From mboxrd@z Thu Jan 1 00:00:00 1970 From: will.deacon@arm.com (Will Deacon) Date: Wed, 26 Aug 2015 12:12:05 +0100 Subject: [PATCH] arm64: fix bug for reloading FPSIMD state after execve on cpu 0. In-Reply-To: <1440556841-17219-1-git-send-email-chunyan.zhang@spreadtrum.com> References: <1440556841-17219-1-git-send-email-chunyan.zhang@spreadtrum.com> Message-ID: <20150826111205.GD30466@arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hello, [adding Ard] On Wed, Aug 26, 2015 at 03:40:41AM +0100, Chunyan Zhang wrote: > From: Janet Liu > > If process A is running on CPU 0 and do execve syscall and after sched_exec, > dest_cpu is 0, fpsimd_state.cpu is 0. If at the time Process A get scheduled > out and after some kernel threads running on CPU 0, process A is back in CPU 0, > A's fpsimd_state.cpu is current cpu id "0", and per_cpu(fpsimd_last_state) > points A's fpsimd_state, TIF_FOREIGN_FPSTATE will be clear, kernel will not > reload the context during it return to userspace. so set the cpu's > fpsimd_last_state to NULL to avoid this. AFAICT, this is only a problem if one of the kernel threads uses the fpsimd registers, right? However, kernel_neon_begin_partial clobbers fpsimd_last_state, so I'm struggling to see the problem. Are you seeing an issue in practice? Will > Signed-off-by: Janet Liu > Signed-off-by: Chunyan Zhang > --- > arch/arm64/kernel/fpsimd.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c > index 44d6f75..ec58d94 100644 > --- a/arch/arm64/kernel/fpsimd.c > +++ b/arch/arm64/kernel/fpsimd.c > @@ -159,6 +159,7 @@ void fpsimd_flush_thread(void) > { > memset(¤t->thread.fpsimd_state, 0, sizeof(struct fpsimd_state)); > set_thread_flag(TIF_FOREIGN_FPSTATE); > + this_cpu_write(fpsimd_last_state, NULL); > } > > /* > -- > 1.7.9.5 >