On Mon, Jul 01, 2024 at 12:16:52PM -0700, Vishal Moola wrote: > On Mon, Jul 01, 2024 at 05:49:34PM +0200, Dan Carpenter wrote: > > tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master > > head: df9574a57d02b265322e77fb8628d4d33641dda9 > > commit: 1cb6271d927cdb448a6a2794291c5405f1effa76 [7944/8232] hugetlb: force allocating surplus hugepages on mempolicy allowed nodes > > config: i386-randconfig-141-20240627 (https://download.01.org/0day-ci/archive/20240627/202406270727.F4yNrBsh-lkp@xxxxxxxxx/config) > > compiler: gcc-9 (Ubuntu 9.5.0-4ubuntu2) 9.5.0 > > > > If you fix the issue in a separate patch/commit (i.e. not just a new version of > > the same patch/commit), kindly add following tags > > | Reported-by: kernel test robot <lkp@xxxxxxxxx> > > | Reported-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx> > > | Closes: https://lore.kernel.org/r/202406270727.F4yNrBsh-lkp@xxxxxxxxx/ > > > > smatch warnings: > > mm/hugetlb.c:2677 gather_surplus_pages() error: uninitialized symbol 'folio'. > > > > vim +/folio +2677 mm/hugetlb.c > > > > 0a4f3d1bb91cac Liu Xiang 2020-12-14 2644 static int gather_surplus_pages(struct hstate *h, long delta) > > 1b2a1e7bb9ce99 Jules Irenge 2020-04-06 2645 __must_hold(&hugetlb_lock) > > e4e574b767ba63 Adam Litke 2007-10-16 2646 { > > 3466534131b28e Miaohe Lin 2022-09-01 2647 LIST_HEAD(surplus_list); > > 454a00c40a21c5 Matthew Wilcox (Oracle 2023-08-16 2648) struct folio *folio, *tmp; > > 0a4f3d1bb91cac Liu Xiang 2020-12-14 2649 int ret; > > 0a4f3d1bb91cac Liu Xiang 2020-12-14 2650 long i; > > 0a4f3d1bb91cac Liu Xiang 2020-12-14 2651 long needed, allocated; > > 28073b02bfaaed Hillf Danton 2012-03-21 2652 bool alloc_ok = true; > > 1cb6271d927cdb Aristeu Rozanski 2024-06-21 2653 int node; > > 1cb6271d927cdb Aristeu Rozanski 2024-06-21 2654 nodemask_t *mbind_nodemask = policy_mbind_nodemask(htlb_alloc_mask(h)); > > e4e574b767ba63 Adam Litke 2007-10-16 2655 > > 9487ca60fd7fa2 Mike Kravetz 2021-05-04 2656 lockdep_assert_held(&hugetlb_lock); > > a5516438959d90 Andi Kleen 2008-07-23 2657 needed = (h->resv_huge_pages + delta) - h->free_huge_pages; > > ac09b3a15154af Adam Litke 2008-03-04 2658 if (needed <= 0) { > > a5516438959d90 Andi Kleen 2008-07-23 2659 h->resv_huge_pages += delta; > > e4e574b767ba63 Adam Litke 2007-10-16 2660 return 0; > > ac09b3a15154af Adam Litke 2008-03-04 2661 } > > e4e574b767ba63 Adam Litke 2007-10-16 2662 > > e4e574b767ba63 Adam Litke 2007-10-16 2663 allocated = 0; > > e4e574b767ba63 Adam Litke 2007-10-16 2664 > > e4e574b767ba63 Adam Litke 2007-10-16 2665 ret = -ENOMEM; > > e4e574b767ba63 Adam Litke 2007-10-16 2666 retry: > > db71ef79b59bb2 Mike Kravetz 2021-05-04 2667 spin_unlock_irq(&hugetlb_lock); > > e4e574b767ba63 Adam Litke 2007-10-16 2668 for (i = 0; i < needed; i++) { > > 1cb6271d927cdb Aristeu Rozanski 2024-06-21 2669 for_each_node_mask(node, cpuset_current_mems_allowed) { > > Smatch might be concerned about us skipping over this > for_each_node_mask()? It appears to be possible if we have 1 non-empty > Numa node. > Oh, yes, good point. That too. Thanks for looking at this. regards, dan carpenter
