Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Oleg Nesterov
Sorry for noise, but just in case... On 09/02, Oleg Nesterov wrote: > > Do you think this can work? Of course, even if this can work, we should do this later. Let's start with the simple changes, then we will see if we can actually remove all other checks. Oleg. -- To unsubscribe from this

Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Oleg Nesterov
On 09/02, Suresh Siddha wrote: > > On Fri, Aug 29, 2014 at 11:17 AM, Oleg Nesterov wrote: > > ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. > > this patch I think needs more thought for sure. please see below. Of course. > > interrupted_kernel_fpu_idle() does: > > > > if

Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Suresh Siddha
On Fri, Aug 29, 2014 at 11:17 AM, Oleg Nesterov wrote: > ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. this patch I think needs more thought for sure. please see below. > > interrupted_kernel_fpu_idle() does: > > if (use_eager_fpu()) > return true; > >

Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Suresh Siddha
On Fri, Aug 29, 2014 at 11:17 AM, Oleg Nesterov o...@redhat.com wrote: ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. this patch I think needs more thought for sure. please see below. interrupted_kernel_fpu_idle() does: if (use_eager_fpu()) return true;

Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Oleg Nesterov
On 09/02, Suresh Siddha wrote: On Fri, Aug 29, 2014 at 11:17 AM, Oleg Nesterov o...@redhat.com wrote: ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. this patch I think needs more thought for sure. please see below. Of course. interrupted_kernel_fpu_idle() does: if

Re: [PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-09-02 Thread Oleg Nesterov
Sorry for noise, but just in case... On 09/02, Oleg Nesterov wrote: Do you think this can work? Of course, even if this can work, we should do this later. Let's start with the simple changes, then we will see if we can actually remove all other checks. Oleg. -- To unsubscribe from this list:

[PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-08-29 Thread Oleg Nesterov
ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. interrupted_kernel_fpu_idle() does: if (use_eager_fpu()) return true; return !__thread_has_fpu(current) && (read_cr0() & X86_CR0_TS); and it is absolutely not clear why these 2 cases differ so

[PATCH 4/4] x86, fpu: irq_fpu_usable: kill all checks except !in_kernel_fpu

2014-08-29 Thread Oleg Nesterov
ONCE AGAIN, THIS IS MORE THE QUESTION THAN THE PATCH. interrupted_kernel_fpu_idle() does: if (use_eager_fpu()) return true; return !__thread_has_fpu(current) (read_cr0() X86_CR0_TS); and it is absolutely not clear why these 2 cases differ so