Messages in this thread | | | Date | Fri, 13 Mar 2015 16:51:21 +0100 | From | Borislav Petkov <> | Subject | Re: [PATCH 3/4] x86/fpu: use restore_init_xstate() instead of math_state_restore() on kthread exec |
| |
On Fri, Mar 13, 2015 at 03:45:14PM +0100, Oleg Nesterov wrote: > Because in math_state_restore() case kernel_fpu_begin()->__save_init_fpu() > will overwrite (corrupt) the same fpu->state buffer we need to restore. > Without kernel_fpu_disable().
Yes.
> restore_init_xstate() obviously differs because it reads init_xstate_buf, > we do not care at all if kernel_fpu_begin() in between overwrites ->state.
Ah yes, so we're on the thread exit path and we restore init xstate. Sure, of course, that makes sense.
> And note! this is the yet another proof that init_fpu()->fpu_finit() is > pointless. This (and almost all) users need fpu_alloc() only.
Right, applying.
Thanks for explaining!
:-)
-- Regards/Gruss, Boris.
ECO tip #101: Trim your mails when you reply. --
| |