| Message ID | 1591321039-22141-1-git-send-email-naoya.horiguchi@nec.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | hwpoison: fixes signaling on memory error | expand |
Sorry, recipients are not enough. I meant to send/CC this to reviewers. https://lore.kernel.org/linux-mm/1591321039-22141-1-git-send-email-naoya.horiguchi@nec.com/T/#t Any comment is welcomed. Thanks, Naoya Horiguchi On Fri, Jun 05, 2020 at 10:37:17AM +0900, Naoya Horiguchi wrote: > Hi, > > This is a small patchset to solve issues in memory error handler to > send SIGBUS to proper process/thread as expected in configuration. > Please see descriptions in individual patches for more details. > > This patchset depends on commit 872e9a205c84 ("mm, memory_failure: don't send > BUS_MCEERR_AO for action required error") currently in upstream. > > Thanks, > Naoya Horiguchi > --- > Summary: > > Naoya Horiguchi (2): > mm/memory-failure: prioritize prctl(PR_MCE_KILL) over vm.memory_failure_early_kill > mm/memory-failure: send SIGBUS(BUS_MCEERR_AR) only to current thread > > mm/memory-failure.c | 43 ++++++++++++++++++++++++++----------------- > 1 file changed, 26 insertions(+), 17 deletions(-) >
On Fri, 5 Jun 2020 01:46:46 +0000 HORIGUCHI NAOYA(堀口 直也) <naoya.horiguchi@nec.com> wrote: > Sorry, recipients are not enough. I meant to send/CC this to reviewers. > > https://lore.kernel.org/linux-mm/1591321039-22141-1-git-send-email-naoya.horiguchi@nec.com/T/#t > > Any comment is welcomed. How serious are these issues? Do you think serious enough to justify adding these to 5.8?
On Sun, Jun 07, 2020 at 05:53:18PM -0700, Andrew Morton wrote: > On Fri, 5 Jun 2020 01:46:46 +0000 HORIGUCHI NAOYA(堀口 直也) <naoya.horiguchi@nec.com> wrote: > > > Sorry, recipients are not enough. I meant to send/CC this to reviewers. > > > > https://lore.kernel.org/linux-mm/1591321039-22141-1-git-send-email-naoya.horiguchi@nec.com/T/#t > > > > Any comment is welcomed. > > How serious are these issues? Do you think serious enough to justify > adding these to 5.8? The issue of 1/2 is not serious. But the issue of 2/2 is a little severer, where we could wrongly kill the main thread when the subthread access to the error in multithread applications. So I want it to be merged for 5.8. Thanks, Naoya Horiguchi
Hi, This is a small patchset to solve issues in memory error handler to send SIGBUS to proper process/thread as expected in configuration. Please see descriptions in individual patches for more details. This patchset depends on commit 872e9a205c84 ("mm, memory_failure: don't send BUS_MCEERR_AO for action required error") currently in upstream. Thanks, Naoya Horiguchi --- Summary: Naoya Horiguchi (2): mm/memory-failure: prioritize prctl(PR_MCE_KILL) over vm.memory_failure_early_kill mm/memory-failure: send SIGBUS(BUS_MCEERR_AR) only to current thread mm/memory-failure.c | 43 ++++++++++++++++++++++++++----------------- 1 file changed, 26 insertions(+), 17 deletions(-)