| Commit message (Collapse) | Author | Age | Files | Lines |
|\
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 seccomp changes from Ingo Molnar:
"This tree includes x86 seccomp filter speedups and related preparatory
work, which touches core seccomp facilities as well.
The main idea is to split seccomp into two phases, to be able to enter
a simple fast path for syscalls with ptrace side effects.
There's no substantial user-visible (and ABI) effects expected from
this, except a change in how we emit a better audit record for
SECCOMP_RET_TRACE events"
* 'x86-seccomp-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86_64, entry: Use split-phase syscall_trace_enter for 64-bit syscalls
x86_64, entry: Treat regs->ax the same in fastpath and slowpath syscalls
x86: Split syscall_trace_enter into two phases
x86, entry: Only call user_exit if TIF_NOHZ
x86, x32, audit: Fix x32's AUDIT_ARCH wrt audit
seccomp: Document two-phase seccomp and arch-provided seccomp_data
seccomp: Allow arch code to provide seccomp_data
seccomp: Refactor the filter callback and the API
seccomp,x86,arm,mips,s390: Remove nr parameter from secure_computing
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
On KVM on my box, this reduces the overhead from an always-accept
seccomp filter from ~130ns to ~17ns. Most of that comes from
avoiding IRET on every syscall when seccomp is enabled.
In extremely approximate hacked-up benchmarking, just bypassing IRET
saves about 80ns, so there's another 43ns of savings here from
simplifying the seccomp path.
The diffstat is also rather nice :)
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/a3dbd267ee990110478d349f78cccfdac5497a84.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
For slowpath syscalls, we initialize regs->ax to -ENOSYS and stick
the syscall number into regs->orig_ax prior to any possible tracing
and syscall execution. This is user-visible ABI used by ptrace
syscall emulation and seccomp.
For fastpath syscalls, there's no good reason not to do the same
thing. It's even slightly simpler than what we're currently doing.
It probably has no measureable performance impact. It should have
no user-visible effect.
The purpose of this patch is to prepare for two-phase syscall
tracing, in which the first phase might modify the saved RAX without
leaving the fast path. This change is just subtle enough that I'm
keeping it separate.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/01218b493f12ae2f98034b78c9ae085e38e94350.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This splits syscall_trace_enter into syscall_trace_enter_phase1 and
syscall_trace_enter_phase2. Only phase 2 has full pt_regs, and only
phase 2 is permitted to modify any of pt_regs except for orig_ax.
The intent is that phase 1 can be called from the syscall fast path.
In this implementation, phase1 can handle any combination of
TIF_NOHZ (RCU context tracking), TIF_SECCOMP, and TIF_SYSCALL_AUDIT,
unless seccomp requests a ptrace event, in which case phase2 is
forced.
In principle, this could yield a big speedup for TIF_NOHZ as well as
for TIF_SECCOMP if syscall exit work were similarly split up.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/2df320a600020fda055fccf2b668145729dd0c04.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The RCU context tracking code requires that arch code call
user_exit() on any entry into kernel code if TIF_NOHZ is set. This
patch adds a check for TIF_NOHZ and a comment to the syscall entry
tracing code.
The main purpose of this patch is to make the code easier to follow:
one can read the body of user_exit and of every function it calls
without finding any explanation of why it's called for traced
syscalls but not for untraced syscalls. This makes it clear when
user_exit() is necessary.
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/0b13e0e24ec0307d67ab7a23b58764f6b1270116.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
is_compat_task() is the wrong check for audit arch; the check should
be is_ia32_task(): x32 syscalls should be AUDIT_ARCH_X86_64, not
AUDIT_ARCH_I386.
CONFIG_AUDITSYSCALL is currently incompatible with x32, so this has
no visible effect.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Link: http://lkml.kernel.org/r/a0138ed8c709882aec06e4acc30bfa9b623b8717.1409954077.git.luto@amacapital.net
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The description of how archs should implement seccomp filters was
still strictly correct, but it failed to describe the newly
available optimizations.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Kees Cook <keescook@chromium.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
populate_seccomp_data is expensive: it works by inspecting
task_pt_regs and various other bits to piece together all the
information, and it's does so in multiple partially redundant steps.
Arch-specific code in the syscall entry path can do much better.
Admittedly this adds a bit of additional room for error, but the
speedup should be worth it.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Kees Cook <keescook@chromium.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The reason I did this is to add a seccomp API that will be usable
for an x86 fast path. The x86 entry code needs to use a rather
expensive slow path for a syscall that might be visible to things
like ptrace. By splitting seccomp into two phases, we can check
whether we need the slow path and then use the fast path in if the
filter allows the syscall or just returns some errno.
As a side effect, I think the new code is much easier to understand
than the old code.
This has one user-visible effect: the audit record written for
SECCOMP_RET_TRACE is now a simple indication that SECCOMP_RET_TRACE
happened. It used to depend in a complicated way on what the tracer
did. I couldn't make much sense of it.
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Kees Cook <keescook@chromium.org>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The secure_computing function took a syscall number parameter, but
it only paid any attention to that parameter if seccomp mode 1 was
enabled. Rather than coming up with a kludge to get the parameter
to work in mode 2, just remove the parameter.
To avoid churn in arches that don't have seccomp filters (and may
not even support syscall_get_nr right now), this leaves the
parameter in secure_computing_strict, which is now a real function.
For ARM, this is a bit ugly due to the fact that ARM conditionally
supports seccomp filters. Fixing that would probably only be a
couple of lines of code, but it should be coordinated with the audit
maintainers.
This will be a slight slowdown on some arches. The right fix is to
pass in all of seccomp_data instead of trying to make just the
syscall nr part be fast.
This is a prerequisite for making two-phase seccomp work cleanly.
Cc: Russell King <linux@arm.linux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: linux-mips@linux-mips.org
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: linux-s390@vger.kernel.org
Cc: x86@kernel.org
Cc: Kees Cook <keescook@chromium.org>
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Kees Cook <keescook@chromium.org>
|
|\ \
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 platform updates from Ingo Molnar:
"The main changes in this tree are:
- fix and update Intel Quark [Galileo] SoC platform support
- update IOSF chipset side band interface and make it available via
debugfs
- enable HPETs on Soekris net6501 and other e6xx based systems"
* 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86: Add cpu_detect_cache_sizes to init_intel() add Quark legacy_cache()
x86: Quark: Comment setup_arch() to document TLB/PGE bug
x86/intel/quark: Switch off CR4.PGE so TLB flush uses CR3 instead
x86/platform/intel/iosf: Add debugfs config option for IOSF
x86/platform/intel/iosf: Add better description of IOSF driver in config
x86/platform/intel/iosf: Add Braswell PCI ID
x86/platform/pmc_atom: Fix warning when CONFIG_DEBUG_FS=n
x86: HPET force enable for e6xx based systems
x86/iosf: Add debugfs support
x86/iosf: Add Kconfig prompt for IOSF_MBI selection
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Intel processors which don't report cache information via cpuid(2)
or cpuid(4) need quirk code in the legacy_cache_size callback to
report this data. For Intel that callback is is intel_size_cache().
This patch enables calling of cpu_detect_cache_sizes() inside of
init_intel() and hence the calling of the legacy_cache callback in
intel_size_cache(). Adding this call will ensure that PIII Tualatin
currently in intel_size_cache() and Quark SoC X1000 being added to
intel_size_cache() in this patch will report their respective cache
sizes.
This model of calling cpu_detect_cache_sizes() is consistent with
AMD/Via/Cirix/Transmeta and Centaur.
Also added is a string to idenitfy the Quark as Quark SoC X1000
giving better and more descriptive output via /proc/cpuinfo
Adding cpu_detect_cache_sizes to init_intel() will enable calling
of intel_size_cache() on Intel processors which currently no code
can reach. Therefore this patch will also re-enable reporting
of PIII Tualatin cache size information as well as add
Quark SoC X1000 support.
Comment text and cache flow logic suggested by Thomas Gleixner
Signed-off-by: Bryan O'Donoghue <pure.logic@nexus-software.ie>
Cc: davej@redhat.com
Cc: hmh@hmh.eng.br
Link: http://lkml.kernel.org/r/1412641189-12415-3-git-send-email-pure.logic@nexus-software.ie
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Quark SoC X1000 advertises Page Global Enable for it's
Translation Lookaside Buffer via cpuid. The silicon does not
in fact support PGE and hence will not flush the TLB when CR4.PGE
is rewritten. The Quark documentation makes clear the necessity to
instead rewrite CR3 in order to flush any TLB entries, irrespective
of the state of CR4.PGE or an individual PTE.PGE
See Intel Quark Core DevMan_001.pdf section 6.4.11
In setup.c setup_arch() the code will load_cr3() and then do a
__flush_tlb_all().
On Quark the entire TLB will be flushed at the load_cr3().
The __flush_tlb_all() have no effect and can be safely ignored.
Later on in the boot process we switch off the flag for cpu_has_pge()
which means that subsequent calls to __flush_tlb_all() will
call __flush_tlb() not __flush_tlb_global() flushing the TLB in the
correct way via load_cr3() not CR4.PGE rewrite
This patch documents the behaviour of flushing the TLB for Quark in
setup_arch()
Comment text suggested by Thomas Gleixner
Signed-off-by: Bryan O'Donoghue <pure.logic@nexus-software.ie>
Cc: davej@redhat.com
Cc: hmh@hmh.eng.br
Link: http://lkml.kernel.org/r/1412641189-12415-2-git-send-email-pure.logic@nexus-software.ie
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Quark x1000 advertises PGE via the standard CPUID method
PGE bits exist in Quark X1000's PTEs. In order to flush
an individual PTE it is necessary to reload CR3 irrespective
of the PTE.PGE bit.
See Quark Core_DevMan_001.pdf section 6.4.11
This bug was fixed in Galileo kernels, unfixed vanilla kernels are expected to
crash and burn on this platform.
Signed-off-by: Bryan O'Donoghue <pure.logic@nexus-software.ie>
Cc: Borislav Petkov <bp@alien8.de>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/1411514784-14885-1-git-send-email-pure.logic@nexus-software.ie
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Makes the IOSF sideband available through debugfs. Allows
developers to experiment with using the sideband to provide
debug and analytical tools for units on the SoC.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1411017231-20807-4-git-send-email-david.e.box@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Adds better description of IOSF driver to determine when it
should be enabled. Also moves the Kconfig option to "Processor
type and features" menu from main configuration menu.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1411017231-20807-3-git-send-email-david.e.box@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Add Braswell PCI ID to list of supported ID's for the IOSF driver.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1411017231-20807-2-git-send-email-david.e.box@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
When compiling with CONFIG_DEBUG_FS=n, GCC emits an unused
variable warning for pmc_atom.c because "ret" is used only
within the CONFIG_DEBUG_FS block.
This patch adds a dummy #ifdef for pmc_dbgfs_register() when
CONFIG_DEBUG_FS=n to simplify the code and remove the warning.
Signed-off-by: Martin Kelly <martkell@amazon.com>
Acked-by: "Li, Aubrey" <aubrey.li@linux.intel.com>
Cc: vishwesh.m.rudramuni@intel.com
Link: http://lkml.kernel.org/r/1410963476-8360-1-git-send-email-martin@martingkelly.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
As the Soekris net6501 and other e6xx based systems do not have
any ACPI implementation, HPET won't get enabled.
This patch enables HPET on such platforms.
[ 0.430149] pci 0000:00:01.0: Force enabled HPET at 0xfed00000
[ 0.644838] HPET: 3 timers in total, 0 timers will be used for per-cpu timer
Original patch by Peter Neubauer (http://www.mail-archive.com/soekris-tech@lists.soekris.com/msg06462.html)
slightly modified by Conrad Kostecki <ck@conrad-kostecki.de> and massaged
accoring to Thomas Gleixners <tglx@linutronix.de> by me.
Suggested-by: Conrad Kostecki <ck@conrad-kostecki.de>
Signed-off-by: Eric Sesterhenn <eric.sesterhenn@lsexperts.de>
Cc: Peter Neubauer <pneubauer@bluerwhite.org>
Link: http://lkml.kernel.org/r/5412D3A5.2030909@lsexperts.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Allows access to the iosf sideband through debugfs.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1409175640-32426-3-git-send-email-david.e.box@linux.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Fixes an error in having the iosf build as 'default m'. On X86 SoC's the iosf
sideband is the only way to access information for some registers, as opposed to
through MSR's on other Intel architectures. While selecting IOSF_MBI is
preferred, it does mean carrying extra code on non-SoC architectures. This
exports the selection to the user, allowing those driver writers to compile out
iosf code if it's not being built.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: http://lkml.kernel.org/r/1409175640-32426-2-git-send-email-david.e.box@linux.intel.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
|\ \ \
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 mm updates from Ingo Molnar:
"This tree includes the following changes:
- fix memory hotplug
- fix hibernation bootup memory layout assumptions
- fix hyperv numa guest kernel messages
- remove dead code
- update documentation"
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Update memory map description to list hypervisor-reserved area
x86/mm, hibernate: Do not assume the first e820 area to be RAM
x86/mm/numa: Drop dead code and rename setup_node_data() to setup_alloc_data()
x86/mm/hotplug: Modify PGD entry when removing memory
x86/mm/hotplug: Pass sync_global_pgds() a correct argument in remove_pagetable()
x86: Remove set_pmd_pfn
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Peter Anvin says:
> 0xffff880000000000 is the lowest usable address because we have
> agreed to leave 0xffff800000000000-0xffff880000000000 for the
> hypervisor or other non-OS uses.
Let's call this out in the documentation.
This came up during the kernel address sanitizer discussions
where it was proposed to use this area for other kernel things.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Link: http://lkml.kernel.org/r/20140918195606.841389D2@viggo.jf.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
In arch/x86/kernel/setup.c::trim_bios_range(), the codes
introduced by 1b5576e6 (base on d8a9e6a5), it updates the first
4Kb of memory to be E820_RESERVED region. That's because it's a
BIOS owned area but generally not listed in the E820 table:
e820: BIOS-provided physical RAM map:
BIOS-e820: [mem 0x0000000000000000-0x0000000000096fff] usable
BIOS-e820: [mem 0x0000000000097000-0x0000000000097fff] reserved
...
e820: update [mem 0x00000000-0x00000fff] usable ==> reserved
e820: remove [mem 0x000a0000-0x000fffff] usable
But the region of first 4Kb didn't register to nosave memory:
PM: Registered nosave memory: [mem 0x00097000-0x00097fff]
PM: Registered nosave memory: [mem 0x000a0000-0x000fffff]
The code in e820_mark_nosave_regions() assumes the first e820
area to be RAM, so it causes the first 4Kb E820_RESERVED region
ignored when register to nosave. This patch removed assumption
of the first e820 area.
Signed-off-by: Lee, Chun-Yi <jlee@suse.com>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Cc: Len Brown <len.brown@intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Takashi Iwai <tiwai@suse.de>
Link: http://lkml.kernel.org/r/1410491038-17576-1-git-send-email-jlee@suse.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The setup_node_data() function allocates a pg_data_t object,
inserts it into the node_data[] array and initializes the
following fields: node_id, node_start_pfn and
node_spanned_pages.
However, a few function calls later during the kernel boot,
free_area_init_node() re-initializes those fields, possibly with
setup_node_data() is not used.
This causes a small glitch when running Linux as a hyperv numa
guest:
SRAT: PXM 0 -> APIC 0x00 -> Node 0
SRAT: PXM 0 -> APIC 0x01 -> Node 0
SRAT: PXM 1 -> APIC 0x02 -> Node 1
SRAT: PXM 1 -> APIC 0x03 -> Node 1
SRAT: Node 0 PXM 0 [mem 0x00000000-0x7fffffff]
SRAT: Node 1 PXM 1 [mem 0x80200000-0xf7ffffff]
SRAT: Node 1 PXM 1 [mem 0x100000000-0x1081fffff]
NUMA: Node 1 [mem 0x80200000-0xf7ffffff] + [mem 0x100000000-0x1081fffff] -> [mem 0x80200000-0x1081fffff]
Initmem setup node 0 [mem 0x00000000-0x7fffffff]
NODE_DATA [mem 0x7ffdc000-0x7ffeffff]
Initmem setup node 1 [mem 0x80800000-0x1081fffff]
NODE_DATA [mem 0x1081ea000-0x1081fdfff]
crashkernel: memory value expected
[ffffea0000000000-ffffea0001ffffff] PMD -> [ffff88007de00000-ffff88007fdfffff] on node 0
[ffffea0002000000-ffffea00043fffff] PMD -> [ffff880105600000-ffff8801077fffff] on node 1
Zone ranges:
DMA [mem 0x00001000-0x00ffffff]
DMA32 [mem 0x01000000-0xffffffff]
Normal [mem 0x100000000-0x1081fffff]
Movable zone start for each node
Early memory node ranges
node 0: [mem 0x00001000-0x0009efff]
node 0: [mem 0x00100000-0x7ffeffff]
node 1: [mem 0x80200000-0xf7ffffff]
node 1: [mem 0x100000000-0x1081fffff]
On node 0 totalpages: 524174
DMA zone: 64 pages used for memmap
DMA zone: 21 pages reserved
DMA zone: 3998 pages, LIFO batch:0
DMA32 zone: 8128 pages used for memmap
DMA32 zone: 520176 pages, LIFO batch:31
On node 1 totalpages: 524288
DMA32 zone: 7672 pages used for memmap
DMA32 zone: 491008 pages, LIFO batch:31
Normal zone: 520 pages used for memmap
Normal zone: 33280 pages, LIFO batch:7
In this dmesg, the SRAT table reports that the memory range for
node 1 starts at 0x80200000. However, the line starting with
"Initmem" reports that node 1 memory range starts at 0x80800000.
The "Initmem" line is reported by setup_node_data() and is
wrong, because the kernel ends up using the range as reported in
the SRAT table.
This commit drops all that dead code from setup_node_data(),
renames it to alloc_node_data() and adds a printk() to
free_area_init_node() so that we report a node's memory range
accurately.
Here's the same dmesg section with this patch applied:
SRAT: PXM 0 -> APIC 0x00 -> Node 0
SRAT: PXM 0 -> APIC 0x01 -> Node 0
SRAT: PXM 1 -> APIC 0x02 -> Node 1
SRAT: PXM 1 -> APIC 0x03 -> Node 1
SRAT: Node 0 PXM 0 [mem 0x00000000-0x7fffffff]
SRAT: Node 1 PXM 1 [mem 0x80200000-0xf7ffffff]
SRAT: Node 1 PXM 1 [mem 0x100000000-0x1081fffff]
NUMA: Node 1 [mem 0x80200000-0xf7ffffff] + [mem 0x100000000-0x1081fffff] -> [mem 0x80200000-0x1081fffff]
NODE_DATA(0) allocated [mem 0x7ffdc000-0x7ffeffff]
NODE_DATA(1) allocated [mem 0x1081ea000-0x1081fdfff]
crashkernel: memory value expected
[ffffea0000000000-ffffea0001ffffff] PMD -> [ffff88007de00000-ffff88007fdfffff] on node 0
[ffffea0002000000-ffffea00043fffff] PMD -> [ffff880105600000-ffff8801077fffff] on node 1
Zone ranges:
DMA [mem 0x00001000-0x00ffffff]
DMA32 [mem 0x01000000-0xffffffff]
Normal [mem 0x100000000-0x1081fffff]
Movable zone start for each node
Early memory node ranges
node 0: [mem 0x00001000-0x0009efff]
node 0: [mem 0x00100000-0x7ffeffff]
node 1: [mem 0x80200000-0xf7ffffff]
node 1: [mem 0x100000000-0x1081fffff]
Initmem setup node 0 [mem 0x00001000-0x7ffeffff]
On node 0 totalpages: 524174
DMA zone: 64 pages used for memmap
DMA zone: 21 pages reserved
DMA zone: 3998 pages, LIFO batch:0
DMA32 zone: 8128 pages used for memmap
DMA32 zone: 520176 pages, LIFO batch:31
Initmem setup node 1 [mem 0x80200000-0x1081fffff]
On node 1 totalpages: 524288
DMA32 zone: 7672 pages used for memmap
DMA32 zone: 491008 pages, LIFO batch:31
Normal zone: 520 pages used for memmap
Normal zone: 33280 pages, LIFO batch:7
This commit was tested on a two node bare-metal NUMA machine and
Linux as a numa guest on hyperv and qemu/kvm.
PS: The wrong memory range reported by setup_node_data() seems to be
harmless in the current kernel because it's just not used. However,
that bad range is used in kernel 2.6.32 to initialize the old boot
memory allocator, which causes a crash during boot.
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
When hot-adding/removing memory, sync_global_pgds() is called
for synchronizing PGD to PGD entries of all processes MM. But
when hot-removing memory, sync_global_pgds() does not work
correctly.
At first, sync_global_pgds() checks whether target PGD is none
or not. And if PGD is none, the PGD is skipped. But when
hot-removing memory, PGD may be none since PGD may be cleared by
free_pud_table(). So when sync_global_pgds() is called after
hot-removing memory, sync_global_pgds() should not skip PGD even
if the PGD is none. And sync_global_pgds() must clear PGD
entries of all processes MM.
Currently sync_global_pgds() does not clear PGD entries of all
processes MM when hot-removing memory. So when hot adding
memory which is same memory range as removed memory after
hot-removing memory, following call traces are shown:
kernel BUG at arch/x86/mm/init_64.c:206!
...
[<ffffffff815e0c80>] kernel_physical_mapping_init+0x1b2/0x1d2
[<ffffffff815ced94>] init_memory_mapping+0x1d4/0x380
[<ffffffff8104aebd>] arch_add_memory+0x3d/0xd0
[<ffffffff815d03d9>] add_memory+0xb9/0x1b0
[<ffffffff81352415>] acpi_memory_device_add+0x1af/0x28e
[<ffffffff81325dc4>] acpi_bus_device_attach+0x8c/0xf0
[<ffffffff813413b9>] acpi_ns_walk_namespace+0xc8/0x17f
[<ffffffff81325d38>] ? acpi_bus_type_and_status+0xb7/0xb7
[<ffffffff81325d38>] ? acpi_bus_type_and_status+0xb7/0xb7
[<ffffffff813418ed>] acpi_walk_namespace+0x95/0xc5
[<ffffffff81326b4c>] acpi_bus_scan+0x9a/0xc2
[<ffffffff81326bff>] acpi_scan_bus_device_check+0x8b/0x12e
[<ffffffff81326cb5>] acpi_scan_device_check+0x13/0x15
[<ffffffff81320122>] acpi_os_execute_deferred+0x25/0x32
[<ffffffff8107e02b>] process_one_work+0x17b/0x460
[<ffffffff8107edfb>] worker_thread+0x11b/0x400
[<ffffffff8107ece0>] ? rescuer_thread+0x400/0x400
[<ffffffff81085aef>] kthread+0xcf/0xe0
[<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
[<ffffffff815fc76c>] ret_from_fork+0x7c/0xb0
[<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
This patch clears PGD entries of all processes MM when
sync_global_pgds() is called after hot-removing memory
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Acked-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Gu Zheng <guz.fnst@cn.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
When hot-adding memory after hot-removing memory, following call
traces are shown:
kernel BUG at arch/x86/mm/init_64.c:206!
...
[<ffffffff815e0c80>] kernel_physical_mapping_init+0x1b2/0x1d2
[<ffffffff815ced94>] init_memory_mapping+0x1d4/0x380
[<ffffffff8104aebd>] arch_add_memory+0x3d/0xd0
[<ffffffff815d03d9>] add_memory+0xb9/0x1b0
[<ffffffff81352415>] acpi_memory_device_add+0x1af/0x28e
[<ffffffff81325dc4>] acpi_bus_device_attach+0x8c/0xf0
[<ffffffff813413b9>] acpi_ns_walk_namespace+0xc8/0x17f
[<ffffffff81325d38>] ? acpi_bus_type_and_status+0xb7/0xb7
[<ffffffff81325d38>] ? acpi_bus_type_and_status+0xb7/0xb7
[<ffffffff813418ed>] acpi_walk_namespace+0x95/0xc5
[<ffffffff81326b4c>] acpi_bus_scan+0x9a/0xc2
[<ffffffff81326bff>] acpi_scan_bus_device_check+0x8b/0x12e
[<ffffffff81326cb5>] acpi_scan_device_check+0x13/0x15
[<ffffffff81320122>] acpi_os_execute_deferred+0x25/0x32
[<ffffffff8107e02b>] process_one_work+0x17b/0x460
[<ffffffff8107edfb>] worker_thread+0x11b/0x400
[<ffffffff8107ece0>] ? rescuer_thread+0x400/0x400
[<ffffffff81085aef>] kthread+0xcf/0xe0
[<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
[<ffffffff815fc76c>] ret_from_fork+0x7c/0xb0
[<ffffffff81085a20>] ? kthread_create_on_node+0x140/0x140
The patch-set fixes the issue.
This patch (of 2):
remove_pagetable() gets start argument and passes the argument
to sync_global_pgds(). In this case, the argument must not be
modified. If the argument is modified and passed to
sync_global_pgds(), sync_global_pgds() does not correctly
synchronize PGD to PGD entries of all processes MM since
synchronized range of memory [start, end] is wrong.
Unfortunately the start argument is modified in
remove_pagetable(). So this patch fixes the issue.
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Acked-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Tang Chen <tangchen@cn.fujitsu.com>
Cc: Gu Zheng <guz.fnst@cn.fujitsu.com>
Cc: Zhang Yanfei <zhangyanfei@cn.fujitsu.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The last user of set_pmd_pfn() went away in commit f03574f2d5b2, so this
has been dead code for over a year.
Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/x86/include/asm/pgtable_32.h | 3 ---
arch/x86/mm/pgtable_32.c | 35 -----------------------------------
2 files changed, 38 deletions(-)
|
|\ \ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 microcode loading updates from Ingo Molnar:
"Misc smaller cleanups"
* 'x86-microcode-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, microcode, intel: Fix total_size computation
x86, microcode, intel: Rename apply_microcode and declare it static
x86, microcode, intel: Fix typos
x86, microcode, intel: Add missing static declarations
x86, microcode, amd: Fix missing static declaration
|
| |\ \ \ \
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp into x86/microcode
Pull x86/microcode updates from Borislav Petkov:
"A bunch of cleanups from Henrique."
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
According to the Intel SDM vol 3A (order code 253668-051US, June 2014),
on section 9.11.1, page 9-28:
"For microcode updates with a data size field equal to 00000000H, the
size of the microcode update is 2048 bytes. The first 48 bytes contain
the microcode update header. The remaining 2000 bytes contain encrypted
data."
"For microcode updates with a data size not equal to 00000000H, the total
size field specifies the size of the microcode update."
Up to 2002/2003, Intel used an "old format" for the microcode update
containers that was always 2048 bytes in size. That old format did not
have Data Size and Total Size fields, the quadwords at those positions
in the microcode container header were "reserved". The microcode header
of the "old format" microcode container has a hrdver of 0x01. You can
hunt down an old copy of the Intel SDM to validate this through its
order number (#243192). I found one from 1999 through a Google search.
Sometime in 2002/2003 (AFAICT, for the Prescott processors), Intel
documented a new format for the microcode containers and contributed in
2003 some code to the Linux kernel microcode driver implementing support
for the new format. This new format has Data Size and Total Size fields,
as well as the optional extended signature table. However, it reuses the
same hrdver as the old format (0x01), and it can only be told apart from
the old format by a non-zero Data Size field.
In fact, the only reason we can even trust a Data Size of zero to mean
that the microcode container is in the old format, is because Intel
reatroatively promised that the old format would always have a zero
there when they wrote the documentation for the _new_ format.
This is a very old bug, dating back to 2003. It has been dormant
ever since, as Intel seems to set all reserved fields to zero on the
microcode updates they distribute: I could not find a public microcode
update that would trigger this bug.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Link: http://lkml.kernel.org/r/1406146251-8540-1-git-send-email-hmh@hmh.eng.br
Signed-off-by: Borislav Petkov <bp@suse.de>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Rename apply_microcode() in microcode/intel.c to
apply_microcode_intel(), and declare it as static. This is a cosmetic
fix to silence a warning issued by sparse.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Link: http://lkml.kernel.org/r/1406146251-8540-1-git-send-email-hmh@hmh.eng.br
Signed-off-by: Borislav Petkov <bp@suse.de>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Fix some typos. One of them was in a struct name, fortunately harmless
because it happened on a "sizeof(struct foo*)" construction.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Link: http://lkml.kernel.org/r/1406146251-8540-1-git-send-email-hmh@hmh.eng.br
Signed-off-by: Borislav Petkov <bp@suse.de>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
gcc reports that a few declarations are missing.
Fix two obvious ones.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Link: http://lkml.kernel.org/r/1406146251-8540-1-git-send-email-hmh@hmh.eng.br
Signed-off-by: Borislav Petkov <bp@suse.de>
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
Make locally used variable static.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Link: http://lkml.kernel.org/r/1406146251-8540-1-git-send-email-hmh@hmh.eng.br
Signed-off-by: Borislav Petkov <bp@suse.de>
|
|\ \ \ \ \ \
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 FPU updates from Ingo Molnar:
"x86 FPU handling fixes, cleanups and enhancements from Oleg.
The signal handling race fix and the __restore_xstate_sig() preemption
fix for eager-mode is marked for -stable as well"
* 'x86-fpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86: copy_thread: Don't nullify ->ptrace_bps twice
x86, fpu: Shift "fpu_counter = 0" from copy_thread() to arch_dup_task_struct()
x86, fpu: copy_process: Sanitize fpu->last_cpu initialization
x86, fpu: copy_process: Avoid fpu_alloc/copy if !used_math()
x86, fpu: Change __thread_fpu_begin() to use use_eager_fpu()
x86, fpu: __restore_xstate_sig()->math_state_restore() needs preempt_disable()
x86, fpu: shift drop_init_fpu() from save_xstate_sig() to handle_signal()
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Both 32bit and 64bit versions of copy_thread() do memset(ptrace_bps)
twice for no reason, kill the 2nd memset().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175733.GA21676@redhat.com
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Cosmetic, but I think thread.fpu_counter should be initialized in
arch_dup_task_struct() too, along with other "fpu" variables. And
probably it make sense to turn it into thread.fpu->counter.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175730.GA21669@redhat.com
Reviewed-by: Suresh Siddha <sbsiddha@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Cosmetic, but imho memset(&dst->thread.fpu, 0) is not good simply
because it hides the (important) usage of ->has_fpu/etc from grep.
Change this code to initialize the members explicitly.
And note that ->last_cpu = 0 looks simply wrong, this can confuse
fpu_lazy_restore() if per_cpu(fpu_owner_task, 0) has already exited
and copy_process() re-allocated the same task_struct. Fortunately
this is not actually possible because child->fpu_counter == 0 and
thus fpu_lazy_restore() will not be called, but still this is not
clean/robust.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175727.GA21666@redhat.com
Reviewed-by: Suresh Siddha <sbsiddha@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
arch_dup_task_struct() copies thread.fpu if fpu_allocated(), this
looks suboptimal and misleading. Say, a forking process could use
FPU only once in a signal handler but now tsk_used_math(src) == F,
in this case the child gets a copy of fpu->state for no reason. The
child won't use the saved registers anyway even if it starts to use
FPU, this can only avoid fpu_alloc() in do_device_not_available().
Change this code to check tsk_used_math(current) instead. We still
need to clear fpu->has_fpu/state, we could do this memset(0) under
fpu_allocated() check but I think this doesn't make sense. See also
the next change.
use_eager_fpu() assumes that fpu_allocated() is always true, but a
forking task (and thus its child) must always have PF_USED_MATH set,
otherwise the child can either use FPU without used_math() (note that
switch_fpu_prepare() doesn't do stts() in this case), or it will be
killed by do_device_not_available()->BUG_ON(use_eager_fpu).
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175723.GA21659@redhat.com
Reviewed-by: Suresh Siddha <sbsiddha@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
__thread_fpu_begin() checks X86_FEATURE_EAGER_FPU by hand, we have
a helper for that.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175720.GA21656@redhat.com
Reviewed-by: Suresh Siddha <sbsiddha@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
Add preempt_disable() + preempt_enable() around math_state_restore() in
__restore_xstate_sig(). Otherwise __switch_to() after __thread_fpu_begin()
can overwrite fpu->state we are going to restore.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175717.GA21649@redhat.com
Cc: <stable@vger.kernel.org> # v3.7+
Reviewed-by: Suresh Siddha <sbsiddha@gmail.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | |
| | | | | | | |
save_xstate_sig()->drop_init_fpu() doesn't look right. setup_rt_frame()
can fail after that, in this case the next setup_rt_frame() triggered
by SIGSEGV won't save fpu simply because the old state was lost. This
obviously mean that fpu won't be restored after sys_rt_sigreturn() from
SIGSEGV handler.
Shift drop_init_fpu() into !failed branch in handle_signal().
Test-case (needs -O2):
#include <stdio.h>
#include <signal.h>
#include <unistd.h>
#include <sys/syscall.h>
#include <sys/mman.h>
#include <pthread.h>
#include <assert.h>
volatile double D;
void test(double d)
{
int pid = getpid();
for (D = d; D == d; ) {
/* sys_tkill(pid, SIGHUP); asm to avoid save/reload
* fp regs around "C" call */
asm ("" : : "a"(200), "D"(pid), "S"(1));
asm ("syscall" : : : "ax");
}
printf("ERR!!\n");
}
void sigh(int sig)
{
}
char altstack[4096 * 10] __attribute__((aligned(4096)));
void *tfunc(void *arg)
{
for (;;) {
mprotect(altstack, sizeof(altstack), PROT_READ);
mprotect(altstack, sizeof(altstack), PROT_READ|PROT_WRITE);
}
}
int main(void)
{
stack_t st = {
.ss_sp = altstack,
.ss_size = sizeof(altstack),
.ss_flags = SS_ONSTACK,
};
struct sigaction sa = {
.sa_handler = sigh,
};
pthread_t pt;
sigaction(SIGSEGV, &sa, NULL);
sigaltstack(&st, NULL);
sa.sa_flags = SA_ONSTACK;
sigaction(SIGHUP, &sa, NULL);
pthread_create(&pt, NULL, tfunc, NULL);
test(123.456);
return 0;
}
Reported-by: Bean Anderson <bean@azulsystems.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20140902175713.GA21646@redhat.com
Cc: <stable@kernel.org> # v3.7+
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
|\ \ \ \ \ \ \
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 cpufeature updates from Ingo Molnar:
"This tree includes the following changes:
- Introduce DISABLED_MASK to list disabled CPU features, to simplify
CPU feature handling and avoid excessive #ifdefs
- Remove the lightly used cpu_has_pae() primitive"
* 'x86-cpufeature-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86: Add more disabled features
x86: Introduce disabled-features
x86: Axe the lightly-used cpu_has_pae
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
The original motivation for these patches was for an Intel CPU
feature called MPX. The patch to add a disabled feature for it
will go in with the other parts of the support.
But, in the meantime, there are a few other features than MPX
that we can make assumptions about at compile-time based on
compile options. Add them to disabled-features.h and check them
with cpu_feature_enabled().
Note that this gets rid of the last things that needed an #ifdef
CONFIG_X86_64 in cpufeature.h. Yay!
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: http://lkml.kernel.org/r/20140911211524.C0EC332A@viggo.jf.intel.com
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
I believe the REQUIRED_MASK aproach was taken so that it was
easier to consult in assembly (arch/x86/kernel/verify_cpu.S).
DISABLED_MASK does not have the same restriction, but I
implemented it the same way for consistency.
We have a REQUIRED_MASK... which does two things:
1. Keeps a list of cpuid bits to check in very early boot and
refuse to boot if those are not present.
2. Consulted during cpu_has() checks, which allows us to
optimize out things at compile-time. In other words, if we
*KNOW* we will not boot with the feature off, then we can
safely assume that it will be present forever.
But, we don't have a similar mechanism for CPU features which
may be present but that we know we will not use. We simply
use our existing mechanisms to repeatedly check the status of
the bit at runtime (well, the alternatives patching helps here
but it does not provide compile-time optimization).
Adding a feature to disabled-features.h allows the bit to be
checked via a new macro: cpu_feature_enabled(). Note that
for features in DISABLED_MASK, checks with this macro have
all of the benefits of an #ifdef. Before, we would have done
this in a header:
#ifdef CONFIG_X86_INTEL_MPX
#define cpu_has_mpx cpu_has(X86_FEATURE_MPX)
#else
#define cpu_has_mpx 0
#endif
and this in the code:
if (cpu_has_mpx)
do_some_mpx_thing();
Now, just add your feature to DISABLED_MASK and you can do this
everywhere, and get the same benefits you would have from
#ifdefs:
if (cpu_feature_enabled(X86_FEATURE_MPX))
do_some_mpx_thing();
We need a new function and *not* a modification to cpu_has()
because there are cases where we actually need to check the CPU
itself, despite what features the kernel supports. The best
example of this is a hypervisor which has no control over what
features its guests are using and where the guest does not depend
on the host for support.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: http://lkml.kernel.org/r/20140911211513.9E35E931@viggo.jf.intel.com
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | |
| | | | | | | | |
cpu_has_pae is only referenced in one place: the X86_32 kexec
code (in a file not even built on 64-bit). It hardly warrants
its own macro, or the trouble we go to ensuring that it can't
be called in X86_64 code.
Axe the macro and replace it with a direct cpu feature check.
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Link: http://lkml.kernel.org/r/20140911211511.AD76E774@viggo.jf.intel.com
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
|
|\ \ \ \ \ \ \ \
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 cpu offlining patch from Ingo Molnar:
"This tree includes a single commit that speeds up x86 suspend/resume
by replacing a naive 100msec sleep based polling loop with proper
completion notification.
This gives some real suspend/resume benefit on servers with larger
core counts"
* 'x86-cpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/smpboot: Speed up suspend/resume by avoiding 100ms sleep for CPU offline during S3
|
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | |
| | | | | | | | | |
during S3
With certain kernel configurations, CPU offline consumes more than
100ms during S3.
It's a timing related issue: native_cpu_die() would occasionally fall
into a 100ms sleep when the CPU idle loop thread marked the CPU state
to DEAD too slowly.
What native_cpu_die() does is that it polls the CPU state and waits
for 100ms if CPU state hasn't been marked to DEAD. The 100ms sleep
doesn't make sense and is purely historic.
To avoid such long sleeping, this patch adds a 'struct completion'
to each CPU, waits for the completion in native_cpu_die() and wakes
up the completion when the CPU state is marked to DEAD.
Tested on an Intel Xeon server with 48 cores, Ivybridge and on
Haswell laptops. The CPU offlining cost on these machines is
reduced from more than 100ms to less than 5ms. The system
suspend time is reduced by 2.3s on the servers.
Borislav and Prarit also helped to test the patch on an AMD
machine and a few systems of various sizes and configurations
(multi-socket, single-socket, no hyper threading, etc.). No
issues were seen.
Tested-by: Prarit Bhargava <prarit@redhat.com>
Signed-off-by: Lan Tianyu <tianyu.lan@intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: srostedt@redhat.com
Cc: toshi.kani@hp.com
Cc: imammedo@redhat.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/1409039025-32310-1-git-send-email-tianyu.lan@intel.com
[ Improved a few minor details in the code, cleaned up the changelog. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|\ \ \ \ \ \ \ \ \
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | | |
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 cleanups from Ingo Molnar:
"Three small cleanups"
* 'x86-cleanups-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/tty/serial/8250: Clean up the asm/serial.h include file a bit
x86/tty/serial/8250: Resolve missing-field-initializers warnings
x86: Remove obsolete comment in uapi/e820.h
|