Messages in this thread | | | From | Paul Mackerras <> | Date | Tue, 21 May 2002 22:18:42 +1000 (EST) | Subject | Re: Make 2.5.17 TLB even more friendlier |
| |
David S. Miller writes:
> The next part is allowing for a "full_mm_flush" state such that > tlb_flush() can make decisions based upon that. Then we move > the tlb_{start,end}_vma() invocations one level up. So for > the exit_mmap case it is: > > tlb_gather_mmu(mm, 1); > > flush_cache_mm(mm); > > for each vma { > ... > unmap_page_range(...); > } > > clear_page_tables(); > tlb_finish_mmu();
I still need tlb_end_vma in this case - or at least I need a hook that gets called after all the tlb_remove_tlb_entry calls are done but before clear_page_tables is called. If I had that hook (called in both the exit_mmap and unmap cases) then I would not need the tlb_start/end_vma hooks.
Regards, Paul. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |