f2fs-tools: fix mkfs.f2fs on big-endian systems

Fixes: FS#749

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
This commit is contained in:
Stijn Tintel 2017-08-24 10:04:15 +03:00
parent 252c8ddf14
commit cdb494fdc2
2 changed files with 67 additions and 1 deletions

View file

@ -9,7 +9,7 @@ include $(TOPDIR)/rules.mk
PKG_NAME:=f2fs-tools
PKG_VERSION:=1.8.0
PKG_RELEASE:=2
PKG_RELEASE:=3
PKG_LICENSE:=GPLv2

View file

@ -0,0 +1,66 @@
From 31873d5cdf8a97d5f7921451c54f6d293293c6cc Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Wed, 23 Aug 2017 13:33:00 -0700
Subject: [PATCH] mkfs.f2fs: fix missing cpu_to_le64 for checkpoint version
The error on mips was reported by Stijn as follow.
Commit 8399a29df92d6867d226df362edbf2e0efa527c1 in f2fs-tools
("mkfs.f2fs: give random checkpoint version") causes a bug when mounting
a filesystem created with it on a MIPS64 device running a 4.4(.83)
kernel. The following kernel warning appears several times per second,
for 30 seconds:
[ 23.837262] ------------[ cut here ]------------
[ 23.842039] WARNING: CPU: 0 PID: 935 at fs/f2fs/segment.c:718
update_sit_entry+0x1c0/0x2b0()
[ 23.850507] Modules linked in: pppoe ppp_async l2tp_ppp iptable_nat
[ 24.174064] Call Trace:
[ 24.176527] [<ffffffff81126e14>] show_stack+0x68/0xb4
[ 24.181595] [<ffffffff81321fc4>] dump_stack+0x8c/0xc4
[ 24.186660] [<ffffffff8113d004>] warn_slowpath_common+0xa0/0xd0
[ 24.192597] [<ffffffff812e0148>] update_sit_entry+0x1c0/0x2b0
[ 24.198353] [<ffffffff812e0a70>] refresh_sit_entry+0x70/0xf8
[ 24.204022] [<ffffffff812e251c>] allocate_data_block+0x1f0/0x310
[ 24.210038] [<ffffffff812e28d8>] do_write_page+0x29c/0x2bc
[ 24.215532] [<ffffffff812e2a88>] write_data_page+0xa0/0xd8
[ 24.221028] [<ffffffff812d844c>] do_write_data_page+0xe4/0x384
[ 24.226870] [<ffffffff812d88f4>] f2fs_write_data_page+0x208/0x464
[ 24.232972] [<ffffffff812d5184>] __f2fs_writepage+0x1c/0x74
[ 24.238553] [<ffffffff812d54dc>]
f2fs_write_cache_pages.constprop.7+0x250/0x394
[ 24.245869] [<ffffffff812d57f4>] f2fs_write_data_pages+0x130/0x1b0
[ 24.252066] [<ffffffff811a9f80>] __filemap_fdatawrite_range+0xa0/0xd4
[ 24.258515] [<ffffffff812d2338>] sync_dirty_dir_inodes+0x94/0xd8
[ 24.264530] [<ffffffff812d2484>] write_checkpoint+0x108/0xb9c
[ 24.270283] [<ffffffff812cc398>] f2fs_sync_fs+0x68/0xb0
[ 24.275526] [<ffffffff812c641c>] f2fs_sync_file+0x2e8/0x518
[ 24.281107] [<ffffffff81213ff4>] do_fsync+0x38/0x70
[ 24.285992] [<ffffffff812142e8>] SyS_fsync+0x14/0x20
[ 24.290972] [<ffffffff81103950>] syscall_common+0x34/0x58
[ 24.296372]
[ 24.298096] ---[ end trace fd3ac44449b218ab ]---
Fix: 8399a29df92d68 ("mkfs.f2fs: give random checkpoint version")
Reported-And-Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
mkfs/f2fs_format.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mkfs/f2fs_format.c b/mkfs/f2fs_format.c
index 92876b8..b379e80 100644
--- a/mkfs/f2fs_format.c
+++ b/mkfs/f2fs_format.c
@@ -546,7 +546,7 @@ static int f2fs_write_check_point_pack(void)
}
/* 1. cp page 1 of checkpoint pack 1 */
- cp->checkpoint_ver = rand() | 0x1;
+ cp->checkpoint_ver = cpu_to_le64(rand() | 0x1);
set_cp(cur_node_segno[0], c.cur_seg[CURSEG_HOT_NODE]);
set_cp(cur_node_segno[1], c.cur_seg[CURSEG_WARM_NODE]);
set_cp(cur_node_segno[2], c.cur_seg[CURSEG_COLD_NODE]);
--
2.13.5