Messages in this thread | | | Date | Tue, 09 Jun 2015 09:49:13 -0700 | From | Dave Hansen <> | Subject | Re: [PATCH 0/3] TLB flush multiple pages per IPI v5 |
| |
On 06/09/2015 08:34 AM, Dave Hansen wrote: > 2. We should measure flushing of ascending, adjacent virtual addresses > mapped with 4k pages since that is the normal case. Perhaps > vmalloc(16MB) or something.
Now that I think about this a bit more, we really have two different patterns here: 1. flush_tlb_mm_range() style, which is contiguous virtual address flushing from an unmap-style operation. 2. Mel's new try_to_unmap_one() style, which are going to be relatively random virtual addresses.
So, the "ascending adjacent" case is still interesting, but it wouldn't cover Mel's new case.
| |