Re: [patch 1/2] x86, fpu: split FPU state from task struct - v5

[Suresh Siddha]

On Tue, Mar 11, 2008 at 11:11:26AM +0100, Ingo Molnar wrote:
> 
> * Alexey Dobriyan <adobriyan@xxxxxxxxx> wrote:
> 
> > On 3/11/08, Ingo Molnar <mingo@xxxxxxx> wrote:
> > >
> > > * Alexey Dobriyan <adobriyan@xxxxxxxxx> wrote:
> > >
> > > > > -	extern void __bad_fxsave_alignment(void);
> > > > > -
> > > > > -	if (offsetof(struct task_struct, thread.i387.fxsave) & 15)
> > > > > -	__bad_fxsave_alignment();
> > > >
> > > > I think removal of such checks needs giving necessary alignment to
> > > > cache. Previously it worked because of __aligned((16)) and
> > > > L1_CACHE_SHIFT combo.
> > >
> > > it's still checked, just elsewhere:
> > >
> > > > + WARN_ON((unsigned long)dst->thread.xstate & 15);
> > >
> > > i'd not expect us to ever align worse than 16 bytes in the future
> > 
> > But sysfs tells me alignment of "xstate" is 8 now. Object size is 512 
> > which is masking it.
> 
> ok - you are right, that needs to be solved then. (object size could 
> change anytime someone adds a few fields for debugging or whatever)

Which sysfs file are we referring here?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/