Messages in this thread | | | From | Tony Luck <> | Date | Tue, 3 Jan 2012 11:49:55 -0800 | Subject | [PATCH 0/6] x86, mce: machine check recovery for applications |
| |
This series adds code to recognise the machine check signature for a recoverable error in the data path (Advanced SKUs of "Sandy Bridge" server processors are the first to be able to allow s/w recovery for this case), save the required information in the machine check handler and then call to the generic memory_failure() code to try for graceful error recovery (sending SIGBUS to affected process(es)).
Updates since last version (December 15th)
Part1-4: unchanged
Part5: Changed stub function for CONFIG_MEMORY_FAILURE=n case to BUG_ON if it is handed an MF_ACTION_REQUIRED case (this indicates an error in severity calculation). Drop "Memory error recovered" message (enough chatter already).
Part6: Only pass back an ACTION_REQUIRED severity to a kernel if it is built with CONFIG_MEMORY_FAILURE=y (i.e. has the code to take the action).
Whole series is available in:
git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras.git mce-recovery
Tony Luck (6): HWPOISON: clean up memory_failure() vs. __memory_failure() HWPOISON: Add code to handle "action required" errors. x86, mce: create helper function to save addr/misc when needed x86, mce: Add mechanism to safely save information in MCE handler x86, mce: handle "action required" errors x86, mce: Recognise machine check bank signature for data path error
arch/x86/kernel/cpu/mcheck/mce-severity.c | 16 +++- arch/x86/kernel/cpu/mcheck/mce.c | 179 ++++++++++++++++++++--------- drivers/base/memory.c | 2 +- include/linux/mm.h | 4 +- mm/hwpoison-inject.c | 4 +- mm/madvise.c | 2 +- mm/memory-failure.c | 96 ++++++++-------- 7 files changed, 197 insertions(+), 106 deletions(-)
-- 1.7.3.1
| |