Re: [patch 3/3] x86/process: Optimize TIF_NOTSC switch

From	Andy Lutomirski <luto@amacapital.net>
Newsgroups	linux.kernel
Subject	Re: [patch 3/3] x86/process: Optimize TIF_NOTSC switch
Date	2016-12-16 19:40 +0100
Message-ID	<sP5Ae-7fU-31@gated-at.bofh.it> (permalink)
References	<sOHoe-87a-13@gated-at.bofh.it> <sOHoe-87a-15@gated-at.bofh.it> <sOIaC-cX-45@gated-at.bofh.it> <sOWwV-19z-17@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Fri, Dec 16, 2016 at 12:50 AM, Thomas Gleixner <tglx@linutronix.de> wrote:
> On Thu, 15 Dec 2016, Andy Lutomirski wrote:
>> On Thu, Dec 15, 2016 at 8:44 AM, Thomas Gleixner <tglx@linutronix.de> wrote:
>> > +static inline void cr4_toggle_bits(unsigned long mask)
>> > +{
>> > +       unsigned long cr4;
>> > +
>> > +       cr4 = this_cpu_read(cpu_tlbstate.cr4);
>> > +       cr4 ^= mask;
>> > +       this_cpu_write(cpu_tlbstate.cr4, cr4);
>> > +       __write_cr4(cr4);
>> > +}
>>
>> This scares me for the same reason as BTF, although this should at
>> least be less fragile.  But how about:
>
> If that is fragile then all cr4 manipulation code is fragile because it
> relies on cpu_tlbstate.cr4. The TIF flag and that per cpu thing are kept in
> sync.

True.

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

[patch 3/3] x86/process: Optimize TIF_NOTSC switch Thomas Gleixner <tglx@linutronix.de> - 2016-12-15 17:50 +0100
  Re: [patch 3/3] x86/process: Optimize TIF_NOTSC switch Andy Lutomirski <luto@amacapital.net> - 2016-12-15 18:40 +0100
    Re: [patch 3/3] x86/process: Optimize TIF_NOTSC switch Thomas Gleixner <tglx@linutronix.de> - 2016-12-16 10:00 +0100
      Re: [patch 3/3] x86/process: Optimize TIF_NOTSC switch Andy Lutomirski <luto@amacapital.net> - 2016-12-16 19:40 +0100

csiph-web