From mboxrd@z Thu Jan 1 00:00:00 1970 From: hanjun.guo@linaro.org (Hanjun Guo) Date: Thu, 20 Aug 2015 16:42:12 +0800 Subject: [PATCH] arm64: entry: always restore x0 from the stack on syscall return In-Reply-To: <20150819165643.GD10297@arm.com> References: <1439996989-5772-1-git-send-email-will.deacon@arm.com> <20150819160320.GI24062@e104818-lin.cambridge.arm.com> <20150819162359.GA10297@arm.com> <20150819163504.GL24062@e104818-lin.cambridge.arm.com> <20150819165643.GD10297@arm.com> Message-ID: <55D592E4.50609@linaro.org> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 08/20/2015 12:56 AM, Will Deacon wrote: > On Wed, Aug 19, 2015 at 05:35:05PM +0100, Catalin Marinas wrote: >> On Wed, Aug 19, 2015 at 05:23:59PM +0100, Will Deacon wrote: >>> On Wed, Aug 19, 2015 at 05:03:20PM +0100, Catalin Marinas wrote: >>>> There is another str x0 in __sys_trace_return which I think we could >>>> remove. >>> >>> Hmm, I don't think we can remove that. It's needed on the slowpath to >>> update the pt_regs with either -ENOSYS (for __ni_sys_trace) or the >>> syscall return value from the blr in __sys_trace. >>> >>> What we can do instead is change the branch above to branch to >>> __sys_trace_return_skipped. Patch below. >>> >>> Will >>> >>> --->8 >>> >>> diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S >>> index 2a5e64ccc991..088322ff1ba0 100644 >>> --- a/arch/arm64/kernel/entry.S >>> +++ b/arch/arm64/kernel/entry.S >>> @@ -619,7 +619,7 @@ ret_fast_syscall: >>> kernel_exit 0 >>> ret_fast_syscall_trace: >>> enable_irq // enable interrupts >>> - b __sys_trace_return >>> + b __sys_trace_return_skipped // we already saved x0 >> >> That would do. With this added: >> >> Reviewed-by: Catalin Marinas >> >> (or I can take it for 4.2 but I'd like more testing like LTP) > > Yeah, I'll run some tests overnight and see how it holds up. How is the stress test going on? I didn't do some stress test but when I applied this patch (along with above additions), the problem I reported is gone, Tested-by: Hanjun Guo Thanks Hanjun