The following commit:
27f6c573e0 ("x86, CMCI: Add proper detection of end of CMCI storms")
Added two preemption bugs:
- machine_check_poll() does a get_cpu_var() without a matching
put_cpu_var(), which causes preemption imbalance and crashes upon
bootup.
- it does percpu ops without disabling preemption. Preemption is not
disabled due to the mistaken use of a raw spinlock.
To fix these bugs fix the imbalance and change
cmci_discover_lock to a regular spinlock.
Reported-by: Owen Kibel <qmewlo@gmail.com>
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Chen, Gong <gong.chen@linux.intel.com>
Cc: Josh Boyer <jwboyer@fedoraproject.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Alexander Todorov <atodorov@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Link: http://lkml.kernel.org/n/tip-jtjptvgigpfkpvtQxpEk1at2@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
--
arch/x86/kernel/cpu/mcheck/mce.c | 4 +---
arch/x86/kernel/cpu/mcheck/mce_intel.c | 18 +++++++++---------
2 files changed, 10 insertions(+), 12 deletions(-)
When CMCI storm persists for a long time(at least beyond predefined
threshold. It's 30 seconds for now), we can watch CMCI storm is
detected immediately after it subsides.
...
Dec 10 22:04:29 kernel: CMCI storm detected: switching to poll mode
Dec 10 22:04:59 kernel: CMCI storm subsided: switching to interrupt mode
Dec 10 22:04:59 kernel: CMCI storm detected: switching to poll mode
Dec 10 22:05:29 kernel: CMCI storm subsided: switching to interrupt mode
...
The problem is that our logic that determines that the storm has
ended is incorrect. We announce the end, re-enable interrupts and
realize that the storm is still going on, so we switch back to
polling mode. Rinse, repeat.
When a storm happens we disable signaling of errors via CMCI and begin
polling machine check banks instead. If we find any logged errors,
then we need to set a per-cpu flag so that our per-cpu tests that
check whether the storm is ongoing will see that errors are still
being logged independently of whether mce_notify_irq() says that the
error has been fully processed.
cmci_clear() is not the right tool to disable a bank. It disables the
interrupt for the bank as desired, but it also clears the bit for
this bank in "mce_banks_owned" so we will skip the bank when polling
(so we fail to see that the storm continues because we stop looking).
New cmci_storm_disable_banks() just disables the interrupt while
allowing polling to continue.
Reported-by: William Dauchy <wdauchy@gmail.com>
Signed-off-by: Chen, Gong <gong.chen@linux.intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
None of these files are actually using any __init type directives
and hence don't need to include <linux/init.h>. Most are just a
left over from __devinit and __cpuinit removal, or simply due to
code getting copied from one driver to the next.
[ hpa: undid incorrect removal from arch/x86/kernel/head_32.S ]
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Link: http://lkml.kernel.org/r/1389054026-12947-1-git-send-email-paul.gortmaker@windriver.com
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The Corrected Machine Check structure (CMC) in HEST has a flag which can be
set by the firmware to indicate to the OS that it prefers to process the
corrected error events first. In this scenario, the OS is expected to not
monitor for corrected errors (through CMCI/polling). Instead, the firmware
notifies the OS on corrected error events through GHES.
Linux already has support for GHES. This patch adds support for parsing CMC
structure and to disable CMCI/polling if the firmware first flag is set.
Further, the list of machine check bank structures at the end of CMC is used
to determine which MCA banks function in FF mode, so that we continue to
monitor error events on the other banks.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
There is some confusion about the 'mce_poll_banks' and 'mce_banks_owned'
per-cpu bitmaps. Provide comments so that we all know exactly what these
are used for, and why.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Acked-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Dave Jones reports that offlining a CPU leads to this trace:
numa_remove_cpu cpu 1 node 0: mask now 0,2-3
smpboot: CPU 1 is now offline
BUG: using smp_processor_id() in preemptible [00000000] code:
cpu-offline.sh/10591
caller is cmci_rediscover+0x6a/0xe0
Pid: 10591, comm: cpu-offline.sh Not tainted 3.9.0-rc3+ #2
Call Trace:
[<ffffffff81333bbd>] debug_smp_processor_id+0xdd/0x100
[<ffffffff8101edba>] cmci_rediscover+0x6a/0xe0
[<ffffffff815f5b9f>] mce_cpu_callback+0x19d/0x1ae
[<ffffffff8160ea66>] notifier_call_chain+0x66/0x150
[<ffffffff8107ad7e>] __raw_notifier_call_chain+0xe/0x10
[<ffffffff8104c2e3>] cpu_notify+0x23/0x50
[<ffffffff8104c31e>] cpu_notify_nofail+0xe/0x20
[<ffffffff815ef082>] _cpu_down+0x302/0x350
[<ffffffff815ef106>] cpu_down+0x36/0x50
[<ffffffff815f1c9d>] store_online+0x8d/0xd0
[<ffffffff813edc48>] dev_attr_store+0x18/0x30
[<ffffffff81226eeb>] sysfs_write_file+0xdb/0x150
[<ffffffff811adfb2>] vfs_write+0xa2/0x170
[<ffffffff811ae16c>] sys_write+0x4c/0xa0
[<ffffffff81613019>] system_call_fastpath+0x16/0x1b
However, a look at cmci_rediscover shows that it can be simplified quite
a bit, apart from solving the above issue. It invokes functions that
take spin locks with interrupts disabled, and hence it can run in atomic
context. Also, it is run in the CPU_POST_DEAD phase, so the dying CPU
is already dead and out of the cpu_online_mask. So take these points into
account and simplify the code, and thereby also fix the above issue.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Pull x86 RAS update from Ingo Molnar:
"Rework all config variables used throughout the MCA code and collect
them together into a mca_config struct. This keeps them tightly and
neatly packed together instead of spilled all over the place.
Then, convert those which are used as booleans into real booleans and
save some space. These bits are exposed via
/sys/devices/system/machinecheck/machinecheck*/"
* 'x86-ras-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, MCA: Finish mca_config conversion
x86, MCA: Convert the next three variables batch
x86, MCA: Convert rip_msr, mce_bootlog, monarch_timeout
x86, MCA: Convert dont_log_ce, banks and tolerant
drivers/base: Add a DEVICE_BOOL_ATTR macro
mce_ser, mce_bios_cmci_threshold and mce_disabled are the last three
bools which need conversion. Move them to the mca_config struct and
adjust usage sites accordingly.
Signed-off-by: Borislav Petkov <bp@alien8.de>
Acked-by: Tony Luck <tony.luck@intel.com>
Move them into the mca_config struct and adjust code touching them
accordingly.
Signed-off-by: Borislav Petkov <bp@alien8.de>
Acked-by: Tony Luck <tony.luck@intel.com>
The ACPI spec doesn't provide for a way for the bios to pass down
recommended thresholds to the OS on a _per-bank_ basis. This patch adds
a new boot option, which if passed, tells Linux to use CMCI thresholds
set by the bios.
As fail-safe, we initialize threshold to 1 if some banks have not been
initialized by the bios and warn the user.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
On Intel systems corrected machine check interrupts (CMCI) may be sent to
multiple logical processors; possibly to all processors on the affected
socket (SDM Volume 3B "15.5.1 CMCI Local APIC Interface"). This means
that a persistent error (such as a stuck bit in ECC memory) may cause
a storm of interrupts that greatly hinders or prevents forward progress
(probably on many processors).
To solve this we keep track of the rate at which each processor sees
CMCI. If we exceed a threshold, we disable CMCI delivery and switch to
polling the machine check banks. If the storm subsides (none of the
affected processors see any more errors for a complete poll interval) we
re-enable CMCI.
[Tony: Added console messages when storm begins/ends and increased storm
threshold from 5 to 15 so we have a few more logged entries before we
disable interrupts and start dropping reports]
Signed-off-by: Chen Gong <gong.chen@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Chen Gong <gong.chen@linux.intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
cmci_discover() works out which machine check banks support CMCI, and
which of those are shared by multiple logical processors. It uses this
information to ensure that exactly one cpu is designated the owner of
each bank so that when interrupts are broadcast to multiple cpus, only one
of them will look in a shared bank to log the error and clear the bank.
At boot time cmci_discover() performs this task silently. But during
certain cpu hotplug operations it prints out a set of summary lines
like this:
CPU 35 MCA banks CMCI:0 CMCI:1 CMCI:3 CMCI:5 CMCI:6 CMCI:7 CMCI:8 CMCI:9 CMCI:10 CMCI:11
CPU 1 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 39 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 38 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 32 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 37 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 36 MCA banks CMCI:0 CMCI:1 CMCI:3
CPU 34 MCA banks CMCI:0 CMCI:1 CMCI:3
The value of these messages seems very low. A user might painstakingly
cross-check against the data sheet for a processor to ensure that all
CMCI supported banks are correctly reported, but this seems improbable.
If users really wanted to do this, we should print the information at
boot time too.
Remove the messages.
Signed-off-by: Tony Luck <tony.luck@intel.com>
The cmci_discover_lock can be taken in atomic context (cpu bring
up sequence) and therefore cannot be preempted on -rt.
In mainline this change documents the low level nature of
the lock - otherwise there's no functional difference. Lockdep
and Sparse checking will work as usual.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Replace all uses of current_cpu_data with this_cpu operations on the
per cpu structure cpu_info. The scala accesses are replaced with the
matching this_cpu ops which results in smaller and more efficient
code.
In the long run, it might be a good idea to remove cpu_data() macro
too and use per_cpu macro directly.
tj: updated description
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
It is reported that CMCI is not raised when number of corrected error
reaches preset threshold. After inspection, it is found that
MSR_IA32_MCI_CTL2 threshold field is not setup properly. This patch
fixed it.
Value of MCI_CTL2_CMCI_THRESHOLD_MASK is fixed according to x86_64
Software Developer's Manual too.
Reported-by: Shaohui Zheng <shaohui.zheng@intel.com>
Signed-off-by: Huang Ying <ying.huang@intel.com>
LKML-Reference: <1275977350.3444.660.camel@yhuang-dev.sh.intel.com>
Reviewed-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Rename CMCI_EN to MCI_CTL2_CMCI_EN and CMCI_THRESHOLD_MASK to
MCI_CTL2_CMCI_THRESHOLD_MASK to make naming consistent.
Signed-off-by: Huang Ying <ying.huang@intel.com>
LKML-Reference: <1275977348.3444.659.camel@yhuang-dev.sh.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
After m68k's task_thread_info() doesn't refer to current,
it's possible to remove sched.h from interrupt.h and not break m68k!
Many thanks to Heiko Carstens for allowing this.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Instead of open coded calculations for bank MSRs hide the indexing of higher
banks MCE register MSRs in new macros.
No semantic changes.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
mce_intel.c uses apic_write() and lapic_get_maxlvt(), and so it needs
<asm/apic.h>.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Rename files that are no longer 64bit specific:
mce_amd_64.c => mce_amd.c
mce_intel_64.c => mce_intel.c
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
move intel_init_thermal() into therm_throt.c
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
There are 2 headers:
arch/x86/include/asm/mce.h
arch/x86/kernel/cpu/mcheck/mce.h
and in the latter small header:
#include <asm/mce.h>
This patch move all contents in the latter header into the former,
and fix all files using the latter to include the former instead.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Decode magic constants and turn them into symbols.
[ Cleanup to use symbols already exists - HS ]
[ Impact: cleanup ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
