| Message ID | 1409540991-7573-1-git-send-email-leoy@marvell.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 1 September 2014 05:09, Leo Yan <leoy@marvell.com> wrote: > Now arm64 defers reloading FPSIMD state, but this optimization also > introduces the bug after cpu resume back from low power mode. > > The reason is after the cpu has been powered off, s/w need set the > cpu's fpsimd_last_state to NULL so that it will force to reload > FPSIMD state for the thread, otherwise there has the chance to meet > the condition for both the task's fpsimd_state.cpu field contains the > id of the current cpu, and the cpu's fpsimd_last_state per-cpu variable > points to the task's fpsimd_state, so finally kernel will skip to reload > the context during it return back to userland. > > Signed-off-by: Leo Yan <leoy@marvell.com> Acked-by: Ard Biesheuvel <ard.biesheuvel@linaro.org> > --- > 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 ad8aebb..3dca156 100644 > --- a/arch/arm64/kernel/fpsimd.c > +++ b/arch/arm64/kernel/fpsimd.c > @@ -270,6 +270,7 @@ static int fpsimd_cpu_pm_notifier(struct notifier_block *self, > case CPU_PM_ENTER: > if (current->mm && !test_thread_flag(TIF_FOREIGN_FPSTATE)) > fpsimd_save_state(¤t->thread.fpsimd_state); > + this_cpu_write(fpsimd_last_state, NULL); > break; > case CPU_PM_EXIT: > if (current->mm) > -- > 1.9.1 >
On Mon, Sep 01, 2014 at 04:09:51AM +0100, Leo Yan wrote: > Now arm64 defers reloading FPSIMD state, but this optimization also > introduces the bug after cpu resume back from low power mode. > > The reason is after the cpu has been powered off, s/w need set the > cpu's fpsimd_last_state to NULL so that it will force to reload > FPSIMD state for the thread, otherwise there has the chance to meet > the condition for both the task's fpsimd_state.cpu field contains the > id of the current cpu, and the cpu's fpsimd_last_state per-cpu variable > points to the task's fpsimd_state, so finally kernel will skip to reload > the context during it return back to userland. > > Signed-off-by: Leo Yan <leoy@marvell.com> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> (and cc'ing Will to pick it up since he's handling the 3.17 release) Thanks.
On Mon, Sep 01, 2014 at 10:34:28AM +0100, Catalin Marinas wrote: > On Mon, Sep 01, 2014 at 04:09:51AM +0100, Leo Yan wrote: > > Now arm64 defers reloading FPSIMD state, but this optimization also > > introduces the bug after cpu resume back from low power mode. > > > > The reason is after the cpu has been powered off, s/w need set the > > cpu's fpsimd_last_state to NULL so that it will force to reload > > FPSIMD state for the thread, otherwise there has the chance to meet > > the condition for both the task's fpsimd_state.cpu field contains the > > id of the current cpu, and the cpu's fpsimd_last_state per-cpu variable > > points to the task's fpsimd_state, so finally kernel will skip to reload > > the context during it return back to userland. > > > > Signed-off-by: Leo Yan <leoy@marvell.com> > > Reviewed-by: Catalin Marinas <catalin.marinas@arm.com> > > (and cc'ing Will to pick it up since he's handling the 3.17 release) Cheers for the CC; applied to fixes/core (I'll push out once I've run the regression tests). I aim to tag that branch tonight if we don't get any more fixes today. Will
diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index ad8aebb..3dca156 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -270,6 +270,7 @@ static int fpsimd_cpu_pm_notifier(struct notifier_block *self, case CPU_PM_ENTER: if (current->mm && !test_thread_flag(TIF_FOREIGN_FPSTATE)) fpsimd_save_state(¤t->thread.fpsimd_state); + this_cpu_write(fpsimd_last_state, NULL); break; case CPU_PM_EXIT: if (current->mm)
Now arm64 defers reloading FPSIMD state, but this optimization also introduces the bug after cpu resume back from low power mode. The reason is after the cpu has been powered off, s/w need set the cpu's fpsimd_last_state to NULL so that it will force to reload FPSIMD state for the thread, otherwise there has the chance to meet the condition for both the task's fpsimd_state.cpu field contains the id of the current cpu, and the cpu's fpsimd_last_state per-cpu variable points to the task's fpsimd_state, so finally kernel will skip to reload the context during it return back to userland. Signed-off-by: Leo Yan <leoy@marvell.com> --- arch/arm64/kernel/fpsimd.c | 1 + 1 file changed, 1 insertion(+)