On Wed, Sep 26, 2018 at 01:53:35PM +0100, Will Deacon wrote:
> On Wed, Sep 26, 2018 at 01:36:28PM +0200, Peter Zijlstra wrote:
> > +#ifndef tlb_flush
> > +
> > +#if defined(tlb_start_vma) || defined(tlb_end_vma)
> > +#error Default tlb_flush() relies on default tlb_start_vma() and tlb_end_vma()
> > +#endif
> > +
> > +#define tlb_flush tlb_flush
>
> Do we need this #define?
Probably not, that was just my fingers doing the normal #ifndef #define
pattern. I'll take em out back for a 'hug' :-)
> > +static inline void tlb_start_vma(struct mmu_gather *tlb, struct vm_area_struct *vma)
> > +{
> > + if (tlb->fullmm)
> > + return;
> > +
> > + /*
> > + * flush_tlb_range() implementations that look at VM_HUGETLB (tile,
> > + * mips-4k) flush only large pages.
> > + *
> > + * flush_tlb_range() implementations that flush I-TLB also flush D-TLB
> > + * (tile, xtensa, arm), so it's ok to just add VM_EXEC to an existing
> > + * range.
> > + *
> > + * We rely on tlb_end_vma() to issue a flush, such that when we reset
> > + * these values the batch is empty.
> > + */
> > + tlb->vma_huge = !!(vma->vm_flags & VM_HUGETLB);
> > + tlb->vma_exec = !!(vma->vm_flags & VM_EXEC);
>
> Hmm, does this result in code generation for archs that don't care about the
> vm_flags?
Yes. It's not much code, but if you deeply care we could frob things to
get rid of it.
