Messages in this thread | | | From | "Aneesh Kumar K.V" <> | Subject | Re: [PATCHv2] Add per-process flag to control thp | Date | Sun, 11 Aug 2013 22:26:21 +0530 |
| |
Alex Thorlton <athorlton@sgi.com> writes:
> This patch implements functionality to allow processes to disable the use of > transparent hugepages through the prctl syscall. > > We've determined that some jobs perform significantly better with thp disabled, > and we needed a way to control thp on a per-process basis, without relying on > madvise. >
Is that because of hugepage alloc on fault ? I understand, that could be a real issue when applications touch one byte per hugepage. In that case we would endup allocating more memory than the application footprint. I have observed that system almost hang and start hitting OOM when run with THP enabled. Now we could possibly handle that by not allowing alloc on fault and always depend on coallpse hugepage to instantiate a hugepage ?
-aneesh
| |