kernel.git/arch/x86/kernel/cpu/mce, branch linux-5.7.y

x86/mce/inject: Fix a wrong assignment of i_mce.status

2020-08-19T06:23:25Z

[ Upstream commit 5d7f7d1d5e01c22894dee7c9c9266500478dca99 ] The original code is a nop as i_mce.status is or'ed with part of itself, fix it. Fixes: a1300e505297 ("x86/ras/mce_amd_inj: Trigger deferred and thresholding errors interrupts") Signed-off-by: Zhenzhong Duan Signed-off-by: Borislav Petkov Acked-by: Yazen Ghannam Link: https://lkml.kernel.org/r/20200611023238.3830-1-zhenzhong.duan@gmail.com Signed-off-by: Sasha Levin

x86/mce/dev-mcelog: Fix -Wstringop-truncation warning about strncpy()

2020-06-24T15:49:10Z

[ Upstream commit 7ccddc4613db446dc3cbb69a3763ba60ec651d13 ] The kbuild test robot reported this warning: arch/x86/kernel/cpu/mce/dev-mcelog.c: In function 'dev_mcelog_init_device': arch/x86/kernel/cpu/mce/dev-mcelog.c:346:2: warning: 'strncpy' output \ truncated before terminating nul copying 12 bytes from a string of the \ same length [-Wstringop-truncation] This is accurate, but I don't care that the trailing NUL character isn't copied. The string being copied is just a magic number signature so that crash dump tools can be sure they are decoding the right blob of memory. Use memcpy() instead of strncpy(). Fixes: d8ecca4043f2 ("x86/mce/dev-mcelog: Dynamically allocate space for machine check records") Reported-by: kbuild test robot Signed-off-by: Tony Luck Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Link: https://lkml.kernel.org/r/20200527182808.27737-1-tony.luck@intel.com Signed-off-by: Sasha Levin

x86/{mce,mm}: Unmap the entire page if the whole page is affected and poisoned

2020-06-17T14:42:56Z

commit 17fae1294ad9d711b2c3dd0edef479d40c76a5e8 upstream. An interesting thing happened when a guest Linux instance took a machine check. The VMM unmapped the bad page from guest physical space and passed the machine check to the guest. Linux took all the normal actions to offline the page from the process that was using it. But then guest Linux crashed because it said there was a second machine check inside the kernel with this stack trace: do_memory_failure set_mce_nospec set_memory_uc _set_memory_uc change_page_attr_set_clr cpa_flush clflush_cache_range_opt This was odd, because a CLFLUSH instruction shouldn't raise a machine check (it isn't consuming the data). Further investigation showed that the VMM had passed in another machine check because is appeared that the guest was accessing the bad page. Fix is to check the scope of the poison by checking the MCi_MISC register. If the entire page is affected, then unmap the page. If only part of the page is affected, then mark the page as uncacheable. This assumes that VMMs will do the logical thing and pass in the "whole page scope" via the MCi_MISC register (since they unmapped the entire page). [ bp: Adjust to x86/entry changes. ] Fixes: 284ce4011ba6 ("x86/memory_failure: Introduce {set, clear}_mce_nospec()") Reported-by: Jue Wang Signed-off-by: Tony Luck Signed-off-by: Borislav Petkov Signed-off-by: Thomas Gleixner Tested-by: Jue Wang Cc: Link: https://lkml.kernel.org/r/20200520163546.GA7977@agluck-desk2.amr.corp.intel.com Signed-off-by: Greg Kroah-Hartman

Merge tag 'x86-entry-2020-03-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2020-03-31T02:14:28Z

Pull x86 entry code updates from Thomas Gleixner: - Convert the 32bit syscalls to be pt_regs based which removes the requirement to push all 6 potential arguments onto the stack and consolidates the interface with the 64bit variant - The first small portion of the exception and syscall related entry code consolidation which aims to address the recently discovered issues vs. RCU, int3, NMI and some other exceptions which can interrupt any context. The bulk of the changes is still work in progress and aimed for 5.8. - A few lockdep namespace cleanups which have been applied into this branch to keep the prerequisites for the ongoing work confined. * tag 'x86-entry-2020-03-30' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (35 commits) x86/entry: Fix build error x86 with !CONFIG_POSIX_TIMERS lockdep: Rename trace_{hard,soft}{irq_context,irqs_enabled}() lockdep: Rename trace_softirqs_{on,off}() lockdep: Rename trace_hardirq_{enter,exit}() x86/entry: Rename ___preempt_schedule x86: Remove unneeded includes x86/entry: Drop asmlinkage from syscalls x86/entry/32: Enable pt_regs based syscalls x86/entry/32: Use IA32-specific wrappers for syscalls taking 64-bit arguments x86/entry/32: Rename 32-bit specific syscalls x86/entry/32: Clean up syscall_32.tbl x86/entry: Remove ABI prefixes from functions in syscall tables x86/entry/64: Add __SYSCALL_COMMON() x86/entry: Remove syscall qualifier support x86/entry/64: Remove ptregs qualifier from syscall table x86/entry: Move max syscall number calculation to syscallhdr.sh x86/entry/64: Split X32 syscall table into its own file x86/entry/64: Move sys_ni_syscall stub to common.c x86/entry/64: Use syscall wrappers for x32_rt_sigreturn x86/entry: Refactor SYS_NI macros ...

Merge tag 'ras_updates_for_5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2020-03-30T20:17:50Z

Pull RAS updates from Borislav Petkov: - Do not report spurious MCEs on some Intel platforms caused by errata; by Prarit Bhargava. - Change dev-mcelog's hardcoded limit of 32 error records to a dynamic one, controlled by the number of logical CPUs, by Tony Luck. - Add support for the processor identification number (PPIN) on AMD, by Wei Huang. * tag 'ras_updates_for_5.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/mce/amd: Add PPIN support for AMD MCE x86/mce/dev-mcelog: Dynamically allocate space for machine check records x86/mce: Do not log spurious corrected mce errors

x86/mce/amd: Add PPIN support for AMD MCE

2020-03-22T10:03:47Z

Newer AMD CPUs support a feature called protected processor identification number (PPIN). This feature can be detected via CPUID_Fn80000008_EBX[23]. However, CPUID alone is not enough to read the processor identification number - MSR_AMD_PPIN_CTL also needs to be configured properly. If, for any reason, MSR_AMD_PPIN_CTL[PPIN_EN] can not be turned on, such as disabled in BIOS, the CPU capability bit X86_FEATURE_AMD_PPIN needs to be cleared. When the X86_FEATURE_AMD_PPIN capability is available, the identification number is issued together with the MCE error info in order to keep track of the source of MCE errors. [ bp: Massage. ] Co-developed-by: Smita Koralahalli Channabasappa Signed-off-by: Smita Koralahalli Channabasappa Signed-off-by: Wei Huang Signed-off-by: Borislav Petkov Acked-by: Tony Luck Link: https://lkml.kernel.org/r/20200321193800.3666964-1-wei.huang2@amd.com

x86/mce/dev-mcelog: Dynamically allocate space for machine check records

2020-03-10T09:25:14Z

We have had a hard coded limit of 32 machine check records since the dawn of time. But as numbers of cores increase, it is possible for more than 32 errors to be reported before a user process reads from /dev/mcelog. In this case the additional errors are lost. Keep 32 as the minimum. But tune the maximum value up based on the number of processors. Signed-off-by: Tony Luck Signed-off-by: Borislav Petkov Link: https://lkml.kernel.org/r/20200218184408.GA23048@agluck-desk2.amr.corp.intel.com

x86/mce: Fix logic and comments around MSR_PPIN_CTL

2020-02-27T20:36:42Z

There are two implemented bits in the PPIN_CTL MSR: Bit 0: LockOut (R/WO) Set 1 to prevent further writes to MSR_PPIN_CTL. Bit 1: Enable_PPIN (R/W) If 1, enables MSR_PPIN to be accessible using RDMSR. If 0, an attempt to read MSR_PPIN will cause #GP. So there are four defined values: 0: PPIN is disabled, PPIN_CTL may be updated 1: PPIN is disabled. PPIN_CTL is locked against updates 2: PPIN is enabled. PPIN_CTL may be updated 3: PPIN is enabled. PPIN_CTL is locked against updates Code would only enable the X86_FEATURE_INTEL_PPIN feature for case "2". When it should have done so for both case "2" and case "3". Fix the final test to just check for the enable bit. Also fix some of the other comments in this function. Fixes: 3f5a7896a509 ("x86/mce: Include the PPIN in MCE records when available") Signed-off-by: Tony Luck Signed-off-by: Borislav Petkov Cc: Link: https://lkml.kernel.org/r/20200226011737.9958-1-tony.luck@intel.com

x86/entry/32: Force MCE through do_mce()

2020-02-27T13:48:39Z

Remove the pointless difference between 32 and 64 bit to make further unifications simpler. Signed-off-by: Thomas Gleixner Reviewed-by: Frederic Weisbecker Reviewed-by: Alexandre Chartre Reviewed-by: Andy Lutomirski Link: https://lkml.kernel.org/r/20200225220216.428188397@linutronix.de

x86/mce: Disable tracing and kprobes on do_machine_check()

2020-02-27T13:48:39Z

do_machine_check() can be raised in almost any context including the most fragile ones. Prevent kprobes and tracing. Signed-off-by: Andy Lutomirski Signed-off-by: Thomas Gleixner Reviewed-by: Borislav Petkov Reviewed-by: Alexandre Chartre Reviewed-by: Andy Lutomirski Link: https://lkml.kernel.org/r/20200225220216.315548935@linutronix.de