| Message ID | 20181015202311.7209-1-aarcange@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/1] mm: thp: relocate flush_cache_range() in migrate_misplaced_transhuge_page() | expand |
On Mon, 15 Oct 2018 16:23:11 -0400 Andrea Arcangeli <aarcange@redhat.com> wrote: > There should be no cache left by the time we overwrite the old > transhuge pmd with the new one. It's already too late to flush through > the virtual address because we already copied the page data to the new > physical address. > > So flush the cache before the data copy. > > Also delete the "end" variable to shutoff a "unused variable" warning > on x86 where flush_cache_range() is a noop. What will be the runtime effects of this change?
On Mon, 15 Oct 2018 16:23:11 -0400 Andrea Arcangeli <aarcange@redhat.com> wrote: > There should be no cache left by the time we overwrite the old > transhuge pmd with the new one. It's already too late to flush through > the virtual address because we already copied the page data to the new > physical address. > > So flush the cache before the data copy. > > Also delete the "end" variable to shutoff a "unused variable" warning > on x86 where flush_cache_range() is a noop. migrate_misplaced_transhuge_page() has changed a bit. This is how I figure the patch should be. Please check: --- a/mm/migrate.c~mm-thp-relocate-flush_cache_range-in-migrate_misplaced_transhuge_page +++ a/mm/migrate.c @@ -1999,6 +1999,8 @@ int migrate_misplaced_transhuge_page(str /* anon mapping, we can simply copy page->mapping to the new page: */ new_page->mapping = page->mapping; new_page->index = page->index; + /* flush the cache before copying using the kernel virtual address */ + flush_cache_range(vma, mmun_start, mmun_end); migrate_page_copy(new_page, page); WARN_ON(PageLRU(new_page)); @@ -2037,7 +2039,6 @@ int migrate_misplaced_transhuge_page(str * The SetPageUptodate on the new page and page_add_new_anon_rmap * guarantee the copy is visible before the pagetable update. */ - flush_cache_range(vma, mmun_start, mmun_end); page_add_anon_rmap(new_page, vma, mmun_start, true); pmdp_huge_clear_flush_notify(vma, mmun_start, pmd); set_pmd_at(mm, mmun_start, pmd, entry);
On Mon, 15 Oct 2018 15:52:49 -0700 Andrew Morton <akpm@linux-foundation.org> wrote: > On Mon, 15 Oct 2018 16:23:11 -0400 Andrea Arcangeli <aarcange@redhat.com> wrote: > > > There should be no cache left by the time we overwrite the old > > transhuge pmd with the new one. It's already too late to flush through > > the virtual address because we already copied the page data to the new > > physical address. > > > > So flush the cache before the data copy. > > > > Also delete the "end" variable to shutoff a "unused variable" warning > > on x86 where flush_cache_range() is a noop. > > migrate_misplaced_transhuge_page() has changed a bit. Is OK, I figured it out :)
diff --git a/mm/migrate.c b/mm/migrate.c index c9e9b7db8b6d..8afb41167641 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -2019,7 +2019,6 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, struct page *new_page = NULL; int page_lru = page_is_file_cache(page); unsigned long start = address & HPAGE_PMD_MASK; - unsigned long end = start + HPAGE_PMD_SIZE; /* * Rate-limit the amount of data that is being migrated to a node. @@ -2050,6 +2049,8 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, /* anon mapping, we can simply copy page->mapping to the new page: */ new_page->mapping = page->mapping; new_page->index = page->index; + /* flush the cache before copying using the kernel virtual address */ + flush_cache_range(vma, start, start + HPAGE_PMD_SIZE); migrate_page_copy(new_page, page); WARN_ON(PageLRU(new_page)); @@ -2087,7 +2088,6 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm, * new page and page_add_new_anon_rmap guarantee the copy is * visible before the pagetable update. */ - flush_cache_range(vma, start, end); page_add_anon_rmap(new_page, vma, start, true); /* * At this point the pmd is numa/protnone (i.e. non present)
There should be no cache left by the time we overwrite the old transhuge pmd with the new one. It's already too late to flush through the virtual address because we already copied the page data to the new physical address. So flush the cache before the data copy. Also delete the "end" variable to shutoff a "unused variable" warning on x86 where flush_cache_range() is a noop. Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> --- mm/migrate.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)