mm/gup: speed up check_and_migrate_cma_pages() on huge page
Subsystems (2)
PATCH v3 1/1 mm/gup: speed up check_and_migrate_cma_pages() on huge page
Both hugetlb and thp locate on the same migration type of pageblock, since they are allocated from a free_list[]. Based on this fact, it is enough to check on a single subpage to decide the migration type of the whole huge page. By this way, it saves (2M/4K - 1) times loop for pmd_huge on x86, similar on other archs.
Furthermore, when executing isolate_huge_page(), it avoid taking global hugetlb_lock many times, and meanless remove/add to the local link list cma_page_list.
---
v2 -> v3: fix page order to size convertion
mm/gup.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)
mm/gup.c +12 -7
@@ -1342,19 +1342,22 @@ static long check_and_migrate_cma_pages(struct task_struct *tsk,LIST_HEAD (cma_page_list); check_again: -for (i =0 ; i < nr_pages; i++) { +for (i =0 ; i < nr_pages;) { + +struct page *head =compound_head (pages[i]); +long step =1 ; + +if (PageCompound (head)) + step =1 <<compound_order (head) - (pages[i] - head);/* * If we get a page from the CMA zone, since we are going to * be pinning these entries, we might as well move them out * of the CMA zone if possible. */ -if (is_migrate_cma_page (pages[i])) { - -struct page *head =compound_head (pages[i]); - -if (PageHuge (head)) { +if (is_migrate_cma_page (head)) { +if (PageHuge (head))isolate_huge_page (head, &cma_page_list); - }else { +else {if (!PageLRU (head) && drain_allow) {lru_add_drain_all (); drain_allow = false;
@@ -1369,6 +1372,8 @@ static long check_and_migrate_cma_pages(struct task_struct *tsk, } } } + + i += step; }if (!list_empty (&cma_page_list)) {
Signature · 1 line
2.7.5
Quoted 34 lines · Pingfan Liu
Both hugetlb and thp locate on the same migration type of pageblock, since they are allocated from a free_list[]. Based on this fact, it is enough to check on a single subpage to decide the migration type of the whole huge page. By this way, it saves (2M/4K - 1) times loop for
pmd_hugeon x86, similar on other archs.Furthermore, when executing
isolate_huge_page(), it avoid taking globalhugetlb_lockmany times, and meanless remove/add to the local link listcma_page_list.Signed-off-by Pingfan Liu <[email protected]>Cc Andrew Morton <[email protected]>Cc Ira Weiny <[email protected]>Cc Mike Rapoport <[email protected]>Cc "Kirill A. Shutemov" <[email protected]>Cc Thomas Gleixner <[email protected]>Cc John Hubbard <[email protected]>Cc "Aneesh Kumar K.V" <[email protected]>Cc Christoph Hellwig <[email protected]>Cc Keith Busch <[email protected]>Cc Mike Kravetz <[email protected]>---
v2 -> v3: fix page order to size convertionmm/gup.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)mm/gup.c +0
-0@@ -1342,19 +1342,22 @@ static long check_and_migrate_cma_pages(struct task_struct *tsk,LIST_HEAD (cma_page_list);
mm/gup.c static long check_and_migrate_cma_pages(struct task_struct *tsk, · line 1351
Quoted 10 lines
mm/gup.c +7
-1check_again: -for (i =0 ; i < nr_pages; i++) { +for (i =0 ; i < nr_pages;) { + +struct page *head =compound_head (pages[i]); +long step =1 ; + +if (PageCompound (head)) + step =1 <<compound_order (head) - (pages[i] - head);
Check your precedence here.
step = (1 << compound_order(head)) - (pages[i] - head);
Ira
Quoted 30 lines
mm/gup.c +5
-6/* * If we get a page from the CMA zone, since we are going to * be pinning these entries, we might as well move them out * of the CMA zone if possible. */ -if (is_migrate_cma_page (pages[i])) { - -struct page *head =compound_head (pages[i]); - -if (PageHuge (head)) { +if (is_migrate_cma_page (head)) { +if (PageHuge (head))isolate_huge_page (head, &cma_page_list); - }else { +else {if (!PageLRU (head) && drain_allow) {lru_add_drain_all (); drain_allow = false;@@ -1369,6 +1372,8 @@ static long check_and_migrate_cma_pages(struct task_struct *tsk, } } } + + i += step; }if (!list_empty (&cma_page_list)) {Signature · 1 line
2.7.5
Quoted 49 lines · Ira Weiny
Quoted 34 lines · Pingfan Liu
Both hugetlb and thp locate on the same migration type of pageblock, since they are allocated from a free_list[]. Based on this fact, it is enough to check on a single subpage to decide the migration type of the whole huge page. By this way, it saves (2M/4K - 1) times loop for
pmd_hugeon x86, similar on other archs.Furthermore, when executing
isolate_huge_page(), it avoid taking globalhugetlb_lockmany times, and meanless remove/add to the local link listcma_page_list.Signed-off-by Pingfan Liu <[email protected]>Cc Andrew Morton <[email protected]>Cc Ira Weiny <[email protected]>Cc Mike Rapoport <[email protected]>Cc "Kirill A. Shutemov" <[email protected]>Cc Thomas Gleixner <[email protected]>Cc John Hubbard <[email protected]>Cc "Aneesh Kumar K.V" <[email protected]>Cc Christoph Hellwig <[email protected]>Cc Keith Busch <[email protected]>Cc Mike Kravetz <[email protected]>---
v2 -> v3: fix page order to size convertionmm/gup.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)mm/gup.c +0
-0@@ -1342,19 +1342,22 @@ static long check_and_migrate_cma_pages(struct task_struct *tsk,LIST_HEAD (cma_page_list);mm/gup.c static long check_and_migrate_cma_pages(struct task_struct *tsk, · line 1351
Quoted 10 lines
mm/gup.c +7
-1check_again: -for (i =0 ; i < nr_pages; i++) { +for (i =0 ; i < nr_pages;) { + +struct page *head =compound_head (pages[i]); +long step =1 ; + +if (PageCompound (head)) + step =1 <<compound_order (head) - (pages[i] - head);Check your precedence here.
Quoted 1 line
step = (1 << compound_order(head)) - (pages[i] - head);
OK.
Thanks,
Pingfan