On Tue, 2013-06-11 at 11:44 -0400, Sasha Levin wrote: > On 06/11/2013 11:23 AM, Christoph Lameter wrote: > > On Tue, 11 Jun 2013, Pekka Enberg wrote: > > > >> So you're OK with going forward with Sasha's patch? It's needed > >> because __GFP_NOWARN was specifically added there to fix this > >> issue earlier. > > > > Why dont we fix the call site to use vmalloc instead for larger allocs? > > > > We should probably be doing both. Allowing a pipe to store thousands of page refs seems quite useless and dangerous. Having to use vmalloc()/vfree() for every splice()/vmsplice() would be a performance loss anyway. (fs/splice.c splice_grow_spd() will also want to allocate big kmalloc() chunks) -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>