linux-kernel - [PATCH] erofs: support multi-page folios for erofs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20240723073024.875290-1-hsiangkao@linux.alibaba.com>
Date: Tue, 23 Jul 2024 15:30:24 +0800
From: Gao Xiang <hsiangkao@...ux.alibaba.com>
To: linux-erofs@...ts.ozlabs.org
Cc: LKML <linux-kernel@...r.kernel.org>,
	Gao Xiang <hsiangkao@...ux.alibaba.com>,
	Matthew Wilcox <willy@...radead.org>
Subject: [PATCH] erofs: support multi-page folios for erofs_bread()

If the requested page is part of the previous multi-page folio, there
is no need to call read_mapping_folio() again.

Also, get rid of the remaining one of page->index [1] in our codebase.

[1] https://lore.kernel.org/r/Zp8fgUSIBGQ1TN0D@casper.infradead.org
Cc: Matthew Wilcox <willy@...radead.org>
Signed-off-by: Gao Xiang <hsiangkao@...ux.alibaba.com>
---
(I plan to test now and take this for this cycle since this patch is
 not quite complex.)

 fs/erofs/data.c | 29 ++++++++++++-----------------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/fs/erofs/data.c b/fs/erofs/data.c
index 8be60797ea2f..4b8b2b004453 100644
--- a/fs/erofs/data.c
+++ b/fs/erofs/data.c
@@ -21,38 +21,33 @@ void erofs_put_metabuf(struct erofs_buf *buf)
 	if (!buf->page)
 		return;
 	erofs_unmap_metabuf(buf);
-	put_page(buf->page);
+	folio_put(page_folio(buf->page));
 	buf->page = NULL;
 }
 
-/*
- * Derive the block size from inode->i_blkbits to make compatible with
- * anonymous inode in fscache mode.
- */
 void *erofs_bread(struct erofs_buf *buf, erofs_off_t offset,
 		  enum erofs_kmap_type type)
 {
 	pgoff_t index = offset >> PAGE_SHIFT;
-	struct page *page = buf->page;
-	struct folio *folio;
-	unsigned int nofs_flag;
+	struct folio *folio = NULL;
 
-	if (!page || page->index != index) {
+	if (buf->page) {
+		folio = page_folio(buf->page);
+		if (folio_file_page(folio, index) != buf->page)
+			erofs_unmap_metabuf(buf);
+	}
+	if (!folio || !folio_contains(folio, index)) {
 		erofs_put_metabuf(buf);
 
-		nofs_flag = memalloc_nofs_save();
-		folio = read_cache_folio(buf->mapping, index, NULL, NULL);
-		memalloc_nofs_restore(nofs_flag);
+		folio = read_mapping_folio(buf->mapping, index, NULL);
 		if (IS_ERR(folio))
 			return folio;
-
-		/* should already be PageUptodate, no need to lock page */
-		page = folio_file_page(folio, index);
-		buf->page = page;
 	}
+	buf->page = folio_file_page(folio, index);
+
 	if (buf->kmap_type == EROFS_NO_KMAP) {
 		if (type == EROFS_KMAP)
-			buf->base = kmap_local_page(page);
+			buf->base = kmap_local_page(buf->page);
 		buf->kmap_type = type;
 	} else if (buf->kmap_type != type) {
 		DBG_BUGON(1);
-- 
2.43.5