git/linux-cryptodev-2.6
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git synced 2026-04-18 03:23:53 -04:00
Code Issues Packages Projects Releases Wiki Activity

filemap: optimize folio refount update in filemap_map_pages

Browse Source 
There are two meaningless folio refcount update for order0 folio in
filemap_map_pages().  First, filemap_map_order0_folio() adds folio
refcount after the folio is mapped to pte.  And then, filemap_map_pages()
drops a refcount grabbed by next_uptodate_folio().  We could remain the
refcount unchanged in this case.

As Matthew metenioned in [1], it is safe to call folio_unlock() before
calling folio_put() here, because the folio is in page cache with refcount
held, and truncation will wait for the unlock.

Optimize filemap_map_folio_range() with the same method too.

With this patch, we can get 8% performance gain for lmbench testcase
'lat_pagefault -P 1 file' in order0 folio case, the size of file is 512M.


Link: https://lkml.kernel.org/r/20250904132737.1250368-1-tujinjiang@huawei.com
Link: https://lore.kernel.org/all/aKcU-fzxeW3xT5Wv@casper.infradead.org/ [1]
Signed-off-by: Jinjiang Tu <tujinjiang@huawei.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Jinjiang Tu
2025-09-04 21:27:37 +08:00
committed by Andrew Morton
parent 24a3c7af3b
commit 0faa77afe7
1 changed files with 14 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
mm/filemap.c
View File

@@ -3665,6 +3665,7 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf,
unsigned long addr, unsigned int nr_pages,
unsigned long *rss, unsigned short *mmap_miss)
{
unsigned int ref_from_caller = 1;
vm_fault_t ret = 0;
struct page *page = folio_page(folio, start);
unsigned int count = 0;
@@ -3698,7 +3699,8 @@ skip:
if (count) {
set_pte_range(vmf, folio, page, count, addr);
*rss += count;
folio_ref_add(folio, count);
folio_ref_add(folio, count - ref_from_caller);
ref_from_caller = 0;
if (in_range(vmf->address, addr, count * PAGE_SIZE))
ret = VM_FAULT_NOPAGE;
}
@@ -3713,12 +3715,16 @@ skip:
if (count) {
set_pte_range(vmf, folio, page, count, addr);
*rss += count;
folio_ref_add(folio, count);
folio_ref_add(folio, count - ref_from_caller);
ref_from_caller = 0;
if (in_range(vmf->address, addr, count * PAGE_SIZE))
ret = VM_FAULT_NOPAGE;
}
vmf->pte = old_ptep;
if (ref_from_caller)
/* Locked folios cannot get truncated. */
folio_ref_dec(folio);
return ret;
}
@@ -3731,7 +3737,7 @@ static vm_fault_t filemap_map_order0_folio(struct vm_fault *vmf,
struct page *page = &folio->page;
if (PageHWPoison(page))
return ret;
goto out;
/* See comment of filemap_map_folio_range() */
if (!folio_test_workingset(folio))
@@ -3743,15 +3749,18 @@ static vm_fault_t filemap_map_order0_folio(struct vm_fault *vmf,
* the fault-around logic.
*/
if (!pte_none(ptep_get(vmf->pte)))
return ret;
goto out;
if (vmf->address == addr)
ret = VM_FAULT_NOPAGE;
set_pte_range(vmf, folio, page, 1, addr);
(*rss)++;
folio_ref_inc(folio);
return ret;
out:
/* Locked folios cannot get truncated. */
folio_ref_dec(folio);
return ret;
}
@@ -3811,7 +3820,6 @@ vm_fault_t filemap_map_pages(struct vm_fault *vmf,
nr_pages, &rss, &mmap_miss);
folio_unlock(folio);
folio_put(folio);
} while ((folio = next_uptodate_folio(&xas, mapping, end_pgoff)) != NULL);
add_mm_counter(vma->vm_mm, folio_type, rss);
pte_unmap_unlock(vmf->pte, vmf->ptl);