Re: [PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
On 2018/12/22 19:08, Martin Blumenstingl wrote: > On Sat, Dec 22, 2018 at 8:00 AM kbuild test robot wrote: >> >> Hi Martin, >> >> Thank you for the patch! Perhaps something to improve: >> >> [auto build test WARNING on f2fs/dev-test] >> [also build test WARNING on v4.20-rc7 next-20181221] >> [if your patch is applied to the wrong git tree, please drop us a note to >> help improve the system] >> >> url: >> https://github.com/0day-ci/linux/commits/Martin-Blumenstingl/f2fs-fix-sanity_check_raw_super-on-big-endian-machines/20181222-142616 >> base: https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git >> dev-test >> config: i386-randconfig-x006-201850 (attached as .config) >> compiler: gcc-7 (Debian 7.3.0-1) 7.3.0 >> reproduce: >> # save the attached .config to linux build tree >> make ARCH=i386 >> >> All warnings (new ones prefixed by >>): >> >>fs/f2fs/super.c: In function 'sanity_check_raw_super': fs/f2fs/super.c:2498:46: warning: format '%u' expects argument of type 'unsigned int', but argument 5 has type 'long long unsigned int' [-Wformat=] >>"Wrong segment_count / block_count (%u > %u)", >> ~^ >> %llu > I sent a v2 with this fixed and Chao Yu's reviewed by included Sorry, I didn't catch previous print issue on v1, anyway, I'm okay with v2. :) Thanks, > > . >
Re: [PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
On Sat, Dec 22, 2018 at 8:00 AM kbuild test robot wrote: > > Hi Martin, > > Thank you for the patch! Perhaps something to improve: > > [auto build test WARNING on f2fs/dev-test] > [also build test WARNING on v4.20-rc7 next-20181221] > [if your patch is applied to the wrong git tree, please drop us a note to > help improve the system] > > url: > https://github.com/0day-ci/linux/commits/Martin-Blumenstingl/f2fs-fix-sanity_check_raw_super-on-big-endian-machines/20181222-142616 > base: https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git > dev-test > config: i386-randconfig-x006-201850 (attached as .config) > compiler: gcc-7 (Debian 7.3.0-1) 7.3.0 > reproduce: > # save the attached .config to linux build tree > make ARCH=i386 > > All warnings (new ones prefixed by >>): > >fs/f2fs/super.c: In function 'sanity_check_raw_super': > >> fs/f2fs/super.c:2498:46: warning: format '%u' expects argument of type > >> 'unsigned int', but argument 5 has type 'long long unsigned int' > >> [-Wformat=] >"Wrong segment_count / block_count (%u > %u)", > ~^ > %llu I sent a v2 with this fixed and Chao Yu's reviewed by included
Re: [PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
Hi Martin,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on f2fs/dev-test]
[also build test WARNING on v4.20-rc7 next-20181221]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Martin-Blumenstingl/f2fs-fix-sanity_check_raw_super-on-big-endian-machines/20181222-142616
base: https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git
dev-test
config: i386-randconfig-a3-12211242 (attached as .config)
compiler: gcc-4.9 (Debian 4.9.4-2) 4.9.4
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All warnings (new ones prefixed by >>):
fs/f2fs/super.c: In function 'sanity_check_raw_super':
>> fs/f2fs/super.c:2499:4: warning: format '%u' expects argument of type
>> 'unsigned int', but argument 5 has type '__le64' [-Wformat=]
segment_count, le64_to_cpu(raw_super->block_count));
^
vim +2499 fs/f2fs/super.c
2382
2383 static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
2384 struct buffer_head *bh)
2385 {
2386 block_t segment_count, segs_per_sec, secs_per_zone;
2387 block_t total_sections, blocks_per_seg;
2388 struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
2389 (bh->b_data +
F2FS_SUPER_OFFSET);
2390 struct super_block *sb = sbi->sb;
2391 unsigned int blocksize;
2392 size_t crc_offset = 0;
2393 __u32 crc = 0;
2394
2395 /* Check checksum_offset and crc in superblock */
2396 if (__F2FS_HAS_FEATURE(raw_super, F2FS_FEATURE_SB_CHKSUM)) {
2397 crc_offset = le32_to_cpu(raw_super->checksum_offset);
2398 if (crc_offset !=
2399 offsetof(struct f2fs_super_block, crc)) {
2400 f2fs_msg(sb, KERN_INFO,
2401 "Invalid SB checksum offset: %zu",
2402 crc_offset);
2403 return 1;
2404 }
2405 crc = le32_to_cpu(raw_super->crc);
2406 if (!f2fs_crc_valid(sbi, crc, raw_super, crc_offset)) {
2407 f2fs_msg(sb, KERN_INFO,
2408 "Invalid SB checksum value: %u", crc);
2409 return 1;
2410 }
2411 }
2412
2413 if (F2FS_SUPER_MAGIC != le32_to_cpu(raw_super->magic)) {
2414 f2fs_msg(sb, KERN_INFO,
2415 "Magic Mismatch, valid(0x%x) - read(0x%x)",
2416 F2FS_SUPER_MAGIC,
le32_to_cpu(raw_super->magic));
2417 return 1;
2418 }
2419
2420 /* Currently, support only 4KB page cache size */
2421 if (F2FS_BLKSIZE != PAGE_SIZE) {
2422 f2fs_msg(sb, KERN_INFO,
2423 "Invalid page_cache_size (%lu), supports only
4KB\n",
2424 PAGE_SIZE);
2425 return 1;
2426 }
2427
2428 /* Currently, support only 4KB block size */
2429 blocksize = 1 << le32_to_cpu(raw_super->log_blocksize);
2430 if (blocksize != F2FS_BLKSIZE) {
2431 f2fs_msg(sb, KERN_INFO,
2432 "Invalid blocksize (%u), supports only 4KB\n",
2433 blocksize);
2434 return 1;
2435 }
2436
2437 /* check log blocks per segment */
2438 if (le32_to_cpu(raw_super->log_blocks_per_seg) != 9) {
2439 f2fs_msg(sb, KERN_INFO,
2440 "Invalid log blocks per segment (%u)\n",
2441 le32_to_cpu(raw_super->log_blocks_per_seg));
2442 return 1;
2443 }
2444
2445 /* Currently, support 512/1024/2048/4096 bytes sector size */
2446 if (le32_to_cpu(raw_super->log_sectorsize) >
2447 F2FS_MAX_LOG_SECTOR_SIZE ||
2448 le32_to_cpu(raw_super->log_sectorsize) <
2449 F2FS_MIN_LOG_SECTOR_SIZE) {
2450 f2fs_msg(sb, KERN_INFO, "Invalid log sectorsize (%u)",
2451 le32_to_cpu(raw_super->log_sectorsize));
2452 return 1;
2453 }
2454 if (le32_to_cpu(raw_super->log_sectors_per_block) +
2455 le32_to_cpu(raw_super->log_sectorsize) !=
2456 F2FS_MAX_LOG_SECTOR_SIZE) {
2457 f2fs_msg(sb, KERN_INFO,
2458 "Invalid log sectors per block(%u) log
sectorsize(%u)",
2459
Re: [PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
Hi Martin,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on f2fs/dev-test]
[also build test WARNING on v4.20-rc7 next-20181221]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system]
url:
https://github.com/0day-ci/linux/commits/Martin-Blumenstingl/f2fs-fix-sanity_check_raw_super-on-big-endian-machines/20181222-142616
base: https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git
dev-test
config: i386-randconfig-x006-201850 (attached as .config)
compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All warnings (new ones prefixed by >>):
fs/f2fs/super.c: In function 'sanity_check_raw_super':
>> fs/f2fs/super.c:2498:46: warning: format '%u' expects argument of type
>> 'unsigned int', but argument 5 has type 'long long unsigned int' [-Wformat=]
"Wrong segment_count / block_count (%u > %u)",
~^
%llu
vim +2498 fs/f2fs/super.c
9a59b62fd Chao Yu 2015-12-15 2382
df728b0f6 Jaegeuk Kim 2016-03-23 2383 static int
sanity_check_raw_super(struct f2fs_sb_info *sbi,
fd694733d Jaegeuk Kim 2016-03-20 2384
struct buffer_head *bh)
aff063e26 Jaegeuk Kim 2012-11-02 2385 {
0cfe75c5b Jaegeuk Kim 2018-04-27 2386 block_t segment_count,
segs_per_sec, secs_per_zone;
0cfe75c5b Jaegeuk Kim 2018-04-27 2387 block_t total_sections,
blocks_per_seg;
fd694733d Jaegeuk Kim 2016-03-20 2388 struct f2fs_super_block
*raw_super = (struct f2fs_super_block *)
fd694733d Jaegeuk Kim 2016-03-20 2389
(bh->b_data + F2FS_SUPER_OFFSET);
df728b0f6 Jaegeuk Kim 2016-03-23 2390 struct super_block *sb
= sbi->sb;
aff063e26 Jaegeuk Kim 2012-11-02 2391 unsigned int blocksize;
d440c52d3 Junling Zheng 2018-09-28 2392 size_t crc_offset = 0;
d440c52d3 Junling Zheng 2018-09-28 2393 __u32 crc = 0;
d440c52d3 Junling Zheng 2018-09-28 2394
d440c52d3 Junling Zheng 2018-09-28 2395 /* Check
checksum_offset and crc in superblock */
7beb01f74 Chao Yu 2018-10-24 2396 if
(__F2FS_HAS_FEATURE(raw_super, F2FS_FEATURE_SB_CHKSUM)) {
d440c52d3 Junling Zheng 2018-09-28 2397 crc_offset =
le32_to_cpu(raw_super->checksum_offset);
d440c52d3 Junling Zheng 2018-09-28 2398 if (crc_offset
!=
d440c52d3 Junling Zheng 2018-09-28 2399
offsetof(struct f2fs_super_block, crc)) {
d440c52d3 Junling Zheng 2018-09-28 2400
f2fs_msg(sb, KERN_INFO,
d440c52d3 Junling Zheng 2018-09-28 2401
"Invalid SB checksum offset: %zu",
d440c52d3 Junling Zheng 2018-09-28 2402
crc_offset);
d440c52d3 Junling Zheng 2018-09-28 2403 return
1;
d440c52d3 Junling Zheng 2018-09-28 2404 }
d440c52d3 Junling Zheng 2018-09-28 2405 crc =
le32_to_cpu(raw_super->crc);
d440c52d3 Junling Zheng 2018-09-28 2406 if
(!f2fs_crc_valid(sbi, crc, raw_super, crc_offset)) {
d440c52d3 Junling Zheng 2018-09-28 2407
f2fs_msg(sb, KERN_INFO,
d440c52d3 Junling Zheng 2018-09-28 2408
"Invalid SB checksum value: %u", crc);
d440c52d3 Junling Zheng 2018-09-28 2409 return
1;
d440c52d3 Junling Zheng 2018-09-28 2410 }
d440c52d3 Junling Zheng 2018-09-28 2411 }
aff063e26 Jaegeuk Kim 2012-11-02 2412
a07ef7843 Namjae Jeon 2012-12-30 2413 if (F2FS_SUPER_MAGIC !=
le32_to_cpu(raw_super->magic)) {
a07ef7843 Namjae Jeon 2012-12-30 2414 f2fs_msg(sb,
KERN_INFO,
a07ef7843 Namjae Jeon 2012-12-30 2415 "Magic
Mismatch, valid(0x%x) - read(0x%x)",
a07ef7843 Namjae Jeon 2012-12-30 2416
F2FS_SUPER_MAGIC, le32_to_cpu(raw_super->magic));
aff063e26 Jaegeuk Kim 2012-11-02 2417 return 1;
a07ef7843 Namjae Jeon 2012-12-30 2418 }
aff063e26 Jaegeuk Kim 2012-11-02 2419
5c9b46929 majianpeng 2013-02-01 2420 /* Currently, support
only 4KB page cache size */
09cbfeaf1 Kirill A. Shutemov 2016-04-01 2421 if (F2FS_BLKSIZE !=
PAGE_SIZE) {
5c9b46929 majianpeng 2013-02-01 2422 f2fs_msg(sb,
KERN_INFO,
14d7e9de0 majianpeng 2013-02-01 2423
"Invalid page_cache_size (%lu), supports only 4KB\n",
09cbfeaf1 Kirill
Re: [PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
On 2018/12/22 3:28, Martin Blumenstingl wrote:
> Treat "block_count" from struct f2fs_super_block as 64-bit little endian
> value in sanity_check_raw_super() because struct f2fs_super_block
> declares "block_count" as "__le64".
>
> This fixes a bug where the superblock validation fails on big endian
> devices with the following error:
> F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
> F2FS-fs (sda1): Can't find valid F2FS filesystem in 1th superblock
> F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
> F2FS-fs (sda1): Can't find valid F2FS filesystem in 2th superblock
> As result of this the partition cannot be mounted.
>
> With this patch applied the superblock validation works fine and the
> partition can be mounted again:
> F2FS-fs (sda1): Mounted with checkpoint version = 7c84
>
> My little endian x86-64 hardware was able to mount the partition without
> this fix.
> To confirm that mounting f2fs filesystems works on big endian machines
> again I tested this on a 32-bit MIPS big endian (lantiq) device.
>
> Fixes: 0cfe75c5b01199 ("f2fs: enhance sanity_check_raw_super() to avoid
> potential overflows")
> Cc: sta...@vger.kernel.org
> Signed-off-by: Martin Blumenstingl
Reviewed-by: Chao Yu
Thanks,
[PATCH 1/1] f2fs: fix validation of the block count in sanity_check_raw_super
Treat "block_count" from struct f2fs_super_block as 64-bit little endian
value in sanity_check_raw_super() because struct f2fs_super_block
declares "block_count" as "__le64".
This fixes a bug where the superblock validation fails on big endian
devices with the following error:
F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
F2FS-fs (sda1): Can't find valid F2FS filesystem in 1th superblock
F2FS-fs (sda1): Wrong segment_count / block_count (61439 > 0)
F2FS-fs (sda1): Can't find valid F2FS filesystem in 2th superblock
As result of this the partition cannot be mounted.
With this patch applied the superblock validation works fine and the
partition can be mounted again:
F2FS-fs (sda1): Mounted with checkpoint version = 7c84
My little endian x86-64 hardware was able to mount the partition without
this fix.
To confirm that mounting f2fs filesystems works on big endian machines
again I tested this on a 32-bit MIPS big endian (lantiq) device.
Fixes: 0cfe75c5b01199 ("f2fs: enhance sanity_check_raw_super() to avoid
potential overflows")
Cc: sta...@vger.kernel.org
Signed-off-by: Martin Blumenstingl
---
fs/f2fs/super.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index af58b2cc21b8..3a65d5af122a 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -2496,10 +2496,10 @@ static int sanity_check_raw_super(struct f2fs_sb_info
*sbi,
return 1;
}
- if (segment_count > (le32_to_cpu(raw_super->block_count) >> 9)) {
+ if (segment_count > (le64_to_cpu(raw_super->block_count) >> 9)) {
f2fs_msg(sb, KERN_INFO,
"Wrong segment_count / block_count (%u > %u)",
- segment_count, le32_to_cpu(raw_super->block_count));
+ segment_count, le64_to_cpu(raw_super->block_count));
return 1;
}
--
2.20.1
