Linux Btrfs filesystem development
 help / color / mirror / Atom feed
* btrfs-6.9 regression tests fail
@ 2024-06-09 23:53 Bruce Dubbs
  2024-06-10  0:53 ` Qu Wenruo
  0 siblings, 1 reply; 5+ messages in thread
From: Bruce Dubbs @ 2024-06-09 23:53 UTC (permalink / raw)
  To: linux-btrfs

The convert and misc tests fail pretty early for me and I can't figure out why.
The other btrfs tests complete normally.

The significant output for convert-tests-results.txt is:

====== RUN CHECK mount -t btrfs -o loop /build/btrfs/btrfs-progs-v6.9/tests/test.img 
/build/btrfs/btrfs-progs-v6.9/tests/mnt
mount: /build/btrfs/btrfs-progs-v6.9/tests/mnt: fsconfig system call failed: No such 
file or directory.
        dmesg(1) may have more information after failed mount system call.
failed: mount -t btrfs -o loop /build/btrfs/btrfs-progs-v6.9/tests/test.img 
/build/btrfs/btrfs-progs-v6.9/tests/mnt
test failed for case 003-ext4-basic

dmesg gives me:

[ 3807.421836] loop0: detected capacity change from 0 to 4194304
[ 3807.421933] BTRFS: device fsid 4f1a8440-1a8f-45d1-9789-72080ddd9917 devid 1 
transid 7 /dev/loop0 (7:0) scanned by mount (3326)
[ 3807.423458] BTRFS info (device loop0): first mount of filesystem 
4f1a8440-1a8f-45d1-9789-72080ddd9917
[ 3807.423469] BTRFS info (device loop0): using crc32c (crc32c-generic) checksum 
algorithm
[ 3807.423477] BTRFS info (device loop0): using free-space-tree
[ 3807.423911] BTRFS warning (device loop0): failed to read root (objectid=12): -2
[ 3807.424407] BTRFS error (device loop0): open_ctree failed

My related kernel options are:

CONFIG_BTRFS_FS=m
CONFIG_BTRFS_FS_POSIX_ACL=y
CONFIG_BTRFS_FS_RUN_SANITY_TESTS=y
CONFIG_BTRFS_DEBUG=y
CONFIG_BTRFS_ASSERT=y
CONFIG_BTRFS_FS_REF_VERIFY=y

and the btrfs kernel module is loaded.

The kernel is version 6.9.3 on an AMD Ryzen 9 3900X

I can format a btrfs partition and mount it.

The system is linuxfromscratch so it may be something I've overlooked, but earlier 
versions worked OK.

Any suggestions will be appreciated.

   -- Bruce Dubbs
      linuxfromscratch.org

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: btrfs-6.9 regression tests fail
  2024-06-09 23:53 btrfs-6.9 regression tests fail Bruce Dubbs
@ 2024-06-10  0:53 ` Qu Wenruo
  2024-06-10  7:12   ` Bruce Dubbs
  0 siblings, 1 reply; 5+ messages in thread
From: Qu Wenruo @ 2024-06-10  0:53 UTC (permalink / raw)
  To: Bruce Dubbs, linux-btrfs



在 2024/6/10 09:23, Bruce Dubbs 写道:
> The convert and misc tests fail pretty early for me and I can't figure
> out why.
> The other btrfs tests complete normally.
>
> The significant output for convert-tests-results.txt is:
>
> ====== RUN CHECK mount -t btrfs -o loop
> /build/btrfs/btrfs-progs-v6.9/tests/test.img
> /build/btrfs/btrfs-progs-v6.9/tests/mnt
> mount: /build/btrfs/btrfs-progs-v6.9/tests/mnt: fsconfig system call
> failed: No such file or directory.
>         dmesg(1) may have more information after failed mount system call.
> failed: mount -t btrfs -o loop
> /build/btrfs/btrfs-progs-v6.9/tests/test.img
> /build/btrfs/btrfs-progs-v6.9/tests/mnt
> test failed for case 003-ext4-basic
>
> dmesg gives me:
>
> [ 3807.421836] loop0: detected capacity change from 0 to 4194304
> [ 3807.421933] BTRFS: device fsid 4f1a8440-1a8f-45d1-9789-72080ddd9917
> devid 1 transid 7 /dev/loop0 (7:0) scanned by mount (3326)
> [ 3807.423458] BTRFS info (device loop0): first mount of filesystem
> 4f1a8440-1a8f-45d1-9789-72080ddd9917
> [ 3807.423469] BTRFS info (device loop0): using crc32c (crc32c-generic)
> checksum algorithm
> [ 3807.423477] BTRFS info (device loop0): using free-space-tree
> [ 3807.423911] BTRFS warning (device loop0): failed to read root
> (objectid=12): -2

That's objectid is for RST, which shouldn't even be enabled unless you
have enabled experimental features for btrfs-progs.

Mind to dump the superblock of that test image?
($BTRFS_PROGS_SOURCE/tests/test.img)

Thanks,
Qu

> [ 3807.424407] BTRFS error (device loop0): open_ctree failed
>
> My related kernel options are:
>
> CONFIG_BTRFS_FS=m
> CONFIG_BTRFS_FS_POSIX_ACL=y
> CONFIG_BTRFS_FS_RUN_SANITY_TESTS=y
> CONFIG_BTRFS_DEBUG=y
> CONFIG_BTRFS_ASSERT=y
> CONFIG_BTRFS_FS_REF_VERIFY=y
>
> and the btrfs kernel module is loaded.
>
> The kernel is version 6.9.3 on an AMD Ryzen 9 3900X
>
> I can format a btrfs partition and mount it.
>
> The system is linuxfromscratch so it may be something I've overlooked,
> but earlier versions worked OK.
>
> Any suggestions will be appreciated.
>
>    -- Bruce Dubbs
>       linuxfromscratch.org
>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: btrfs-6.9 regression tests fail
  2024-06-10  0:53 ` Qu Wenruo
@ 2024-06-10  7:12   ` Bruce Dubbs
  2024-06-10  7:49     ` Qu Wenruo
  0 siblings, 1 reply; 5+ messages in thread
From: Bruce Dubbs @ 2024-06-10  7:12 UTC (permalink / raw)
  To: Qu Wenruo, linux-btrfs

On 6/9/24 19:53, Qu Wenruo wrote:
> 
> 
> 在 2024/6/10 09:23, Bruce Dubbs 写道:
>> The convert and misc tests fail pretty early for me and I can't figure
>> out why.
>> The other btrfs tests complete normally.
>>
>> The significant output for convert-tests-results.txt is:
>>
>> ====== RUN CHECK mount -t btrfs -o loop
>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>> mount: /build/btrfs/btrfs-progs-v6.9/tests/mnt: fsconfig system call
>> failed: No such file or directory.
>>         dmesg(1) may have more information after failed mount system call.
>> failed: mount -t btrfs -o loop
>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>> test failed for case 003-ext4-basic
>>
>> dmesg gives me:
>>
>> [ 3807.421836] loop0: detected capacity change from 0 to 4194304
>> [ 3807.421933] BTRFS: device fsid 4f1a8440-1a8f-45d1-9789-72080ddd9917
>> devid 1 transid 7 /dev/loop0 (7:0) scanned by mount (3326)
>> [ 3807.423458] BTRFS info (device loop0): first mount of filesystem
>> 4f1a8440-1a8f-45d1-9789-72080ddd9917
>> [ 3807.423469] BTRFS info (device loop0): using crc32c (crc32c-generic)
>> checksum algorithm
>> [ 3807.423477] BTRFS info (device loop0): using free-space-tree
>> [ 3807.423911] BTRFS warning (device loop0): failed to read root
>> (objectid=12): -2
> 
> That's objectid is for RST, which shouldn't even be enabled unless you
> have enabled experimental features for btrfs-progs.

I have not.  I used:

./configure --prefix=/usr           \
             --disable-static        \
             --disable-documentation &&
make
make fssum

cd tests
./convert-tests.sh
...

> Mind to dump the superblock of that test image?
> ($BTRFS_PROGS_SOURCE/tests/test.img)

I haven't done that before, but this is what I have:

$ btrfs inspect-internal dump-super test.img
superblock: bytenr=65536, device=test.img
---------------------------------------------------------
csum_type               0 (crc32c)
csum_size               4
csum                    0x3da36db9 [match]
bytenr                  65536
flags                   0x1
                         ( WRITTEN )
magic                   _BHRfS_M [match]
fsid                    c90faada-13cd-4e77-bd4a-a32018ad3b93
metadata_uuid           c90faada-13cd-4e77-bd4a-a32018ad3b93
label
generation              7
root                    67416064
sys_array_size          97
chunk_root_generation   2
root_level              0
chunk_root              35852288
chunk_root_level        0
log_root                0
log_root_transid (deprecated)   0
log_root_level          0
total_bytes             2147483648
bytes_used              177639424
sectorsize              4096
nodesize                4096
leafsize (deprecated)   4096
stripesize              4096
root_dir                6
num_devices             1
compat_flags            0x0
compat_ro_flags         0x3
                         ( FREE_SPACE_TREE |
                           FREE_SPACE_TREE_VALID )
incompat_flags          0x4341
                         ( MIXED_BACKREF |
                           EXTENDED_IREF |
                           SKINNY_METADATA |
                           NO_HOLES |
                           RAID_STRIPE_TREE )
cache_generation        0
uuid_tree_generation    0
dev_item.uuid           3cb58644-ed73-46ab-be03-68ebacc4a459
dev_item.fsid           c90faada-13cd-4e77-bd4a-a32018ad3b93 [match]
dev_item.type           0
dev_item.total_bytes    2147483648
dev_item.bytes_used     441581568
dev_item.io_align       4096
dev_item.io_width       4096
dev_item.sector_size    4096
dev_item.devid          1
dev_item.dev_group      0
dev_item.seek_speed     0
dev_item.bandwidth      0
dev_item.generation     0

AFAICT test.img is generated by the tests.  I really do not know what any of that 
output means.

Other info: I am building with gcc-14.1.0.

   -- Bruce



^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: btrfs-6.9 regression tests fail
  2024-06-10  7:12   ` Bruce Dubbs
@ 2024-06-10  7:49     ` Qu Wenruo
  2024-06-10 19:45       ` Bruce Dubbs
  0 siblings, 1 reply; 5+ messages in thread
From: Qu Wenruo @ 2024-06-10  7:49 UTC (permalink / raw)
  To: Bruce Dubbs, linux-btrfs, David Sterba



在 2024/6/10 16:42, Bruce Dubbs 写道:
> On 6/9/24 19:53, Qu Wenruo wrote:
>>
>>
>> 在 2024/6/10 09:23, Bruce Dubbs 写道:
>>> The convert and misc tests fail pretty early for me and I can't figure
>>> out why.
>>> The other btrfs tests complete normally.
>>>
>>> The significant output for convert-tests-results.txt is:
>>>
>>> ====== RUN CHECK mount -t btrfs -o loop
>>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>>> mount: /build/btrfs/btrfs-progs-v6.9/tests/mnt: fsconfig system call
>>> failed: No such file or directory.
>>>         dmesg(1) may have more information after failed mount system 
>>> call.
>>> failed: mount -t btrfs -o loop
>>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>>> test failed for case 003-ext4-basic
>>>
>>> dmesg gives me:
>>>
>>> [ 3807.421836] loop0: detected capacity change from 0 to 4194304
>>> [ 3807.421933] BTRFS: device fsid 4f1a8440-1a8f-45d1-9789-72080ddd9917
>>> devid 1 transid 7 /dev/loop0 (7:0) scanned by mount (3326)
>>> [ 3807.423458] BTRFS info (device loop0): first mount of filesystem
>>> 4f1a8440-1a8f-45d1-9789-72080ddd9917
>>> [ 3807.423469] BTRFS info (device loop0): using crc32c (crc32c-generic)
>>> checksum algorithm
>>> [ 3807.423477] BTRFS info (device loop0): using free-space-tree
>>> [ 3807.423911] BTRFS warning (device loop0): failed to read root
>>> (objectid=12): -2
>>
>> That's objectid is for RST, which shouldn't even be enabled unless you
>> have enabled experimental features for btrfs-progs.
> 
> I have not.  I used:
> 
> ./configure --prefix=/usr           \
>              --disable-static        \
>              --disable-documentation &&
> make
> make fssum
> 
> cd tests
> ./convert-tests.sh
> ...
> 
>> Mind to dump the superblock of that test image?
>> ($BTRFS_PROGS_SOURCE/tests/test.img)
> 
> I haven't done that before, but this is what I have:
> 
> $ btrfs inspect-internal dump-super test.img
> superblock: bytenr=65536, device=test.img
> ---------------------------------------------------------
> csum_type               0 (crc32c)
> csum_size               4
> csum                    0x3da36db9 [match]
> bytenr                  65536
> flags                   0x1
>                          ( WRITTEN )
> magic                   _BHRfS_M [match]
> fsid                    c90faada-13cd-4e77-bd4a-a32018ad3b93
> metadata_uuid           c90faada-13cd-4e77-bd4a-a32018ad3b93
> label
> generation              7
> root                    67416064
> sys_array_size          97
> chunk_root_generation   2
> root_level              0
> chunk_root              35852288
> chunk_root_level        0
> log_root                0
> log_root_transid (deprecated)   0
> log_root_level          0
> total_bytes             2147483648
> bytes_used              177639424
> sectorsize              4096
> nodesize                4096
> leafsize (deprecated)   4096
> stripesize              4096
> root_dir                6
> num_devices             1
> compat_flags            0x0
> compat_ro_flags         0x3
>                          ( FREE_SPACE_TREE |
>                            FREE_SPACE_TREE_VALID )
> incompat_flags          0x4341
>                          ( MIXED_BACKREF |
>                            EXTENDED_IREF |
>                            SKINNY_METADATA |
>                            NO_HOLES |
>                            RAID_STRIPE_TREE )

OK, it's a fix that didn't get merged by David:

[PATCH 1/3] Revert "btrfs-progs: convert: add raid-stripe-tree to 
allowed features"

To David: any reason why that revert is not merged?

Thanks,
Qu
T
> cache_generation        0
> uuid_tree_generation    0
> dev_item.uuid           3cb58644-ed73-46ab-be03-68ebacc4a459
> dev_item.fsid           c90faada-13cd-4e77-bd4a-a32018ad3b93 [match]
> dev_item.type           0
> dev_item.total_bytes    2147483648
> dev_item.bytes_used     441581568
> dev_item.io_align       4096
> dev_item.io_width       4096
> dev_item.sector_size    4096
> dev_item.devid          1
> dev_item.dev_group      0
> dev_item.seek_speed     0
> dev_item.bandwidth      0
> dev_item.generation     0
> 
> AFAICT test.img is generated by the tests.  I really do not know what 
> any of that output means.
> 
> Other info: I am building with gcc-14.1.0.
> 
>    -- Bruce
> 
> 
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: btrfs-6.9 regression tests fail
  2024-06-10  7:49     ` Qu Wenruo
@ 2024-06-10 19:45       ` Bruce Dubbs
  0 siblings, 0 replies; 5+ messages in thread
From: Bruce Dubbs @ 2024-06-10 19:45 UTC (permalink / raw)
  To: Qu Wenruo, linux-btrfs, David Sterba

On 6/10/24 02:49, Qu Wenruo wrote:
> 
> 
> 在 2024/6/10 16:42, Bruce Dubbs 写道:
>> On 6/9/24 19:53, Qu Wenruo wrote:
>>>
>>>
>>> 在 2024/6/10 09:23, Bruce Dubbs 写道:
>>>> The convert and misc tests fail pretty early for me and I can't figure
>>>> out why.
>>>> The other btrfs tests complete normally.
>>>>
>>>> The significant output for convert-tests-results.txt is:
>>>>
>>>> ====== RUN CHECK mount -t btrfs -o loop
>>>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>>>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>>>> mount: /build/btrfs/btrfs-progs-v6.9/tests/mnt: fsconfig system call
>>>> failed: No such file or directory.
>>>>         dmesg(1) may have more information after failed mount system call.
>>>> failed: mount -t btrfs -o loop
>>>> /build/btrfs/btrfs-progs-v6.9/tests/test.img
>>>> /build/btrfs/btrfs-progs-v6.9/tests/mnt
>>>> test failed for case 003-ext4-basic
>>>>
>>>> dmesg gives me:
>>>>
>>>> [ 3807.421836] loop0: detected capacity change from 0 to 4194304
>>>> [ 3807.421933] BTRFS: device fsid 4f1a8440-1a8f-45d1-9789-72080ddd9917
>>>> devid 1 transid 7 /dev/loop0 (7:0) scanned by mount (3326)
>>>> [ 3807.423458] BTRFS info (device loop0): first mount of filesystem
>>>> 4f1a8440-1a8f-45d1-9789-72080ddd9917
>>>> [ 3807.423469] BTRFS info (device loop0): using crc32c (crc32c-generic)
>>>> checksum algorithm
>>>> [ 3807.423477] BTRFS info (device loop0): using free-space-tree
>>>> [ 3807.423911] BTRFS warning (device loop0): failed to read root
>>>> (objectid=12): -2
>>>
>>> That's objectid is for RST, which shouldn't even be enabled unless you
>>> have enabled experimental features for btrfs-progs.
>>
>> I have not.  I used:
>>
>> ./configure --prefix=/usr           \
>>              --disable-static        \
>>              --disable-documentation &&
>> make
>> make fssum
>>
>> cd tests
>> ./convert-tests.sh
>> ...
>>
>>> Mind to dump the superblock of that test image?
>>> ($BTRFS_PROGS_SOURCE/tests/test.img)
>>
>> I haven't done that before, but this is what I have:
>>
>> $ btrfs inspect-internal dump-super test.img
>> superblock: bytenr=65536, device=test.img
>> ---------------------------------------------------------
>> csum_type               0 (crc32c)
>> csum_size               4
>> csum                    0x3da36db9 [match]
>> bytenr                  65536
>> flags                   0x1
>>                          ( WRITTEN )
>> magic                   _BHRfS_M [match]
>> fsid                    c90faada-13cd-4e77-bd4a-a32018ad3b93
>> metadata_uuid           c90faada-13cd-4e77-bd4a-a32018ad3b93
>> label
>> generation              7
>> root                    67416064
>> sys_array_size          97
>> chunk_root_generation   2
>> root_level              0
>> chunk_root              35852288
>> chunk_root_level        0
>> log_root                0
>> log_root_transid (deprecated)   0
>> log_root_level          0
>> total_bytes             2147483648
>> bytes_used              177639424
>> sectorsize              4096
>> nodesize                4096
>> leafsize (deprecated)   4096
>> stripesize              4096
>> root_dir                6
>> num_devices             1
>> compat_flags            0x0
>> compat_ro_flags         0x3
>>                          ( FREE_SPACE_TREE |
>>                            FREE_SPACE_TREE_VALID )
>> incompat_flags          0x4341
>>                          ( MIXED_BACKREF |
>>                            EXTENDED_IREF |
>>                            SKINNY_METADATA |
>>                            NO_HOLES |
>>                            RAID_STRIPE_TREE )
> 
> OK, it's a fix that didn't get merged by David:
> 
> [PATCH 1/3] Revert "btrfs-progs: convert: add raid-stripe-tree to allowed features"
> 
> To David: any reason why that revert is not merged?

Just a note on kernel configuration.

On the original system above with the convert test failures the kernel had:

CONFIG_BTRFS_FS=m
CONFIG_BTRFS_FS_POSIX_ACL=y
CONFIG_BTRFS_FS_RUN_SANITY_TESTS=y
CONFIG_BTRFS_DEBUG=y
CONFIG_BTRFS_ASSERT=y
CONFIG_BTRFS_FS_REF_VERIFY=y

But on another system with configuration:

CONFIG_BTRFS_FS=m
CONFIG_BTRFS_FS_POSIX_ACL=y
# CONFIG_BTRFS_FS_RUN_SANITY_TESTS is not set
# CONFIG_BTRFS_DEBUG is not set
# CONFIG_BTRFS_ASSERT is not set
# CONFIG_BTRFS_FS_REF_VERIFY is not set

All tests convert tests pass.

   -- Bruce



^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2024-06-10 19:45 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-06-09 23:53 btrfs-6.9 regression tests fail Bruce Dubbs
2024-06-10  0:53 ` Qu Wenruo
2024-06-10  7:12   ` Bruce Dubbs
2024-06-10  7:49     ` Qu Wenruo
2024-06-10 19:45       ` Bruce Dubbs

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox