This series is based on tip/perf/urgent. It should apply to any recent kernel, but my test case [1] will OOPS without the fix in tip/perf/urgent.
The first patch or something like it is needed for 3.16, regardless of any lret hackery -- I think it's making its way toward a pull request this week, but I haven't seen it land in -tip yet. I'm repeating it here because the LRET stuff won't work on paravirt, and patch 1 puts the machinery for that in place. tl;dr Read patch 2. To test, apply patch 1 and consider basing on tip/perf/urgent. [1] sigreturn_32 from https://gitorious.org/linux-test-utils/linux-clock-tests/ Andy Lutomirski (2): x86_64,entry,xen: Do not invoke espfix64 on Xen x86_64,entry: Use lret to return to userspace when possible arch/x86/include/asm/irqflags.h | 3 +- arch/x86/include/asm/paravirt.h | 4 ++ arch/x86/include/asm/traps.h | 6 ++ arch/x86/kernel/cpu/mcheck/mce.c | 2 + arch/x86/kernel/entry_64.S | 121 ++++++++++++++++++++++++++++++------ arch/x86/kernel/nmi.c | 21 +++++++ arch/x86/kernel/paravirt_patch_64.c | 2 - 7 files changed, 136 insertions(+), 23 deletions(-) -- 1.9.3 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
