Le lundi 21 novembre 2011 à 17:52 +0100, Eric Dumazet a écrit : > Le lundi 21 novembre 2011 à 17:10 +0100, Markus Trippelsdorf a écrit : > > > Sure. This one happend with CONFIG_DEBUG_PAGEALLOC=y: > > > > [drm] Initialized radeon 2.11.0 20080528 for 0000:01:05.0 on minor 0 > > loop: module loaded > > ahci 0000:00:11.0: version 3.0 > > ahci 0000:00:11.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22 > > ahci 0000:00:11.0: AHCI 0001.0100 32 slots 6 ports 3 Gbps 0x3f impl SATA mode > > ahci 0000:00:11.0: flags: 64bit ncq sntf ilck pm led clo pmp pio slum part ccc > > scsi0 : ahci > > scsi1 : ahci > > ============================================================================= > > BUG task_struct: Poison overwritten > > ----------------------------------------------------------------------------- > > Unfortunately thats the same problem, not catched by DEBUG_PAGEALLOC > because freed page is immediately reused. > > We should keep pages in free list longer, to have a bigger window. > > Hmm... > > Please try following patch : > > diff --git a/mm/page_alloc.c b/mm/page_alloc.c > index 9dd443d..b8932a6 100644 > --- a/mm/page_alloc.c > +++ b/mm/page_alloc.c > @@ -1196,7 +1196,7 @@ void free_hot_cold_page(struct page *page, int cold) > } > > pcp = &this_cpu_ptr(zone->pageset)->pcp; > - if (cold) > + if (IS_ENABLED(CONFIG_DEBUG_PAGEALLOC) || cold) > list_add_tail(&page->lru, &pcp->lists[migratetype]); > else > list_add(&page->lru, &pcp->lists[migratetype]); > Also add "slub_max_order=0" to your boot command, since it will make the pool larger... -- 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>