Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Mel Gorman
On (01/07/08 10:29), David Gibson didst pronounce: > On Mon, Jun 30, 2008 at 10:29:01PM +0100, Mel Gorman wrote: > > On (30/06/08 14:23), Nishanth Aravamudan didst pronounce: > > > On 30.06.2008 [16:12:20 -0500], Jon Tollefson wrote: > > > > Mel Gorman wrote: > > > > > HUGETLB_MORECORE currently ex

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread David Gibson
On Mon, Jun 30, 2008 at 10:29:01PM +0100, Mel Gorman wrote: > On (30/06/08 14:23), Nishanth Aravamudan didst pronounce: > > On 30.06.2008 [16:12:20 -0500], Jon Tollefson wrote: > > > Mel Gorman wrote: > > > > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > > > > large pages

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread David Gibson
On Mon, Jun 30, 2008 at 02:21:38PM -0700, Nishanth Aravamudan wrote: > On 30.06.2008 [16:53:31 +0100], Mel Gorman wrote: > > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > > large pages instead of small pages. However, not all applications use glibc > > malloc() nor is it

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Mel Gorman
On (30/06/08 14:23), Nishanth Aravamudan didst pronounce: > On 30.06.2008 [16:12:20 -0500], Jon Tollefson wrote: > > Mel Gorman wrote: > > > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > > > large pages instead of small pages. However, not all applications use > > > glib

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Nishanth Aravamudan
On 30.06.2008 [16:12:20 -0500], Jon Tollefson wrote: > Mel Gorman wrote: > > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > > large pages instead of small pages. However, not all applications use glibc > > malloc() nor is it always desirable to back malloc() with huge page

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Nishanth Aravamudan
On 30.06.2008 [16:53:31 +0100], Mel Gorman wrote: > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > large pages instead of small pages. However, not all applications use glibc > malloc() nor is it always desirable to back malloc() with huge pages. There > exists a requireme

Re: [Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Jon Tollefson
Mel Gorman wrote: > HUGETLB_MORECORE currently exists to allow glibc to back malloc() with > large pages instead of small pages. However, not all applications use glibc > malloc() nor is it always desirable to back malloc() with huge pages. There > exists a requirement that a hugepage-aware applica

[Libhugetlbfs-devel] [PATCH] Provide a direct allocator API for huge pages V3

2008-06-30 Thread Mel Gorman
HUGETLB_MORECORE currently exists to allow glibc to back malloc() with large pages instead of small pages. However, not all applications use glibc malloc() nor is it always desirable to back malloc() with huge pages. There exists a requirement that a hugepage-aware application be able to allocate h