threads; + struct perf_evsel *evsel; + unsigned int nr_open_calls = 111, i; + int id = trace_event__id("sys_enter_open"); + + if (id < 0) { + pr_debug("is debugfs mounted on /sys/kernel/debug?\n"); + return -1; + } + + threads = thread_map__new(-1, getpid()); + if (threads == NULL) { + pr_debug("thread_map__new\n"); + return -1; + } + + evsel = perf_evsel__new(PERF_TYPE_TRACEPOINT, id, 0); + if (evsel == NULL) { + pr_debug("perf_evsel__new\n"); + goto out_thread_map_delete; + } + + if (perf_evsel__open_per_thread(evsel, threads) < 0) { + pr_debug("failed to open counter: %s, " + "tweak /proc/sys/kernel/perf_event_paranoid?\n", + strerror(errno)); + goto out_evsel_delete; + } + + for (i = 0; i < nr_open_calls; ++i) { + fd = open("/etc/passwd", O_RDONLY); + close(fd); + } + + if (perf_evsel__read_on_cpu(evsel, 0, 0) < 0) { + pr_debug("perf_evsel__open_read_on_cpu\n"); + goto out_close_fd; + } + + if (evsel->counts->cpu[0].val != nr_open_calls) { + pr_debug("perf_evsel__read_on_cpu: expected to intercept %d calls, got %Ld\n", + nr_open_calls, evsel->counts->cpu[0].val); + goto out_close_fd; + } + + err = 0; +out_close_fd: + perf_evsel__close_fd(evsel, 1, threads->nr); +out_evsel_delete: + perf_evsel__delete(evsel); +out_thread_map_delete: + thread_map__delete(threads); + return err; +} + static struct test { const char *desc; int (*func)(void); @@ -228,6 +326,10 @@ static struct test { .func = test__vmlinux_matches_kallsyms, }, { + .desc = "detect open syscall event", + .func = test__open_syscall_event, + }, + { .func = NULL, }, }; diff --git a/tools/perf/builtin-timechart.c b/tools/perf/builtin-timechart.c index 9bcc38f0b706..746cf03cb05d 100644 --- a/tools/perf/builtin-timechart.c +++ b/tools/perf/builtin-timechart.c @@ -32,6 +32,10 @@ #include "util/session.h" #include "util/svghelper.h" +#define SUPPORT_OLD_POWER_EVENTS 1 +#define PWR_EVENT_EXIT -1 + + static char const *input_name = "perf.data"; static char const *output_name = "output.svg"; @@ -272,19 +276,22 @@ static int cpus_cstate_state[MAX_CPUS]; static u64 cpus_pstate_start_times[MAX_CPUS]; static u64 cpus_pstate_state[MAX_CPUS]; -static int process_comm_event(event_t *event, struct perf_session *session __used) +static int process_comm_event(event_t *event, struct sample_data *sample __used, + struct perf_session *session __used) { pid_set_comm(event->comm.tid, event->comm.comm); return 0; } -static int process_fork_event(event_t *event, struct perf_session *session __used) +static int process_fork_event(event_t *event, struct sample_data *sample __used, + struct perf_session *session __used) { pid_fork(event->fork.pid, event->fork.ppid, event->fork.time); return 0; } -static int process_exit_event(event_t *event, struct perf_session *session __used) +static int process_exit_event(event_t *event, struct sample_data *sample __used, + struct perf_session *session __used) { pid_exit(event->fork.pid, event->fork.time); return 0; @@ -298,12 +305,21 @@ struct trace_entry { int lock_depth; }; -struct power_entry { +#ifdef SUPPORT_OLD_POWER_EVENTS +static int use_old_power_events; +struct power_entry_old { struct trace_entry te; u64 type; u64 value; u64 cpu_id; }; +#endif + +struct power_processor_entry { + struct trace_entry te; + u32 state; + u32 cpu_id; +}; #define TASK_COMM_LEN 16 struct wakeup_entry { @@ -470,48 +486,65 @@ static void sched_switch(int cpu, u64 timestamp, struct trace_entry *te) } -static int process_sample_event(event_t *event, struct perf_session *session) +static int process_sample_event(event_t *event __used, + struct sample_data *sample, + struct perf_session *session) { - struct sample_data data; struct trace_entry *te; - memset(&data, 0, sizeof(data)); - - event__parse_sample(event, session->sample_type, &data); - if (session->sample_type & PERF_SAMPLE_TIME) { - if (!first_time || first_time > data.time) - first_time = data.time; - if (last_time < data.time) - last_time = data.time; + if (!first_time || first_time > sample->time) + first_time = sample->time; + if (last_time < sample->time) + last_time = sample->time; } - te = (void *)data.raw_data; - if (session->sample_type & PERF_SAMPLE_RAW && data.raw_size > 0) { + te = (void *)sample->raw_data; + if (session->sample_type & PERF_SAMPLE_RAW && sample->raw_size > 0) { char *event_str; - struct power_entry *pe; - - pe = (void *)te; - +#ifdef SUPPORT_OLD_POWER_EVENTS + struct power_entry_old *peo; + peo = (void *)te; +#endif event_str = perf_header__find_event(te->type); if (!event_str) return 0; - if (strcmp(event_str, "power:power_start") == 0) - c_state_start(pe->cpu_id, data.time, pe->value); + if (strcmp(event_str, "power:cpu_idle") == 0) { + struct power_processor_entry *ppe = (void *)te; + if (ppe->state == (u32)PWR_EVENT_EXIT) + c_state_end(ppe->cpu_id, sample->time); + else + c_state_start(ppe->cpu_id, sample->time, + ppe->state); + } + else if (strcmp(event_str, "power:cpu_frequency") == 0) { + struct power_processor_entry *ppe = (void *)te; + p_state_change(ppe->cpu_id, sample->time, ppe->state); + } + + else if (strcmp(event_str, "sched:sched_wakeup") == 0) + sched_wakeup(sample->cpu, sample->time, sample->pid, te); - if (strcmp(event_str, "power:power_end") == 0) - c_state_end(pe->cpu_id, data.time); + else if (strcmp(event_str, "sched:sched_switch") == 0) + sched_switch(sample->cpu, sample->time, te); - if (strcmp(event_str, "power:power_frequency") == 0) - p_state_change(pe->cpu_id, data.time, pe->value); +#ifdef SUPPORT_OLD_POWER_EVENTS + if (use_old_power_events) { + if (strcmp(event_str, "power:power_start") == 0) + c_state_start(peo->cpu_id, sample->time, + peo->value); - if (strcmp(event_str, "sched:sched_wakeup") == 0) - sched_wakeup(data.cpu, data.time, data.pid, te); + else if (strcmp(event_str, "power:power_end") == 0) + c_state_end(sample->cpu, sample->time); - if (strcmp(event_str, "sched:sched_switch") == 0) - sched_switch(data.cpu, data.time, te); + else if (strcmp(event_str, + "power:power_frequency") == 0) + p_state_change(peo->cpu_id, sample->time, + peo->value); + } +#endif } return 0; } @@ -937,7 +970,8 @@ static struct perf_event_ops event_ops = { static int __cmd_timechart(void) { - struct perf_session *session = perf_session__new(input_name, O_RDONLY, 0, false); + struct perf_session *session = perf_session__new(input_name, O_RDONLY, + 0, false, &event_ops); int ret = -EINVAL; if (session == NULL) @@ -968,7 +1002,8 @@ static const char * const timechart_usage[] = { NULL }; -static const char *record_args[] = { +#ifdef SUPPORT_OLD_POWER_EVENTS +static const char * const record_old_args[] = { "record", "-a", "-R", @@ -980,16 +1015,43 @@ static const char *record_args[] = { "-e", "sched:sched_wakeup", "-e", "sched:sched_switch", }; +#endif + +static const char * const record_new_args[] = { + "record", + "-a", + "-R", + "-f", + "-c", "1", + "-e", "power:cpu_frequency", + "-e", "power:cpu_idle", + "-e", "sched:sched_wakeup", + "-e", "sched:sched_switch", +}; static int __cmd_record(int argc, const char **argv) { unsigned int rec_argc, i, j; const char **rec_argv; + const char * const *record_args = record_new_args; + unsigned int record_elems = ARRAY_SIZE(record_new_args); + +#ifdef SUPPORT_OLD_POWER_EVENTS + if (!is_valid_tracepoint("power:cpu_idle") && + is_valid_tracepoint("power:power_start")) { + use_old_power_events = 1; + record_args = record_old_args; + record_elems = ARRAY_SIZE(record_old_args); + } +#endif - rec_argc = ARRAY_SIZE(record_args) + argc - 1; + rec_argc = record_elems + argc - 1; rec_argv = calloc(rec_argc + 1, sizeof(char *)); - for (i = 0; i < ARRAY_SIZE(record_args); i++) + if (rec_argv == NULL) + return -ENOMEM; + + for (i = 0; i < record_elems; i++) rec_argv[i] = strdup(record_args[i]); for (j = 1; j < (unsigned int)argc; j++, i++) @@ -1018,6 +1080,8 @@ static const struct option options[] = { OPT_CALLBACK('p', "process", NULL, "process", "process selector. Pass a pid or process name.", parse_process), + OPT_STRING(0, "symfs", &symbol_conf.symfs, "directory", + "Look for files with symbols relative to this directory"), OPT_END() }; diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index dd625808c2a5..1e67ab9c7ebc 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -21,6 +21,7 @@ #include "perf.h" #include "util/color.h" +#include "util/evsel.h" #include "util/session.h" #include "util/symbol.h" #include "util/thread.h" @@ -29,6 +30,7 @@ #include "util/parse-options.h" #include "util/parse-events.h" #include "util/cpumap.h" +#include "util/xyarray.h" #include "util/debug.h" @@ -55,7 +57,7 @@ #include <linux/unistd.h> #include <linux/types.h> -static int *fd[MAX_NR_CPUS][MAX_COUNTERS]; +#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y)) static bool system_wide = false; @@ -66,10 +68,9 @@ static int print_entries; static int target_pid = -1; static int target_tid = -1; -static pid_t *all_tids = NULL; -static int thread_num = 0; +static struct thread_map *threads; static bool inherit = false; -static int nr_cpus = 0; +static struct cpu_map *cpus; static int realtime_prio = 0; static bool group = false; static unsigned int page_size; @@ -100,6 +101,7 @@ struct sym_entry *sym_filter_entry = NULL; struct sym_entry *sym_filter_entry_sched = NULL; static int sym_pcnt_filter = 5; static int sym_counter = 0; +static struct perf_evsel *sym_evsel = NULL; static int display_weighted = -1; static const char *cpu_list; @@ -353,7 +355,7 @@ static void show_details(struct sym_entry *syme) return; symbol = sym_entry__symbol(syme); - printf("Showing %s for %s\n", event_name(sym_counter), symbol->name); + printf("Showing %s for %s\n", event_name(sym_evsel), symbol->name); printf(" Events Pcnt (>=%d%%)\n", sym_pcnt_filter); pthread_mutex_lock(&syme->src->lock); @@ -460,7 +462,8 @@ static void rb_insert_active_sym(struct rb_root *tree, struct sym_entry *se) static void print_sym_table(void) { int printed = 0, j; - int counter, snap = !display_weighted ? sym_counter : 0; + struct perf_evsel *counter; + int snap = !display_weighted ? sym_counter : 0; float samples_per_sec = samples/delay_secs; float ksamples_per_sec = kernel_samples/delay_secs; float us_samples_per_sec = (us_samples)/delay_secs; @@ -532,7 +535,9 @@ static void print_sym_table(void) } if (nr_counters == 1 || !display_weighted) { - printf("%Ld", (u64)attrs[0].sample_period); + struct perf_evsel *first; + first = list_entry(evsel_list.next, struct perf_evsel, node); + printf("%Ld", first->attr.sample_period); if (freq) printf("Hz "); else @@ -540,9 +545,9 @@ static void print_sym_table(void) } if (!display_weighted) - printf("%s", event_name(sym_counter)); - else for (counter = 0; counter < nr_counters; counter++) { - if (counter) + printf("%s", event_name(sym_evsel)); + else list_for_each_entry(counter, &evsel_list, node) { + if (counter->idx) printf("/"); printf("%s", event_name(counter)); @@ -558,12 +563,12 @@ static void print_sym_table(void) printf(" (all"); if (cpu_list) - printf(", CPU%s: %s)\n", nr_cpus > 1 ? "s" : "", cpu_list); + printf(", CPU%s: %s)\n", cpus->nr > 1 ? "s" : "", cpu_list); else { if (target_tid != -1) printf(")\n"); else - printf(", %d CPU%s)\n", nr_cpus, nr_cpus > 1 ? "s" : ""); + printf(", %d CPU%s)\n", cpus->nr, cpus->nr > 1 ? "s" : ""); } printf("%-*.*s\n", win_width, win_width, graph_dotted_line); @@ -739,7 +744,7 @@ static void print_mapped_keys(void) fprintf(stdout, "\t[e] display entries (lines). \t(%d)\n", print_entries); if (nr_counters > 1) - fprintf(stdout, "\t[E] active event counter. \t(%s)\n", event_name(sym_counter)); + fprintf(stdout, "\t[E] active event counter. \t(%s)\n", event_name(sym_evsel)); fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", count_filter); @@ -826,19 +831,23 @@ static void handle_keypress(struct perf_session *session, int c) break; case 'E': if (nr_counters > 1) { - int i; - fprintf(stderr, "\nAvailable events:"); - for (i = 0; i < nr_counters; i++) - fprintf(stderr, "\n\t%d %s", i, event_name(i)); + + list_for_each_entry(sym_evsel, &evsel_list, node) + fprintf(stderr, "\n\t%d %s", sym_evsel->idx, event_name(sym_evsel)); prompt_integer(&sym_counter, "Enter details event counter"); if (sym_counter >= nr_counters) { - fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(0)); + sym_evsel = list_entry(evsel_list.next, struct perf_evsel, node); sym_counter = 0; + fprintf(stderr, "Sorry, no such event, using %s.\n", event_name(sym_evsel)); sleep(1); + break; } + list_for_each_entry(sym_evsel, &evsel_list, node) + if (sym_evsel->idx == sym_counter) + break; } else sym_counter = 0; break; case 'f': @@ -977,12 +986,13 @@ static int symbol_filter(struct map *map, struct symbol *sym) } static void event__process_sample(const event_t *self, - struct perf_session *session, int counter) + struct sample_data *sample, + struct perf_session *session, + struct perf_evsel *evsel) { u64 ip = self->ip.ip; struct sym_entry *syme; struct addr_location al; - struct sample_data data; struct machine *machine; u8 origin = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; @@ -1025,7 +1035,7 @@ static void event__process_sample(const event_t *self, if (self->header.misc & PERF_RECORD_MISC_EXACT_IP) exact_samples++; - if (event__preprocess_sample(self, session, &al, &data, + if (event__preprocess_sample(self, session, &al, sample, symbol_filter) < 0 || al.filtered) return; @@ -1071,9 +1081,9 @@ static void event__process_sample(const event_t *self, syme = symbol__priv(al.sym); if (!syme->skip) { - syme->count[counter]++; + syme->count[evsel->idx]++; syme->origin = origin; - record_precise_ip(syme, counter, ip); + record_precise_ip(syme, evsel->idx, ip); pthread_mutex_lock(&active_symbols_lock); if (list_empty(&syme->node) || !syme->node.next) __list_insert_active_sym(syme); @@ -1082,12 +1092,24 @@ static void event__process_sample(const event_t *self, } struct mmap_data { - int counter; void *base; int mask; unsigned int prev; }; +static int perf_evsel__alloc_mmap_per_thread(struct perf_evsel *evsel, + int ncpus, int nthreads) +{ + evsel->priv = xyarray__new(ncpus, nthreads, sizeof(struct mmap_data)); + return evsel->priv != NULL ? 0 : -ENOMEM; +} + +static void perf_evsel__free_mmap(struct perf_evsel *evsel) +{ + xyarray__delete(evsel->priv); + evsel->priv = NULL; +} + static unsigned int mmap_read_head(struct mmap_data *md) { struct perf_event_mmap_page *pc = md->base; @@ -1100,11 +1122,15 @@ static unsigned int mmap_read_head(struct mmap_data *md) } static void perf_session__mmap_read_counter(struct perf_session *self, - struct mmap_data *md) + struct perf_evsel *evsel, + int cpu, int thread_idx) { + struct xyarray *mmap_array = evsel->priv; + struct mmap_data *md = xyarray__entry(mmap_array, cpu, thread_idx); unsigned int head = mmap_read_head(md); unsigned int old = md->prev; unsigned char *data = md->base + page_size; + struct sample_data sample; int diff; /* @@ -1152,10 +1178,11 @@ static void perf_session__mmap_read_counter(struct perf_session *self, event = &event_copy; } + event__parse_sample(event, self, &sample); if (event->header.type == PERF_RECORD_SAMPLE) - event__process_sample(event, self, md->counter); + event__process_sample(event, &sample, self, evsel); else - event__process(event, self); + event__process(event, &sample, self); old += size; } @@ -1163,36 +1190,39 @@ static void perf_session__mmap_read_counter(struct perf_session *self, } static struct pollfd *event_array; -static struct mmap_data *mmap_array[MAX_NR_CPUS][MAX_COUNTERS]; static void perf_session__mmap_read(struct perf_session *self) { - int i, counter, thread_index; + struct perf_evsel *counter; + int i, thread_index; - for (i = 0; i < nr_cpus; i++) { - for (counter = 0; counter < nr_counters; counter++) + for (i = 0; i < cpus->nr; i++) { + list_for_each_entry(counter, &evsel_list, node) { for (thread_index = 0; - thread_index < thread_num; + thread_index < threads->nr; thread_index++) { perf_session__mmap_read_counter(self, - &mmap_array[i][counter][thread_index]); + counter, i, thread_index); } + } } } int nr_poll; int group_fd; -static void start_counter(int i, int counter) +static void start_counter(int i, struct perf_evsel *evsel) { + struct xyarray *mmap_array = evsel->priv; + struct mmap_data *mm; struct perf_event_attr *attr; int cpu = -1; int thread_index; if (target_tid == -1) - cpu = cpumap[i]; + cpu = cpus->map[i]; - attr = attrs + counter; + attr = &evsel->attr; attr->sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_TID; @@ -1205,16 +1235,18 @@ static void start_counter(int i, int counter) attr->inherit = (cpu < 0) && inherit; attr->mmap = 1; - for (thread_index = 0; thread_index < thread_num; thread_index++) { + for (thread_index = 0; thread_index < threads->nr; thread_index++) { try_again: - fd[i][counter][thread_index] = sys_perf_event_open(attr, - all_tids[thread_index], cpu, group_fd, 0); + FD(evsel, i, thread_index) = sys_perf_event_open(attr, + threads->map[thread_index], cpu, group_fd, 0); - if (fd[i][counter][thread_index] < 0) { + if (FD(evsel, i, thread_index) < 0) { int err = errno; if (err == EPERM || err == EACCES) - die("No permission - are you root?\n"); + die("Permission error - are you root?\n" + "\t Consider tweaking" + " /proc/sys/kernel/perf_event_paranoid.\n"); /* * If it's cycles then fall back to hrtimer * based cpu-clock-tick sw counter, which @@ -1231,30 +1263,30 @@ try_again: goto try_again; } printf("\n"); - error("perfcounter syscall returned with %d (%s)\n", - fd[i][counter][thread_index], strerror(err)); + error("sys_perf_event_open() syscall returned with %d (%s). /bin/dmesg may provide additional information.\n", + FD(evsel, i, thread_index), strerror(err)); die("No CONFIG_PERF_EVENTS=y kernel support configured?\n"); exit(-1); } - assert(fd[i][counter][thread_index] >= 0); - fcntl(fd[i][counter][thread_index], F_SETFL, O_NONBLOCK); + assert(FD(evsel, i, thread_index) >= 0); + fcntl(FD(evsel, i, thread_index), F_SETFL, O_NONBLOCK); /* * First counter acts as the group leader: */ if (group && group_fd == -1) - group_fd = fd[i][counter][thread_index]; + group_fd = FD(evsel, i, thread_index); - event_array[nr_poll].fd = fd[i][counter][thread_index]; + event_array[nr_poll].fd = FD(evsel, i, thread_index); event_array[nr_poll].events = POLLIN; nr_poll++; - mmap_array[i][counter][thread_index].counter = counter; - mmap_array[i][counter][thread_index].prev = 0; - mmap_array[i][counter][thread_index].mask = mmap_pages*page_size - 1; - mmap_array[i][counter][thread_index].base = mmap(NULL, (mmap_pages+1)*page_size, - PROT_READ, MAP_SHARED, fd[i][counter][thread_index], 0); - if (mmap_array[i][counter][thread_index].base == MAP_FAILED) + mm = xyarray__entry(mmap_array, i, thread_index); + mm->prev = 0; + mm->mask = mmap_pages*page_size - 1; + mm->base = mmap(NULL, (mmap_pages+1)*page_size, + PROT_READ, MAP_SHARED, FD(evsel, i, thread_index), 0); + if (mm->base == MAP_FAILED) die("failed to mmap with %d (%s)\n", errno, strerror(errno)); } } @@ -1262,13 +1294,13 @@ try_again: static int __cmd_top(void) { pthread_t thread; - int i, counter; - int ret; + struct perf_evsel *counter; + int i, ret; /* * FIXME: perf_session__new should allow passing a O_MMAP, so that all this * mmap reading, etc is encapsulated in it. Use O_WRONLY for now. */ - struct perf_session *session = perf_session__new(NULL, O_WRONLY, false, false); + struct perf_session *session = perf_session__new(NULL, O_WRONLY, false, false, NULL); if (session == NULL) return -ENOMEM; @@ -1277,9 +1309,9 @@ static int __cmd_top(void) else event__synthesize_threads(event__process, session); - for (i = 0; i < nr_cpus; i++) { + for (i = 0; i < cpus->nr; i++) { group_fd = -1; - for (counter = 0; counter < nr_counters; counter++) + list_for_each_entry(counter, &evsel_list, node) start_counter(i, counter); } @@ -1368,8 +1400,8 @@ static const struct option options[] = { int cmd_top(int argc, const char **argv, const char *prefix __used) { - int counter; - int i,j; + struct perf_evsel *pos; + int status = -ENOMEM; page_size = sysconf(_SC_PAGE_SIZE); @@ -1377,34 +1409,17 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) if (argc) usage_with_options(top_usage, options); - if (target_pid != -1) { + if (target_pid != -1) target_tid = target_pid; - thread_num = find_all_tid(target_pid, &all_tids); - if (thread_num <= 0) { - fprintf(stderr, "Can't find all threads of pid %d\n", - target_pid); - usage_with_options(top_usage, options); - } - } else { - all_tids=malloc(sizeof(pid_t)); - if (!all_tids) - return -ENOMEM; - all_tids[0] = target_tid; - thread_num = 1; + threads = thread_map__new(target_pid, target_tid); + if (threads == NULL) { + pr_err("Problems finding threads of monitor\n"); + usage_with_options(top_usage, options); } - for (i = 0; i < MAX_NR_CPUS; i++) { - for (j = 0; j < MAX_COUNTERS; j++) { - fd[i][j] = malloc(sizeof(int)*thread_num); - mmap_array[i][j] = zalloc( - sizeof(struct mmap_data)*thread_num); - if (!fd[i][j] || !mmap_array[i][j]) - return -ENOMEM; - } - } - event_array = malloc( - sizeof(struct pollfd)*MAX_NR_CPUS*MAX_COUNTERS*thread_num); + event_array = malloc((sizeof(struct pollfd) * + MAX_NR_CPUS * MAX_COUNTERS * threads->nr)); if (!event_array) return -ENOMEM; @@ -1415,15 +1430,10 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) cpu_list = NULL; } - if (!nr_counters) - nr_counters = 1; - - symbol_conf.priv_size = (sizeof(struct sym_entry) + - (nr_counters + 1) * sizeof(unsigned long)); - - symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL); - if (symbol__init() < 0) - return -1; + if (!nr_counters && perf_evsel_list__create_default() < 0) { + pr_err("Not enough memory for event selector list\n"); + return -ENOMEM; + } if (delay_secs < 1) delay_secs = 1; @@ -1440,23 +1450,33 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) exit(EXIT_FAILURE); } - /* - * Fill in the ones not specifically initialized via -c: - */ - for (counter = 0; counter < nr_counters; counter++) { - if (attrs[counter].sample_period) + if (target_tid != -1) + cpus = cpu_map__dummy_new(); + else + cpus = cpu_map__new(cpu_list); + + if (cpus == NULL) + usage_with_options(top_usage, options); + + list_for_each_entry(pos, &evsel_list, node) { + if (perf_evsel__alloc_mmap_per_thread(pos, cpus->nr, threads->nr) < 0 || + perf_evsel__alloc_fd(pos, cpus->nr, threads->nr) < 0) + goto out_free_fd; + /* + * Fill in the ones not specifically initialized via -c: + */ + if (pos->attr.sample_period) continue; - attrs[counter].sample_period = default_interval; + pos->attr.sample_period = default_interval; } - if (target_tid != -1) - nr_cpus = 1; - else - nr_cpus = read_cpu_map(cpu_list); + symbol_conf.priv_size = (sizeof(struct sym_entry) + + (nr_counters + 1) * sizeof(unsigned long)); - if (nr_cpus < 1) - usage_with_options(top_usage, options); + symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL); + if (symbol__init() < 0) + return -1; get_term_dimensions(&winsize); if (print_entries == 0) { @@ -1464,5 +1484,10 @@ int cmd_top(int argc, const char **argv, const char *prefix __used) signal(SIGWINCH, sig_winch_handler); } - return __cmd_top(); + status = __cmd_top(); +out_free_fd: + list_for_each_entry(pos, &evsel_list, node) + perf_evsel__free_mmap(pos); + + return status; } diff --git a/tools/perf/builtin.h b/tools/perf/builtin.h index 921245b28583..c7798c7f24ed 100644 --- a/tools/perf/builtin.h +++ b/tools/perf/builtin.h @@ -27,7 +27,7 @@ extern int cmd_report(int argc, const char **argv, const char *prefix); extern int cmd_stat(int argc, const char **argv, const char *prefix); extern int cmd_timechart(int argc, const char **argv, const char *prefix); extern int cmd_top(int argc, const char **argv, const char *prefix); -extern int cmd_trace(int argc, const char **argv, const char *prefix); +extern int cmd_script(int argc, const char **argv, const char *prefix); extern int cmd_version(int argc, const char **argv, const char *prefix); extern int cmd_probe(int argc, const char **argv, const char *prefix); extern int cmd_kmem(int argc, const char **argv, const char *prefix); diff --git a/tools/perf/command-list.txt b/tools/perf/command-list.txt index 949d77fc0b97..16b5088cf8f4 100644 --- a/tools/perf/command-list.txt +++ b/tools/perf/command-list.txt @@ -16,7 +16,7 @@ perf-report mainporcelain common perf-stat mainporcelain common perf-timechart mainporcelain common perf-top mainporcelain common -perf-trace mainporcelain common +perf-script mainporcelain common perf-probe mainporcelain common perf-kmem mainporcelain common perf-lock mainporcelain common diff --git a/tools/perf/feature-tests.mak b/tools/perf/feature-tests.mak index b253db634f04..b041ca67a2cb 100644 --- a/tools/perf/feature-tests.mak +++ b/tools/perf/feature-tests.mak @@ -9,8 +9,8 @@ endef ifndef NO_DWARF define SOURCE_DWARF #include <dwarf.h> -#include <libdw.h> -#include <version.h> +#include <elfutils/libdw.h> +#include <elfutils/version.h> #ifndef _ELFUTILS_PREREQ #error #endif diff --git a/tools/perf/perf.c b/tools/perf/perf.c index cdd6c03f1e14..5b1ecd66bb36 100644 --- a/tools/perf/perf.c +++ b/tools/perf/perf.c @@ -286,6 +286,8 @@ static int run_builtin(struct cmd_struct *p, int argc, const char **argv) status = p->fn(argc, argv, prefix); exit_browser(status); + perf_evsel_list__delete(); + if (status) return status & 0xff; @@ -323,7 +325,7 @@ static void handle_internal_command(int argc, const char **argv) { "top", cmd_top, 0 }, { "annotate", cmd_annotate, 0 }, { "version", cmd_version, 0 }, - { "trace", cmd_trace, 0 }, + { "script", cmd_script, 0 }, { "sched", cmd_sched, 0 }, { "probe", cmd_probe, 0 }, { "kmem", cmd_kmem, 0 }, diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.c b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c index 01a64ad693f2..790ceba6ad3f 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/Context.c +++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.c @@ -8,7 +8,7 @@ #line 1 "Context.xs" /* - * Context.xs. XS interfaces for perf trace. + * Context.xs. XS interfaces for perf script. * * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> * diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs index 549cf0467d30..c1e2ed1ed34e 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs +++ b/tools/perf/scripts/perl/Perf-Trace-Util/Context.xs @@ -1,5 +1,5 @@ /* - * Context.xs. XS interfaces for perf trace. + * Context.xs. XS interfaces for perf script. * * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> * @@ -23,7 +23,7 @@ #include "perl.h" #include "XSUB.h" #include "../../../perf.h" -#include "../../../util/trace-event.h" +#include "../../../util/script-event.h" MODULE = Perf::Trace::Context PACKAGE = Perf::Trace::Context PROTOTYPES: ENABLE diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/README b/tools/perf/scripts/perl/Perf-Trace-Util/README index 9a9707630791..2f0c7f3043ee 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/README +++ b/tools/perf/scripts/perl/Perf-Trace-Util/README @@ -1,7 +1,7 @@ Perf-Trace-Util version 0.01 ============================ -This module contains utility functions for use with perf trace. +This module contains utility functions for use with perf script. Core.pm and Util.pm are pure Perl modules; Core.pm contains routines that the core perf support for Perl calls on and should always be @@ -33,7 +33,7 @@ After you do that: INSTALLATION -Building perf with perf trace Perl scripting should install this +Building perf with perf script Perl scripting should install this module in the right place. You should make sure libperl and ExtUtils/Embed.pm are installed first diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm index 6c7f3659cb17..4e2f6039ac92 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Context.pm @@ -34,7 +34,7 @@ Perf::Trace::Context - Perl extension for accessing functions in perf. =head1 SEE ALSO -Perf (trace) documentation +Perf (script) documentation =head1 AUTHOR diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm index 9df376a9f629..9158458d3eeb 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Core.pm @@ -163,7 +163,7 @@ sub dump_symbolic_fields __END__ =head1 NAME -Perf::Trace::Core - Perl extension for perf trace +Perf::Trace::Core - Perl extension for perf script =head1 SYNOPSIS @@ -171,7 +171,7 @@ Perf::Trace::Core - Perl extension for perf trace =head1 SEE ALSO -Perf (trace) documentation +Perf (script) documentation =head1 AUTHOR diff --git a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm index d94b40c8ac85..053500114625 100644 --- a/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm +++ b/tools/perf/scripts/perl/Perf-Trace-Util/lib/Perf/Trace/Util.pm @@ -65,7 +65,7 @@ sub clear_term __END__ =head1 NAME -Perf::Trace::Util - Perl extension for perf trace +Perf::Trace::Util - Perl extension for perf script =head1 SYNOPSIS @@ -73,7 +73,7 @@ Perf::Trace::Util - Perl extension for perf trace =head1 SEE ALSO -Perf (trace) documentation +Perf (script) documentation =head1 AUTHOR diff --git a/tools/perf/scripts/perl/bin/failed-syscalls-report b/tools/perf/scripts/perl/bin/failed-syscalls-report index 4028d92dc4ae..9f83cc1ad8ba 100644 --- a/tools/perf/scripts/perl/bin/failed-syscalls-report +++ b/tools/perf/scripts/perl/bin/failed-syscalls-report @@ -7,4 +7,4 @@ if [ $# -gt 0 ] ; then shift fi fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/failed-syscalls.pl $comm +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/failed-syscalls.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-file-report b/tools/perf/scripts/perl/bin/rw-by-file-report index ba25f4d41fb0..77200b3f3100 100644 --- a/tools/perf/scripts/perl/bin/rw-by-file-report +++ b/tools/perf/scripts/perl/bin/rw-by-file-report @@ -7,7 +7,4 @@ if [ $# -lt 1 ] ; then fi comm=$1 shift -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-file.pl $comm - - - +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-file.pl $comm diff --git a/tools/perf/scripts/perl/bin/rw-by-pid-report b/tools/perf/scripts/perl/bin/rw-by-pid-report index 641a3f5d085c..a27b9f311f95 100644 --- a/tools/perf/scripts/perl/bin/rw-by-pid-report +++ b/tools/perf/scripts/perl/bin/rw-by-pid-report @@ -1,6 +1,3 @@ #!/bin/bash # description: system-wide r/w activity -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-pid.pl - - - +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rw-by-pid.pl diff --git a/tools/perf/scripts/perl/bin/rwtop-report b/tools/perf/scripts/perl/bin/rwtop-report index 4918dba77021..83e11ec2e190 100644 --- a/tools/perf/scripts/perl/bin/rwtop-report +++ b/tools/perf/scripts/perl/bin/rwtop-report @@ -17,7 +17,4 @@ if [ "$n_args" -gt 0 ] ; then interval=$1 shift fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/rwtop.pl $interval - - - +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/rwtop.pl $interval diff --git a/tools/perf/scripts/perl/bin/wakeup-latency-report b/tools/perf/scripts/perl/bin/wakeup-latency-report index 49052ebcb632..889e8130cca5 100644 --- a/tools/perf/scripts/perl/bin/wakeup-latency-report +++ b/tools/perf/scripts/perl/bin/wakeup-latency-report @@ -1,6 +1,3 @@ #!/bin/bash # description: system-wide min/max/avg wakeup latency -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/wakeup-latency.pl - - - +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/wakeup-latency.pl diff --git a/tools/perf/scripts/perl/bin/workqueue-stats-report b/tools/perf/scripts/perl/bin/workqueue-stats-report index df0c65f4ca93..6d91411d248c 100644 --- a/tools/perf/scripts/perl/bin/workqueue-stats-report +++ b/tools/perf/scripts/perl/bin/workqueue-stats-report @@ -1,7 +1,3 @@ #!/bin/bash # description: workqueue stats (ins/exe/create/destroy) -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/perl/workqueue-stats.pl - - - - +perf script $@ -s "$PERF_EXEC_PATH"/scripts/perl/workqueue-stats.pl diff --git a/tools/perf/scripts/perl/check-perf-trace.pl b/tools/perf/scripts/perl/check-perf-trace.pl index 4e7dc0a407a5..4e7076c20616 100644 --- a/tools/perf/scripts/perl/check-perf-trace.pl +++ b/tools/perf/scripts/perl/check-perf-trace.pl @@ -1,4 +1,4 @@ -# perf trace event handlers, generated by perf trace -g perl +# perf script event handlers, generated by perf script -g perl # (c) 2009, Tom Zanussi <tzanussi@gmail.com> # Licensed under the terms of the GNU GPL License version 2 diff --git a/tools/perf/scripts/perl/rw-by-file.pl b/tools/perf/scripts/perl/rw-by-file.pl index 2a39097687b9..74844ee2be3e 100644 --- a/tools/perf/scripts/perl/rw-by-file.pl +++ b/tools/perf/scripts/perl/rw-by-file.pl @@ -18,7 +18,7 @@ use lib "./Perf-Trace-Util/lib"; use Perf::Trace::Core; use Perf::Trace::Util; -my $usage = "perf trace -s rw-by-file.pl <comm>\n"; +my $usage = "perf script -s rw-by-file.pl <comm>\n"; my $for_comm = shift or die $usage; diff --git a/tools/perf/scripts/perl/workqueue-stats.pl b/tools/perf/scripts/perl/workqueue-stats.pl index b84b12699b70..a8eaff5119e0 100644 --- a/tools/perf/scripts/perl/workqueue-stats.pl +++ b/tools/perf/scripts/perl/workqueue-stats.pl @@ -10,7 +10,7 @@ # workqueue:workqueue_destruction -e workqueue:workqueue_execution # -e workqueue:workqueue_insertion # -# perf trace -p -s tools/perf/scripts/perl/workqueue-stats.pl +# perf script -p -s tools/perf/scripts/perl/workqueue-stats.pl use 5.010000; use strict; diff --git a/tools/perf/scripts/python/Perf-Trace-Util/Context.c b/tools/perf/scripts/python/Perf-Trace-Util/Context.c index 957085dd5d8d..315067b8f552 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/Context.c +++ b/tools/perf/scripts/python/Perf-Trace-Util/Context.c @@ -1,5 +1,5 @@ /* - * Context.c. Python interfaces for perf trace. + * Context.c. Python interfaces for perf script. * * Copyright (C) 2010 Tom Zanussi <tzanussi@gmail.com> * diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py index aad7525bca1d..de7211e4fa47 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py @@ -1,4 +1,4 @@ -# Core.py - Python extension for perf trace, core functions +# Core.py - Python extension for perf script, core functions # # Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com> # diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py index ae9a56e43e05..fdd92f699055 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/SchedGui.py @@ -1,4 +1,4 @@ -# SchedGui.py - Python extension for perf trace, basic GUI code for +# SchedGui.py - Python extension for perf script, basic GUI code for # traces drawing and overview. # # Copyright (C) 2010 by Frederic Weisbecker <fweisbec@gmail.com> diff --git a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py index 13cc02b5893a..15c8400240fd 100644 --- a/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py +++ b/tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Util.py @@ -1,4 +1,4 @@ -# Util.py - Python extension for perf trace, miscellaneous utility code +# Util.py - Python extension for perf script, miscellaneous utility code # # Copyright (C) 2010 by Tom Zanussi <tzanussi@gmail.com> # diff --git a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report index 03587021463d..fda5096d0cbf 100644 --- a/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report +++ b/tools/perf/scripts/python/bin/failed-syscalls-by-pid-report @@ -7,4 +7,4 @@ if [ $# -gt 0 ] ; then shift fi fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/failed-syscalls-by-pid.py $comm +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/failed-syscalls-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/futex-contention-report b/tools/perf/scripts/python/bin/futex-contention-report index c8268138fb7e..6c44271091ab 100644 --- a/tools/perf/scripts/python/bin/futex-contention-report +++ b/tools/perf/scripts/python/bin/futex-contention-report @@ -1,4 +1,4 @@ #!/bin/bash # description: futext contention measurement -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/futex-contention.py +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/futex-contention.py diff --git a/tools/perf/scripts/python/bin/netdev-times-report b/tools/perf/scripts/python/bin/netdev-times-report index 4ad361b31249..8f759291da86 100644 --- a/tools/perf/scripts/python/bin/netdev-times-report +++ b/tools/perf/scripts/python/bin/netdev-times-report @@ -2,4 +2,4 @@ # description: display a process of packet and processing time # args: [tx] [rx] [dev=] [debug] -perf trace -s "$PERF_EXEC_PATH"/scripts/python/netdev-times.py $@ +perf script -s "$PERF_EXEC_PATH"/scripts/python/netdev-times.py $@ diff --git a/tools/perf/scripts/python/bin/sched-migration-report b/tools/perf/scripts/python/bin/sched-migration-report index df1791f07c24..68b037a1849b 100644 --- a/tools/perf/scripts/python/bin/sched-migration-report +++ b/tools/perf/scripts/python/bin/sched-migration-report @@ -1,3 +1,3 @@ #!/bin/bash # description: sched migration overview -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/sched-migration.py +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sched-migration.py diff --git a/tools/perf/scripts/python/bin/sctop-report b/tools/perf/scripts/python/bin/sctop-report index 36b409c05e50..c32db294124d 100644 --- a/tools/perf/scripts/python/bin/sctop-report +++ b/tools/perf/scripts/python/bin/sctop-report @@ -21,4 +21,4 @@ elif [ "$n_args" -gt 0 ] ; then interval=$1 shift fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/sctop.py $comm $interval +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/sctop.py $comm $interval diff --git a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report index 4eb88c9fc83c..16eb8d65c543 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-by-pid-report +++ b/tools/perf/scripts/python/bin/syscall-counts-by-pid-report @@ -7,4 +7,4 @@ if [ $# -gt 0 ] ; then shift fi fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts-by-pid.py $comm +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts-by-pid.py $comm diff --git a/tools/perf/scripts/python/bin/syscall-counts-report b/tools/perf/scripts/python/bin/syscall-counts-report index cb2f9c5cf17e..0f0e9d453bb4 100644 --- a/tools/perf/scripts/python/bin/syscall-counts-report +++ b/tools/perf/scripts/python/bin/syscall-counts-report @@ -7,4 +7,4 @@ if [ $# -gt 0 ] ; then shift fi fi -perf trace $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts.py $comm +perf script $@ -s "$PERF_EXEC_PATH"/scripts/python/syscall-counts.py $comm diff --git a/tools/perf/scripts/python/check-perf-trace.py b/tools/perf/scripts/python/check-perf-trace.py index d9f7893e315c..4647a7694cf6 100644 --- a/tools/perf/scripts/python/check-perf-trace.py +++ b/tools/perf/scripts/python/check-perf-trace.py @@ -1,4 +1,4 @@ -# perf trace event handlers, generated by perf trace -g python +# perf script event handlers, generated by perf script -g python # (c) 2010, Tom Zanussi <tzanussi@gmail.com> # Licensed under the terms of the GNU GPL License version 2 # diff --git a/tools/perf/scripts/python/failed-syscalls-by-pid.py b/tools/perf/scripts/python/failed-syscalls-by-pid.py index acd7848717b3..85805fac4116 100644 --- a/tools/perf/scripts/python/failed-syscalls-by-pid.py +++ b/tools/perf/scripts/python/failed-syscalls-by-pid.py @@ -15,7 +15,7 @@ from perf_trace_context import * from Core import * from Util import * -usage = "perf trace -s syscall-counts-by-pid.py [comm|pid]\n"; +usage = "perf script -s syscall-counts-by-pid.py [comm|pid]\n"; for_comm = None for_pid = None diff --git a/tools/perf/scripts/python/sched-migration.py b/tools/perf/scripts/python/sched-migration.py index b934383c3364..74d55ec08aed 100644 --- a/tools/perf/scripts/python/sched-migration.py +++ b/tools/perf/scripts/python/sched-migration.py @@ -4,7 +4,7 @@ # # Copyright (C) 2010 Frederic Weisbecker <fweisbec@gmail.com> # -# perf trace event handlers have been generated by perf trace -g python +# perf script event handlers have been generated by perf script -g python # # This software is distributed under the terms of the GNU General # Public License ("GPL") version 2 as published by the Free Software diff --git a/tools/perf/scripts/python/sctop.py b/tools/perf/scripts/python/sctop.py index 7a6ec2c7d8ab..42c267e292fa 100644 --- a/tools/perf/scripts/python/sctop.py +++ b/tools/perf/scripts/python/sctop.py @@ -17,7 +17,7 @@ from perf_trace_context import * from Core import * from Util import * -usage = "perf trace -s sctop.py [comm] [interval]\n"; +usage = "perf script -s sctop.py [comm] [interval]\n"; for_comm = None default_interval = 3 diff --git a/tools/perf/scripts/python/syscall-counts-by-pid.py b/tools/perf/scripts/python/syscall-counts-by-pid.py index d1ee3ec10cf2..c64d1c55d745 100644 --- a/tools/perf/scripts/python/syscall-counts-by-pid.py +++ b/tools/perf/scripts/python/syscall-counts-by-pid.py @@ -14,7 +14,7 @@ from perf_trace_context import * from Core import * from Util import syscall_name -usage = "perf trace -s syscall-counts-by-pid.py [comm]\n"; +usage = "perf script -s syscall-counts-by-pid.py [comm]\n"; for_comm = None for_pid = None diff --git a/tools/perf/scripts/python/syscall-counts.py b/tools/perf/scripts/python/syscall-counts.py index ea183dc82d29..b435d3f188e8 100644 --- a/tools/perf/scripts/python/syscall-counts.py +++ b/tools/perf/scripts/python/syscall-counts.py @@ -15,7 +15,7 @@ from perf_trace_context import * from Core import * from Util import syscall_name -usage = "perf trace -s syscall-counts.py [comm]\n"; +usage = "perf script -s syscall-counts.py [comm]\n"; for_comm = None diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c index e437edb72417..deffb8c96071 100644 --- a/tools/perf/util/build-id.c +++ b/tools/perf/util/build-id.c @@ -14,7 +14,9 @@ #include <linux/kernel.h> #include "debug.h" -static int build_id__mark_dso_hit(event_t *event, struct perf_session *session) +static int build_id__mark_dso_hit(event_t *event, + struct sample_data *sample __used, + struct perf_session *session) { struct addr_location al; u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; @@ -35,7 +37,8 @@ static int build_id__mark_dso_hit(event_t *event, struct perf_session *session) return 0; } -static int event__exit_del_thread(event_t *self, struct perf_session *session) +static int event__exit_del_thread(event_t *self, struct sample_data *sample __used, + struct perf_session *session) { struct thread *thread = perf_session__findnew(session, self->fork.tid); diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 0f9b8d7a7d7e..3ccaa1043383 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -4,32 +4,53 @@ #include <assert.h> #include <stdio.h> -int cpumap[MAX_NR_CPUS]; - -static int default_cpu_map(void) +static struct cpu_map *cpu_map__default_new(void) { - int nr_cpus, i; + struct cpu_map *cpus; + int nr_cpus; nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); - assert(nr_cpus <= MAX_NR_CPUS); - assert((int)nr_cpus >= 0); + if (nr_cpus < 0) + return NULL; + + cpus = malloc(sizeof(*cpus) + nr_cpus * sizeof(int)); + if (cpus != NULL) { + int i; + for (i = 0; i < nr_cpus; ++i) + cpus->map[i] = i; - for (i = 0; i < nr_cpus; ++i) - cpumap[i] = i; + cpus->nr = nr_cpus; + } - return nr_cpus; + return cpus; } -static int read_all_cpu_map(void) +static struct cpu_map *cpu_map__trim_new(int nr_cpus, int *tmp_cpus) { + size_t payload_size = nr_cpus * sizeof(int); + struct cpu_map *cpus = malloc(sizeof(*cpus) + payload_size); + + if (cpus != NULL) { + cpus->nr = nr_cpus; + memcpy(cpus->map, tmp_cpus, payload_size); + } + + return cpus; +} + +static struct cpu_map *cpu_map__read_all_cpu_map(void) +{ + struct cpu_map *cpus = NULL; FILE *onlnf; int nr_cpus = 0; + int *tmp_cpus = NULL, *tmp; + int max_entries = 0; int n, cpu, prev; char sep; onlnf = fopen("/sys/devices/system/cpu/online", "r"); if (!onlnf) - return default_cpu_map(); + return cpu_map__default_new(); sep = 0; prev = -1; @@ -38,12 +59,28 @@ static int read_all_cpu_map(void) if (n <= 0) break; if (prev >= 0) { - assert(nr_cpus + cpu - prev - 1 < MAX_NR_CPUS); + int new_max = nr_cpus + cpu - prev - 1; + + if (new_max >= max_entries) { + max_entries = new_max + MAX_NR_CPUS / 2; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto out_free_tmp; + tmp_cpus = tmp; + } + while (++prev < cpu) - cpumap[nr_cpus++] = prev; + tmp_cpus[nr_cpus++] = prev; + } + if (nr_cpus == max_entries) { + max_entries += MAX_NR_CPUS; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto out_free_tmp; + tmp_cpus = tmp; } - assert (nr_cpus < MAX_NR_CPUS); - cpumap[nr_cpus++] = cpu; + + tmp_cpus[nr_cpus++] = cpu; if (n == 2 && sep == '-') prev = cpu; else @@ -51,24 +88,31 @@ static int read_all_cpu_map(void) if (n == 1 || sep == '\n') break; } - fclose(onlnf); - if (nr_cpus > 0) - return nr_cpus; - return default_cpu_map(); + if (nr_cpus > 0) + cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); + else + cpus = cpu_map__default_new(); +out_free_tmp: + free(tmp_cpus); + fclose(onlnf); + return cpus; } -int read_cpu_map(const char *cpu_list) +struct cpu_map *cpu_map__new(const char *cpu_list) { + struct cpu_map *cpus = NULL; unsigned long start_cpu, end_cpu = 0; char *p = NULL; int i, nr_cpus = 0; + int *tmp_cpus = NULL, *tmp; + int max_entries = 0; if (!cpu_list) - return read_all_cpu_map(); + return cpu_map__read_all_cpu_map(); if (!isdigit(*cpu_list)) - goto invalid; + goto out; while (isdigit(*cpu_list)) { p = NULL; @@ -94,21 +138,42 @@ int read_cpu_map(const char *cpu_list) for (; start_cpu <= end_cpu; start_cpu++) { /* check for duplicates */ for (i = 0; i < nr_cpus; i++) - if (cpumap[i] == (int)start_cpu) + if (tmp_cpus[i] == (int)start_cpu) goto invalid; - assert(nr_cpus < MAX_NR_CPUS); - cpumap[nr_cpus++] = (int)start_cpu; + if (nr_cpus == max_entries) { + max_entries += MAX_NR_CPUS; + tmp = realloc(tmp_cpus, max_entries * sizeof(int)); + if (tmp == NULL) + goto invalid; + tmp_cpus = tmp; + } + tmp_cpus[nr_cpus++] = (int)start_cpu; } if (*p) ++p; cpu_list = p; } - if (nr_cpus > 0) - return nr_cpus; - return default_cpu_map(); + if (nr_cpus > 0) + cpus = cpu_map__trim_new(nr_cpus, tmp_cpus); + else + cpus = cpu_map__default_new(); invalid: - return -1; + free(tmp_cpus); +out: + return cpus; +} + +struct cpu_map *cpu_map__dummy_new(void) +{ + struct cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int)); + + if (cpus != NULL) { + cpus->nr = 1; + cpus->map[0] = -1; + } + + return cpus; } diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index 3e60f56e490e..f7a4f42f6307 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -1,7 +1,13 @@ #ifndef __PERF_CPUMAP_H #define __PERF_CPUMAP_H -extern int read_cpu_map(const char *cpu_list); -extern int cpumap[]; +struct cpu_map { + int nr; + int map[]; +}; + +struct cpu_map *cpu_map__new(const char *cpu_list); +struct cpu_map *cpu_map__dummy_new(void); +void *cpu_map__delete(struct cpu_map *map); #endif /* __PERF_CPUMAP_H */ diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c index c8d81b00089d..01bbe8ecec3f 100644 --- a/tools/perf/util/debug.c +++ b/tools/perf/util/debug.c @@ -46,20 +46,16 @@ int dump_printf(const char *fmt, ...) return ret; } -static int dump_printf_color(const char *fmt, const char *color, ...) +#ifdef NO_NEWT_SUPPORT +void ui__warning(const char *format, ...) { va_list args; - int ret = 0; - if (dump_trace) { - va_start(args, color); - ret = color_vfprintf(stdout, color, fmt, args); - va_end(args); - } - - return ret; + va_start(args, format); + vfprintf(stderr, format, args); + va_end(args); } - +#endif void trace_event(event_t *event) { @@ -70,29 +66,29 @@ void trace_event(event_t *event) if (!dump_trace) return; - dump_printf("."); - dump_printf_color("\n. ... raw event: size %d bytes\n", color, - event->header.size); + printf("."); + color_fprintf(stdout, color, "\n. ... raw event: size %d bytes\n", + event->header.size); for (i = 0; i < event->header.size; i++) { if ((i & 15) == 0) { - dump_printf("."); - dump_printf_color(" %04x: ", color, i); + printf("."); + color_fprintf(stdout, color, " %04x: ", i); } - dump_printf_color(" %02x", color, raw_event[i]); + color_fprintf(stdout, color, " %02x", raw_event[i]); if (((i & 15) == 15) || i == event->header.size-1) { - dump_printf_color(" ", color); + color_fprintf(stdout, color, " "); for (j = 0; j < 15-(i & 15); j++) - dump_printf_color(" ", color); + color_fprintf(stdout, color, " "); for (j = i & ~15; j <= i; j++) { - dump_printf_color("%c", color, - isprint(raw_event[j]) ? - raw_event[j] : '.'); + color_fprintf(stdout, color, "%c", + isprint(raw_event[j]) ? + raw_event[j] : '.'); } - dump_printf_color("\n", color); + color_fprintf(stdout, color, "\n"); } } - dump_printf(".\n"); + printf(".\n"); } diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h index 7b514082bbaf..ca35fd66b5df 100644 --- a/tools/perf/util/debug.h +++ b/tools/perf/util/debug.h @@ -35,4 +35,6 @@ int ui_helpline__show_help(const char *format, va_list ap); #include "ui/progress.h" #endif +void ui__warning(const char *format, ...) __attribute__((format(printf, 1, 2))); + #endif /* __PERF_DEBUG_H */ diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index dab9e754a281..2302ec051bb4 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -7,7 +7,7 @@ #include "strlist.h" #include "thread.h" -const char *event__name[] = { +static const char *event__name[] = { [0] = "TOTAL", [PERF_RECORD_MMAP] = "MMAP", [PERF_RECORD_LOST] = "LOST", @@ -22,13 +22,31 @@ const char *event__name[] = { [PERF_RECORD_HEADER_EVENT_TYPE] = "EVENT_TYPE", [PERF_RECORD_HEADER_TRACING_DATA] = "TRACING_DATA", [PERF_RECORD_HEADER_BUILD_ID] = "BUILD_ID", + [PERF_RECORD_FINISHED_ROUND] = "FINISHED_ROUND", }; -static pid_t event__synthesize_comm(pid_t pid, int full, +const char *event__get_event_name(unsigned int id) +{ + if (id >= ARRAY_SIZE(event__name)) + return "INVALID"; + if (!event__name[id]) + return "UNKNOWN"; + return event__name[id]; +} + +static struct sample_data synth_sample = { + .pid = -1, + .tid = -1, + .time = -1, + .stream_id = -1, + .cpu = -1, + .period = 1, +}; + +static pid_t event__synthesize_comm(event_t *event, pid_t pid, int full, event__handler_t process, struct perf_session *session) { - event_t ev; char filename[PATH_MAX]; char bf[BUFSIZ]; FILE *fp; @@ -49,34 +67,39 @@ out_race: return 0; } - memset(&ev.comm, 0, sizeof(ev.comm)); - while (!ev.comm.comm[0] || !ev.comm.pid) { - if (fgets(bf, sizeof(bf), fp) == NULL) - goto out_failure; + memset(&event->comm, 0, sizeof(event->comm)); + + while (!event->comm.comm[0] || !event->comm.pid) { + if (fgets(bf, sizeof(bf), fp) == NULL) { + pr_warning("couldn't get COMM and pgid, malformed %s\n", filename); + goto out; + } if (memcmp(bf, "Name:", 5) == 0) { char *name = bf + 5; while (*name && isspace(*name)) ++name; size = strlen(name) - 1; - memcpy(ev.comm.comm, name, size++); + memcpy(event->comm.comm, name, size++); } else if (memcmp(bf, "Tgid:", 5) == 0) { char *tgids = bf + 5; while (*tgids && isspace(*tgids)) ++tgids; - tgid = ev.comm.pid = atoi(tgids); + tgid = event->comm.pid = atoi(tgids); } } - ev.comm.header.type = PERF_RECORD_COMM; + event->comm.header.type = PERF_RECORD_COMM; size = ALIGN(size, sizeof(u64)); - ev.comm.header.size = sizeof(ev.comm) - (sizeof(ev.comm.comm) - size); - + memset(event->comm.comm + size, 0, session->id_hdr_size); + event->comm.header.size = (sizeof(event->comm) - + (sizeof(event->comm.comm) - size) + + session->id_hdr_size); if (!full) { - ev.comm.tid = pid; + event->comm.tid = pid; - process(&ev, session); - goto out_fclose; + process(event, &synth_sample, session); + goto out; } snprintf(filename, sizeof(filename), "/proc/%d/task", pid); @@ -91,22 +114,19 @@ out_race: if (*end) continue; - ev.comm.tid = pid; + event->comm.tid = pid; - process(&ev, session); + process(event, &synth_sample, session); } - closedir(tasks); -out_fclose: + closedir(tasks); +out: fclose(fp); - return tgid; -out_failure: - pr_warning("couldn't get COMM and pgid, malformed %s\n", filename); - return -1; + return tgid; } -static int event__synthesize_mmap_events(pid_t pid, pid_t tgid, +static int event__synthesize_mmap_events(event_t *event, pid_t pid, pid_t tgid, event__handler_t process, struct perf_session *session) { @@ -124,29 +144,25 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid, return -1; } + event->header.type = PERF_RECORD_MMAP; + /* + * Just like the kernel, see __perf_event_mmap in kernel/perf_event.c + */ + event->header.misc = PERF_RECORD_MISC_USER; + while (1) { char bf[BUFSIZ], *pbf = bf; - event_t ev = { - .header = { - .type = PERF_RECORD_MMAP, - /* - * Just like the kernel, see __perf_event_mmap - * in kernel/perf_event.c - */ - .misc = PERF_RECORD_MISC_USER, - }, - }; int n; size_t size; if (fgets(bf, sizeof(bf), fp) == NULL) break; /* 00400000-0040c000 r-xp 00000000 fd:01 41038 /bin/cat */ - n = hex2u64(pbf, &ev.mmap.start); + n = hex2u64(pbf, &event->mmap.start); if (n < 0) continue; pbf += n + 1; - n = hex2u64(pbf, &ev.mmap.len); + n = hex2u64(pbf, &event->mmap.len); if (n < 0) continue; pbf += n + 3; @@ -161,19 +177,21 @@ static int event__synthesize_mmap_events(pid_t pid, pid_t tgid, continue; pbf += 3; - n = hex2u64(pbf, &ev.mmap.pgoff); + n = hex2u64(pbf, &event->mmap.pgoff); size = strlen(execname); execname[size - 1] = '\0'; /* Remove \n */ - memcpy(ev.mmap.filename, execname, size); + memcpy(event->mmap.filename, execname, size); size = ALIGN(size, sizeof(u64)); - ev.mmap.len -= ev.mmap.start; - ev.mmap.header.size = (sizeof(ev.mmap) - - (sizeof(ev.mmap.filename) - size)); - ev.mmap.pid = tgid; - ev.mmap.tid = pid; - - process(&ev, session); + event->mmap.len -= event->mmap.start; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size)); + memset(event->mmap.filename + size, 0, session->id_hdr_size); + event->mmap.header.size += session->id_hdr_size; + event->mmap.pid = tgid; + event->mmap.tid = pid; + + process(event, &synth_sample, session); } } @@ -187,20 +205,27 @@ int event__synthesize_modules(event__handler_t process, { struct rb_node *nd; struct map_groups *kmaps = &machine->kmaps; - u16 misc; + event_t *event = zalloc(sizeof(event->mmap) + session->id_hdr_size); + + if (event == NULL) { + pr_debug("Not enough memory synthesizing mmap event " + "for kernel modules\n"); + return -1; + } + + event->header.type = PERF_RECORD_MMAP; /* * kernel uses 0 for user space maps, see kernel/perf_event.c * __perf_event_mmap */ if (machine__is_host(machine)) - misc = PERF_RECORD_MISC_KERNEL; + event->header.misc = PERF_RECORD_MISC_KERNEL; else - misc = PERF_RECORD_MISC_GUEST_KERNEL; + event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; for (nd = rb_first(&kmaps->maps[MAP__FUNCTION]); nd; nd = rb_next(nd)) { - event_t ev; size_t size; struct map *pos = rb_entry(nd, struct map, rb_node); @@ -208,39 +233,78 @@ int event__synthesize_modules(event__handler_t process, continue; size = ALIGN(pos->dso->long_name_len + 1, sizeof(u64)); - memset(&ev, 0, sizeof(ev)); - ev.mmap.header.misc = misc; - ev.mmap.header.type = PERF_RECORD_MMAP; - ev.mmap.header.size = (sizeof(ev.mmap) - - (sizeof(ev.mmap.filename) - size)); - ev.mmap.start = pos->start; - ev.mmap.len = pos->end - pos->start; - ev.mmap.pid = machine->pid; - - memcpy(ev.mmap.filename, pos->dso->long_name, + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size)); + memset(event->mmap.filename + size, 0, session->id_hdr_size); + event->mmap.header.size += session->id_hdr_size; + event->mmap.start = pos->start; + event->mmap.len = pos->end - pos->start; + event->mmap.pid = machine->pid; + + memcpy(event->mmap.filename, pos->dso->long_name, pos->dso->long_name_len + 1); - process(&ev, session); + process(event, &synth_sample, session); } + free(event); return 0; } -int event__synthesize_thread(pid_t pid, event__handler_t process, - struct perf_session *session) +static int __event__synthesize_thread(event_t *comm_event, event_t *mmap_event, + pid_t pid, event__handler_t process, + struct perf_session *session) { - pid_t tgid = event__synthesize_comm(pid, 1, process, session); + pid_t tgid = event__synthesize_comm(comm_event, pid, 1, process, + session); if (tgid == -1) return -1; - return event__synthesize_mmap_events(pid, tgid, process, session); + return event__synthesize_mmap_events(mmap_event, pid, tgid, + process, session); +} + +int event__synthesize_thread(pid_t pid, event__handler_t process, + struct perf_session *session) +{ + event_t *comm_event, *mmap_event; + int err = -1; + + comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size); + if (comm_event == NULL) + goto out; + + mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size); + if (mmap_event == NULL) + goto out_free_comm; + + err = __event__synthesize_thread(comm_event, mmap_event, pid, + process, session); + free(mmap_event); +out_free_comm: + free(comm_event); +out: + return err; } -void event__synthesize_threads(event__handler_t process, - struct perf_session *session) +int event__synthesize_threads(event__handler_t process, + struct perf_session *session) { DIR *proc; struct dirent dirent, *next; + event_t *comm_event, *mmap_event; + int err = -1; + + comm_event = malloc(sizeof(comm_event->comm) + session->id_hdr_size); + if (comm_event == NULL) + goto out; + + mmap_event = malloc(sizeof(mmap_event->mmap) + session->id_hdr_size); + if (mmap_event == NULL) + goto out_free_comm; proc = opendir("/proc"); + if (proc == NULL) + goto out_free_mmap; while (!readdir_r(proc, &dirent, &next) && next) { char *end; @@ -249,10 +313,18 @@ void event__synthesize_threads(event__handler_t process, if (*end) /* only interested in proper numerical dirents */ continue; - event__synthesize_thread(pid, process, session); + __event__synthesize_thread(comm_event, mmap_event, pid, + process, session); } closedir(proc); + err = 0; +out_free_mmap: + free(mmap_event); +out_free_comm: + free(comm_event); +out: + return err; } struct process_symbol_args { @@ -260,7 +332,8 @@ struct process_symbol_args { u64 start; }; -static int find_symbol_cb(void *arg, const char *name, char type, u64 start) +static int find_symbol_cb(void *arg, const char *name, char type, + u64 start, u64 end __used) { struct process_symbol_args *args = arg; @@ -286,18 +359,20 @@ int event__synthesize_kernel_mmap(event__handler_t process, char path[PATH_MAX]; char name_buff[PATH_MAX]; struct map *map; - - event_t ev = { - .header = { - .type = PERF_RECORD_MMAP, - }, - }; + int err; /* * We should get this from /sys/kernel/sections/.text, but till that is * available use this, and after it is use this as a fallback for older * kernels. */ struct process_symbol_args args = { .name = symbol_name, }; + event_t *event = zalloc(sizeof(event->mmap) + session->id_hdr_size); + + if (event == NULL) { + pr_debug("Not enough memory synthesizing mmap event " + "for kernel modules\n"); + return -1; + } mmap_name = machine__mmap_name(machine, name_buff, sizeof(name_buff)); if (machine__is_host(machine)) { @@ -305,10 +380,10 @@ int event__synthesize_kernel_mmap(event__handler_t process, * kernel uses PERF_RECORD_MISC_USER for user space maps, * see kernel/perf_event.c __perf_event_mmap */ - ev.header.misc = PERF_RECORD_MISC_KERNEL; + event->header.misc = PERF_RECORD_MISC_KERNEL; filename = "/proc/kallsyms"; } else { - ev.header.misc = PERF_RECORD_MISC_GUEST_KERNEL; + event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; if (machine__is_default_guest(machine)) filename = (char *) symbol_conf.default_guest_kallsyms; else { @@ -321,17 +396,21 @@ int event__synthesize_kernel_mmap(event__handler_t process, return -ENOENT; map = machine->vmlinux_maps[MAP__FUNCTION]; - size = snprintf(ev.mmap.filename, sizeof(ev.mmap.filename), + size = snprintf(event->mmap.filename, sizeof(event->mmap.filename), "%s%s", mmap_name, symbol_name) + 1; size = ALIGN(size, sizeof(u64)); - ev.mmap.header.size = (sizeof(ev.mmap) - - (sizeof(ev.mmap.filename) - size)); - ev.mmap.pgoff = args.start; - ev.mmap.start = map->start; - ev.mmap.len = map->end - ev.mmap.start; - ev.mmap.pid = machine->pid; - - return process(&ev, session); + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size) + session->id_hdr_size); + event->mmap.pgoff = args.start; + event->mmap.start = map->start; + event->mmap.len = map->end - event->mmap.start; + event->mmap.pid = machine->pid; + + err = process(event, &synth_sample, session); + free(event); + + return err; } static void thread__comm_adjust(struct thread *self, struct hists *hists) @@ -361,7 +440,8 @@ static int thread__set_comm_adjust(struct thread *self, const char *comm, return 0; } -int event__process_comm(event_t *self, struct perf_session *session) +int event__process_comm(event_t *self, struct sample_data *sample __used, + struct perf_session *session) { struct thread *thread = perf_session__findnew(session, self->comm.tid); @@ -376,7 +456,8 @@ int event__process_comm(event_t *self, struct perf_session *session) return 0; } -int event__process_lost(event_t *self, struct perf_session *session) +int event__process_lost(event_t *self, struct sample_data *sample __used, + struct perf_session *session) { dump_printf(": id:%Ld: lost:%Ld\n", self->lost.id, self->lost.lost); session->hists.stats.total_lost += self->lost.lost; @@ -392,7 +473,7 @@ static void event_set_kernel_mmap_len(struct map **maps, event_t *self) * a zero sized synthesized MMAP event for the kernel. */ if (maps[MAP__FUNCTION]->end == 0) - maps[MAP__FUNCTION]->end = ~0UL; + maps[MAP__FUNCTION]->end = ~0ULL; } static int event__process_kernel_mmap(event_t *self, @@ -485,7 +566,8 @@ out_problem: return -1; } -int event__process_mmap(event_t *self, struct perf_session *session) +int event__process_mmap(event_t *self, struct sample_data *sample __used, + struct perf_session *session) { struct machine *machine; struct thread *thread; @@ -526,7 +608,8 @@ out_problem: return 0; } -int event__process_task(event_t *self, struct perf_session *session) +int event__process_task(event_t *self, struct sample_data *sample __used, + struct perf_session *session) { struct thread *thread = perf_session__findnew(session, self->fork.tid); struct thread *parent = perf_session__findnew(session, self->fork.ptid); @@ -548,18 +631,19 @@ int event__process_task(event_t *self, struct perf_session *session) return 0; } -int event__process(event_t *event, struct perf_session *session) +int event__process(event_t *event, struct sample_data *sample, + struct perf_session *session) { switch (event->header.type) { case PERF_RECORD_COMM: - event__process_comm(event, session); + event__process_comm(event, sample, session); break; case PERF_RECORD_MMAP: - event__process_mmap(event, session); + event__process_mmap(event, sample, session); break; case PERF_RECORD_FORK: case PERF_RECORD_EXIT: - event__process_task(event, session); + event__process_task(event, sample, session); break; default: break; @@ -674,32 +758,8 @@ int event__preprocess_sample(const event_t *self, struct perf_session *session, symbol_filter_t filter) { u8 cpumode = self->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; - struct thread *thread; - - event__parse_sample(self, session->sample_type, data); - - dump_printf("(IP, %d): %d/%d: %#Lx period: %Ld cpu:%d\n", - self->header.misc, data->pid, data->tid, data->ip, - data->period, data->cpu); - - if (session->sample_type & PERF_SAMPLE_CALLCHAIN) { - unsigned int i; - - dump_printf("... chain: nr:%Lu\n", data->callchain->nr); + struct thread *thread = perf_session__findnew(session, self->ip.pid); - if (!ip_callchain__valid(data->callchain, self)) { - pr_debug("call-chain problem with event, " - "skipping it.\n"); - goto out_filtered; - } - - if (dump_trace) { - for (i = 0; i < data->callchain->nr; i++) - dump_printf("..... %2d: %016Lx\n", - i, data->callchain->ips[i]); - } - } - thread = perf_session__findnew(session, self->ip.pid); if (thread == NULL) return -1; @@ -766,9 +826,65 @@ out_filtered: return 0; } -int event__parse_sample(const event_t *event, u64 type, struct sample_data *data) +static int event__parse_id_sample(const event_t *event, + struct perf_session *session, + struct sample_data *sample) { - const u64 *array = event->sample.array; + const u64 *array; + u64 type; + + sample->cpu = sample->pid = sample->tid = -1; + sample->stream_id = sample->id = sample->time = -1ULL; + + if (!session->sample_id_all) + return 0; + + array = event->sample.array; + array += ((event->header.size - + sizeof(event->header)) / sizeof(u64)) - 1; + type = session->sample_type; + + if (type & PERF_SAMPLE_CPU) { + u32 *p = (u32 *)array; + sample->cpu = *p; + array--; + } + + if (type & PERF_SAMPLE_STREAM_ID) { + sample->stream_id = *array; + array--; + } + + if (type & PERF_SAMPLE_ID) { + sample->id = *array; + array--; + } + + if (type & PERF_SAMPLE_TIME) { + sample->time = *array; + array--; + } + + if (type & PERF_SAMPLE_TID) { + u32 *p = (u32 *)array; + sample->pid = p[0]; + sample->tid = p[1]; + } + + return 0; +} + +int event__parse_sample(const event_t *event, struct perf_session *session, + struct sample_data *data) +{ + const u64 *array; + u64 type; + + if (event->header.type != PERF_RECORD_SAMPLE) + return event__parse_id_sample(event, session, data); + + array = event->sample.array; + type = session->sample_type; if (type & PERF_SAMPLE_IP) { data->ip = event->ip.ip; diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 8e790dae7026..2b7e91902f10 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -85,6 +85,7 @@ struct build_id_event { }; enum perf_user_event_type { /* above any possible kernel type */ + PERF_RECORD_USER_TYPE_START = 64, PERF_RECORD_HEADER_ATTR = 64, PERF_RECORD_HEADER_EVENT_TYPE = 65, PERF_RECORD_HEADER_TRACING_DATA = 66, @@ -135,12 +136,15 @@ void event__print_totals(void); struct perf_session; -typedef int (*event__handler_t)(event_t *event, struct perf_session *session); +typedef int (*event__handler_synth_t)(event_t *event, + struct perf_session *session); +typedef int (*event__handler_t)(event_t *event, struct sample_data *sample, + struct perf_session *session); int event__synthesize_thread(pid_t pid, event__handler_t process, struct perf_session *session); -void event__synthesize_threads(event__handler_t process, - struct perf_session *session); +int event__synthesize_threads(event__handler_t process, + struct perf_session *session); int event__synthesize_kernel_mmap(event__handler_t process, struct perf_session *session, struct machine *machine, @@ -150,18 +154,24 @@ int event__synthesize_modules(event__handler_t process, struct perf_session *session, struct machine *machine); -int event__process_comm(event_t *self, struct perf_session *session); -int event__process_lost(event_t *self, struct perf_session *session); -int event__process_mmap(event_t *self, struct perf_session *session); -int event__process_task(event_t *self, struct perf_session *session); -int event__process(event_t *event, struct perf_session *session); +int event__process_comm(event_t *self, struct sample_data *sample, + struct perf_session *session); +int event__process_lost(event_t *self, struct sample_data *sample, + struct perf_session *session); +int event__process_mmap(event_t *self, struct sample_data *sample, + struct perf_session *session); +int event__process_task(event_t *self, struct sample_data *sample, + struct perf_session *session); +int event__process(event_t *event, struct sample_data *sample, + struct perf_session *session); struct addr_location; int event__preprocess_sample(const event_t *self, struct perf_session *session, struct addr_location *al, struct sample_data *data, symbol_filter_t filter); -int event__parse_sample(const event_t *event, u64 type, struct sample_data *data); +int event__parse_sample(const event_t *event, struct perf_session *session, + struct sample_data *sample); -extern const char *event__name[]; +const char *event__get_event_name(unsigned int id); #endif /* __PERF_RECORD_H */ diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c new file mode 100644 index 000000000000..c95267e63c5b --- /dev/null +++ b/tools/perf/util/evsel.c @@ -0,0 +1,186 @@ +#include "evsel.h" +#include "../perf.h" +#include "util.h" +#include "cpumap.h" +#include "thread.h" + +#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y)) + +struct perf_evsel *perf_evsel__new(u32 type, u64 config, int idx) +{ + struct perf_evsel *evsel = zalloc(sizeof(*evsel)); + + if (evsel != NULL) { + evsel->idx = idx; + evsel->attr.type = type; + evsel->attr.config = config; + INIT_LIST_HEAD(&evsel->node); + } + + return evsel; +} + +int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int)); + return evsel->fd != NULL ? 0 : -ENOMEM; +} + +int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus) +{ + evsel->counts = zalloc((sizeof(*evsel->counts) + + (ncpus * sizeof(struct perf_counts_values)))); + return evsel->counts != NULL ? 0 : -ENOMEM; +} + +void perf_evsel__free_fd(struct perf_evsel *evsel) +{ + xyarray__delete(evsel->fd); + evsel->fd = NULL; +} + +void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads) +{ + int cpu, thread; + + for (cpu = 0; cpu < ncpus; cpu++) + for (thread = 0; thread < nthreads; ++thread) { + close(FD(evsel, cpu, thread)); + FD(evsel, cpu, thread) = -1; + } +} + +void perf_evsel__delete(struct perf_evsel *evsel) +{ + assert(list_empty(&evsel->node)); + xyarray__delete(evsel->fd); + free(evsel); +} + +int __perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread, bool scale) +{ + struct perf_counts_values count; + size_t nv = scale ? 3 : 1; + + if (FD(evsel, cpu, thread) < 0) + return -EINVAL; + + if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0) + return -ENOMEM; + + if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0) + return -errno; + + if (scale) { + if (count.run == 0) + count.val = 0; + else if (count.run < count.ena) + count.val = (u64)((double)count.val * count.ena / count.run + 0.5); + } else + count.ena = count.run = 0; + + evsel->counts->cpu[cpu] = count; + return 0; +} + +int __perf_evsel__read(struct perf_evsel *evsel, + int ncpus, int nthreads, bool scale) +{ + size_t nv = scale ? 3 : 1; + int cpu, thread; + struct perf_counts_values *aggr = &evsel->counts->aggr, count; + + aggr->val = 0; + + for (cpu = 0; cpu < ncpus; cpu++) { + for (thread = 0; thread < nthreads; thread++) { + if (FD(evsel, cpu, thread) < 0) + continue; + + if (readn(FD(evsel, cpu, thread), + &count, nv * sizeof(u64)) < 0) + return -errno; + + aggr->val += count.val; + if (scale) { + aggr->ena += count.ena; + aggr->run += count.run; + } + } + } + + evsel->counts->scaled = 0; + if (scale) { + if (aggr->run == 0) { + evsel->counts->scaled = -1; + aggr->val = 0; + return 0; + } + + if (aggr->run < aggr->ena) { + evsel->counts->scaled = 1; + aggr->val = (u64)((double)aggr->val * aggr->ena / aggr->run + 0.5); + } + } else + aggr->ena = aggr->run = 0; + + return 0; +} + +int perf_evsel__open_per_cpu(struct perf_evsel *evsel, struct cpu_map *cpus) +{ + int cpu; + + if (evsel->fd == NULL && perf_evsel__alloc_fd(evsel, cpus->nr, 1) < 0) + return -1; + + for (cpu = 0; cpu < cpus->nr; cpu++) { + FD(evsel, cpu, 0) = sys_perf_event_open(&evsel->attr, -1, + cpus->map[cpu], -1, 0); + if (FD(evsel, cpu, 0) < 0) + goto out_close; + } + + return 0; + +out_close: + while (--cpu >= 0) { + close(FD(evsel, cpu, 0)); + FD(evsel, cpu, 0) = -1; + } + return -1; +} + +int perf_evsel__open_per_thread(struct perf_evsel *evsel, struct thread_map *threads) +{ + int thread; + + if (evsel->fd == NULL && perf_evsel__alloc_fd(evsel, 1, threads->nr)) + return -1; + + for (thread = 0; thread < threads->nr; thread++) { + FD(evsel, 0, thread) = sys_perf_event_open(&evsel->attr, + threads->map[thread], -1, -1, 0); + if (FD(evsel, 0, thread) < 0) + goto out_close; + } + + return 0; + +out_close: + while (--thread >= 0) { + close(FD(evsel, 0, thread)); + FD(evsel, 0, thread) = -1; + } + return -1; +} + +int perf_evsel__open(struct perf_evsel *evsel, + struct cpu_map *cpus, struct thread_map *threads) +{ + if (threads == NULL) + return perf_evsel__open_per_cpu(evsel, cpus); + + return perf_evsel__open_per_thread(evsel, threads); +} diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h new file mode 100644 index 000000000000..a0ccd69c3fc2 --- /dev/null +++ b/tools/perf/util/evsel.h @@ -0,0 +1,115 @@ +#ifndef __PERF_EVSEL_H +#define __PERF_EVSEL_H 1 + +#include <linux/list.h> +#include <stdbool.h> +#include "../../../include/linux/perf_event.h" +#include "types.h" +#include "xyarray.h" + +struct perf_counts_values { + union { + struct { + u64 val; + u64 ena; + u64 run; + }; + u64 values[3]; + }; +}; + +struct perf_counts { + s8 scaled; + struct perf_counts_values aggr; + struct perf_counts_values cpu[]; +}; + +struct perf_evsel { + struct list_head node; + struct perf_event_attr attr; + char *filter; + struct xyarray *fd; + struct perf_counts *counts; + int idx; + void *priv; +}; + +struct cpu_map; +struct thread_map; + +struct perf_evsel *perf_evsel__new(u32 type, u64 config, int idx); +void perf_evsel__delete(struct perf_evsel *evsel); + +int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads); +int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus); +void perf_evsel__free_fd(struct perf_evsel *evsel); +void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads); + +int perf_evsel__open_per_cpu(struct perf_evsel *evsel, struct cpu_map *cpus); +int perf_evsel__open_per_thread(struct perf_evsel *evsel, struct thread_map *threads); +int perf_evsel__open(struct perf_evsel *evsel, + struct cpu_map *cpus, struct thread_map *threads); + +#define perf_evsel__match(evsel, t, c) \ + (evsel->attr.type == PERF_TYPE_##t && \ + evsel->attr.config == PERF_COUNT_##c) + +int __perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread, bool scale); + +/** + * perf_evsel__read_on_cpu - Read out the results on a CPU and thread + * + * @evsel - event selector to read value + * @cpu - CPU of interest + * @thread - thread of interest + */ +static inline int perf_evsel__read_on_cpu(struct perf_evsel *evsel, + int cpu, int thread) +{ + return __perf_evsel__read_on_cpu(evsel, cpu, thread, false); +} + +/** + * perf_evsel__read_on_cpu_scaled - Read out the results on a CPU and thread, scaled + * + * @evsel - event selector to read value + * @cpu - CPU of interest + * @thread - thread of interest + */ +static inline int perf_evsel__read_on_cpu_scaled(struct perf_evsel *evsel, + int cpu, int thread) +{ + return __perf_evsel__read_on_cpu(evsel, cpu, thread, true); +} + +int __perf_evsel__read(struct perf_evsel *evsel, int ncpus, int nthreads, + bool scale); + +/** + * perf_evsel__read - Read the aggregate results on all CPUs + * + * @evsel - event selector to read value + * @ncpus - Number of cpus affected, from zero + * @nthreads - Number of threads affected, from zero + */ +static inline int perf_evsel__read(struct perf_evsel *evsel, + int ncpus, int nthreads) +{ + return __perf_evsel__read(evsel, ncpus, nthreads, false); +} + +/** + * perf_evsel__read_scaled - Read the aggregate results on all CPUs, scaled + * + * @evsel - event selector to read value + * @ncpus - Number of cpus affected, from zero + * @nthreads - Number of threads affected, from zero + */ +static inline int perf_evsel__read_scaled(struct perf_evsel *evsel, + int ncpus, int nthreads) +{ + return __perf_evsel__read(evsel, ncpus, nthreads, true); +} + +#endif /* __PERF_EVSEL_H */ diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index 7cba0551a565..989fa2dee2fd 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -152,6 +152,11 @@ void perf_header__set_feat(struct perf_header *self, int feat) set_bit(feat, self->adds_features); } +void perf_header__clear_feat(struct perf_header *self, int feat) +{ + clear_bit(feat, self->adds_features); +} + bool perf_header__has_feat(const struct perf_header *self, int feat) { return test_bit(feat, self->adds_features); @@ -433,8 +438,10 @@ static int perf_header__adds_write(struct perf_header *self, int fd) int idx = 0, err; session = container_of(self, struct perf_session, header); - if (perf_session__read_build_ids(session, true)) - perf_header__set_feat(self, HEADER_BUILD_ID); + + if (perf_header__has_feat(self, HEADER_BUILD_ID && + !perf_session__read_build_ids(session, true))) + perf_header__clear_feat(self, HEADER_BUILD_ID); nr_sections = bitmap_weight(self->adds_features, HEADER_FEAT_BITS); if (!nr_sections) @@ -456,7 +463,7 @@ static int perf_header__adds_write(struct perf_header *self, int fd) /* Write trace info */ trace_sec->offset = lseek(fd, 0, SEEK_CUR); - read_tracing_data(fd, attrs, nr_counters); + read_tracing_data(fd, &evsel_list); trace_sec->size = lseek(fd, 0, SEEK_CUR) - trace_sec->offset; } @@ -599,7 +606,7 @@ int perf_header__write(struct perf_header *self, int fd, bool at_exit) static int perf_header__getbuffer64(struct perf_header *self, int fd, void *buf, size_t size) { - if (do_read(fd, buf, size) <= 0) + if (readn(fd, buf, size) <= 0) return -1; if (self->needs_swap) @@ -655,7 +662,7 @@ int perf_file_header__read(struct perf_file_header *self, { lseek(fd, 0, SEEK_SET); - if (do_read(fd, self, sizeof(*self)) <= 0 || + if (readn(fd, self, sizeof(*self)) <= 0 || memcmp(&self->magic, __perf_magic, sizeof(self->magic))) return -1; @@ -816,7 +823,7 @@ static int perf_file_header__read_pipe(struct perf_pipe_file_header *self, struct perf_header *ph, int fd, bool repipe) { - if (do_read(fd, self, sizeof(*self)) <= 0 || + if (readn(fd, self, sizeof(*self)) <= 0 || memcmp(&self->magic, __perf_magic, sizeof(self->magic))) return -1; @@ -941,6 +948,24 @@ u64 perf_header__sample_type(struct perf_header *header) return type; } +bool perf_header__sample_id_all(const struct perf_header *header) +{ + bool value = false, first = true; + int i; + + for (i = 0; i < header->attrs; i++) { + struct perf_header_attr *attr = header->attr[i]; + + if (first) { + value = attr->attr.sample_id_all; + first = false; + } else if (value != attr->attr.sample_id_all) + die("non matching sample_id_all"); + } + + return value; +} + struct perf_event_attr * perf_header__find_attr(u64 id, struct perf_header *header) { @@ -987,21 +1012,23 @@ int event__synthesize_attr(struct perf_event_attr *attr, u16 ids, u64 *id, ev = malloc(size); + if (ev == NULL) + return -ENOMEM; + ev->attr.attr = *attr; memcpy(ev->attr.id, id, ids * sizeof(u64)); ev->attr.header.type = PERF_RECORD_HEADER_ATTR; ev->attr.header.size = size; - err = process(ev, session); + err = process(ev, NULL, session); free(ev); return err; } -int event__synthesize_attrs(struct perf_header *self, - event__handler_t process, +int event__synthesize_attrs(struct perf_header *self, event__handler_t process, struct perf_session *session) { struct perf_header_attr *attr; @@ -1071,7 +1098,7 @@ int event__synthesize_event_type(u64 event_id, char *name, ev.event_type.header.size = sizeof(ev.event_type) - (sizeof(ev.event_type.event_type.name) - size); - err = process(&ev, session); + err = process(&ev, NULL, session); return err; } @@ -1106,8 +1133,7 @@ int event__process_event_type(event_t *self, return 0; } -int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, - int nb_events, +int event__synthesize_tracing_data(int fd, struct list_head *pattrs, event__handler_t process, struct perf_session *session __unused) { @@ -1118,7 +1144,7 @@ int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, memset(&ev, 0, sizeof(ev)); ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA; - size = read_tracing_data_size(fd, pattrs, nb_events); + size = read_tracing_data_size(fd, pattrs); if (size <= 0) return size; aligned_size = ALIGN(size, sizeof(u64)); @@ -1126,9 +1152,9 @@ int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, ev.tracing_data.header.size = sizeof(ev.tracing_data); ev.tracing_data.size = aligned_size; - process(&ev, session); + process(&ev, NULL, session); - err = read_tracing_data(fd, pattrs, nb_events); + err = read_tracing_data(fd, pattrs); write_padded(fd, NULL, 0, padding); return aligned_size; @@ -1186,7 +1212,7 @@ int event__synthesize_build_id(struct dso *pos, u16 misc, ev.build_id.header.size = sizeof(ev.build_id) + len; memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len); - err = process(&ev, session); + err = process(&ev, NULL, session); return err; } diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h index 402ac2454cf8..33f16be7b72f 100644 --- a/tools/perf/util/header.h +++ b/tools/perf/util/header.h @@ -81,9 +81,11 @@ void perf_header_attr__delete(struct perf_header_attr *self); int perf_header_attr__add_id(struct perf_header_attr *self, u64 id); u64 perf_header__sample_type(struct perf_header *header); +bool perf_header__sample_id_all(const struct perf_header *header); struct perf_event_attr * perf_header__find_attr(u64 id, struct perf_header *header); void perf_header__set_feat(struct perf_header *self, int feat); +void perf_header__clear_feat(struct perf_header *self, int feat); bool perf_header__has_feat(const struct perf_header *self, int feat); int perf_header__process_sections(struct perf_header *self, int fd, @@ -111,8 +113,7 @@ int event__synthesize_event_types(event__handler_t process, int event__process_event_type(event_t *self, struct perf_session *session); -int event__synthesize_tracing_data(int fd, struct perf_event_attr *pattrs, - int nb_events, +int event__synthesize_tracing_data(int fd, struct list_head *pattrs, event__handler_t process, struct perf_session *session); int event__process_tracing_data(event_t *self, diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index 76bcc35cf9b1..c749ba6136a0 100644 --- a/tools/perf/util/hist.c +++ b/tools/perf/util/hist.c @@ -1092,6 +1092,12 @@ int hist_entry__annotate(struct hist_entry *self, struct list_head *head, FILE *file; int err = 0; u64 len; + char symfs_filename[PATH_MAX]; + + if (filename) { + snprintf(symfs_filename, sizeof(symfs_filename), "%s%s", + symbol_conf.symfs, filename); + } if (filename == NULL) { if (dso->has_build_id) { @@ -1100,9 +1106,9 @@ int hist_entry__annotate(struct hist_entry *self, struct list_head *head, return -ENOMEM; } goto fallback; - } else if (readlink(filename, command, sizeof(command)) < 0 || + } else if (readlink(symfs_filename, command, sizeof(command)) < 0 || strstr(command, "[kernel.kallsyms]") || - access(filename, R_OK)) { + access(symfs_filename, R_OK)) { free(filename); fallback: /* @@ -1111,6 +1117,8 @@ fallback: * DSO is the same as when 'perf record' ran. */ filename = dso->long_name; + snprintf(symfs_filename, sizeof(symfs_filename), "%s%s", + symbol_conf.symfs, filename); free_filename = false; } @@ -1137,7 +1145,7 @@ fallback: "objdump --start-address=0x%016Lx --stop-address=0x%016Lx -dS -C %s|grep -v %s|expand", map__rip_2objdump(map, sym->start), map__rip_2objdump(map, sym->end), - filename, filename); + symfs_filename, filename); pr_debug("Executing: %s\n", command); @@ -1168,10 +1176,13 @@ size_t hists__fprintf_nr_events(struct hists *self, FILE *fp) size_t ret = 0; for (i = 0; i < PERF_RECORD_HEADER_MAX; ++i) { - if (!event__name[i]) + const char *name = event__get_event_name(i); + + if (!strcmp(name, "UNKNOWN")) continue; - ret += fprintf(fp, "%10s events: %10d\n", - event__name[i], self->stats.nr_events[i]); + + ret += fprintf(fp, "%16s events: %10d\n", name, + self->stats.nr_events[i]); } return ret; diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h index 587d375d3430..ee789856a8c9 100644 --- a/tools/perf/util/hist.h +++ b/tools/perf/util/hist.h @@ -52,8 +52,10 @@ struct sym_priv { struct events_stats { u64 total_period; u64 total_lost; + u64 total_invalid_chains; u32 nr_events[PERF_RECORD_HEADER_MAX]; u32 nr_unknown_events; + u32 nr_invalid_chains; }; enum hist_column { diff --git a/tools/perf/util/include/asm/cpufeature.h b/tools/perf/util/include/asm/cpufeature.h new file mode 100644 index 000000000000..acffd5e4d1d4 --- /dev/null +++ b/tools/perf/util/include/asm/cpufeature.h @@ -0,0 +1,9 @@ + +#ifndef PERF_CPUFEATURE_H +#define PERF_CPUFEATURE_H + +/* cpufeature.h ... dummy header file for including arch/x86/lib/memcpy_64.S */ + +#define X86_FEATURE_REP_GOOD 0 + +#endif /* PERF_CPUFEATURE_H */ diff --git a/tools/perf/util/include/asm/dwarf2.h b/tools/perf/util/include/asm/dwarf2.h new file mode 100644 index 000000000000..bb4198e7837a --- /dev/null +++ b/tools/perf/util/include/asm/dwarf2.h @@ -0,0 +1,11 @@ + +#ifndef PERF_DWARF2_H +#define PERF_DWARF2_H + +/* dwarf2.h ... dummy header file for including arch/x86/lib/memcpy_64.S */ + +#define CFI_STARTPROC +#define CFI_ENDPROC + +#endif /* PERF_DWARF2_H */ + diff --git a/tools/perf/util/include/linux/bitops.h b/tools/perf/util/include/linux/bitops.h index bb4ac2e05385..8be0b968ca0b 100644 --- a/tools/perf/util/include/linux/bitops.h +++ b/tools/perf/util/include/linux/bitops.h @@ -13,6 +13,11 @@ static inline void set_bit(int nr, unsigned long *addr) addr[nr / BITS_PER_LONG] |= 1UL << (nr % BITS_PER_LONG); } +static inline void clear_bit(int nr, unsigned long *addr) +{ + addr[nr / BITS_PER_LONG] &= ~(1UL << (nr % BITS_PER_LONG)); +} + static __always_inline int test_bit(unsigned int nr, const unsigned long *addr) { return ((1UL << (nr % BITS_PER_LONG)) & diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h new file mode 100644 index 000000000000..06387cffe125 --- /dev/null +++ b/tools/perf/util/include/linux/linkage.h @@ -0,0 +1,13 @@ + +#ifndef PERF_LINUX_LINKAGE_H_ +#define PERF_LINUX_LINKAGE_H_ + +/* linkage.h ... for including arch/x86/lib/memcpy_64.S */ + +#define ENTRY(name) \ + .globl name; \ + name: + +#define ENDPROC(name) + +#endif /* PERF_LINUX_LINKAGE_H_ */ diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 4af5bd59cfd1..649083f27e08 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -1,6 +1,7 @@ #include "../../../include/linux/hw_breakpoint.h" #include "util.h" #include "../perf.h" +#include "evsel.h" #include "parse-options.h" #include "parse-events.h" #include "exec_cmd.h" @@ -12,8 +13,7 @@ int nr_counters; -struct perf_event_attr attrs[MAX_COUNTERS]; -char *filters[MAX_COUNTERS]; +LIST_HEAD(evsel_list); struct event_symbol { u8 type; @@ -266,10 +266,10 @@ static char *event_cache_name(u8 cache_type, u8 cache_op, u8 cache_result) return name; } -const char *event_name(int counter) +const char *event_name(struct perf_evsel *evsel) { - u64 config = attrs[counter].config; - int type = attrs[counter].type; + u64 config = evsel->attr.config; + int type = evsel->attr.type; return __event_name(type, config); } @@ -434,7 +434,7 @@ parse_single_tracepoint_event(char *sys_name, id = atoll(id_buf); attr->config = id; attr->type = PERF_TYPE_TRACEPOINT; - *strp = evt_name + evt_length; + *strp += strlen(sys_name) + evt_length + 1; /* + 1 for the ':' */ attr->sample_type |= PERF_SAMPLE_RAW; attr->sample_type |= PERF_SAMPLE_TIME; @@ -495,7 +495,7 @@ static enum event_result parse_tracepoint_event(const char **strp, struct perf_event_attr *attr) { const char *evt_name; - char *flags; + char *flags = NULL, *comma_loc; char sys_name[MAX_EVENT_LENGTH]; unsigned int sys_length, evt_length; @@ -514,6 +514,11 @@ static enum event_result parse_tracepoint_event(const char **strp, sys_name[sys_length] = '\0'; evt_name = evt_name + 1; + comma_loc = strchr(evt_name, ','); + if (comma_loc) { + /* take the event name up to the comma */ + evt_name = strndup(evt_name, comma_loc - evt_name); + } flags = strchr(evt_name, ':'); if (flags) { /* split it out: */ @@ -524,9 +529,8 @@ static enum event_result parse_tracepoint_event(const char **strp, evt_length = strlen(evt_name); if (evt_length >= MAX_EVENT_LENGTH) return EVT_FAILED; - if (strpbrk(evt_name, "*?")) { - *strp = evt_name + evt_length; + *strp += strlen(sys_name) + evt_length; return parse_multiple_tracepoint_event(sys_name, evt_name, flags); } else @@ -810,9 +814,6 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u return -1; for (;;) { - if (nr_counters == MAX_COUNTERS) - return -1; - memset(&attr, 0, sizeof(attr)); ret = parse_event_symbols(&str, &attr); if (ret == EVT_FAILED) @@ -822,8 +823,13 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u return -1; if (ret != EVT_HANDLED_ALL) { - attrs[nr_counters] = attr; - nr_counters++; + struct perf_evsel *evsel; + evsel = perf_evsel__new(attr.type, attr.config, + nr_counters); + if (evsel == NULL) + return -1; + list_add_tail(&evsel->node, &evsel_list); + ++nr_counters; } if (*str == 0) @@ -840,21 +846,22 @@ int parse_events(const struct option *opt __used, const char *str, int unset __u int parse_filter(const struct option *opt __used, const char *str, int unset __used) { - int i = nr_counters - 1; - int len = strlen(str); + struct perf_evsel *last = NULL; - if (i < 0 || attrs[i].type != PERF_TYPE_TRACEPOINT) { + if (!list_empty(&evsel_list)) + last = list_entry(evsel_list.prev, struct perf_evsel, node); + + if (last == NULL || last->attr.type != PERF_TYPE_TRACEPOINT) { fprintf(stderr, "-F option should follow a -e tracepoint option\n"); return -1; } - filters[i] = malloc(len + 1); - if (!filters[i]) { + last->filter = strdup(str); + if (last->filter == NULL) { fprintf(stderr, "not enough memory to hold filter string\n"); return -1; } - strcpy(filters[i], str); return 0; } @@ -906,6 +913,47 @@ static void print_tracepoint_events(void) } /* + * Check whether event is in <debugfs_mount_point>/tracing/events + */ + +int is_valid_tracepoint(const char *event_string) +{ + DIR *sys_dir, *evt_dir; + struct dirent *sys_next, *evt_next, sys_dirent, evt_dirent; + char evt_path[MAXPATHLEN]; + char dir_path[MAXPATHLEN]; + + if (debugfs_valid_mountpoint(debugfs_path)) + return 0; + + sys_dir = opendir(debugfs_path); + if (!sys_dir) + return 0; + + for_each_subsystem(sys_dir, sys_dirent, sys_next) { + + snprintf(dir_path, MAXPATHLEN, "%s/%s", debugfs_path, + sys_dirent.d_name); + evt_dir = opendir(dir_path); + if (!evt_dir) + continue; + + for_each_event(sys_dirent, evt_dir, evt_dirent, evt_next) { + snprintf(evt_path, MAXPATHLEN, "%s:%s", + sys_dirent.d_name, evt_dirent.d_name); + if (!strcmp(evt_path, event_string)) { + closedir(evt_dir); + closedir(sys_dir); + return 1; + } + } + closedir(evt_dir); + } + closedir(sys_dir); + return 0; +} + +/* * Print the help text for the event symbols: */ void print_events(void) @@ -963,3 +1011,26 @@ void print_events(void) exit(129); } + +int perf_evsel_list__create_default(void) +{ + struct perf_evsel *evsel = perf_evsel__new(PERF_TYPE_HARDWARE, + PERF_COUNT_HW_CPU_CYCLES, 0); + if (evsel == NULL) + return -ENOMEM; + + list_add(&evsel->node, &evsel_list); + ++nr_counters; + return 0; +} + +void perf_evsel_list__delete(void) +{ + struct perf_evsel *pos, *n; + + list_for_each_entry_safe(pos, n, &evsel_list, node) { + list_del_init(&pos->node); + perf_evsel__delete(pos); + } + nr_counters = 0; +} diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index fc4ab3fe877a..b82cafb83772 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -4,6 +4,16 @@ * Parse symbolic events/counts passed in as options: */ +#include "../../../include/linux/perf_event.h" + +struct list_head; +struct perf_evsel; + +extern struct list_head evsel_list; + +int perf_evsel_list__create_default(void); +void perf_evsel_list__delete(void); + struct option; struct tracepoint_path { @@ -13,14 +23,11 @@ struct tracepoint_path { }; extern struct tracepoint_path *tracepoint_id_to_path(u64 config); -extern bool have_tracepoints(struct perf_event_attr *pattrs, int nb_events); +extern bool have_tracepoints(struct list_head *evsel_list); extern int nr_counters; -extern struct perf_event_attr attrs[MAX_COUNTERS]; -extern char *filters[MAX_COUNTERS]; - -extern const char *event_name(int ctr); +const char *event_name(struct perf_evsel *event); extern const char *__event_name(int type, u64 config); extern int parse_events(const struct option *opt, const char *str, int unset); @@ -29,9 +36,9 @@ extern int parse_filter(const struct option *opt, const char *str, int unset); #define EVENTS_HELP_MAX (128*1024) extern void print_events(void); +extern int is_valid_tracepoint(const char *event_string); extern char debugfs_path[]; extern int valid_debugfs_mount(const char *debugfs); - #endif /* __PERF_PARSE_EVENTS_H */ diff --git a/tools/perf/util/parse-options.h b/tools/perf/util/parse-options.h index c7d72dce54b2..abc31a1dac1a 100644 --- a/tools/perf/util/parse-options.h +++ b/tools/perf/util/parse-options.h @@ -119,6 +119,10 @@ struct option { { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .flags = PARSE_OPT_NOARG } #define OPT_CALLBACK_DEFAULT(s, l, v, a, h, f, d) \ { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l), .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d, .flags = PARSE_OPT_LASTARG_DEFAULT } +#define OPT_CALLBACK_DEFAULT_NOOPT(s, l, v, a, h, f, d) \ + { .type = OPTION_CALLBACK, .short_name = (s), .long_name = (l),\ + .value = (v), (a), .help = (h), .callback = (f), .defval = (intptr_t)d,\ + .flags = PARSE_OPT_LASTARG_DEFAULT | PARSE_OPT_NOARG} /* parse_options() will filter out the processed options and leave the * non-option argments in argv[]. diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 61191c6cbe7a..128aaab0aeda 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -95,7 +95,7 @@ static int init_vmlinux(void) goto out; if (machine__create_kernel_maps(&machine) < 0) { - pr_debug("machine__create_kernel_maps "); + pr_debug("machine__create_kernel_maps() failed.\n"); goto out; } out: @@ -149,7 +149,8 @@ static int open_vmlinux(const char *module) { const char *path = kernel_get_module_path(module); if (!path) { - pr_err("Failed to find path of %s module", module ?: "kernel"); + pr_err("Failed to find path of %s module.\n", + module ?: "kernel"); return -ENOENT; } pr_debug("Try to open %s\n", path); @@ -226,7 +227,7 @@ static int try_to_find_probe_trace_events(struct perf_probe_event *pev, pr_warning("Warning: No dwarf info found in the vmlinux - " "please rebuild kernel with CONFIG_DEBUG_INFO=y.\n"); if (!need_dwarf) { - pr_debug("Trying to use symbols.\nn"); + pr_debug("Trying to use symbols.\n"); return 0; } } @@ -295,42 +296,49 @@ static int get_real_path(const char *raw_path, const char *comp_dir, #define LINEBUF_SIZE 256 #define NR_ADDITIONAL_LINES 2 -static int show_one_line(FILE *fp, int l, bool skip, bool show_num) +static int __show_one_line(FILE *fp, int l, bool skip, bool show_num) { char buf[LINEBUF_SIZE]; - const char *color = PERF_COLOR_BLUE; + const char *color = show_num ? "" : PERF_COLOR_BLUE; + const char *prefix = NULL; - if (fgets(buf, LINEBUF_SIZE, fp) == NULL) - goto error; - if (!skip) { - if (show_num) - fprintf(stdout, "%7d %s", l, buf); - else - color_fprintf(stdout, color, " %s", buf); - } - - while (strlen(buf) == LINEBUF_SIZE - 1 && - buf[LINEBUF_SIZE - 2] != '\n') { + do { if (fgets(buf, LINEBUF_SIZE, fp) == NULL) goto error; - if (!skip) { - if (show_num) - fprintf(stdout, "%s", buf); - else - color_fprintf(stdout, color, "%s", buf); + if (skip) + continue; + if (!prefix) { + prefix = show_num ? "%7d " : " "; + color_fprintf(stdout, color, prefix, l); } - } + color_fprintf(stdout, color, "%s", buf); - return 0; + } while (strchr(buf, '\n') == NULL); + + return 1; error: - if (feof(fp)) - pr_warning("Source file is shorter than expected.\n"); - else + if (ferror(fp)) { pr_warning("File read error: %s\n", strerror(errno)); + return -1; + } + return 0; +} - return -1; +static int _show_one_line(FILE *fp, int l, bool skip, bool show_num) +{ + int rv = __show_one_line(fp, l, skip, show_num); + if (rv == 0) { + pr_warning("Source file is shorter than expected.\n"); + rv = -1; + } + return rv; } +#define show_one_line_with_num(f,l) _show_one_line(f,l,false,true) +#define show_one_line(f,l) _show_one_line(f,l,false,false) +#define skip_one_line(f,l) _show_one_line(f,l,true,false) +#define show_one_line_or_eof(f,l) __show_one_line(f,l,false,false) + /* * Show line-range always requires debuginfo to find source file and * line number. @@ -379,7 +387,7 @@ int show_line_range(struct line_range *lr, const char *module) fprintf(stdout, "<%s:%d>\n", lr->function, lr->start - lr->offset); else - fprintf(stdout, "<%s:%d>\n", lr->file, lr->start); + fprintf(stdout, "<%s:%d>\n", lr->path, lr->start); fp = fopen(lr->path, "r"); if (fp == NULL) { @@ -388,26 +396,30 @@ int show_line_range(struct line_range *lr, const char *module) return -errno; } /* Skip to starting line number */ - while (l < lr->start && ret >= 0) - ret = show_one_line(fp, l++, true, false); - if (ret < 0) - goto end; + while (l < lr->start) { + ret = skip_one_line(fp, l++); + if (ret < 0) + goto end; + } list_for_each_entry(ln, &lr->line_list, list) { - while (ln->line > l && ret >= 0) - ret = show_one_line(fp, (l++) - lr->offset, - false, false); - if (ret >= 0) - ret = show_one_line(fp, (l++) - lr->offset, - false, true); + for (; ln->line > l; l++) { + ret = show_one_line(fp, l - lr->offset); + if (ret < 0) + goto end; + } + ret = show_one_line_with_num(fp, l++ - lr->offset); if (ret < 0) goto end; } if (lr->end == INT_MAX) lr->end = l + NR_ADDITIONAL_LINES; - while (l <= lr->end && !feof(fp) && ret >= 0) - ret = show_one_line(fp, (l++) - lr->offset, false, false); + while (l <= lr->end) { + ret = show_one_line_or_eof(fp, l++ - lr->offset); + if (ret <= 0) + break; + } end: fclose(fp); return ret; @@ -466,7 +478,7 @@ int show_available_vars(struct perf_probe_event *pevs, int npevs, fd = open_vmlinux(module); if (fd < 0) { - pr_warning("Failed to open debuginfo file.\n"); + pr_warning("Failed to open debug information file.\n"); return fd; } @@ -526,56 +538,87 @@ int show_available_vars(struct perf_probe_event *pevs __unused, } #endif +static int parse_line_num(char **ptr, int *val, const char *what) +{ + const char *start = *ptr; + + errno = 0; + *val = strtol(*ptr, ptr, 0); + if (errno || *ptr == start) { + semantic_error("'%s' is not a valid number.\n", what); + return -EINVAL; + } + return 0; +} + +/* + * Stuff 'lr' according to the line range described by 'arg'. + * The line range syntax is described by: + * + * SRC[:SLN[+NUM|-ELN]] + * FNC[:SLN[+NUM|-ELN]] + */ int parse_line_range_desc(const char *arg, struct line_range *lr) { - const char *ptr; - char *tmp; - /* - * <Syntax> - * SRC:SLN[+NUM|-ELN] - * FUNC[:SLN[+NUM|-ELN]] - */ - ptr = strchr(arg, ':'); - if (ptr) { - lr->start = (int)strtoul(ptr + 1, &tmp, 0); - if (*tmp == '+') { - lr->end = lr->start + (int)strtoul(tmp + 1, &tmp, 0); - lr->end--; /* - * Adjust the number of lines here. - * If the number of lines == 1, the - * the end of line should be equal to - * the start of line. - */ - } else if (*tmp == '-') - lr->end = (int)strtoul(tmp + 1, &tmp, 0); - else - lr->end = INT_MAX; + char *range, *name = strdup(arg); + int err; + + if (!name) + return -ENOMEM; + + lr->start = 0; + lr->end = INT_MAX; + + range = strchr(name, ':'); + if (range) { + *range++ = '\0'; + + err = parse_line_num(&range, &lr->start, "start line"); + if (err) + goto err; + + if (*range == '+' || *range == '-') { + const char c = *range++; + + err = parse_line_num(&range, &lr->end, "end line"); + if (err) + goto err; + + if (c == '+') { + lr->end += lr->start; + /* + * Adjust the number of lines here. + * If the number of lines == 1, the + * the end of line should be equal to + * the start of line. + */ + lr->end--; + } + } + pr_debug("Line range is %d to %d\n", lr->start, lr->end); + + err = -EINVAL; if (lr->start > lr->end) { semantic_error("Start line must be smaller" " than end line.\n"); - return -EINVAL; + goto err; } - if (*tmp != '\0') { - semantic_error("Tailing with invalid character '%d'.\n", - *tmp); - return -EINVAL; + if (*range != '\0') { + semantic_error("Tailing with invalid str '%s'.\n", range); + goto err; } - tmp = strndup(arg, (ptr - arg)); - } else { - tmp = strdup(arg); - lr->end = INT_MAX; } - if (tmp == NULL) - return -ENOMEM; - - if (strchr(tmp, '.')) - lr->file = tmp; + if (strchr(name, '.')) + lr->file = name; else - lr->function = tmp; + lr->function = name; return 0; +err: + free(name); + return err; } /* Check the name is good for event/group */ @@ -699,39 +742,40 @@ static int parse_perf_probe_point(char *arg, struct perf_probe_event *pev) /* Exclusion check */ if (pp->lazy_line && pp->line) { - semantic_error("Lazy pattern can't be used with line number."); + semantic_error("Lazy pattern can't be used with" + " line number.\n"); return -EINVAL; } if (pp->lazy_line && pp->offset) { - semantic_error("Lazy pattern can't be used with offset."); + semantic_error("Lazy pattern can't be used with offset.\n"); return -EINVAL; } if (pp->line && pp->offset) { - semantic_error("Offset can't be used with line number."); + semantic_error("Offset can't be used with line number.\n"); return -EINVAL; } if (!pp->line && !pp->lazy_line && pp->file && !pp->function) { semantic_error("File always requires line number or " - "lazy pattern."); + "lazy pattern.\n"); return -EINVAL; } if (pp->offset && !pp->function) { - semantic_error("Offset requires an entry function."); + semantic_error("Offset requires an entry function.\n"); return -EINVAL; } if (pp->retprobe && !pp->function) { - semantic_error("Return probe requires an entry function."); + semantic_error("Return probe requires an entry function.\n"); return -EINVAL; } if ((pp->offset || pp->line || pp->lazy_line) && pp->retprobe) { semantic_error("Offset/Line/Lazy pattern can't be used with " - "return probe."); + "return probe.\n"); return -EINVAL; } @@ -1005,7 +1049,7 @@ int synthesize_perf_probe_arg(struct perf_probe_arg *pa, char *buf, size_t len) return tmp - buf; error: - pr_debug("Failed to synthesize perf probe argument: %s", + pr_debug("Failed to synthesize perf probe argument: %s\n", strerror(-ret)); return ret; } @@ -1033,13 +1077,13 @@ static char *synthesize_perf_probe_point(struct perf_probe_point *pp) goto error; } if (pp->file) { - len = strlen(pp->file) - 31; - if (len < 0) - len = 0; - tmp = strchr(pp->file + len, '/'); - if (!tmp) - tmp = pp->file + len; - ret = e_snprintf(file, 32, "@%s", tmp + 1); + tmp = pp->file; + len = strlen(tmp); + if (len > 30) { + tmp = strchr(pp->file + len - 30, '/'); + tmp = tmp ? tmp + 1 : pp->file + len - 30; + } + ret = e_snprintf(file, 32, "@%s", tmp); if (ret <= 0) goto error; } @@ -1055,7 +1099,7 @@ static char *synthesize_perf_probe_point(struct perf_probe_point *pp) return buf; error: - pr_debug("Failed to synthesize perf probe point: %s", + pr_debug("Failed to synthesize perf probe point: %s\n", strerror(-ret)); if (buf) free(buf); @@ -1796,7 +1840,7 @@ static int del_trace_probe_event(int fd, const char *group, ret = e_snprintf(buf, 128, "%s:%s", group, event); if (ret < 0) { - pr_err("Failed to copy event."); + pr_err("Failed to copy event.\n"); return ret; } diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index ddf4d4556321..ab83b6ac5d65 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -652,8 +652,8 @@ static_var: regs = get_arch_regstr(regn); if (!regs) { /* This should be a bug in DWARF or this tool */ - pr_warning("Mapping for DWARF register number %u " - "missing on this architecture.", regn); + pr_warning("Mapping for the register number %u " + "missing on this architecture.\n", regn); return -ERANGE; } @@ -699,13 +699,14 @@ static int convert_variable_type(Dwarf_Die *vr_die, if (ret != DW_TAG_pointer_type && ret != DW_TAG_array_type) { pr_warning("Failed to cast into string: " - "%s(%s) is not a pointer nor array.", + "%s(%s) is not a pointer nor array.\n", dwarf_diename(vr_die), dwarf_diename(&type)); return -EINVAL; } if (ret == DW_TAG_pointer_type) { if (die_get_real_type(&type, &type) == NULL) { - pr_warning("Failed to get a type information."); + pr_warning("Failed to get a type" + " information.\n"); return -ENOENT; } while (*ref_ptr) @@ -720,7 +721,7 @@ static int convert_variable_type(Dwarf_Die *vr_die, if (!die_compare_name(&type, "char") && !die_compare_name(&type, "unsigned char")) { pr_warning("Failed to cast into string: " - "%s is not (unsigned) char *.", + "%s is not (unsigned) char *.\n", dwarf_diename(vr_die)); return -EINVAL; } @@ -830,8 +831,8 @@ static int convert_variable_fields(Dwarf_Die *vr_die, const char *varname, return -EINVAL; } if (field->name[0] == '[') { - pr_err("Semantic error: %s is not a pointor nor array.", - varname); + pr_err("Semantic error: %s is not a pointor" + " nor array.\n", varname); return -EINVAL; } if (field->ref) { @@ -978,7 +979,7 @@ static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr, name = dwarf_diename(sp_die); if (name) { if (dwarf_entrypc(sp_die, &eaddr) != 0) { - pr_warning("Failed to get entry pc of %s\n", + pr_warning("Failed to get entry address of %s\n", dwarf_diename(sp_die)); return -ENOENT; } @@ -994,7 +995,7 @@ static int convert_to_trace_point(Dwarf_Die *sp_die, Dwarf_Addr paddr, if (retprobe) { if (eaddr != paddr) { pr_warning("Return probe must be on the head of" - " a real function\n"); + " a real function.\n"); return -EINVAL; } tp->retprobe = true; @@ -1033,7 +1034,7 @@ static int call_probe_finder(Dwarf_Die *sp_die, struct probe_finder *pf) Dwarf_Frame *frame; if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 || dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) { - pr_warning("Failed to get CFA on 0x%jx\n", + pr_warning("Failed to get call frame on 0x%jx\n", (uintmax_t)pf->addr); return -ENOENT; } @@ -1060,7 +1061,7 @@ static int find_probe_point_by_line(struct probe_finder *pf) int ret = 0; if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) { - pr_warning("No source lines found in this CU.\n"); + pr_warning("No source lines found.\n"); return -ENOENT; } @@ -1162,7 +1163,7 @@ static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) } if (dwarf_getsrclines(&pf->cu_die, &lines, &nlines) != 0) { - pr_warning("No source lines found in this CU.\n"); + pr_warning("No source lines found.\n"); return -ENOENT; } @@ -1220,7 +1221,7 @@ static int probe_point_inline_cb(Dwarf_Die *in_die, void *data) else { /* Get probe address */ if (dwarf_entrypc(in_die, &addr) != 0) { - pr_warning("Failed to get entry pc of %s.\n", + pr_warning("Failed to get entry address of %s.\n", dwarf_diename(in_die)); param->retval = -ENOENT; return DWARF_CB_ABORT; @@ -1261,8 +1262,8 @@ static int probe_point_search_cb(Dwarf_Die *sp_die, void *data) param->retval = find_probe_point_lazy(sp_die, pf); else { if (dwarf_entrypc(sp_die, &pf->addr) != 0) { - pr_warning("Failed to get entry pc of %s.\n", - dwarf_diename(sp_die)); + pr_warning("Failed to get entry address of " + "%s.\n", dwarf_diename(sp_die)); param->retval = -ENOENT; return DWARF_CB_ABORT; } @@ -1304,7 +1305,7 @@ static int find_probes(int fd, struct probe_finder *pf) dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias); if (!dbg) { - pr_warning("No dwarf info found in the vmlinux - " + pr_warning("No debug information found in the vmlinux - " "please rebuild with CONFIG_DEBUG_INFO=y.\n"); return -EBADF; } @@ -1549,7 +1550,7 @@ int find_perf_probe_point(unsigned long addr, struct perf_probe_point *ppt) /* Open the live linux kernel */ dbg = dwfl_init_live_kernel_dwarf(addr, &dwfl, &bias); if (!dbg) { - pr_warning("No dwarf info found in the vmlinux - " + pr_warning("No debug information found in the vmlinux - " "please rebuild with CONFIG_DEBUG_INFO=y.\n"); ret = -EINVAL; goto end; @@ -1559,7 +1560,8 @@ int find_perf_probe_point(unsigned long addr, struct perf_probe_point *ppt) addr += bias; /* Find cu die */ if (!dwarf_addrdie(dbg, (Dwarf_Addr)addr - bias, &cudie)) { - pr_warning("No CU DIE is found at %lx\n", addr); + pr_warning("Failed to find debug information for address %lx\n", + addr); ret = -EINVAL; goto end; } @@ -1684,7 +1686,7 @@ static int find_line_range_by_line(Dwarf_Die *sp_die, struct line_finder *lf) line_list__init(&lf->lr->line_list); if (dwarf_getsrclines(&lf->cu_die, &lines, &nlines) != 0) { - pr_warning("No source lines found in this CU.\n"); + pr_warning("No source lines found.\n"); return -ENOENT; } @@ -1809,7 +1811,7 @@ int find_line_range(int fd, struct line_range *lr) dbg = dwfl_init_offline_dwarf(fd, &dwfl, &bias); if (!dbg) { - pr_warning("No dwarf info found in the vmlinux - " + pr_warning("No debug information found in the vmlinux - " "please rebuild with CONFIG_DEBUG_INFO=y.\n"); return -EBADF; } diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h index bba69d455699..beaefc3c1223 100644 --- a/tools/perf/util/probe-finder.h +++ b/tools/perf/util/probe-finder.h @@ -34,9 +34,9 @@ extern int find_available_vars_at(int fd, struct perf_probe_event *pev, bool externs); #include <dwarf.h> -#include <libdw.h> -#include <libdwfl.h> -#include <version.h> +#include <elfutils/libdw.h> +#include <elfutils/libdwfl.h> +#include <elfutils/version.h> struct probe_finder { struct perf_probe_event *pev; /* Target probe event */ diff --git a/tools/perf/util/scripting-engines/trace-event-perl.c b/tools/perf/util/scripting-engines/trace-event-perl.c index b059dc50cc2d..93680818e244 100644 --- a/tools/perf/util/scripting-engines/trace-event-perl.c +++ b/tools/perf/util/scripting-engines/trace-event-perl.c @@ -1,5 +1,5 @@ /* - * trace-event-perl. Feed perf trace events to an embedded Perl interpreter. + * trace-event-perl. Feed perf script events to an embedded Perl interpreter. * * Copyright (C) 2009 Tom Zanussi <tzanussi@gmail.com> * @@ -411,8 +411,8 @@ static int perl_generate_script(const char *outfile) return -1; } - fprintf(ofp, "# perf trace event handlers, " - "generated by perf trace -g perl\n"); + fprintf(ofp, "# perf script event handlers, " + "generated by perf script -g perl\n"); fprintf(ofp, "# Licensed under the terms of the GNU GPL" " License version 2\n\n"); diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 33a632523743..c6d99334bdfa 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -442,8 +442,8 @@ static int python_generate_script(const char *outfile) fprintf(stderr, "couldn't open %s\n", fname); return -1; } - fprintf(ofp, "# perf trace event handlers, " - "generated by perf trace -g python\n"); + fprintf(ofp, "# perf script event handlers, " + "generated by perf script -g python\n"); fprintf(ofp, "# Licensed under the terms of the GNU GPL" " License version 2\n\n"); diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index fa9d652c2dc3..6fb4694d05fa 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -65,9 +65,49 @@ out_close: return -1; } +static void perf_session__id_header_size(struct perf_session *session) +{ + struct sample_data *data; + u64 sample_type = session->sample_type; + u16 size = 0; + + if (!session->sample_id_all) + goto out; + + if (sample_type & PERF_SAMPLE_TID) + size += sizeof(data->tid) * 2; + + if (sample_type & PERF_SAMPLE_TIME) + size += sizeof(data->time); + + if (sample_type & PERF_SAMPLE_ID) + size += sizeof(data->id); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + size += sizeof(data->stream_id); + + if (sample_type & PERF_SAMPLE_CPU) + size += sizeof(data->cpu) * 2; +out: + session->id_hdr_size = size; +} + +void perf_session__set_sample_id_all(struct perf_session *session, bool value) +{ + session->sample_id_all = value; + perf_session__id_header_size(session); +} + +void perf_session__set_sample_type(struct perf_session *session, u64 type) +{ + session->sample_type = type; +} + void perf_session__update_sample_type(struct perf_session *self) { self->sample_type = perf_header__sample_type(&self->header); + self->sample_id_all = perf_header__sample_id_all(&self->header); + perf_session__id_header_size(self); } int perf_session__create_kernel_maps(struct perf_session *self) @@ -85,7 +125,9 @@ static void perf_session__destroy_kernel_maps(struct perf_session *self) machines__destroy_guest_kernel_maps(&self->machines); } -struct perf_session *perf_session__new(const char *filename, int mode, bool force, bool repipe) +struct perf_session *perf_session__new(const char *filename, int mode, + bool force, bool repipe, + struct perf_event_ops *ops) { size_t len = filename ? strlen(filename) + 1 : 0; struct perf_session *self = zalloc(sizeof(*self) + len); @@ -101,10 +143,20 @@ struct perf_session *perf_session__new(const char *filename, int mode, bool forc INIT_LIST_HEAD(&self->dead_threads); self->hists_tree = RB_ROOT; self->last_match = NULL; - self->mmap_window = 32; + /* + * On 64bit we can mmap the data file in one go. No need for tiny mmap + * slices. On 32bit we use 32MB. + */ +#if BITS_PER_LONG == 64 + self->mmap_window = ULLONG_MAX; +#else + self->mmap_window = 32 * 1024 * 1024ULL; +#endif self->machines = RB_ROOT; self->repipe = repipe; - INIT_LIST_HEAD(&self->ordered_samples.samples_head); + INIT_LIST_HEAD(&self->ordered_samples.samples); + INIT_LIST_HEAD(&self->ordered_samples.sample_cache); + INIT_LIST_HEAD(&self->ordered_samples.to_free); machine__init(&self->host_machine, "", HOST_KERNEL_ID); if (mode == O_RDONLY) { @@ -120,6 +172,13 @@ struct perf_session *perf_session__new(const char *filename, int mode, bool forc } perf_session__update_sample_type(self); + + if (ops && ops->ordering_requires_timestamps && + ops->ordered_samples && !self->sample_id_all) { + dump_printf("WARNING: No sample_id_all support, falling back to unordered processing\n"); + ops->ordered_samples = false; + } + out: return self; out_free: @@ -230,7 +289,15 @@ struct map_symbol *perf_session__resolve_callchain(struct perf_session *self, return syms; } +static int process_event_synth_stub(event_t *event __used, + struct perf_session *session __used) +{ + dump_printf(": unhandled!\n"); + return 0; +} + static int process_event_stub(event_t *event __used, + struct sample_data *sample __used, struct perf_session *session __used) { dump_printf(": unhandled!\n"); @@ -262,7 +329,7 @@ static void perf_event_ops__fill_defaults(struct perf_event_ops *handler) if (handler->exit == NULL) handler->exit = process_event_stub; if (handler->lost == NULL) - handler->lost = process_event_stub; + handler->lost = event__process_lost; if (handler->read == NULL) handler->read = process_event_stub; if (handler->throttle == NULL) @@ -270,13 +337,13 @@ static void perf_event_ops__fill_defaults(struct perf_event_ops *handler) if (handler->unthrottle == NULL) handler->unthrottle = process_event_stub; if (handler->attr == NULL) - handler->attr = process_event_stub; + handler->attr = process_event_synth_stub; if (handler->event_type == NULL) - handler->event_type = process_event_stub; + handler->event_type = process_event_synth_stub; if (handler->tracing_data == NULL) - handler->tracing_data = process_event_stub; + handler->tracing_data = process_event_synth_stub; if (handler->build_id == NULL) - handler->build_id = process_event_stub; + handler->build_id = process_event_synth_stub; if (handler->finished_round == NULL) { if (handler->ordered_samples) handler->finished_round = process_finished_round; @@ -386,33 +453,61 @@ static event__swap_op event__swap_ops[] = { struct sample_queue { u64 timestamp; - struct sample_event *event; + u64 file_offset; + event_t *event; struct list_head list; }; +static void perf_session_free_sample_buffers(struct perf_session *session) +{ + struct ordered_samples *os = &session->ordered_samples; + + while (!list_empty(&os->to_free)) { + struct sample_queue *sq; + + sq = list_entry(os->to_free.next, struct sample_queue, list); + list_del(&sq->list); + free(sq); + } +} + +static int perf_session_deliver_event(struct perf_session *session, + event_t *event, + struct sample_data *sample, + struct perf_event_ops *ops, + u64 file_offset); + static void flush_sample_queue(struct perf_session *s, struct perf_event_ops *ops) { - struct list_head *head = &s->ordered_samples.samples_head; - u64 limit = s->ordered_samples.next_flush; + struct ordered_samples *os = &s->ordered_samples; + struct list_head *head = &os->samples; struct sample_queue *tmp, *iter; + struct sample_data sample; + u64 limit = os->next_flush; + u64 last_ts = os->last_sample ? os->last_sample->timestamp : 0ULL; if (!ops->ordered_samples || !limit) return; list_for_each_entry_safe(iter, tmp, head, list) { if (iter->timestamp > limit) - return; + break; - if (iter == s->ordered_samples.last_inserted) - s->ordered_samples.last_inserted = NULL; + event__parse_sample(iter->event, s, &sample); + perf_session_deliver_event(s, iter->event, &sample, ops, + iter->file_offset); - ops->sample((event_t *)iter->event, s); - - s->ordered_samples.last_flush = iter->timestamp; + os->last_flush = iter->timestamp; list_del(&iter->list); - free(iter->event); - free(iter); + list_add(&iter->list, &os->sample_cache); + } + + if (list_empty(head)) { + os->last_sample = NULL; + } else if (last_ts <= limit) { + os->last_sample = + list_entry(head->prev, struct sample_queue, list); } } @@ -465,178 +560,265 @@ static int process_finished_round(event_t *event __used, return 0; } -static void __queue_sample_end(struct sample_queue *new, struct list_head *head) -{ - struct sample_queue *iter; - - list_for_each_entry_reverse(iter, head, list) { - if (iter->timestamp < new->timestamp) { - list_add(&new->list, &iter->list); - return; - } - } - - list_add(&new->list, head); -} - -static void __queue_sample_before(struct sample_queue *new, - struct sample_queue *iter, - struct list_head *head) -{ - list_for_each_entry_continue_reverse(iter, head, list) { - if (iter->timestamp < new->timestamp) { - list_add(&new->list, &iter->list); - return; - } - } - - list_add(&new->list, head); -} - -static void __queue_sample_after(struct sample_queue *new, - struct sample_queue *iter, - struct list_head *head) -{ - list_for_each_entry_continue(iter, head, list) { - if (iter->timestamp > new->timestamp) { - list_add_tail(&new->list, &iter->list); - return; - } - } - list_add_tail(&new->list, head); -} - /* The queue is ordered by time */ -static void __queue_sample_event(struct sample_queue *new, - struct perf_session *s) +static void __queue_event(struct sample_queue *new, struct perf_session *s) { - struct sample_queue *last_inserted = s->ordered_samples.last_inserted; - struct list_head *head = &s->ordered_samples.samples_head; + struct ordered_samples *os = &s->ordered_samples; + struct sample_queue *sample = os->last_sample; + u64 timestamp = new->timestamp; + struct list_head *p; + os->last_sample = new; - if (!last_inserted) { - __queue_sample_end(new, head); + if (!sample) { + list_add(&new->list, &os->samples); + os->max_timestamp = timestamp; return; } /* - * Most of the time the current event has a timestamp - * very close to the last event inserted, unless we just switched - * to another event buffer. Having a sorting based on a list and - * on the last inserted event that is close to the current one is - * probably more efficient than an rbtree based sorting. + * last_sample might point to some random place in the list as it's + * the last queued event. We expect that the new event is close to + * this. */ - if (last_inserted->timestamp >= new->timestamp) - __queue_sample_before(new, last_inserted, head); - else - __queue_sample_after(new, last_inserted, head); + if (sample->timestamp <= timestamp) { + while (sample->timestamp <= timestamp) { + p = sample->list.next; + if (p == &os->samples) { + list_add_tail(&new->list, &os->samples); + os->max_timestamp = timestamp; + return; + } + sample = list_entry(p, struct sample_queue, list); + } + list_add_tail(&new->list, &sample->list); + } else { + while (sample->timestamp > timestamp) { + p = sample->list.prev; + if (p == &os->samples) { + list_add(&new->list, &os->samples); + return; + } + sample = list_entry(p, struct sample_queue, list); + } + list_add(&new->list, &sample->list); + } } -static int queue_sample_event(event_t *event, struct sample_data *data, - struct perf_session *s) +#define MAX_SAMPLE_BUFFER (64 * 1024 / sizeof(struct sample_queue)) + +static int perf_session_queue_event(struct perf_session *s, event_t *event, + struct sample_data *data, u64 file_offset) { + struct ordered_samples *os = &s->ordered_samples; + struct list_head *sc = &os->sample_cache; u64 timestamp = data->time; struct sample_queue *new; + if (!timestamp || timestamp == ~0ULL) + return -ETIME; if (timestamp < s->ordered_samples.last_flush) { printf("Warning: Timestamp below last timeslice flush\n"); return -EINVAL; } - new = malloc(sizeof(*new)); - if (!new) - return -ENOMEM; + if (!list_empty(sc)) { + new = list_entry(sc->next, struct sample_queue, list); + list_del(&new->list); + } else if (os->sample_buffer) { + new = os->sample_buffer + os->sample_buffer_idx; + if (++os->sample_buffer_idx == MAX_SAMPLE_BUFFER) + os->sample_buffer = NULL; + } else { + os->sample_buffer = malloc(MAX_SAMPLE_BUFFER * sizeof(*new)); + if (!os->sample_buffer) + return -ENOMEM; + list_add(&os->sample_buffer->list, &os->to_free); + os->sample_buffer_idx = 2; + new = os->sample_buffer + 1; + } new->timestamp = timestamp; + new->file_offset = file_offset; + new->event = event; - new->event = malloc(event->header.size); - if (!new->event) { - free(new); - return -ENOMEM; - } + __queue_event(new, s); - memcpy(new->event, event, event->header.size); + return 0; +} - __queue_sample_event(new, s); - s->ordered_samples.last_inserted = new; +static void callchain__printf(struct sample_data *sample) +{ + unsigned int i; - if (new->timestamp > s->ordered_samples.max_timestamp) - s->ordered_samples.max_timestamp = new->timestamp; + printf("... chain: nr:%Lu\n", sample->callchain->nr); - return 0; + for (i = 0; i < sample->callchain->nr; i++) + printf("..... %2d: %016Lx\n", i, sample->callchain->ips[i]); } -static int perf_session__process_sample(event_t *event, struct perf_session *s, - struct perf_event_ops *ops) +static void perf_session__print_tstamp(struct perf_session *session, + event_t *event, + struct sample_data *sample) { - struct sample_data data; + if (event->header.type != PERF_RECORD_SAMPLE && + !session->sample_id_all) { + fputs("-1 -1 ", stdout); + return; + } - if (!ops->ordered_samples) - return ops->sample(event, s); + if ((session->sample_type & PERF_SAMPLE_CPU)) + printf("%u ", sample->cpu); - bzero(&data, sizeof(struct sample_data)); - event__parse_sample(event, s->sample_type, &data); + if (session->sample_type & PERF_SAMPLE_TIME) + printf("%Lu ", sample->time); +} - queue_sample_event(event, &data, s); +static void dump_event(struct perf_session *session, event_t *event, + u64 file_offset, struct sample_data *sample) +{ + if (!dump_trace) + return; - return 0; + printf("\n%#Lx [%#x]: event: %d\n", file_offset, event->header.size, + event->header.type); + + trace_event(event); + + if (sample) + perf_session__print_tstamp(session, event, sample); + + printf("%#Lx [%#x]: PERF_RECORD_%s", file_offset, event->header.size, + event__get_event_name(event->header.type)); } -static int perf_session__process_event(struct perf_session *self, - event_t *event, - struct perf_event_ops *ops, - u64 offset, u64 head) +static void dump_sample(struct perf_session *session, event_t *event, + struct sample_data *sample) { - trace_event(event); + if (!dump_trace) + return; - if (event->header.type < PERF_RECORD_HEADER_MAX) { - dump_printf("%#Lx [%#x]: PERF_RECORD_%s", - offset + head, event->header.size, - event__name[event->header.type]); - hists__inc_nr_events(&self->hists, event->header.type); - } + printf("(IP, %d): %d/%d: %#Lx period: %Ld\n", event->header.misc, + sample->pid, sample->tid, sample->ip, sample->period); - if (self->header.needs_swap && event__swap_ops[event->header.type]) - event__swap_ops[event->header.type](event); + if (session->sample_type & PERF_SAMPLE_CALLCHAIN) + callchain__printf(sample); +} + +static int perf_session_deliver_event(struct perf_session *session, + event_t *event, + struct sample_data *sample, + struct perf_event_ops *ops, + u64 file_offset) +{ + dump_event(session, event, file_offset, sample); switch (event->header.type) { case PERF_RECORD_SAMPLE: - return perf_session__process_sample(event, self, ops); + dump_sample(session, event, sample); + return ops->sample(event, sample, session); case PERF_RECORD_MMAP: - return ops->mmap(event, self); + return ops->mmap(event, sample, session); case PERF_RECORD_COMM: - return ops->comm(event, self); + return ops->comm(event, sample, session); case PERF_RECORD_FORK: - return ops->fork(event, self); + return ops->fork(event, sample, session); case PERF_RECORD_EXIT: - return ops->exit(event, self); + return ops->exit(event, sample, session); case PERF_RECORD_LOST: - return ops->lost(event, self); + return ops->lost(event, sample, session); case PERF_RECORD_READ: - return ops->read(event, self); + return ops->read(event, sample, session); case PERF_RECORD_THROTTLE: - return ops->throttle(event, self); + return ops->throttle(event, sample, session); case PERF_RECORD_UNTHROTTLE: - return ops->unthrottle(event, self); + return ops->unthrottle(event, sample, session); + default: + ++session->hists.stats.nr_unknown_events; + return -1; + } +} + +static int perf_session__preprocess_sample(struct perf_session *session, + event_t *event, struct sample_data *sample) +{ + if (event->header.type != PERF_RECORD_SAMPLE || + !(session->sample_type & PERF_SAMPLE_CALLCHAIN)) + return 0; + + if (!ip_callchain__valid(sample->callchain, event)) { + pr_debug("call-chain problem with event, skipping it.\n"); + ++session->hists.stats.nr_invalid_chains; + session->hists.stats.total_invalid_chains += sample->period; + return -EINVAL; + } + return 0; +} + +static int perf_session__process_user_event(struct perf_session *session, event_t *event, + struct perf_event_ops *ops, u64 file_offset) +{ + dump_event(session, event, file_offset, NULL); + + /* These events are processed right away */ + switch (event->header.type) { case PERF_RECORD_HEADER_ATTR: - return ops->attr(event, self); + return ops->attr(event, session); case PERF_RECORD_HEADER_EVENT_TYPE: - return ops->event_type(event, self); + return ops->event_type(event, session); case PERF_RECORD_HEADER_TRACING_DATA: /* setup for reading amidst mmap */ - lseek(self->fd, offset + head, SEEK_SET); - return ops->tracing_data(event, self); + lseek(session->fd, file_offset, SEEK_SET); + return ops->tracing_data(event, session); case PERF_RECORD_HEADER_BUILD_ID: - return ops->build_id(event, self); + return ops->build_id(event, session); case PERF_RECORD_FINISHED_ROUND: - return ops->finished_round(event, self, ops); + return ops->finished_round(event, session, ops); default: - ++self->hists.stats.nr_unknown_events; - return -1; + return -EINVAL; } } +static int perf_session__process_event(struct perf_session *session, + event_t *event, + struct perf_event_ops *ops, + u64 file_offset) +{ + struct sample_data sample; + int ret; + + if (session->header.needs_swap && event__swap_ops[event->header.type]) + event__swap_ops[event->header.type](event); + + if (event->header.type >= PERF_RECORD_HEADER_MAX) + return -EINVAL; + + hists__inc_nr_events(&session->hists, event->header.type); + + if (event->header.type >= PERF_RECORD_USER_TYPE_START) + return perf_session__process_user_event(session, event, ops, file_offset); + + /* + * For all kernel events we get the sample data + */ + event__parse_sample(event, session, &sample); + + /* Preprocess sample records - precheck callchains */ + if (perf_session__preprocess_sample(session, event, &sample)) + return 0; + + if (ops->ordered_samples) { + ret = perf_session_queue_event(session, event, &sample, + file_offset); + if (ret != -ETIME) + return ret; + } + + return perf_session_deliver_event(session, event, &sample, ops, + file_offset); +} + void perf_event_header__bswap(struct perf_event_header *self) { self->type = bswap_32(self->type); @@ -656,21 +838,33 @@ static struct thread *perf_session__register_idle_thread(struct perf_session *se return thread; } -int do_read(int fd, void *buf, size_t size) +static void perf_session__warn_about_errors(const struct perf_session *session, + const struct perf_event_ops *ops) { - void *buf_start = buf; - - while (size) { - int ret = read(fd, buf, size); - - if (ret <= 0) - return ret; + if (ops->lost == event__process_lost && + session->hists.stats.total_lost != 0) { + ui__warning("Processed %Lu events and LOST %Lu!\n\n" + "Check IO/CPU overload!\n\n", + session->hists.stats.total_period, + session->hists.stats.total_lost); + } - size -= ret; - buf += ret; + if (session->hists.stats.nr_unknown_events != 0) { + ui__warning("Found %u unknown events!\n\n" + "Is this an older tool processing a perf.data " + "file generated by a more recent tool?\n\n" + "If that is not the case, consider " + "reporting to linux-kernel@vger.kernel.org.\n\n", + session->hists.stats.nr_unknown_events); } - return buf - buf_start; + if (session->hists.stats.nr_invalid_chains != 0) { + ui__warning("Found invalid callchains!\n\n" + "%u out of %u events were discarded for this reason.\n\n" + "Consider reporting to linux-kernel@vger.kernel.org.\n\n", + session->hists.stats.nr_invalid_chains, + session->hists.stats.nr_events[PERF_RECORD_SAMPLE]); + } } #define session_done() (*(volatile int *)(&session_done)) @@ -690,7 +884,7 @@ static int __perf_session__process_pipe_events(struct perf_session *self, head = 0; more: - err = do_read(self->fd, &event, sizeof(struct perf_event_header)); + err = readn(self->fd, &event, sizeof(struct perf_event_header)); if (err <= 0) { if (err == 0) goto done; @@ -710,8 +904,7 @@ more: p += sizeof(struct perf_event_header); if (size - sizeof(struct perf_event_header)) { - err = do_read(self->fd, p, - size - sizeof(struct perf_event_header)); + err = readn(self->fd, p, size - sizeof(struct perf_event_header)); if (err <= 0) { if (err == 0) { pr_err("unexpected end of event stream\n"); @@ -724,8 +917,7 @@ more: } if (size == 0 || - (skip = perf_session__process_event(self, &event, ops, - 0, head)) < 0) { + (skip = perf_session__process_event(self, &event, ops, head)) < 0) { dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n", head, event.header.size, event.header.type); /* @@ -740,9 +932,6 @@ more: head += size; - dump_printf("\n%#Lx [%#x]: event: %d\n", - head, event.header.size, event.header.type); - if (skip > 0) head += skip; @@ -751,82 +940,91 @@ more: done: err = 0; out_err: + perf_session__warn_about_errors(self, ops); + perf_session_free_sample_buffers(self); return err; } -int __perf_session__process_events(struct perf_session *self, +int __perf_session__process_events(struct perf_session *session, u64 data_offset, u64 data_size, u64 file_size, struct perf_event_ops *ops) { - int err, mmap_prot, mmap_flags; - u64 head, shift; - u64 offset = 0; - size_t page_size; + u64 head, page_offset, file_offset, file_pos, progress_next; + int err, mmap_prot, mmap_flags, map_idx = 0; + struct ui_progress *progress; + size_t page_size, mmap_size; + char *buf, *mmaps[8]; event_t *event; uint32_t size; - char *buf; - struct ui_progress *progress = ui_progress__new("Processing events...", - self->size); - if (progress == NULL) - return -1; perf_event_ops__fill_defaults(ops); page_size = sysconf(_SC_PAGESIZE); - head = data_offset; - shift = page_size * (head / page_size); - offset += shift; - head -= shift; + page_offset = page_size * (data_offset / page_size); + file_offset = page_offset; + head = data_offset - page_offset; + + if (data_offset + data_size < file_size) + file_size = data_offset + data_size; + + progress_next = file_size / 16; + progress = ui_progress__new("Processing events...", file_size); + if (progress == NULL) + return -1; + + mmap_size = session->mmap_window; + if (mmap_size > file_size) + mmap_size = file_size; + + memset(mmaps, 0, sizeof(mmaps)); mmap_prot = PROT_READ; mmap_flags = MAP_SHARED; - if (self->header.needs_swap) { + if (session->header.needs_swap) { mmap_prot |= PROT_WRITE; mmap_flags = MAP_PRIVATE; } remap: - buf = mmap(NULL, page_size * self->mmap_window, mmap_prot, - mmap_flags, self->fd, offset); + buf = mmap(NULL, mmap_size, mmap_prot, mmap_flags, session->fd, + file_offset); if (buf == MAP_FAILED) { pr_err("failed to mmap file\n"); err = -errno; goto out_err; } + mmaps[map_idx] = buf; + map_idx = (map_idx + 1) & (ARRAY_SIZE(mmaps) - 1); + file_pos = file_offset + head; more: event = (event_t *)(buf + head); - ui_progress__update(progress, offset); - if (self->header.needs_swap) + if (session->header.needs_swap) perf_event_header__bswap(&event->header); size = event->header.size; if (size == 0) size = 8; - if (head + event->header.size >= page_size * self->mmap_window) { - int munmap_ret; - - shift = page_size * (head / page_size); - - munmap_ret = munmap(buf, page_size * self->mmap_window); - assert(munmap_ret == 0); + if (head + event->header.size >= mmap_size) { + if (mmaps[map_idx]) { + munmap(mmaps[map_idx], mmap_size); + mmaps[map_idx] = NULL; + } - offset += shift; - head -= shift; + page_offset = page_size * (head / page_size); + file_offset += page_offset; + head -= page_offset; goto remap; } size = event->header.size; - dump_printf("\n%#Lx [%#x]: event: %d\n", - offset + head, event->header.size, event->header.type); - if (size == 0 || - perf_session__process_event(self, event, ops, offset, head) < 0) { + perf_session__process_event(session, event, ops, file_pos) < 0) { dump_printf("%#Lx [%#x]: skipping unknown header type: %d\n", - offset + head, event->header.size, + file_offset + head, event->header.size, event->header.type); /* * assume we lost track of the stream, check alignment, and @@ -839,19 +1037,24 @@ more: } head += size; + file_pos += size; - if (offset + head >= data_offset + data_size) - goto done; + if (file_pos >= progress_next) { + progress_next += file_size / 16; + ui_progress__update(progress, file_pos); + } - if (offset + head < file_size) + if (file_pos < file_size) goto more; -done: + err = 0; /* do the final flush for ordered samples */ - self->ordered_samples.next_flush = ULLONG_MAX; - flush_sample_queue(self, ops); + session->ordered_samples.next_flush = ULLONG_MAX; + flush_sample_queue(session, ops); out_err: ui_progress__delete(progress); + perf_session__warn_about_errors(session, ops); + perf_session_free_sample_buffers(session); return err; } diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h index 9fa0fc2a863f..decd83f274fd 100644 --- a/tools/perf/util/session.h +++ b/tools/perf/util/session.h @@ -17,8 +17,12 @@ struct ordered_samples { u64 last_flush; u64 next_flush; u64 max_timestamp; - struct list_head samples_head; - struct sample_queue *last_inserted; + struct list_head samples; + struct list_head sample_cache; + struct list_head to_free; + struct sample_queue *sample_buffer; + struct sample_queue *last_sample; + int sample_buffer_idx; }; struct perf_session { @@ -42,6 +46,8 @@ struct perf_session { int fd; bool fd_pipe; bool repipe; + bool sample_id_all; + u16 id_hdr_size; int cwdlen; char *cwd; struct ordered_samples ordered_samples; @@ -50,7 +56,9 @@ struct perf_session { struct perf_event_ops; -typedef int (*event_op)(event_t *self, struct perf_session *session); +typedef int (*event_op)(event_t *self, struct sample_data *sample, + struct perf_session *session); +typedef int (*event_synth_op)(event_t *self, struct perf_session *session); typedef int (*event_op2)(event_t *self, struct perf_session *session, struct perf_event_ops *ops); @@ -63,16 +71,19 @@ struct perf_event_ops { lost, read, throttle, - unthrottle, - attr, + unthrottle; + event_synth_op attr, event_type, tracing_data, build_id; event_op2 finished_round; bool ordered_samples; + bool ordering_requires_timestamps; }; -struct perf_session *perf_session__new(const char *filename, int mode, bool force, bool repipe); +struct perf_session *perf_session__new(const char *filename, int mode, + bool force, bool repipe, + struct perf_event_ops *ops); void perf_session__delete(struct perf_session *self); void perf_event_header__bswap(struct perf_event_header *self); @@ -98,8 +109,9 @@ void mem_bswap_64(void *src, int byte_size); int perf_session__create_kernel_maps(struct perf_session *self); -int do_read(int fd, void *buf, size_t size); void perf_session__update_sample_type(struct perf_session *self); +void perf_session__set_sample_id_all(struct perf_session *session, bool value); +void perf_session__set_sample_type(struct perf_session *session, u64 type); void perf_session__remove_thread(struct perf_session *self, struct thread *th); static inline diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index b62a553cc67d..f44fa541d56e 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -170,7 +170,7 @@ static int hist_entry__dso_snprintf(struct hist_entry *self, char *bf, return repsep_snprintf(bf, size, "%-*s", width, dso_name); } - return repsep_snprintf(bf, size, "%*Lx", width, self->ip); + return repsep_snprintf(bf, size, "%-*s", width, "[unknown]"); } /* --sort symbol */ @@ -196,7 +196,7 @@ static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf, if (verbose) { char o = self->ms.map ? dso__symtab_origin(self->ms.map->dso) : '!'; - ret += repsep_snprintf(bf, size, "%*Lx %c ", + ret += repsep_snprintf(bf, size, "%-#*llx %c ", BITS_PER_LONG / 4, self->ip, o); } @@ -205,7 +205,7 @@ static int hist_entry__sym_snprintf(struct hist_entry *self, char *bf, ret += repsep_snprintf(bf + ret, size - ret, "%s", self->ms.sym->name); else - ret += repsep_snprintf(bf + ret, size - ret, "%*Lx", + ret += repsep_snprintf(bf + ret, size - ret, "%-#*llx", BITS_PER_LONG / 4, self->ip); return ret; diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 439ab947daf4..15ccfba8cdf8 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -22,6 +22,10 @@ #include <limits.h> #include <sys/utsname.h> +#ifndef KSYM_NAME_LEN +#define KSYM_NAME_LEN 128 +#endif + #ifndef NT_GNU_BUILD_ID #define NT_GNU_BUILD_ID 3 #endif @@ -41,6 +45,7 @@ struct symbol_conf symbol_conf = { .exclude_other = true, .use_modules = true, .try_vmlinux_path = true, + .symfs = "", }; int dso__name_len(const struct dso *self) @@ -92,7 +97,7 @@ static void symbols__fixup_end(struct rb_root *self) prev = curr; curr = rb_entry(nd, struct symbol, rb_node); - if (prev->end == prev->start) + if (prev->end == prev->start && prev->end != curr->start) prev->end = curr->start - 1; } @@ -121,7 +126,7 @@ static void __map_groups__fixup_end(struct map_groups *self, enum map_type type) * We still haven't the actual symbols, so guess the * last map final address. */ - curr->end = ~0UL; + curr->end = ~0ULL; } static void map_groups__fixup_end(struct map_groups *self) @@ -425,16 +430,25 @@ size_t dso__fprintf(struct dso *self, enum map_type type, FILE *fp) int kallsyms__parse(const char *filename, void *arg, int (*process_symbol)(void *arg, const char *name, - char type, u64 start)) + char type, u64 start, u64 end)) { char *line = NULL; size_t n; - int err = 0; + int err = -1; + u64 prev_start = 0; + char prev_symbol_type = 0; + char *prev_symbol_name; FILE *file = fopen(filename, "r"); if (file == NULL) goto out_failure; + prev_symbol_name = malloc(KSYM_NAME_LEN); + if (prev_symbol_name == NULL) + goto out_close; + + err = 0; + while (!feof(file)) { u64 start; int line_len, len; @@ -454,14 +468,33 @@ int kallsyms__parse(const char *filename, void *arg, continue; symbol_type = toupper(line[len]); - symbol_name = line + len + 2; + len += 2; + symbol_name = line + len; + len = line_len - len; - err = process_symbol(arg, symbol_name, symbol_type, start); - if (err) + if (len >= KSYM_NAME_LEN) { + err = -1; break; + } + + if (prev_symbol_type) { + u64 end = start; + if (end != prev_start) + --end; + err = process_symbol(arg, prev_symbol_name, + prev_symbol_type, prev_start, end); + if (err) + break; + } + + memcpy(prev_symbol_name, symbol_name, len + 1); + prev_symbol_type = symbol_type; + prev_start = start; } + free(prev_symbol_name); free(line); +out_close: fclose(file); return err; @@ -483,7 +516,7 @@ static u8 kallsyms2elf_type(char type) } static int map__process_kallsym_symbol(void *arg, const char *name, - char type, u64 start) + char type, u64 start, u64 end) { struct symbol *sym; struct process_kallsyms_args *a = arg; @@ -492,11 +525,8 @@ static int map__process_kallsym_symbol(void *arg, const char *name, if (!symbol_type__is_a(type, a->map->type)) return 0; - /* - * Will fix up the end later, when we have all symbols sorted. - */ - sym = symbol__new(start, 0, kallsyms2elf_type(type), name); - + sym = symbol__new(start, end - start + 1, + kallsyms2elf_type(type), name); if (sym == NULL) return -ENOMEM; /* @@ -649,7 +679,6 @@ int dso__load_kallsyms(struct dso *self, const char *filename, if (dso__load_all_kallsyms(self, filename, map) < 0) return -1; - symbols__fixup_end(&self->symbols[map->type]); if (self->kernel == DSO_TYPE_GUEST_KERNEL) self->origin = DSO__ORIG_GUEST_KERNEL; else @@ -839,8 +868,11 @@ static int dso__synthesize_plt_symbols(struct dso *self, struct map *map, char sympltname[1024]; Elf *elf; int nr = 0, symidx, fd, err = 0; + char name[PATH_MAX]; - fd = open(self->long_name, O_RDONLY); + snprintf(name, sizeof(name), "%s%s", + symbol_conf.symfs, self->long_name); + fd = open(name, O_RDONLY); if (fd < 0) goto out; @@ -1452,16 +1484,19 @@ int dso__load(struct dso *self, struct map *map, symbol_filter_t filter) self->origin++) { switch (self->origin) { case DSO__ORIG_BUILD_ID_CACHE: - if (dso__build_id_filename(self, name, size) == NULL) + /* skip the locally configured cache if a symfs is given */ + if (symbol_conf.symfs[0] || + (dso__build_id_filename(self, name, size) == NULL)) { continue; + } break; case DSO__ORIG_FEDORA: - snprintf(name, size, "/usr/lib/debug%s.debug", - self->long_name); + snprintf(name, size, "%s/usr/lib/debug%s.debug", + symbol_conf.symfs, self->long_name); break; case DSO__ORIG_UBUNTU: - snprintf(name, size, "/usr/lib/debug%s", - self->long_name); + snprintf(name, size, "%s/usr/lib/debug%s", + symbol_conf.symfs, self->long_name); break; case DSO__ORIG_BUILDID: { char build_id_hex[BUILD_ID_SIZE * 2 + 1]; @@ -1473,19 +1508,26 @@ int dso__load(struct dso *self, struct map *map, symbol_filter_t filter) sizeof(self->build_id), build_id_hex); snprintf(name, size, - "/usr/lib/debug/.build-id/%.2s/%s.debug", - build_id_hex, build_id_hex + 2); + "%s/usr/lib/debug/.build-id/%.2s/%s.debug", + symbol_conf.symfs, build_id_hex, build_id_hex + 2); } break; case DSO__ORIG_DSO: - snprintf(name, size, "%s", self->long_name); + snprintf(name, size, "%s%s", + symbol_conf.symfs, self->long_name); break; case DSO__ORIG_GUEST_KMODULE: if (map->groups && map->groups->machine) root_dir = map->groups->machine->root_dir; else root_dir = ""; - snprintf(name, size, "%s%s", root_dir, self->long_name); + snprintf(name, size, "%s%s%s", symbol_conf.symfs, + root_dir, self->long_name); + break; + + case DSO__ORIG_KMODULE: + snprintf(name, size, "%s%s", symbol_conf.symfs, + self->long_name); break; default: @@ -1784,17 +1826,20 @@ int dso__load_vmlinux(struct dso *self, struct map *map, const char *vmlinux, symbol_filter_t filter) { int err = -1, fd; + char symfs_vmlinux[PATH_MAX]; - fd = open(vmlinux, O_RDONLY); + snprintf(symfs_vmlinux, sizeof(symfs_vmlinux), "%s/%s", + symbol_conf.symfs, vmlinux); + fd = open(symfs_vmlinux, O_RDONLY); if (fd < 0) return -1; dso__set_loaded(self, map->type); - err = dso__load_sym(self, map, vmlinux, fd, filter, 0, 0); + err = dso__load_sym(self, map, symfs_vmlinux, fd, filter, 0, 0); close(fd); if (err > 0) - pr_debug("Using %s for symbols\n", vmlinux); + pr_debug("Using %s for symbols\n", symfs_vmlinux); return err; } @@ -1836,8 +1881,8 @@ static int dso__load_kernel_sym(struct dso *self, struct map *map, const char *kallsyms_filename = NULL; char *kallsyms_allocated_filename = NULL; /* - * Step 1: if the user specified a vmlinux filename, use it and only - * it, reporting errors to the user if it cannot be used. + * Step 1: if the user specified a kallsyms or vmlinux filename, use + * it and only it, reporting errors to the user if it cannot be used. * * For instance, try to analyse an ARM perf.data file _without_ a * build-id, or if the user specifies the wrong path to the right @@ -1850,6 +1895,11 @@ static int dso__load_kernel_sym(struct dso *self, struct map *map, * validation in dso__load_vmlinux and will bail out if they don't * match. */ + if (symbol_conf.kallsyms_name != NULL) { + kallsyms_filename = symbol_conf.kallsyms_name; + goto do_kallsyms; + } + if (symbol_conf.vmlinux_name != NULL) { err = dso__load_vmlinux(self, map, symbol_conf.vmlinux_name, filter); @@ -1867,6 +1917,10 @@ static int dso__load_kernel_sym(struct dso *self, struct map *map, goto out_fixup; } + /* do not try local files if a symfs was given */ + if (symbol_conf.symfs[0] != 0) + return -1; + /* * Say the kernel DSO was created when processing the build-id header table, * we have a build-id, so check if it is the same as the running kernel, @@ -2136,7 +2190,7 @@ struct process_args { }; static int symbol__in_kernel(void *arg, const char *name, - char type __used, u64 start) + char type __used, u64 start, u64 end __used) { struct process_args *args = arg; @@ -2257,9 +2311,6 @@ static int vmlinux_path__init(void) struct utsname uts; char bf[PATH_MAX]; - if (uname(&uts) < 0) - return -1; - vmlinux_path = malloc(sizeof(char *) * 5); if (vmlinux_path == NULL) return -1; @@ -2272,6 +2323,14 @@ static int vmlinux_path__init(void) if (vmlinux_path[vmlinux_path__nr_entries] == NULL) goto out_fail; ++vmlinux_path__nr_entries; + + /* only try running kernel version if no symfs was given */ + if (symbol_conf.symfs[0] != 0) + return 0; + + if (uname(&uts) < 0) + return -1; + snprintf(bf, sizeof(bf), "/boot/vmlinux-%s", uts.release); vmlinux_path[vmlinux_path__nr_entries] = strdup(bf); if (vmlinux_path[vmlinux_path__nr_entries] == NULL) @@ -2331,6 +2390,8 @@ static int setup_list(struct strlist **list, const char *list_str, int symbol__init(void) { + const char *symfs; + if (symbol_conf.initialized) return 0; @@ -2359,6 +2420,18 @@ int symbol__init(void) symbol_conf.sym_list_str, "symbol") < 0) goto out_free_comm_list; + /* + * A path to symbols of "/" is identical to "" + * reset here for simplicity. + */ + symfs = realpath(symbol_conf.symfs, NULL); + if (symfs == NULL) + symfs = symbol_conf.symfs; + if (strcmp(symfs, "/") == 0) + symbol_conf.symfs = ""; + if (symfs != symbol_conf.symfs) + free((void *)symfs); + symbol_conf.initialized = true; return 0; diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 6c6eafdb932d..670cd1c88f54 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -72,6 +72,7 @@ struct symbol_conf { show_cpu_utilization, initialized; const char *vmlinux_name, + *kallsyms_name, *source_prefix, *field_sep; const char *default_guest_vmlinux_name, @@ -85,6 +86,7 @@ struct symbol_conf { struct strlist *dso_list, *comm_list, *sym_list; + const char *symfs; }; extern struct symbol_conf symbol_conf; @@ -215,7 +217,7 @@ bool __dsos__read_build_ids(struct list_head *head, bool with_hits); int build_id__sprintf(const u8 *self, int len, char *bf); int kallsyms__parse(const char *filename, void *arg, int (*process_symbol)(void *arg, const char *name, - char type, u64 start)); + char type, u64 start, u64 end)); void machine__destroy_kernel_maps(struct machine *self); int __machine__create_kernel_maps(struct machine *self, struct dso *kernel); diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index 8c72d888e449..00f4eade2e3e 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -16,35 +16,50 @@ static int filter(const struct dirent *dir) return 1; } -int find_all_tid(int pid, pid_t ** all_tid) +struct thread_map *thread_map__new_by_pid(pid_t pid) { + struct thread_map *threads; char name[256]; int items; struct dirent **namelist = NULL; - int ret = 0; int i; sprintf(name, "/proc/%d/task", pid); items = scandir(name, &namelist, filter, NULL); if (items <= 0) - return -ENOENT; - *all_tid = malloc(sizeof(pid_t) * items); - if (!*all_tid) { - ret = -ENOMEM; - goto failure; - } - - for (i = 0; i < items; i++) - (*all_tid)[i] = atoi(namelist[i]->d_name); + return NULL; - ret = items; + threads = malloc(sizeof(*threads) + sizeof(pid_t) * items); + if (threads != NULL) { + for (i = 0; i < items; i++) + threads->map[i] = atoi(namelist[i]->d_name); + threads->nr = items; + } -failure: for (i=0; i<items; i++) free(namelist[i]); free(namelist); - return ret; + return threads; +} + +struct thread_map *thread_map__new_by_tid(pid_t tid) +{ + struct thread_map *threads = malloc(sizeof(*threads) + sizeof(pid_t)); + + if (threads != NULL) { + threads->map[0] = tid; + threads->nr = 1; + } + + return threads; +} + +struct thread_map *thread_map__new(pid_t pid, pid_t tid) +{ + if (pid != -1) + return thread_map__new_by_pid(pid); + return thread_map__new_by_tid(tid); } static struct thread *thread__new(pid_t pid) diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index 688500ff826f..d7574101054a 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -18,11 +18,24 @@ struct thread { int comm_len; }; +struct thread_map { + int nr; + int map[]; +}; + struct perf_session; void thread__delete(struct thread *self); -int find_all_tid(int pid, pid_t ** all_tid); +struct thread_map *thread_map__new_by_pid(pid_t pid); +struct thread_map *thread_map__new_by_tid(pid_t tid); +struct thread_map *thread_map__new(pid_t pid, pid_t tid); + +static inline void thread_map__delete(struct thread_map *threads) +{ + free(threads); +} + int thread__set_comm(struct thread *self, const char *comm); int thread__comm_len(struct thread *self); struct thread *perf_session__findnew(struct perf_session *self, pid_t pid); diff --git a/tools/perf/util/trace-event-info.c b/tools/perf/util/trace-event-info.c index b1572601286c..35729f4c40cb 100644 --- a/tools/perf/util/trace-event-info.c +++ b/tools/perf/util/trace-event-info.c @@ -34,11 +34,13 @@ #include <ctype.h> #include <errno.h> #include <stdbool.h> +#include <linux/list.h> #include <linux/kernel.h> #include "../perf.h" #include "trace-event.h" #include "debugfs.h" +#include "evsel.h" #define VERSION "0.5" @@ -469,16 +471,17 @@ out: } static struct tracepoint_path * -get_tracepoints_path(struct perf_event_attr *pattrs, int nb_events) +get_tracepoints_path(struct list_head *pattrs) { struct tracepoint_path path, *ppath = &path; - int i, nr_tracepoints = 0; + struct perf_evsel *pos; + int nr_tracepoints = 0; - for (i = 0; i < nb_events; i++) { - if (pattrs[i].type != PERF_TYPE_TRACEPOINT) + list_for_each_entry(pos, pattrs, node) { + if (pos->attr.type != PERF_TYPE_TRACEPOINT) continue; ++nr_tracepoints; - ppath->next = tracepoint_id_to_path(pattrs[i].config); + ppath->next = tracepoint_id_to_path(pos->attr.config); if (!ppath->next) die("%s\n", "No memory to alloc tracepoints list"); ppath = ppath->next; @@ -487,21 +490,21 @@ get_tracepoints_path(struct perf_event_attr *pattrs, int nb_events) return nr_tracepoints > 0 ? path.next : NULL; } -bool have_tracepoints(struct perf_event_attr *pattrs, int nb_events) +bool have_tracepoints(struct list_head *pattrs) { - int i; + struct perf_evsel *pos; - for (i = 0; i < nb_events; i++) - if (pattrs[i].type == PERF_TYPE_TRACEPOINT) + list_for_each_entry(pos, pattrs, node) + if (pos->attr.type == PERF_TYPE_TRACEPOINT) return true; return false; } -int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events) +int read_tracing_data(int fd, struct list_head *pattrs) { char buf[BUFSIZ]; - struct tracepoint_path *tps = get_tracepoints_path(pattrs, nb_events); + struct tracepoint_path *tps = get_tracepoints_path(pattrs); /* * What? No tracepoints? No sense writing anything here, bail out. @@ -545,14 +548,13 @@ int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events) return 0; } -ssize_t read_tracing_data_size(int fd, struct perf_event_attr *pattrs, - int nb_events) +ssize_t read_tracing_data_size(int fd, struct list_head *pattrs) { ssize_t size; int err = 0; calc_data_size = 1; - err = read_tracing_data(fd, pattrs, nb_events); + err = read_tracing_data(fd, pattrs); size = calc_data_size - 1; calc_data_size = 0; diff --git a/tools/perf/util/trace-event.h b/tools/perf/util/trace-event.h index b3e86b1e4444..b5f12ca24d99 100644 --- a/tools/perf/util/trace-event.h +++ b/tools/perf/util/trace-event.h @@ -262,9 +262,8 @@ raw_field_value(struct event *event, const char *name, void *data); void *raw_field_ptr(struct event *event, const char *name, void *data); unsigned long long eval_flag(const char *flag); -int read_tracing_data(int fd, struct perf_event_attr *pattrs, int nb_events); -ssize_t read_tracing_data_size(int fd, struct perf_event_attr *pattrs, - int nb_events); +int read_tracing_data(int fd, struct list_head *pattrs); +ssize_t read_tracing_data_size(int fd, struct list_head *pattrs); /* taken from kernel/trace/trace.h */ enum trace_flag_type { diff --git a/tools/perf/util/ui/util.c b/tools/perf/util/ui/util.c index 056c69521a38..7b5a8926624e 100644 --- a/tools/perf/util/ui/util.c +++ b/tools/perf/util/ui/util.c @@ -104,10 +104,24 @@ out_destroy_form: return rc; } -static const char yes[] = "Yes", no[] = "No"; +static const char yes[] = "Yes", no[] = "No", + warning_str[] = "Warning!", ok[] = "Ok"; bool ui__dialog_yesno(const char *msg) { /* newtWinChoice should really be accepting const char pointers... */ return newtWinChoice(NULL, (char *)yes, (char *)no, (char *)msg) == 1; } + +void ui__warning(const char *format, ...) +{ + va_list args; + + va_start(args, format); + if (use_browser > 0) + newtWinMessagev((char *)warning_str, (char *)ok, + (char *)format, args); + else + vfprintf(stderr, format, args); + va_end(args); +} diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c index 214265674ddd..5b3ea49aa63e 100644 --- a/tools/perf/util/util.c +++ b/tools/perf/util/util.c @@ -114,3 +114,20 @@ unsigned long convert_unit(unsigned long value, char *unit) return value; } + +int readn(int fd, void *buf, size_t n) +{ + void *buf_start = buf; + + while (n) { + int ret = read(fd, buf, n); + + if (ret <= 0) + return ret; + + n -= ret; + buf += ret; + } + + return buf - buf_start; +} diff --git a/tools/perf/util/util.h b/tools/perf/util/util.h index 7562707ddd1c..e833f26f3bfc 100644 --- a/tools/perf/util/util.h +++ b/tools/perf/util/util.h @@ -265,6 +265,7 @@ void argv_free(char **argv); bool strglobmatch(const char *str, const char *pat); bool strlazymatch(const char *str, const char *pat); unsigned long convert_unit(unsigned long value, char *unit); +int readn(int fd, void *buf, size_t size); #define _STR(x) #x #define STR(x) _STR(x) diff --git a/tools/perf/util/xyarray.c b/tools/perf/util/xyarray.c new file mode 100644 index 000000000000..22afbf6c536a --- /dev/null +++ b/tools/perf/util/xyarray.c @@ -0,0 +1,20 @@ +#include "xyarray.h" +#include "util.h" + +struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size) +{ + size_t row_size = ylen * entry_size; + struct xyarray *xy = zalloc(sizeof(*xy) + xlen * row_size); + + if (xy != NULL) { + xy->entry_size = entry_size; + xy->row_size = row_size; + } + + return xy; +} + +void xyarray__delete(struct xyarray *xy) +{ + free(xy); +} diff --git a/tools/perf/util/xyarray.h b/tools/perf/util/xyarray.h new file mode 100644 index 000000000000..c488a07275dd --- /dev/null +++ b/tools/perf/util/xyarray.h @@ -0,0 +1,20 @@ +#ifndef _PERF_XYARRAY_H_ +#define _PERF_XYARRAY_H_ 1 + +#include <sys/types.h> + +struct xyarray { + size_t row_size; + size_t entry_size; + char contents[]; +}; + +struct xyarray *xyarray__new(int xlen, int ylen, size_t entry_size); +void xyarray__delete(struct xyarray *xy); + +static inline void *xyarray__entry(struct xyarray *xy, int x, int y) +{ + return &xy->contents[x * xy->row_size + y * xy->entry_size]; +} + +#endif /* _PERF_XYARRAY_H_ */ |
