[lkml]   [2014]   [Feb]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCHv3 0/2] mm: map few pages around fault address if they are in page cache
On Thu, Feb 27, 2014 at 6:10 PM, Kirill A. Shutemov
<> wrote:
> Also Matthew noticed that some drivers do ugly hacks like fault in whole
> VMA on first page fault. IIUC, it's for performance reasons. See
> psbfb_vm_fault() or ttm_bo_vm_fault().

I guarantee it's not for performance reasons, it's probably some other breakage.

And if anything really does want to populate things fully, doing so at
fault time is wrong anyway, you should just use MAP_POPULATE at mmap

So I'll believe the shm/tmpfs issue, and that's "mm internal" anyway.
But random crappy drivers? No way in hell.


 \ /
  Last update: 2014-03-01 00:41    [W:0.064 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site