From d546b67a940eb42a99f56b86c5cd8d47c8348c2a Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Tue, 25 Mar 2008 17:39:12 -0700 Subject: x86: fix performance drop for glx fix the 3D performance drop reported at: http://bugzilla.kernel.org/show_bug.cgi?id=10328 fb drivers are using ioremap()/ioremap_nocache(), followed by mtrr_add with WC attribute. Recent changes in page attribute code made both ioremap()/ioremap_nocache() mappings as UC (instead of previous UC-). This breaks the graphics performance, as the effective memory type is UC instead of expected WC. The correct way to fix this is to add ioremap_wc() (which uses UC- in the absence of PAT kernel support and WC with PAT) and change all the fb drivers to use this new ioremap_wc() API. We can take this correct and longer route for post 2.6.25. For now, revert back to the UC- behavior for ioremap/ioremap_nocache. Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Cc: Arjan van de Ven Signed-off-by: Ingo Molnar --- include/asm-x86/pgtable.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h index 174b87738714..9cf472aeb9ce 100644 --- a/include/asm-x86/pgtable.h +++ b/include/asm-x86/pgtable.h @@ -85,6 +85,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC; #define __PAGE_KERNEL_RX (__PAGE_KERNEL_EXEC & ~_PAGE_RW) #define __PAGE_KERNEL_EXEC_NOCACHE (__PAGE_KERNEL_EXEC | _PAGE_PCD | _PAGE_PWT) #define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT) +#define __PAGE_KERNEL_UC_MINUS (__PAGE_KERNEL | _PAGE_PCD) #define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RX | _PAGE_USER) #define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT) #define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE) @@ -101,6 +102,7 @@ extern pteval_t __PAGE_KERNEL, __PAGE_KERNEL_EXEC; #define PAGE_KERNEL_EXEC MAKE_GLOBAL(__PAGE_KERNEL_EXEC) #define PAGE_KERNEL_RX MAKE_GLOBAL(__PAGE_KERNEL_RX) #define PAGE_KERNEL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE) +#define PAGE_KERNEL_UC_MINUS MAKE_GLOBAL(__PAGE_KERNEL_UC_MINUS) #define PAGE_KERNEL_EXEC_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_EXEC_NOCACHE) #define PAGE_KERNEL_LARGE MAKE_GLOBAL(__PAGE_KERNEL_LARGE) #define PAGE_KERNEL_LARGE_EXEC MAKE_GLOBAL(__PAGE_KERNEL_LARGE_EXEC) -- cgit v1.2.1 From b2ef749720a97053d60605a7456772a1752164cc Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 26 Mar 2008 22:39:15 +0100 Subject: rdc321x: GPIO routines bugfixes This patch fixes the use of GPIO routines which are in the PCI configuration space of the RDC321x, therefore reading/writing to this space without spinlock protection can be problematic. We also now request and free GPIOs and support the MGB100 board, previous code was very AR525W-centric. Signed-off-by: Volker Weiss Signed-off-by: Florian Fainelli Signed-off-by: Ingo Molnar --- include/asm-x86/mach-rdc321x/gpio.h | 9 +++++---- include/asm-x86/mach-rdc321x/rdc321x_defs.h | 8 +++++++- 2 files changed, 12 insertions(+), 5 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/mach-rdc321x/gpio.h b/include/asm-x86/mach-rdc321x/gpio.h index db31b929b990..acce0b7d397b 100644 --- a/include/asm-x86/mach-rdc321x/gpio.h +++ b/include/asm-x86/mach-rdc321x/gpio.h @@ -5,19 +5,20 @@ extern int rdc_gpio_get_value(unsigned gpio); extern void rdc_gpio_set_value(unsigned gpio, int value); extern int rdc_gpio_direction_input(unsigned gpio); extern int rdc_gpio_direction_output(unsigned gpio, int value); - +extern int rdc_gpio_request(unsigned gpio, const char *label); +extern void rdc_gpio_free(unsigned gpio); +extern void __init rdc321x_gpio_setup(void); /* Wrappers for the arch-neutral GPIO API */ static inline int gpio_request(unsigned gpio, const char *label) { - /* Not yet implemented */ - return 0; + return rdc_gpio_request(gpio, label); } static inline void gpio_free(unsigned gpio) { - /* Not yet implemented */ + rdc_gpio_free(gpio); } static inline int gpio_direction_input(unsigned gpio) diff --git a/include/asm-x86/mach-rdc321x/rdc321x_defs.h b/include/asm-x86/mach-rdc321x/rdc321x_defs.h index 838ba8f64fd3..c8e9c8bed3d0 100644 --- a/include/asm-x86/mach-rdc321x/rdc321x_defs.h +++ b/include/asm-x86/mach-rdc321x/rdc321x_defs.h @@ -3,4 +3,10 @@ /* General purpose configuration and data registers */ #define RDC3210_CFGREG_ADDR 0x0CF8 #define RDC3210_CFGREG_DATA 0x0CFC -#define RDC_MAX_GPIO 0x3A + +#define RDC321X_GPIO_CTRL_REG1 0x48 +#define RDC321X_GPIO_CTRL_REG2 0x84 +#define RDC321X_GPIO_DATA_REG1 0x4c +#define RDC321X_GPIO_DATA_REG2 0x88 + +#define RDC321X_MAX_GPIO 58 -- cgit v1.2.1 From a6bd8e13034dd7d60b6f14217096efa192d0adc1 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Fri, 28 Mar 2008 11:05:53 -0500 Subject: lguest: comment documentation update. Took some cycles to re-read the Lguest Journey end-to-end, fix some rot and tighten some phrases. Only comments change. No new jokes, but a couple of recycled old jokes. Signed-off-by: Rusty Russell --- include/asm-x86/lguest_hcall.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/lguest_hcall.h b/include/asm-x86/lguest_hcall.h index 758b9a5d4539..f239e7069cab 100644 --- a/include/asm-x86/lguest_hcall.h +++ b/include/asm-x86/lguest_hcall.h @@ -27,7 +27,7 @@ #ifndef __ASSEMBLY__ #include -/*G:031 First, how does our Guest contact the Host to ask for privileged +/*G:031 But first, how does our Guest contact the Host to ask for privileged * operations? There are two ways: the direct way is to make a "hypercall", * to make requests of the Host Itself. * -- cgit v1.2.1 From bae1d2507e44417455eda76d4435352fee14cf51 Mon Sep 17 00:00:00 2001 From: Ravikiran G Thirumalai Date: Fri, 4 Apr 2008 03:06:29 -0700 Subject: x86: fix breakage of vSMP irq operations 25-rc* stopped working with CONFIG_X86_VSMP on vSMP machines. Looks like the vsmp irq ops got accidentally removed during merge of x86_64 pvops in 2.6.25. -- commit 6abcd98ffafbff81f0bfd7ee1d129e634af13245 removed vsmp irq ops. Tested with both CONFIG_X86_VSMP and without CONFIG_X86_VSMP, on vSMP and non vSMP x86_64 machines. Please apply. Signed-off-by: Ravikiran Thirumalai Signed-off-by: Ingo Molnar --- include/asm-x86/irqflags.h | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/irqflags.h b/include/asm-x86/irqflags.h index 92021c1ffa3a..0e2292483b35 100644 --- a/include/asm-x86/irqflags.h +++ b/include/asm-x86/irqflags.h @@ -70,6 +70,26 @@ static inline void raw_local_irq_restore(unsigned long flags) native_restore_fl(flags); } +#ifdef CONFIG_X86_VSMP + +/* + * Interrupt control for the VSMP architecture: + */ + +static inline void raw_local_irq_disable(void) +{ + unsigned long flags = __raw_local_save_flags(); + raw_local_irq_restore((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC); +} + +static inline void raw_local_irq_enable(void) +{ + unsigned long flags = __raw_local_save_flags(); + raw_local_irq_restore((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC)); +} + +#else + static inline void raw_local_irq_disable(void) { native_irq_disable(); @@ -80,6 +100,8 @@ static inline void raw_local_irq_enable(void) native_irq_enable(); } +#endif + /* * Used in the idle loop; sti takes one instruction cycle * to complete: @@ -137,10 +159,17 @@ static inline unsigned long __raw_local_irq_save(void) #define raw_local_irq_save(flags) \ do { (flags) = __raw_local_irq_save(); } while (0) +#ifdef CONFIG_X86_VSMP +static inline int raw_irqs_disabled_flags(unsigned long flags) +{ + return !(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC); +} +#else static inline int raw_irqs_disabled_flags(unsigned long flags) { return !(flags & X86_EFLAGS_IF); } +#endif static inline int raw_irqs_disabled(void) { -- cgit v1.2.1 From 871de939030c903fd5ed50a7c4c88e02998e1cbc Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Mon, 7 Apr 2008 11:56:34 -0700 Subject: x86: fix 64-bit asm NOPS for CONFIG_GENERIC_CPU ASM_NOP's for 64-bit kernel with CONFIG_GENERIC_CPU is broken with the recent x86 nops merge. They were using GENERIC_NOPS which will truncate the upper 32bits of %rsi, because of the missing 64bit rex prefix. For now, fall back ASM NOPS for generic cpu to K8 NOPS, similar to the code before the wrong x86 nop merge. This should resolve the crash seen by Ingo on a test-system: BUG: unable to handle kernel paging request at 00000000d80d8ee8 IP: [] save_i387_ia32+0x61/0xd8 PGD b8e0067 PUD 51490067 PMD 0 Oops: 0000 [1] SMP CPU 2 Modules linked in: Pid: 3871, comm: distcc Not tainted 2.6.25-rc7-sched-devel.git-x86-latest.git #359 RIP: 0010:[] [] save_i387_ia32+0x61/0xd8 RSP: 0000:ffff81003abd3cb8 EFLAGS: 00010246 RAX: ffff810082e93400 RBX: 00000000ffc37f84 RCX: ffff8100d80d8ee0 RDX: 0000000000000000 RSI: 00000000d80d8ee0 RDI: ffff810082e93400 RBP: 00000000ffc37fdc R08: 00000000ffc37f88 R09: 0000000000000008 R10: ffff81003abd2000 R11: 0000000000000000 R12: ffff810082e93400 R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff81011fb12dc0(0063) knlGS:00000000f7f1a6c0 CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 CR2: 00000000d80d8ee8 CR3: 0000000076922000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process distcc (pid: 3871, threadinfo ffff81003abd2000, task ffff8100d80d8ee0) Stack: ffff8100bb670380 ffffffff8026de50 0000000000000118 0000000000000002 0000000000000002 ffff81003abd3e68 ffff81003abd3ed8 ffff81003abd3de8 ffff81003abd3d18 ffffffff80229785 ffff8100d80d8ee0 ffff810001041280 Call Trace: [] ? __generic_file_aio_write_nolock+0x343/0x377 [] ? update_curr+0x54/0x64 [] ? ia32_setup_sigcontext+0x125/0x1d2 [] ? ia32_setup_frame+0x73/0x1a5 [] ? do_notify_resume+0x1aa/0x7db [] ? getnstimeofday+0x31/0x85 [] ? ktime_get_ts+0x17/0x48 [] ? ktime_get+0xc/0x41 [] ? hrtimer_nanosleep+0x75/0xd5 [] ? hrtimer_wakeup+0x0/0x21 [] ? int_signal+0x12/0x17 [] ? dummy_file_free_security+0x0/0x1 Code: a6 08 05 00 00 f6 40 14 01 74 34 4c 89 e7 48 0f ae 07 48 8b 86 08 05 00 00 80 78 02 00 79 02 db e2 90 8d b4 26 00 00 00 00 89 f6 <48> 8b 46 08 83 60 14 fe 0f 20 c0 48 83 c8 08 0f 22 c0 eb 07 c6 Signed-off-by: Suresh Siddha Signed-off-by: Ingo Molnar --- include/asm-x86/nops.h | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/nops.h b/include/asm-x86/nops.h index e3b2bce0aff8..b3930ae539b3 100644 --- a/include/asm-x86/nops.h +++ b/include/asm-x86/nops.h @@ -73,16 +73,7 @@ #define P6_NOP7 ".byte 0x0f,0x1f,0x80,0,0,0,0\n" #define P6_NOP8 ".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n" -#if defined(CONFIG_MK8) -#define ASM_NOP1 K8_NOP1 -#define ASM_NOP2 K8_NOP2 -#define ASM_NOP3 K8_NOP3 -#define ASM_NOP4 K8_NOP4 -#define ASM_NOP5 K8_NOP5 -#define ASM_NOP6 K8_NOP6 -#define ASM_NOP7 K8_NOP7 -#define ASM_NOP8 K8_NOP8 -#elif defined(CONFIG_MK7) +#if defined(CONFIG_MK7) #define ASM_NOP1 K7_NOP1 #define ASM_NOP2 K7_NOP2 #define ASM_NOP3 K7_NOP3 @@ -100,6 +91,15 @@ #define ASM_NOP6 P6_NOP6 #define ASM_NOP7 P6_NOP7 #define ASM_NOP8 P6_NOP8 +#elif defined(CONFIG_X86_64) +#define ASM_NOP1 K8_NOP1 +#define ASM_NOP2 K8_NOP2 +#define ASM_NOP3 K8_NOP3 +#define ASM_NOP4 K8_NOP4 +#define ASM_NOP5 K8_NOP5 +#define ASM_NOP6 K8_NOP6 +#define ASM_NOP7 K8_NOP7 +#define ASM_NOP8 K8_NOP8 #else #define ASM_NOP1 GENERIC_NOP1 #define ASM_NOP2 GENERIC_NOP2 -- cgit v1.2.1 From 54a015104136974262afa4b8ddd943ea70dec8a2 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 10 Apr 2008 15:37:38 -0700 Subject: asmlinkage_protect replaces prevent_tail_call The prevent_tail_call() macro works around the problem of the compiler clobbering argument words on the stack, which for asmlinkage functions is the caller's (user's) struct pt_regs. The tail/sibling-call optimization is not the only way that the compiler can decide to use stack argument words as scratch space, which we have to prevent. Other optimizations can do it too. Until we have new compiler support to make "asmlinkage" binding on the compiler's own use of the stack argument frame, we have work around all the manifestations of this issue that crop up. More cases seem to be prevented by also keeping the incoming argument variables live at the end of the function. This makes their original stack slots attractive places to leave those variables, so the compiler tends not clobber them for something else. It's still no guarantee, but it handles some observed cases that prevent_tail_call() did not. Signed-off-by: Roland McGrath Signed-off-by: Linus Torvalds --- include/asm-x86/linkage.h | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) (limited to 'include/asm-x86') diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index 31739c7d66a9..d605eeba0f70 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -8,12 +8,34 @@ #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) -#define prevent_tail_call(ret) __asm__ ("" : "=r" (ret) : "0" (ret)) /* * For 32-bit UML - mark functions implemented in assembly that use * regparm input parameters: */ #define asmregparm __attribute__((regparm(3))) + +#define asmlinkage_protect(n, ret, args...) \ + __asmlinkage_protect##n(ret, ##args) +#define __asmlinkage_protect_n(ret, args...) \ + __asm__ __volatile__ ("" : "=r" (ret) : "0" (ret), ##args) +#define __asmlinkage_protect0(ret) \ + __asmlinkage_protect_n(ret) +#define __asmlinkage_protect1(ret, arg1) \ + __asmlinkage_protect_n(ret, "g" (arg1)) +#define __asmlinkage_protect2(ret, arg1, arg2) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2)) +#define __asmlinkage_protect3(ret, arg1, arg2, arg3) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3)) +#define __asmlinkage_protect4(ret, arg1, arg2, arg3, arg4) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4)) +#define __asmlinkage_protect5(ret, arg1, arg2, arg3, arg4, arg5) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5)) +#define __asmlinkage_protect6(ret, arg1, arg2, arg3, arg4, arg5, arg6) \ + __asmlinkage_protect_n(ret, "g" (arg1), "g" (arg2), "g" (arg3), \ + "g" (arg4), "g" (arg5), "g" (arg6)) + #endif #ifdef CONFIG_X86_ALIGNMENT_16 -- cgit v1.2.1 From d10d89ec78114f925f63c5126a2b2490f501a462 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 10 Apr 2008 17:35:23 -0700 Subject: Add commentary about the new "asmlinkage_protect()" macro It's really a pretty ugly thing to need, and some day it will hopefully be obviated by teaching gcc about the magic calling conventions for the low-level system call code, but in the meantime we can at least add big honking comments about why we need these insane and strange macros. I took my comments from my version of the macro, but I ended up deciding to just pick Roland's version of the actual code instead (with his prettier syntax that uses vararg macros). Thus the previous two commits that actually implement it. Signed-off-by: Linus Torvalds --- include/asm-x86/linkage.h | 11 +++++++++++ 1 file changed, 11 insertions(+) (limited to 'include/asm-x86') diff --git a/include/asm-x86/linkage.h b/include/asm-x86/linkage.h index d605eeba0f70..c048353f4b85 100644 --- a/include/asm-x86/linkage.h +++ b/include/asm-x86/linkage.h @@ -14,6 +14,17 @@ */ #define asmregparm __attribute__((regparm(3))) +/* + * Make sure the compiler doesn't do anything stupid with the + * arguments on the stack - they are owned by the *caller*, not + * the callee. This just fools gcc into not spilling into them, + * and keeps it from doing tailcall recursion and/or using the + * stack slots for temporaries, since they are live and "used" + * all the way to the end of the function. + * + * NOTE! On x86-64, all the arguments are in registers, so this + * only matters on a 32-bit kernel. + */ #define asmlinkage_protect(n, ret, args...) \ __asmlinkage_protect##n(ret, ##args) #define __asmlinkage_protect_n(ret, args...) \ -- cgit v1.2.1