BMC/Intel-BMC/linux.git/arch/x86/include/asm/mce.h, branch dev

x86/mce: Add infrastructure to support Local MCE

2015-06-07T13:33:14+00:00

Initialize and prepare for handling LMCEs. Add a boot-time option to disable LMCEs. Signed-off-by: Ashok Raj [ Simplify stuff, align statements for better readability, reflow comments; kill unused lmce_clear(); save us an MSR write if LMCE is already enabled. ] Signed-off-by: Borislav Petkov Cc: Andrew Morton Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Tony Luck Cc: linux-edac Link: http://lkml.kernel.org/r/1433436928-31903-16-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar

x86/mce: Add Local MCE definitions

2015-06-07T13:33:13+00:00

Add required definitions to support Local Machine Check Exceptions. Historically, machine check exceptions on Intel x86 processors have been broadcast to all logical processors in the system. Upcoming CPUs will support an opt-in mechanism to request some machine check exceptions be delivered to a single logical processor experiencing the fault. See http://www.intel.com/sdm Volume 3, System Programming Guide, chapter 15 for more information on MSRs and documentation on Local MCE. Signed-off-by: Ashok Raj Signed-off-by: Borislav Petkov Cc: Andrew Morton Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Tony Luck Cc: linux-edac Link: http://lkml.kernel.org/r/1433436928-31903-15-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar

x86/mce/amd: Introduce deferred error interrupt handler

2015-05-07T08:23:32+00:00

Deferred errors indicate error conditions that were not corrected, but require no action from S/W (or action is optional).These errors provide info about a latent UC MCE that can occur when a poisoned data is consumed by the processor. Processors that report these errors can be configured to generate APIC interrupts to notify OS about the error. Provide an interrupt handler in this patch so that OS can catch these errors as and when they happen. Currently, we simply log the errors and exit the handler as S/W action is not mandated. Signed-off-by: Aravind Gopalakrishnan Cc: Tony Luck Cc: x86-ml Cc: linux-edac Link: http://lkml.kernel.org/r/1430913538-1415-5-git-send-email-Aravind.Gopalakrishnan@amd.com Signed-off-by: Borislav Petkov

x86/mce: Add support for deferred errors on AMD

2015-05-06T18:34:31+00:00

Deferred errors indicate error conditions that were not corrected, but those errors have not been consumed yet. They require no action from S/W (or action is optional). These errors provide info about a latent uncorrectable MCE that can occur when a poisoned data is consumed by the processor. Newer AMD processors can generate deferred errors and can be configured to generate APIC interrupts on such events. SUCCOR stands for S/W UnCorrectable error COntainment and Recovery. It indicates support for data poisoning in HW and deferred error interrupts. Add new bitfield to mce_vendor_flags for this. We use this to verify presence of deferred error interrupts before we enable them in mce_amd.c While at it, clarify comments in mce_vendor_flags to provide an indication of usages of the bitfields. Signed-off-by: Aravind Gopalakrishnan Cc: Tony Luck Cc: x86-ml Cc: linux-edac Link: http://lkml.kernel.org/r/1430913538-1415-4-git-send-email-Aravind.Gopalakrishnan@amd.com [ beef up commit message, do CPUID(8000_0007) only once. ] Signed-off-by: Borislav Petkov

x86/mce: Define mce_severity function pointer

2015-03-24T11:14:15+00:00

Rename mce_severity() to mce_severity_intel() and assign the mce_severity function pointer to mce_severity_amd() during init on AMD. This way, we can avoid a test to call mce_severity_amd every time we get into mce_severity(). And it's cleaner to do it this way. Signed-off-by: Aravind Gopalakrishnan Suggested-by: Tony Luck Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: Chen Yucong Cc: Andy Lutomirski Cc: linux-edac Link: http://lkml.kernel.org/r/1427125373-2918-3-git-send-email-Aravind.Gopalakrishnan@amd.com Signed-off-by: Borislav Petkov

x86/mce: Add an AMD severities-grading function

2015-03-24T11:13:34+00:00

Add a severities function that caters to AMD processors. This allows us to do some vendor-specific work within the function if necessary. Also, introduce a vendor flag bitfield for vendor-specific settings. The severities code uses this to define error scope based on the prescence of the flags field. This is based off of work by Boris Petkov. Testing details: Fam10h, Model 9h (Greyhound) Fam15h: Models 0h-0fh (Orochi), 30h-3fh (Kaveri) and 60h-6fh (Carrizo), Fam16h Model 00h-0fh (Kabini) Boris: Intel SNB AMD K8 (JH-E0) Signed-off-by: Aravind Gopalakrishnan Acked-by: Tony Luck Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: Chen Yucong Cc: Andy Lutomirski Cc: linux-edac@vger.kernel.org Link: http://lkml.kernel.org/r/1427125373-2918-2-git-send-email-Aravind.Gopalakrishnan@amd.com [ Fixup build, clean up comments. ] Signed-off-by: Borislav Petkov

Merge tag 'ras_for_3.21' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras into x86/ras

2015-02-19T12:31:33+00:00

Pull RAS updates from Borislav Petkov: "- Enable AMD thresholding IRQ by default if supported. (Aravind Gopalakrishnan) - Unify mce_panic() message pattern. (Derek Che) - A bit more involved simplification of the CMCI logic after yet another report about race condition with the adaptive logic. (Borislav Petkov) - ACPI APEI EINJ fleshing out of the user documentation. (Borislav Petkov) - Minor cleanup. (Jan Beulich.)" Signed-off-by: Ingo Molnar

x86/MCE/intel: Cleanup CMCI storm logic

2015-02-19T12:24:25+00:00

Initially, this started with the yet another report about a race condition in the CMCI storm adaptive period length thing. Yes, we have to admit, it is fragile and error prone. So let's simplify it. The simpler logic is: now, after we enter storm mode, we go straight to polling with CMCI_STORM_INTERVAL, i.e. once a second. We remain in storm mode as long as we see errors being logged while polling. Theoretically, if we see an uninterrupted error stream, we will remain in storm mode indefinitely and keep polling the MSRs. However, when the storm is actually a burst of errors, once we have logged them all, we back out of it after ~5 mins of polling and no more errors logged. If we encounter an error during those 5 minutes, we reset the polling interval to 5 mins. Making machine_check_poll() return a bool and denoting whether it has seen an error or not lets us simplify a bunch of code and move the storm handling private to mce_intel.c. Some minor cleanups while at it. Reported-by: Calvin Owens Tested-by: Tony Luck Link: http://lkml.kernel.org/r/1417746575-23299-1-git-send-email-calvinowens@fb.com Signed-off-by: Borislav Petkov

x86, mce: Get rid of TIF_MCE_NOTIFY and associated mce tricks

2015-01-07T15:47:42+00:00

We now switch to the kernel stack when a machine check interrupts during user mode. This means that we can perform recovery actions in the tail of do_machine_check() Acked-by: Borislav Petkov Signed-off-by: Tony Luck Signed-off-by: Andy Lutomirski

x86, mce, severity: Extend the the mce_severity mechanism to handle UCNA/DEFERRED error

2014-11-19T18:55:43+00:00

Until now, the mce_severity mechanism can only identify the severity of UCNA error as MCE_KEEP_SEVERITY. Meanwhile, it is not able to filter out DEFERRED error for AMD platform. This patch extends the mce_severity mechanism for handling UCNA/DEFERRED error. In order to do this, the patch introduces a new severity level - MCE_UCNA/DEFERRED_SEVERITY. In addition, mce_severity is specific to machine check exception, and it will check MCIP/EIPV/RIPV bits. In order to use mce_severity mechanism in non-exception context, the patch also introduces a new argument (is_excp) for mce_severity. `is_excp' is used to explicitly specify the calling context of mce_severity. Reviewed-by: Aravind Gopalakrishnan Signed-off-by: Chen Yucong Signed-off-by: Tony Luck