On Fri, Jul 29, 2011 at 05:38:47PM +0900, Minchan Kim wrote: > On Wed, Jul 27, 2011 at 04:45:23PM -0700, Andrew Morton wrote: > > Using an additional 44 bytes of stack on that path is also > > significant(ly bad). But we need to fix that problem anyway. One way > > we could improve things in mm/vmscan.c is to move the blk_plug into > > scan_control then get the scan_control off the stack in some manner. > > That's easy for kswapd: allocate one scan_control per kswapd at > > startup. Doing it for direct-reclaim would be a bit trickier... > > Stack diet in direct reclaim... > Of course, it's a matter as I pointed out in this patch > but frankly speaking, it's very annoying to consider stack usage > whenever we add something in direct reclaim path. It's a fact of life that direct reclaim has to live with - memory allocation can occur with a lot of stack already consumed. If you don't want to care about stack usage, then lets increase the default stack size to 16k for x86-64..... > I think better solution is to avoid write in direct reclaim like the approach of Mel. Yeah, and we should probably stop swapping in the direct reclaim path, too, because I've seen the stack usage from memory allocation to swap IO issue exceed 4k on x86-64.... Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>