When there are a large number of processors in a system, there
is an excessive amount of messages sent to the system console.
It's estimated that with 4096 processors in a system, and the
console baudrate set to 56K, the startup messages will take
about 84 minutes to clear the serial port.
This set of patches limits the number of repetitious messages
which contain no additional information. Much of this information
is obtainable from the /proc and /sysfs. Some of the messages
are also sent to the kernel log buffer as KERN_DEBUG messages so
dmesg can be used to examine more closely any details specific to
a problem.
The new cpu bootup sequence for system_state == SYSTEM_BOOTING:
Booting Node 0, Processors #1#2#3#4#5#6#7 Ok.
Booting Node 1, Processors #8#9#10#11#12#13#14#15 Ok.
...
Booting Node 3, Processors #56#57#58#59#60#61#62#63 Ok.
Brought up 64 CPUs
After the system is running, a single line boot message is displayed
when CPU's are hotplugged on:
Booting Node %d Processor %d APIC 0x%x
Status of the following lines:
CPU: Physical Processor ID: printed once (for boot cpu)
CPU: Processor Core ID: printed once (for boot cpu)
CPU: Hyper-Threading is disabled printed once (for boot cpu)
CPU: Thermal monitoring enabled printed once (for boot cpu)
CPU %d/0x%x -> Node %d: removed
CPU %d is now offline: only if system_state == RUNNING
Initializing CPU#%d: KERN_DEBUG
Signed-off-by: Mike Travis <travis@sgi.com>
LKML-Reference: <4B219E28.8080601@sgi.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The per_cpu cpuid4_info shared_map can contain stale data when CPUs are added
and removed.
The stale data can lead to a NULL pointer derefernce panic on a remove of a
CPU that has had siblings previously removed.
This patch resolves the panic by verifying a cpu is actually online before
adding it to the shared_cpu_map, only examining cpus that are part of
the same lower level cache, and by updating other siblings lowest level cache
maps when a cpu is added.
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
LKML-Reference: <20091209183336.17855.98708.sendpatchset@prarit.bos.redhat.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* 'acpica' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux-acpi-2.6:
ACPICA: Update version to 20091112.
ACPICA: Add additional module-level code support
ACPICA: Deploy new create integer interface where appropriate
ACPICA: New internal utility function to create Integer objects
ACPICA: Add repair for predefined methods that must return sorted lists
ACPICA: Fix possible fault if return Package objects contain NULL elements
ACPICA: Add post-order callback to acpi_walk_namespace
ACPICA: Change package length error message to an info message
ACPICA: Reduce severity of predefined repair messages, Warning to Info
ACPICA: Update version to 20091013
ACPICA: Fix possible memory leak for Scope ASL operator
ACPICA: Remove possibility of executing _REG methods twice
ACPICA: Add repair for bad _MAT buffers
ACPICA: Add repair for bad _BIF/_BIX packages
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, mce: don't restart timer if disabled
x86: Use -maccumulate-outgoing-args for sane mcount prologues
x86: Prevent GCC 4.4.x (pentium-mmx et al) function prologue wreckage
x86: AMD Northbridge: Verify NB's node is online
x86 VSDO: Fix Kconfig help
x86: Fix typo in Intel CPU cache size descriptor
x86: Add new Intel CPU cache size descriptors
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (36 commits)
x86, mm: Correct the implementation of is_untracked_pat_range()
x86/pat: Trivial: don't create debugfs for memtype if pat is disabled
x86, mtrr: Fix sorting of mtrr after subtracting
x86: Move find_smp_config() earlier and avoid bootmem usage
x86, platform: Change is_untracked_pat_range() to bool; cleanup init
x86: Change is_ISA_range() into an inline function
x86, mm: is_untracked_pat_range() takes a normal semiclosed range
x86, mm: Call is_untracked_pat_range() rather than is_ISA_range()
x86: UV SGI: Don't track GRU space in PAT
x86: SGI UV: Fix BAU initialization
x86, numa: Use near(er) online node instead of roundrobin for NUMA
x86, numa, bootmem: Only free bootmem on NUMA failure path
x86: Change crash kernel to reserve via reserve_early()
x86: Eliminate redundant/contradicting cache line size config options
x86: When cleaning MTRRs, do not fold WP into UC
x86: remove "extern" from function prototypes in <asm/proto.h>
x86, mm: Report state of NX protections during boot
x86, mm: Clean up and simplify NX enablement
x86, pageattr: Make set_memory_(x|nx) aware of NX support
x86, sleep: Always save the value of EFER
...
Fix up conflicts (added both iommu_shutdown and is_untracked_pat_range)
to 'struct x86_platform_ops') in
arch/x86/include/asm/x86_init.h
arch/x86/kernel/x86_init.c
Commit cebe182033 had an unnecessary,
wrong change: &mce_banks[i].attr is equivalent to the former
bank_attrs[i], not to mce_attrs[i].
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Andi Kleen <andi@firstfloor.org>
LKML-Reference: <4B1E05CC.4040703@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
mce_timer must be passed to setup_timer() in all cases, no
matter whether it is going to be actually used. Otherwise, when
the CPU gets brought down, its call to del_timer_sync() will
never return, as the timer won't have a base associated, and
hence lock_timer_base() will loop infinitely.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: <stable@kernel.org>
LKML-Reference: <4B1DB831.2030801@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'x86-debug-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Limit number of per cpu TSC sync messages
x86: dumpstack, 64-bit: Disable preemption when walking the IRQ/exception stacks
x86: dumpstack: Clean up the x86_stack_ids[][] initalization and other details
x86, cpu: mv display_cacheinfo -> cpu_detect_cache_sizes
x86: Suppress stack overrun message for init_task
x86: Fix cpu_devs[] initialization in early_cpu_init()
x86: Remove CPU cache size output for non-Intel too
x86: Minimise printk spew from per-vendor init code
x86: Remove the CPU cache size printk's
cpumask: Avoid cpumask_t in arch/x86/kernel/apic/nmi.c
x86: Make sure we also print a Code: line for show_regs()
* 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (30 commits)
x86, apic: Enable lapic nmi watchdog on AMD Family 11h
x86: Remove unnecessary mdelay() from cpu_disable_common()
x86, ioapic: Document another case when level irq is seen as an edge
x86, ioapic: Fix the EOI register detection mechanism
x86, io-apic: Move the effort of clearing remoteIRR explicitly before migrating the irq
x86: SGI UV: Map low MMR ranges
x86: apic: Print out SRAT table APIC id in hex
x86: Re-get cfg_new in case reuse/move irq_desc
x86: apic: Remove not needed #ifdef
x86: io-apic: IO-APIC MMIO should not fail on resource insertion
x86: Remove asm/apicnum.h
x86: apic: Do not use stacked physid_mask_t
x86, apic: Get rid of apicid_to_cpu_present assign on 64-bit
x86, ioapic: Use snrpintf while set names for IO-APIC resourses
x86, apic: Use PAGE_SIZE instead of numbers
x86: Remove local_irq_enable()/local_irq_disable() in fixup_irqs()
x86: Use EOI register in io-apic on intel platforms
x86: Force irq complete move during cpu offline
x86: Remove move_cleanup_count from irq_cfg
x86, intr-remap: Avoid irq_chip mask/unmask in fixup_irqs() for intr-remapping
...
The x86 lapic nmi watchdog does not recognize AMD Family 11h,
resulting in:
NMI watchdog: CPU not supported
As far as I can see from available documentation (the BKDM),
family 11h looks identical to family 10h as far as the PMU
is concerned.
Extending the check to accept family 11h results in:
Testing NMI watchdog ... OK.
I've been running with this change on a Turion X2 Ultra ZM-82
laptop for a couple of weeks now without problems.
Signed-off-by: Mikael Pettersson <mikpe@it.uu.se>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Joerg Roedel <joerg.roedel@amd.com>
Cc: <stable@kernel.org>
LKML-Reference: <19223.53436.931768.278021@pilspetsen.it.uu.se>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Even it is in error path unlikely taken, add_timer_on() at
CPU_DOWN_FAILED* needs to be skipped if mce_timer is disabled.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: <stable@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The mce_disable_cpu() and mce_reenable_cpu() are called only
from mce_cpu_callback() which is marked as __cpuinit.
So these functions can be __cpuinit too.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <4B0E3C4E.4090809@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The existing interface only has a pre-order callback. This change
adds an additional parameter for a post-order callback which will
be more useful for bus scans. ACPICA BZ 779.
Also update the external calls to acpi_walk_namespace.
http://www.acpica.org/bugzilla/show_bug.cgi?id=779
Signed-off-by: Lin Ming <ming.m.lin@intel.com>
Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
In some cases we can coalesce MTRR entries after cleanup; this may
allow us to have more entries. As such, introduce clean_sort_range to
to sort and coaelsce the MTRR entries.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4B0BB9A3.5020908@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The validate_event() was failing on valid event combinations. The
function was assuming that if x86_schedule_event() returned 0, it
meant error. But x86_schedule_event() returns the counter index and
0 is a perfectly valid value. An error is returned if the function
returns a negative value.
Furthermore, validate_event() was also failing for event groups
because the event->pmu was not set until after
hw_perf_event_init().
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: peterz@infradead.org
Cc: paulus@samba.org
Cc: perfmon2-devel@lists.sourceforge.net
Cc: eranian@gmail.com
LKML-Reference: <4b0bdf36.1818d00a.07cc.25ae@mx.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
--
arch/x86/kernel/cpu/perf_event.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
display_cacheinfo() doesn't display anything anymore and it is used to
detect CPU cache sizes. Rename it accordingly.
Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
LKML-Reference: <20091121130145.GA31357@liondog.tnic>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Decreases perf overhead when function tracing is enabled,
by about 50%.
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
CPU to node mapping is set via the following sequence:
1. numa_init_array(): Set up roundrobin from cpu to online node
2. init_cpu_to_node(): Set that according to apicid_to_node[]
according to srat only handle the node that
is online, and leave other cpu on node
without ram (aka not online) to still
roundrobin.
3. later call srat_detect_node for Intel/AMD, will use first_online
node or nearby node.
Problem is that setup_per_cpu_areas() is not called between 2 and 3,
the per_cpu for cpu on node with ram is on different node, and could
put that on node with two hops away.
So try to optimize this and add find_near_online_node() and call
init_cpu_to_node().
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: David Rientjes <rientjes@google.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <4B07A739.3030104@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
"[CPUFREQ] cpumask: avoid playing with cpus_allowed in speedstep-ich.c"
changed the code to mistakenly pass the current cpu as the "processor"
argument of speedstep_get_frequency(), whereas it should be the type of
the processor.
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=14340
Based on a patch by Dave Mueller.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Dominik Brodowski <linux@brodo.de>
Reported-by: Dave Mueller <dave.mueller@gmx.ch>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Jones <davej@redhat.com>
Removing the SMT/HT check, since the Errata doesn't mention
Hyper-Threading.
Adding in a printk, so that the user knows why acpi-cpufreq refuses to
load. Also, once system is blacklisted, don't repeat checks to see if
blacklisted. This also causes the message to only be printed once,
rather than for each CPU.
Signed-off-by: John L. Villalovos <john.l.villalovos@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
There is a typo in the longhaul detection code so only Longhaul v1 or Longhaul v3
is selected. The Longhaul v2 is not selected even for CPUs which are capable of.
Tested on PCChips Giga Pro board. Frequency changes work and the Longhaul v2
detects that the board is not capable of changing CPU voltage.
Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Dave Jones <davej@redhat.com>
The current MTRR code treats WP as a form of UC. This really isn't
desirable behaviour, except possibly in the case of severe MTRR
shortage. Disable this, to allow legitimate uses of WP to remain
unmolested.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
The 32- and 64-bit code used very different mechanisms for enabling
NX, but even the 32-bit code was enabling NX in head_32.S if it is
available. Furthermore, we had a bewildering collection of tests for
the available of NX.
This patch:
a) merges the 32-bit set_nx() and the 64-bit check_efer() function
into a single x86_configure_nx() function. EFER control is left
to the head code.
b) eliminates the nx_enabled variable entirely. Things that need to
test for NX enablement can verify __supported_pte_mask directly,
and cpu_has_nx gives the supported status of NX.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Vegard Nossum <vegardno@ifi.uio.no>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Chris Wright <chrisw@sous-sol.org>
LKML-Reference: <1258154897-6770-5-git-send-email-hpa@zytor.com>
Acked-by: Kees Cook <kees.cook@canonical.com>
Yinghai Lu noticed that this commit:
0388423: x86: Minimise printk spew from per-vendor init code
mistakenly left out the initialization of cpu_devs[] in the
!PROCESSOR_SELECT case. Fix it.
Reported-by: Yinghai Lu <yinghai@kernel.org>
Cc: Dave Jones <davej@redhat.com>
LKML-Reference: <20091113203000.GA19160@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
As Dave Jones said about the output in intel_cacheinfo.c: "They
aren't useful, and pollute the dmesg output a lot (especially on
machines with many cores). Also the same information can be
trivially found out from userspace."
Give the generic display_cacheinfo() function the same treatment.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Acked-by: Dave Jones <davej@redhat.com>
Cc: Mike Travis <travis@sgi.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <adaocn6dp99.fsf_-_@roland-alpha.cisco.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In the default case where the kernel supports all CPU vendors,
we currently print out a bunch of not useful messages on every
system.
32-bit:
KERNEL supported cpus:
Intel GenuineIntel
AMD AuthenticAMD
NSC Geode by NSC
Cyrix CyrixInstead
Centaur CentaurHauls
Transmeta GenuineTMx86
Transmeta TransmetaCPU
UMC UMC UMC UMC
64-bit:
KERNEL supported cpus:
Intel GenuineIntel
AMD AuthenticAMD
Centaur CentaurHauls
Given that "what CPUs does the kernel support" isn't useful for
the "support everything" case, we can suppress these printk's.
Signed-off-by: Dave Jones <davej@redhat.com>
LKML-Reference: <20091113203000.GA19160@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
They aren't really useful, and they pollute the dmesg output a lot
(especially on machines with many cores).
Also the same information can be trivially found out from
userspace.
Reported-by: Mike Travis <travis@sgi.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Cc: Tejun Heo <tj@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <20091112231542.GA7129@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The intel_init_thermal() is called from resume path, so it
cannot be marked as __init.
OTOH mce_banks_init() is only called from
__mcheck_cpu_cap_init() which is marked as __cpuinit, so it can
be also marked as __cpuinit.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Yong Wang <yong.y.wang@linux.intel.com>
LKML-Reference: <4AFBB0B8.2070501@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Mark the thermal init functions __init so that the init memory
can be freed.
Signed-off-by: Yong Wang <yong.y.wang@intel.com>
LKML-Reference: <20091111075125.GA17900@ywang-moblin2.bj.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
I double-checked the datasheet. One of the existing
descriptors has a typo: it should be 2MB not 2038 KB.
Signed-off-by: Dave Jones <davej@redhat.com>
Cc: <stable@kernel.org> # .3x.x: 85160b9: x86: Add new Intel CPU cache size descriptors
Cc: <stable@kernel.org> # .3x.x
LKML-Reference: <20091110200120.GA27090@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The latest rev of Intel doc AP-485 details new cache descriptors
that we don't yet support. 12MB, 18MB and 24MB 24-way assoc L3
caches.
Signed-off-by: Dave Jones <davej@redhat.com>
LKML-Reference: <20091110184924.GA20337@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
On platforms where the BIOS handles the thermal monitor interrupt,
APIC_LVTTHMR on each logical CPU is programmed to generate a SMI
and OS must not touch it.
Unfortunately AP bringup sequence using INIT-SIPI-SIPI clears all
the LVT entries except the mask bit. Essentially this results in
all LVT entries including the thermal monitoring interrupt set
to masked (clearing the bios programmed value for APIC_LVTTHMR).
And this leads to kernel take over the thermal monitoring
interrupt on AP's but not on BSP (leaving the bios programmed
value only on BSP).
As a result of this, we have seen system hangs when the thermal
monitoring interrupt is generated.
Fix this by reading the initial value of thermal LVT entry on
BSP and if bios has taken over the control, then program the
same value on all AP's and leave the thermal monitoring
interrupt control on all the logical cpu's to the bios.
Signed-off-by: Yong Wang <yong.y.wang@intel.com>
Reviewed-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Borislav Petkov <borislav.petkov@amd.com>
Cc: Arjan van de Ven <arjan@infradead.org>
LKML-Reference: <20091110013824.GA24940@ywang-moblin2.bj.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: stable@kernel.org
Conflicts:
tools/perf/Makefile
Merge reason: Resolve the conflict, merge to upstream and merge in
perf fixes so we can add a dependent patch.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Prefix global/setup routines with "mcheck_" thus differentiating
from the internal facilities prefixed with "mce_". Also, prefix
the per cpu calls with mcheck_cpu and rename them to reflect the
MCE setup hierarchy of calls better.
There should be no functionality change resulting from this
patch.
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Cc: Andi Kleen <andi@firstfloor.org>
LKML-Reference: <1255689093-26921-1-git-send-email-borislav.petkov@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The MCE initialization code explicitly says it doesn't handle
asymmetric configurations where different CPUs support different
numbers of MCE banks, and it prints a big warning in that case.
Therefore, printing the "mce: CPU supports <x> MCE banks"
message into the kernel log for every CPU is pure redundancy
that clutters the log significantly for systems with lots of
CPUs.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
LKML-Reference: <adaeip473qt.fsf@cisco.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This approach is the first baby step towards solving many of the
structural problems the x86 MCE logging code is having today:
- It has a private ring-buffer implementation that has a number
of limitations and has been historically fragile and buggy.
- It is using a quirky /dev/mcelog ioctl driven ABI that is MCE
specific. /dev/mcelog is not part of any larger logging
framework and hence has remained on the fringes for many years.
- The MCE logging code is still very unclean partly due to its ABI
limitations. Fields are being reused for multiple purposes, and
the whole message structure is limited and x86 specific to begin
with.
All in one, the x86 tree would like to move away from this private
implementation of an event logging facility to a broader framework.
By using perf events we gain the following advantages:
- Multiple user-space agents can access MCE events. We can have an
mcelog daemon running but also a system-wide tracer capturing
important events in flight-recorder mode.
- Sampling support: the kernel and the user-space call-chain of MCE
events can be stored and analyzed as well. This way actual patterns
of bad behavior can be matched to precisely what kind of activity
happened in the kernel (and/or in the app) around that moment in
time.
- Coupling with other hardware and software events: the PMU can track a
number of other anomalies - monitoring software might chose to
monitor those plus the MCE events as well - in one coherent stream of
events.
- Discovery of MCE sources - tracepoints are enumerated and tools can
act upon the existence (or non-existence) of various channels of MCE
information.
- Filtering support: we just subscribe to and act upon the events we
are interested in. Then even on a per event source basis there's
in-kernel filter expressions available that can restrict the amount
of data that hits the event channel.
- Arbitrary deep per cpu buffering of events - we can buffer 32
entries or we can buffer as much as we want, as long as we have
the RAM.
- An NMI-safe ring-buffer implementation - mappable to user-space.
- Built-in support for timestamping of events, PID markers, CPU
markers, etc.
- A rich ABI accessible over system call interface. Per cpu, per task
and per workload monitoring of MCE events can be done this way. The
ABI itself has a nice, meaningful structure.
- Extensible ABI: new fields can be added without breaking tooling.
New tracepoints can be added as the hardware side evolves. There's
various parsers that can be used.
- Lots of scheduling/buffering/batching modes of operandi for MCE
events. poll() support. mmap() support. read() support. You name it.
- Rich tooling support: even without any MCE specific extensions added
the 'perf' tool today offers various views of MCE data: perf report,
perf stat, perf trace can all be used to view logged MCE events and
perhaps correlate them to certain user-space usage patterns. But it
can be used directly as well, for user-space agents and policy action
in mcelog, etc.
With this we hope to achieve significant code cleanup and feature
improvements in the MCE code, and we hope to be able to drop the
/dev/mcelog facility in the end.
This patch is just a plain dumb dump of mce_log() records to
the tracepoints / perf events framework - a first proof of
concept step.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
LKML-Reference: <4AD42A0D.7050104@jp.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
