cur_seg[CURSEG_COLD_DATA] will exceed end boundary of main area when:
device[1]: zone device size = [2 MB ~ 10MB]
So, if there are not enough seq zones for six cursegs, we should still
assign 0 to c.cur_seg[CURSEG_HOT_NODE] or reserve several conv zones for
cursegs.
On 1/29/2024 11:47 PM, Dae
On Tue, Jan 30, 2024 at 8:46 AM Zhiguo Niu wrote:
>
> Hi Jaegeuk,
> Thanks for your suggestion,the following is my comment.
>
> On Tue, Jan 30, 2024 at 3:53 AM Jaegeuk Kim wrote:
> >
> > On 01/23, Zhiguo Niu wrote:
> > > There are some cases of f2fs_is_valid_blkaddr not handled as
> > > ERROR_INV
Hi Jaegeuk,
Thanks for your suggestion,the following is my comment.
On Tue, Jan 30, 2024 at 3:53 AM Jaegeuk Kim wrote:
>
> On 01/23, Zhiguo Niu wrote:
> > There are some cases of f2fs_is_valid_blkaddr not handled as
> > ERROR_INVALID_BLKADDR,so unify the error handling about all of
> > f2fs_is_va
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Now that all callers pass in GFP_KERNEL to blkdev_zone_mgmt() and use
> memalloc_no{io,fs}_{save,restore}() to define the allocation scope, we can
> drop the gfp_mask parameter from blkdev_zone_mgmt() as well as
> blkdev_zone_reset_all() and blkdev_zone
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Guard the calls to blkdev_zone_mgmt() with a memalloc_nofs scope.
> This helps us getting rid of the GFP_NOFS argument to blkdev_zone_mgmt();
>
> Signed-off-by: Johannes Thumshirn
Reviewed-by: Damien Le Moal
--
Damien Le Moal
Western Digital Resea
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Add a memalloc_nofs scope around all calls to blkdev_zone_mgmt(). This
> allows us to further get rid of the GFP_NOFS argument for
> blkdev_zone_mgmt().
>
> Signed-off-by: Johannes Thumshirn
Looks OK to me.
Reviewed-by: Damien Le Moal
--
Damien
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Guard the calls to blkdev_zone_mgmt() with a memalloc_noio scope.
> This helps us getting rid of the GFP_NOIO argument to blkdev_zone_mgmt();
>
> Signed-off-by: Johannes Thumshirn
Reviewed-by: Damien Le Moal
> ---
> drivers/md/dm-zoned-metadata.c
On 1/29/24 16:52, Johannes Thumshirn wrote:
> Pass GFP_KERNEL instead of GFP_NOFS to the blkdev_zone_mgmt() call in
> zonefs_zone_mgmt().
>
> As as zonefs_zone_mgmt() and zonefs_inode_zone_mgmt() are never called
> from a place that can recurse back into the filesystem on memory reclaim,
> it is s
fscrypt now supports configuring dentry operations at dentry-creation
time through the preset sb->s_d_op, instead of at lookup time.
Enable this in ubifs, since the lookup-time mechanism is going away.
Signed-off-by: Gabriel Krisman Bertazi
---
fs/ubifs/dir.c | 1 -
fs/ubifs/super.c | 1 +
2 f
No filesystems depend on it anymore, and it is generally a bad idea.
Since all dentries should have the same set of dentry operations in
case-insensitive filesystems, it should be propagated through ->s_d_op.
Signed-off-by: Gabriel Krisman Bertazi
---
fs/libfs.c | 34
This was already the case for case-insensitive before commit
bb9cd9106b22 ("fscrypt: Have filesystems handle their d_ops"), but it
was changed to set at lookup-time to facilitate the integration with
fscrypt. But it's a problem because dentries that don't get created
through ->lookup() won't have
In preparation to drop the similar helper that sets d_op at lookup time,
add a version to set the right d_op filesystem-wide, through sb->s_d_op.
The operations structures are shared across filesystems supporting
fscrypt and/or casefolding, therefore we can keep it in common libfs
code.
Signed-off
This was already the case for case-insensitive before commit
bb9cd9106b22 ("fscrypt: Have filesystems handle their d_ops"), but it
was changed to set at lookup-time to facilitate the integration with
fscrypt. But it's a problem because dentries that don't get created
through ->lookup() won't have
In preparation to get case-insensitive dentry operations from
sb->s_d_op again, use the same structure for case-insensitive
filesystems with and without fscrypt.
This means that on a casefolded filesystem without fscrypt, we end up
having to call fscrypt_d_revalidate once per dentry, which does th
>From fscrypt perspective, once the key is available, the dentry will
remain valid until evicted for other reasons, since keyed dentries don't
require revalidation and, if the key is removed, the dentry is
forcefully evicted. Therefore, we don't need to keep revalidating them
repeatedly.
Obviousl
Now that we do more than just clear the DCACHE_NOKEY_NAME in
fscrypt_handle_d_move, skip it entirely for plaintext dentries, to avoid
extra costs.
Note that VFS will call this function for any dentry, whether the volume
has fscrypt on not. But, since we only care about DCACHE_NOKEY_NAME, we
can c
In preparation to dropping DCACHE_OP_REVALIDATE for dentries that
don't need it at lookup time, refactor the code to make unencrypted
denties also call fscrypt_prepare_dentry. This makes the
non-inline __fscrypt_prepare_lookup superfulous, so drop it.
Signed-off-by: Gabriel Krisman Bertazi
---
Both fscrypt_prepare_lookup_dentry_partial and
fscrypt_prepare_lookup_dentry will set DCACHE_NOKEY_NAME for dentries
when the key is not available. Extract out a helper to set this flag in
a single place, in preparation to also add the optimization that will
disable ->d_revalidate if possible.
Sig
Hi,
Sorry for the quick respin. The only difference from v4 is that we
change the way we check for relevant dentries during a d_move, as
suggested by Eric.
The v5 of this patchset addresses the issues Eric pointed out in the
previous version. The patch merging the fscrypt lookup helpers was
comp
overlayfs relies on the filesystem setting DCACHE_OP_HASH or
DCACHE_OP_COMPARE to reject mounting over case-insensitive directories.
Since commit bb9cd9106b22 ("fscrypt: Have filesystems handle their
d_ops"), we set ->d_op through a hook in ->d_lookup, which
means the root dentry won't have them,
Unencrypted and encrypted-dentries where the key is available don't need
to be revalidated with regards to fscrypt, since they don't go stale
from under VFS and the key cannot be removed for the encrypted case
without evicting the dentry. Mark them with d_set_always_valid, to
avoid unnecessary rev
Hello:
This series was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Fri, 26 Jan 2024 23:19:16 +0800 you wrote:
> This patch supports using printk_ratelimited() in f2fs_printk(), and
> wrap ratelimited f2fs_printk() into f2fs_{err,warn,info}_ratelimited(),
> then, use these new helps to c
Hello:
The following patches were marked "accepted", because they were applied to
jaegeuk/f2fs.git (dev):
Patch: [f2fs-dev] f2fs: remove unnecessary f2fs_put_page in f2fs_rename
Submitter: Jaegeuk Kim
Committer: Jaegeuk Kim
Patchwork: https://patchwork.kernel.org/project/f2fs/list/?series
Hello:
This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Fri, 12 Jan 2024 09:16:45 -0800 you wrote:
> [1] changed the below condition, which made f2fs_put_page() voided.
> This patch reapplies the AL's resolution in -next from [2].
>
> - if (S_ISDIR(old_inode->i_mode)) {
Hello:
This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Thu, 28 Dec 2023 20:25:07 -0700 you wrote:
> 'f2fs_is_checkpoint_ready()' checks free sections. If there is not
> enough free sections, most f2fs operations will return -ENOSPC when
> checkpoint is disabled.
>
> It would
Hello:
This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Mon, 29 Jan 2024 19:27:40 +0800 you wrote:
> It needs to check last zone_pending_bio and wait IO completion before
> traverse next fio in io->io_list, otherwise, bio in next zone may be
> submitted before all IO completio
Hello:
This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Tue, 16 Jan 2024 22:11:38 +0800 you wrote:
> BUG: kernel NULL pointer dereference, address: 0014
> RIP: 0010:f2fs_submit_page_write+0x6cf/0x780 [f2fs]
> Call Trace:
>
> ? show_regs+0x6e/0x80
> ? __die+0x29/0x
Hello:
This series was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim :
On Sat, 13 Jan 2024 03:41:27 +0800 you wrote:
> If data block in compressed cluster is not persisted with metadata
> during checkpoint, after SPOR, the data may be corrupted, let's
> guarantee to write compressed page by ch
On 01/26, Daniel Rosenberg wrote:
> Since we may not know the block size when initializing sparse files, we
> should assume that the sparse file's blocksize is correct.
>
> Signed-off-by: Daniel Rosenberg
> ---
> fsck/mount.c | 20 +---
> lib/libf2fs_io.c | 11 +++
>
On 01/23, Zhiguo Niu wrote:
> There are some cases of f2fs_is_valid_blkaddr not handled as
> ERROR_INVALID_BLKADDR,so unify the error handling about all of
> f2fs_is_valid_blkaddr.
>
> Signed-off-by: Zhiguo Niu
> ---
> changes of v2: improve patch according Chao's suggestions.
> changes of v3:
>
Eric Biggers writes:
> On Thu, Jan 25, 2024 at 05:20:56PM -0300, Gabriel Krisman Bertazi wrote:
>> Eric Biggers writes:
>>
>> > On Fri, Jan 19, 2024 at 03:47:36PM -0300, Gabriel Krisman Bertazi wrote:
>> >> /*
>> >> * When d_splice_alias() moves a directory's no-key alias to its
>> >> plainte
Reviewed-by: Daeho Jeong
Thanks,
On Mon, Jan 29, 2024 at 3:29 AM Chao Yu wrote:
>
> It needs to check last zone_pending_bio and wait IO completion before
> traverse next fio in io->io_list, otherwise, bio in next zone may be
> submitted before all IO completion in current zone.
>
> Fixes: e067d
On Sun, Jan 28, 2024 at 5:27 PM Chao Yu wrote:
>
> On 2024/1/27 2:17, Daeho Jeong wrote:
> > On Thu, Jan 25, 2024 at 5:27 PM Chao Yu wrote:
> >>
> >> On 2024/1/26 0:25, Daeho Jeong wrote:
> >>> On Wed, Jan 24, 2024 at 7:34 PM Chao Yu wrote:
>
> +Cc Yongpeng Yang
>
> Daeho,
>
It needs to check last zone_pending_bio and wait IO completion before
traverse next fio in io->io_list, otherwise, bio in next zone may be
submitted before all IO completion in current zone.
Fixes: e067dc3c6b9c ("f2fs: maintain six open zones for zoned devices")
Cc: Daeho Jeong
Signed-off-by: Cha
On 2024/1/29 18:27, Chao Yu wrote:
On 2024/1/16 22:11, Wenjie Qi wrote:
BUG: kernel NULL pointer dereference, address: 0014
RIP: 0010:f2fs_submit_page_write+0x6cf/0x780 [f2fs]
Call Trace:
? show_regs+0x6e/0x80
? __die+0x29/0x70
? page_fault_oops+0x154/0x4a0
? prb_read_valid+0x20/0x3
On 2024/1/16 22:11, Wenjie Qi wrote:
BUG: kernel NULL pointer dereference, address: 0014
RIP: 0010:f2fs_submit_page_write+0x6cf/0x780 [f2fs]
Call Trace:
? show_regs+0x6e/0x80
? __die+0x29/0x70
? page_fault_oops+0x154/0x4a0
? prb_read_valid+0x20/0x30
? __irq_work_queue_local+0x39/0xd0
There is a corner scenario on a small-capacity partition with 64MB size:
1. The main area has a total of 24 segments, and there are no free
segments left shown from the free_segmap bitmap and free_secmap in
free_segmap_info.
-
bitm
NULL_SEGNO should also be returned when the blk_addr value is
out-of-bound main area even __is_valid_data_blkaddr return true.
For example, a 64MB partition with total 24 main segments has no
any free segments left, then a new wrtie request use get_new_segment
may get a out-of-bound segno 24 if CO
When CONFIG_F2FS_CHECK_FS is not enabled, f2fs_bug_on just printing
warning, get_new_segment may get an out-of-bounds segment when there
is no free segments. Then a block is allocated from this invalid
segment, update_sit_entry will access the invalid bitmap address,
cause system panic. Just as bel
A panic issue happened in a reboot test in small capacity device
as following:
1.The device size is 64MB, and main area has 24 segments, and
CONFIG_F2FS_CHECK_FS is not enabled.
2.There is no any free segments left shown in free_segmap_info,
then another write request cause get_new_segment get a ou
Now that all callers pass in GFP_KERNEL to blkdev_zone_mgmt() and use
memalloc_no{io,fs}_{save,restore}() to define the allocation scope, we can
drop the gfp_mask parameter from blkdev_zone_mgmt() as well as
blkdev_zone_reset_all() and blkdev_zone_reset_all_emulated().
Signed-off-by: Johannes Thum
Guard the calls to blkdev_zone_mgmt() with a memalloc_noio scope.
This helps us getting rid of the GFP_NOIO argument to blkdev_zone_mgmt();
Signed-off-by: Johannes Thumshirn
---
drivers/md/dm-zoned-metadata.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/md/dm-z
Add a memalloc_nofs scope around all calls to blkdev_zone_mgmt(). This
allows us to further get rid of the GFP_NOFS argument for
blkdev_zone_mgmt().
Signed-off-by: Johannes Thumshirn
---
fs/btrfs/zoned.c | 33 ++---
1 file changed, 26 insertions(+), 7 deletions(-)
di
Pass GFP_KERNEL instead of GFP_NOFS to the blkdev_zone_mgmt() call in
zonefs_zone_mgmt().
As as zonefs_zone_mgmt() and zonefs_inode_zone_mgmt() are never called
from a place that can recurse back into the filesystem on memory reclaim,
it is save to call blkdev_zone_mgmt() with GFP_KERNEL.
Link: h
Fueled by the LSFMM discussion on removing GFP_NOFS initiated by Willy,
I've looked into the sole GFP_NOFS allocation in zonefs. As it turned out,
it is only done for zone management commands and can be removed.
After digging into more callers of blkdev_zone_mgmt() I came to the
conclusion that th
Guard the calls to blkdev_zone_mgmt() with a memalloc_nofs scope.
This helps us getting rid of the GFP_NOFS argument to blkdev_zone_mgmt();
Signed-off-by: Johannes Thumshirn
---
fs/f2fs/segment.c | 15 ---
1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/fs/f2fs/segment
46 matches
Mail list logo
