Now EXT4_FC_TAG_ADD_RANGE uses ext4_extent to track the
newly-added blocks, but the limit on the max value of
ee_len field is ignored, and it can lead to BUG_ON as
shown below when running command "fallocate -l 128M file"
on a fast_commit-enabled fs:
kernel BUG at fs/ext4/ext4_extents.h:199!
invalid opcode: 0000 [#1] SMP PTI
CPU: 3 PID: 624 Comm: fallocate Not tainted 5.14.0-rc6+ #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
RIP: 0010:ext4_fc_write_inode_data+0x1f3/0x200
Call Trace:
? ext4_fc_write_inode+0xf2/0x150
ext4_fc_commit+0x93b/0xa00
? ext4_fallocate+0x1ad/0x10d0
ext4_sync_file+0x157/0x340
? ext4_sync_file+0x157/0x340
vfs_fsync_range+0x49/0x80
do_fsync+0x3d/0x70
__x64_sys_fsync+0x14/0x20
do_syscall_64+0x3b/0xc0
entry_SYSCALL_64_after_hwframe+0x44/0xae
Simply fixing it by limiting the number of blocks
in one EXT4_FC_TAG_ADD_RANGE TLV.
Fixes: aa75f4d3daae ("ext4: main fast-commit commit path")
Signed-off-by: Hou Tao <[email protected]>
---
fs/ext4/fast_commit.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
index e8195229c252..782d05a3f97a 100644
--- a/fs/ext4/fast_commit.c
+++ b/fs/ext4/fast_commit.c
@@ -893,6 +893,12 @@ static int ext4_fc_write_inode_data(struct inode *inode, u32 *crc)
sizeof(lrange), (u8 *)&lrange, crc))
return -ENOSPC;
} else {
+ unsigned int max = (map.m_flags & EXT4_MAP_UNWRITTEN) ?
+ EXT_UNWRITTEN_MAX_LEN : EXT_INIT_MAX_LEN;
+
+ /* Limit the number of blocks in one extent */
+ map.m_len = min(max, map.m_len);
+
fc_ext.fc_ino = cpu_to_le32(inode->i_ino);
ex = (struct ext4_extent *)&fc_ext.fc_ex;
ex->ee_block = cpu_to_le32(map.m_lblk);
--
2.29.2
ping ?
On 8/20/2021 12:45 PM, Hou Tao wrote:
> Now EXT4_FC_TAG_ADD_RANGE uses ext4_extent to track the
> newly-added blocks, but the limit on the max value of
> ee_len field is ignored, and it can lead to BUG_ON as
> shown below when running command "fallocate -l 128M file"
> on a fast_commit-enabled fs:
>
> kernel BUG at fs/ext4/ext4_extents.h:199!
> invalid opcode: 0000 [#1] SMP PTI
> CPU: 3 PID: 624 Comm: fallocate Not tainted 5.14.0-rc6+ #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
> RIP: 0010:ext4_fc_write_inode_data+0x1f3/0x200
> Call Trace:
> ? ext4_fc_write_inode+0xf2/0x150
> ext4_fc_commit+0x93b/0xa00
> ? ext4_fallocate+0x1ad/0x10d0
> ext4_sync_file+0x157/0x340
> ? ext4_sync_file+0x157/0x340
> vfs_fsync_range+0x49/0x80
> do_fsync+0x3d/0x70
> __x64_sys_fsync+0x14/0x20
> do_syscall_64+0x3b/0xc0
> entry_SYSCALL_64_after_hwframe+0x44/0xae
>
> Simply fixing it by limiting the number of blocks
> in one EXT4_FC_TAG_ADD_RANGE TLV.
>
> Fixes: aa75f4d3daae ("ext4: main fast-commit commit path")
> Signed-off-by: Hou Tao <[email protected]>
> ---
> fs/ext4/fast_commit.c | 6 ++++++
> 1 file changed, 6 insertions(+)
>
> diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
> index e8195229c252..782d05a3f97a 100644
> --- a/fs/ext4/fast_commit.c
> +++ b/fs/ext4/fast_commit.c
> @@ -893,6 +893,12 @@ static int ext4_fc_write_inode_data(struct inode *inode, u32 *crc)
> sizeof(lrange), (u8 *)&lrange, crc))
> return -ENOSPC;
> } else {
> + unsigned int max = (map.m_flags & EXT4_MAP_UNWRITTEN) ?
> + EXT_UNWRITTEN_MAX_LEN : EXT_INIT_MAX_LEN;
> +
> + /* Limit the number of blocks in one extent */
> + map.m_len = min(max, map.m_len);
> +
> fc_ext.fc_ino = cpu_to_le32(inode->i_ino);
> ex = (struct ext4_extent *)&fc_ext.fc_ex;
> ex->ee_block = cpu_to_le32(map.m_lblk);
ping ?
On 8/30/2021 3:52 PM, Hou Tao wrote:
> ping ?
>
> On 8/20/2021 12:45 PM, Hou Tao wrote:
>> Now EXT4_FC_TAG_ADD_RANGE uses ext4_extent to track the
>> newly-added blocks, but the limit on the max value of
>> ee_len field is ignored, and it can lead to BUG_ON as
>> shown below when running command "fallocate -l 128M file"
>> on a fast_commit-enabled fs:
>>
>> kernel BUG at fs/ext4/ext4_extents.h:199!
>> invalid opcode: 0000 [#1] SMP PTI
>> CPU: 3 PID: 624 Comm: fallocate Not tainted 5.14.0-rc6+ #1
>> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
>> RIP: 0010:ext4_fc_write_inode_data+0x1f3/0x200
>> Call Trace:
>> ? ext4_fc_write_inode+0xf2/0x150
>> ext4_fc_commit+0x93b/0xa00
>> ? ext4_fallocate+0x1ad/0x10d0
>> ext4_sync_file+0x157/0x340
>> ? ext4_sync_file+0x157/0x340
>> vfs_fsync_range+0x49/0x80
>> do_fsync+0x3d/0x70
>> __x64_sys_fsync+0x14/0x20
>> do_syscall_64+0x3b/0xc0
>> entry_SYSCALL_64_after_hwframe+0x44/0xae
>>
>> Simply fixing it by limiting the number of blocks
>> in one EXT4_FC_TAG_ADD_RANGE TLV.
>>
>> Fixes: aa75f4d3daae ("ext4: main fast-commit commit path")
>> Signed-off-by: Hou Tao <[email protected]>
>> ---
>> fs/ext4/fast_commit.c | 6 ++++++
>> 1 file changed, 6 insertions(+)
>>
>> diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
>> index e8195229c252..782d05a3f97a 100644
>> --- a/fs/ext4/fast_commit.c
>> +++ b/fs/ext4/fast_commit.c
>> @@ -893,6 +893,12 @@ static int ext4_fc_write_inode_data(struct inode *inode, u32 *crc)
>> sizeof(lrange), (u8 *)&lrange, crc))
>> return -ENOSPC;
>> } else {
>> + unsigned int max = (map.m_flags & EXT4_MAP_UNWRITTEN) ?
>> + EXT_UNWRITTEN_MAX_LEN : EXT_INIT_MAX_LEN;
>> +
>> + /* Limit the number of blocks in one extent */
>> + map.m_len = min(max, map.m_len);
>> +
>> fc_ext.fc_ino = cpu_to_le32(inode->i_ino);
>> ex = (struct ext4_extent *)&fc_ext.fc_ex;
>> ex->ee_block = cpu_to_le32(map.m_lblk);
> .
Sorry Hou for getting back to you late. This looks good to me.
Reviewed-by: Harshad Shirwadkar <[email protected]>
On Tue, Sep 7, 2021 at 4:21 AM Hou Tao <[email protected]> wrote:
>
> ping ?
>
> On 8/30/2021 3:52 PM, Hou Tao wrote:
> > ping ?
> >
> > On 8/20/2021 12:45 PM, Hou Tao wrote:
> >> Now EXT4_FC_TAG_ADD_RANGE uses ext4_extent to track the
> >> newly-added blocks, but the limit on the max value of
> >> ee_len field is ignored, and it can lead to BUG_ON as
> >> shown below when running command "fallocate -l 128M file"
> >> on a fast_commit-enabled fs:
> >>
> >> kernel BUG at fs/ext4/ext4_extents.h:199!
> >> invalid opcode: 0000 [#1] SMP PTI
> >> CPU: 3 PID: 624 Comm: fallocate Not tainted 5.14.0-rc6+ #1
> >> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
> >> RIP: 0010:ext4_fc_write_inode_data+0x1f3/0x200
> >> Call Trace:
> >> ? ext4_fc_write_inode+0xf2/0x150
> >> ext4_fc_commit+0x93b/0xa00
> >> ? ext4_fallocate+0x1ad/0x10d0
> >> ext4_sync_file+0x157/0x340
> >> ? ext4_sync_file+0x157/0x340
> >> vfs_fsync_range+0x49/0x80
> >> do_fsync+0x3d/0x70
> >> __x64_sys_fsync+0x14/0x20
> >> do_syscall_64+0x3b/0xc0
> >> entry_SYSCALL_64_after_hwframe+0x44/0xae
> >>
> >> Simply fixing it by limiting the number of blocks
> >> in one EXT4_FC_TAG_ADD_RANGE TLV.
> >>
> >> Fixes: aa75f4d3daae ("ext4: main fast-commit commit path")
> >> Signed-off-by: Hou Tao <[email protected]>
> >> ---
> >> fs/ext4/fast_commit.c | 6 ++++++
> >> 1 file changed, 6 insertions(+)
> >>
> >> diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
> >> index e8195229c252..782d05a3f97a 100644
> >> --- a/fs/ext4/fast_commit.c
> >> +++ b/fs/ext4/fast_commit.c
> >> @@ -893,6 +893,12 @@ static int ext4_fc_write_inode_data(struct inode *inode, u32 *crc)
> >> sizeof(lrange), (u8 *)&lrange, crc))
> >> return -ENOSPC;
> >> } else {
> >> + unsigned int max = (map.m_flags & EXT4_MAP_UNWRITTEN) ?
> >> + EXT_UNWRITTEN_MAX_LEN : EXT_INIT_MAX_LEN;
> >> +
> >> + /* Limit the number of blocks in one extent */
> >> + map.m_len = min(max, map.m_len);
> >> +
> >> fc_ext.fc_ino = cpu_to_le32(inode->i_ino);
> >> ex = (struct ext4_extent *)&fc_ext.fc_ex;
> >> ex->ee_block = cpu_to_le32(map.m_lblk);
> > .
On Fri, 20 Aug 2021 12:45:05 +0800, Hou Tao wrote:
> Now EXT4_FC_TAG_ADD_RANGE uses ext4_extent to track the
> newly-added blocks, but the limit on the max value of
> ee_len field is ignored, and it can lead to BUG_ON as
> shown below when running command "fallocate -l 128M file"
> on a fast_commit-enabled fs:
>
> kernel BUG at fs/ext4/ext4_extents.h:199!
> invalid opcode: 0000 [#1] SMP PTI
> CPU: 3 PID: 624 Comm: fallocate Not tainted 5.14.0-rc6+ #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996)
> RIP: 0010:ext4_fc_write_inode_data+0x1f3/0x200
> Call Trace:
> ? ext4_fc_write_inode+0xf2/0x150
> ext4_fc_commit+0x93b/0xa00
> ? ext4_fallocate+0x1ad/0x10d0
> ext4_sync_file+0x157/0x340
> ? ext4_sync_file+0x157/0x340
> vfs_fsync_range+0x49/0x80
> do_fsync+0x3d/0x70
> __x64_sys_fsync+0x14/0x20
> do_syscall_64+0x3b/0xc0
> entry_SYSCALL_64_after_hwframe+0x44/0xae
>
> [...]
Applied, thanks!
[1/1] ext4: limit the number of blocks in one ADD_RANGE TLV
commit: af8137dff33d975ac84c8a5f63e057b2d0405fc3
Best regards,
--
Theodore Ts'o <[email protected]>