Re: [PATCH V3 2/2] mm: THP page cache support for ppc64
From: Kirill A. Shutemov <hidden>
Date: 2016-11-14 02:11:54
Also in:
linux-mm, lkml
On Sun, Nov 13, 2016 at 08:30:25PM +0530, Aneesh Kumar K.V wrote:
Add arch specific callback in the generic THP page cache code that will deposit and withdarw preallocated page table. Archs like ppc64 use this preallocated table to store the hash pte slot information. Testing: kernel build of the patch series on tmpfs mounted with option huge=always The related thp stat: thp_fault_alloc 72939 thp_fault_fallback 60547 thp_collapse_alloc 603 thp_collapse_alloc_failed 0 thp_file_alloc 253763 thp_file_mapped 4251 thp_split_page 51518 thp_split_page_failed 1 thp_deferred_split_page 73566 thp_split_pmd 665 thp_zero_page_alloc 3 thp_zero_page_alloc_failed 0 Signed-off-by: Aneesh Kumar K.V <redacted>
One nit-pick below, but otherwise Acked-by: Kirill A. Shutemov <redacted>
quoted hunk ↗ jump to hunk
@@ -2975,6 +3004,13 @@ static int do_set_pmd(struct fault_env *fe, struct page *page) ret = 0; count_vm_event(THP_FILE_MAPPED); out: + /* + * If we are going to fallback to pte mapping, do a + * withdraw with pmd lock held. + */ + if (arch_needs_pgtable_deposit() && (ret == VM_FAULT_FALLBACK))
Parenthesis are redundant around ret check.
+ fe->prealloc_pte = pgtable_trans_huge_withdraw(vma->vm_mm, + fe->pmd); spin_unlock(fe->ptl); return ret; }
-- Kirill A. Shutemov