linux-kernel - [RFC PATCH 2/2] erofs: directly traverse pages in z_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-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20210705183253.14833-2-xiang@kernel.org>
Date:   Tue,  6 Jul 2021 02:32:53 +0800
From:   Gao Xiang <xiang@...nel.org>
To:     linux-erofs@...ts.ozlabs.org
Cc:     Matthew Wilcox <willy@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>, Gao Xiang <xiang@...nel.org>
Subject: [RFC PATCH 2/2] erofs: directly traverse pages in z_erofs_readahead()

In that way, pages can be accessed directly with xarray.

Cc: Matthew Wilcox <willy@...radead.org>
Signed-off-by: Gao Xiang <xiang@...nel.org>
---
Just an open-coded PoC one, since that is what EROFS actually needs but
without the iteration API (see [PATCH 1/2] as well.)

 fs/erofs/zdata.c | 32 +++++++++++++-------------------
 1 file changed, 13 insertions(+), 19 deletions(-)

diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index 054b9839e9db..210b2965ecc4 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -1416,7 +1416,8 @@ static void z_erofs_readahead(struct readahead_control *rac)
 	bool sync = (sbi->ctx.readahead_sync_decompress &&
 			nr_pages <= sbi->ctx.max_sync_decompress_pages);
 	struct z_erofs_decompress_frontend f = DECOMPRESS_FRONTEND_INIT(inode);
-	struct page *page, *head = NULL;
+	struct page *page;
+	pgoff_t index;
 	LIST_HEAD(pagepool);
 
 	trace_erofs_readpages(inode, readahead_index(rac), nr_pages, false);
@@ -1434,26 +1435,19 @@ static void z_erofs_readahead(struct readahead_control *rac)
 	 *  2) submission chain can be then in the forward order since
 	 *     pclusters are all inserted at head.
 	 */
-	while ((page = readahead_page(rac))) {
-		prefetchw(&page->flags);
-
-		/*
-		 * A pure asynchronous readahead is indicated if
-		 * a PG_readahead marked page is hitted at first.
-		 * Let's also do asynchronous decompression for this case.
-		 */
-		sync &= !(PageReadahead(page) && !head);
-
-		set_page_private(page, (unsigned long)head);
-		head = page;
-	}
-
-	while (head) {
-		struct page *page = head;
+	index = rac->_index + rac->_nr_pages;
+	while (rac->_nr_pages) {
+		struct page *head;
 		int err;
 
-		/* traversal in reverse order */
-		head = (void *)page_private(page);
+		--rac->_nr_pages;
+		page = xa_load(&rac->mapping->i_pages, --index);
+		/* XXX: very incomplete thp support */
+		head = thp_head(page);
+		if (head != page) {
+			index -= page->index - head->index;
+			page = head;
+		}
 
 		err = z_erofs_do_read_page(&f, page, &pagepool);
 		if (err)
-- 
2.20.1