On Tue, Oct 15, 2013 at 9:58 PM, Chris Zankel <chris@xxxxxxxxxx> wrote: > Hi Max, > > Thanks for the patches. Is my assumption correct that this fixes memory > faults while in double exception? And, instead of panic-ing, it works > now but registers are likely wrong with ptrace? Well, one of them fixes the fixup handler for register spilling. It is not functional in xtensa-2.6.29-smp and I've blindly ported it with a3/excsave changes. If register spilling causes an exception current fixup handler messes up registers, which results in unrecoverable exception with very obscure trace. I don't think the fix has any effect on ptrace. This fix is not related to the issue with delivering signals generated in double exception, I haven't got chance to fix it it properly yet. Related ML thread: http://lists.linux-xtensa.org/pipermail/linux-xtensa/Week-of-Mon-20130909/001198.html http://lists.linux-xtensa.org/pipermail/linux-xtensa/Week-of-Mon-20130916/001209.html > Also, should that go in for 3.12 (rc6?) or only in 3.13? This is for 3.12. I have another (bigger :) pile for 3.13, will send it in a couple of days. > On 10/14/2013 03:22 PM, Max Filippov wrote: >> Hi Chris, >> >> could you please push these two fixes to the 3.12? >> One fixes the regression introduced by the a3/excsave conversion in 3.12-rc1, >> and the other fixes use of nonexistent alternative signal stack by threads. >> >> You can pull these changes from my git tree, >> the following changes since commit 61e6cfa80de5760bbe406f4e815b7739205754d2: >> >> Linux 3.12-rc5 (2013-10-13 15:41:28 -0700) >> >> are available in the git repository at: >> git://github.com/jcmvbkbc/linux-xtensa.git xtensa-fixes-for-upstream >> >> Baruch Siach (1): >> xtensa: don't use alternate signal stack on threads >> >> Max Filippov (1): >> xtensa: fix fast_syscall_spill_registers_fixup >> >> arch/xtensa/kernel/entry.S | 49 ++++++++++++++++++++++++++---------------- >> arch/xtensa/kernel/signal.c | 2 +- >> 2 files changed, 31 insertions(+), 20 deletions(-) >> -- Thanks. -- Max -- To unsubscribe from this list: send the line "unsubscribe linux-arch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
