lkml.org 
[lkml]   [2015]   [May]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [PATCH 054/208] x86/fpu: Get rid of PF_USED_MATH usage, convert it to fpu->fpstate_active
On May 5, 2015 9:59 PM, "Ingo Molnar" <mingo@kernel.org> wrote:
>
> Introduce a simple fpu->fpstate_active flag in the fpu context data structure
> and use that instead of PF_USED_MATH in task->flags.
>
> Testing for this flag byte should be slightly more efficient than
> testing a bit in a bitmask, but the main advantage is that most
> FPU functions can now be performed on a 'struct fpu' alone, they
> don't need access to 'struct task_struct' anymore.
>
> There's a slight linecount increase, mostly due to the 'fpu' local
> variables and due to extra comments. The local variables will go away
> once we move most of the FPU methods to pure 'struct fpu' parameters.
>
> Reviewed-by: Borislav Petkov <bp@alien8.de>
> Cc: Andy Lutomirski <luto@amacapital.net>
> Cc: Dave Hansen <dave.hansen@linux.intel.com>
> Cc: Fenghua Yu <fenghua.yu@intel.com>
> Cc: H. Peter Anvin <hpa@zytor.com>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Oleg Nesterov <oleg@redhat.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Signed-off-by: Ingo Molnar <mingo@kernel.org>
> ---
> arch/x86/ia32/ia32_signal.c | 3 ++-
> arch/x86/include/asm/fpu-internal.h | 4 ++--
> arch/x86/include/asm/fpu/types.h | 6 ++++++
> arch/x86/include/asm/processor.h | 6 ++++--
> arch/x86/kernel/fpu/core.c | 38 +++++++++++++++++++++++++-------------
> arch/x86/kernel/fpu/xsave.c | 11 ++++++-----
> arch/x86/kernel/signal.c | 8 +++++---
> arch/x86/kvm/x86.c | 3 ++-
> arch/x86/math-emu/fpu_entry.c | 3 ++-
> 9 files changed, 54 insertions(+), 28 deletions(-)
>
> diff --git a/arch/x86/ia32/ia32_signal.c b/arch/x86/ia32/ia32_signal.c
> index bffb2c49ceb6..e1ec6f90d09e 100644
> --- a/arch/x86/ia32/ia32_signal.c
> +++ b/arch/x86/ia32/ia32_signal.c
> @@ -307,6 +307,7 @@ static void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
> size_t frame_size,
> void __user **fpstate)
> {
> + struct fpu *fpu = &current->thread.fpu;
> unsigned long sp;
>
> /* Default to using normal stack */
> @@ -321,7 +322,7 @@ static void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs,
> ksig->ka.sa.sa_restorer)
> sp = (unsigned long) ksig->ka.sa.sa_restorer;
>
> - if (current->flags & PF_USED_MATH) {
> + if (fpu->fpstate_active) {
> unsigned long fx_aligned, math_size;
>
> sp = alloc_mathframe(sp, 1, &fx_aligned, &math_size);
> diff --git a/arch/x86/include/asm/fpu-internal.h b/arch/x86/include/asm/fpu-internal.h
> index 2cac49e3b4bd..9311126571ab 100644
> --- a/arch/x86/include/asm/fpu-internal.h
> +++ b/arch/x86/include/asm/fpu-internal.h
> @@ -375,7 +375,7 @@ static inline void drop_fpu(struct task_struct *tsk)
> __thread_fpu_end(fpu);
> }
>
> - tsk->flags &= ~PF_USED_MATH;
> + fpu->fpstate_active = 0;
>
> preempt_enable();
> }
> @@ -424,7 +424,7 @@ static inline fpu_switch_t switch_fpu_prepare(struct task_struct *old, struct ta
> * If the task has used the math, pre-load the FPU on xsave processors
> * or if the past 5 consecutive context-switches used math.
> */
> - fpu.preload = (new->flags & PF_USED_MATH) &&
> + fpu.preload = new_fpu->fpstate_active &&
> (use_eager_fpu() || new->thread.fpu.counter > 5);
>
> if (old_fpu->has_fpu) {
> diff --git a/arch/x86/include/asm/fpu/types.h b/arch/x86/include/asm/fpu/types.h
> index efb520dcf38e..f6317d9aa808 100644
> --- a/arch/x86/include/asm/fpu/types.h
> +++ b/arch/x86/include/asm/fpu/types.h
> @@ -137,6 +137,12 @@ struct fpu {
> * deal with bursty apps that only use the FPU for a short time:
> */
> unsigned char counter;
> + /*
> + * This flag indicates whether this context is fpstate_active: if the task is
> + * not running then we can restore from this context, if the task
> + * is running then we should save into this context.
> + */
> + unsigned char fpstate_active;

I don't understand. What does it mean if !fpstate_active?

--Andy


\
 
 \ /
  Last update: 2015-05-06 03:21    [W:0.273 / U:0.116 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site