Re: [PATCH 00/17 stable-5.15.y] Fix mmap + page fault deadlocks

[Anand Jain <anand.jain@oracle.com>]

On 11/04/2022 15:49, Greg KH wrote:
> On Tue, Apr 05, 2022 at 07:32:11PM +0800, Anand Jain wrote:
>>
>>
>> On 04/04/2022 17:41, Filipe Manana wrote:
>>> On Sat, Apr 02, 2022 at 06:25:37PM +0800, Anand Jain wrote:
>>>> This set fixes a process hang issue in btrfs and gf2 filesystems. When we
>>>> do a direct IO read or write when the buffer given by the user is
>>>> memory-mapped to the file range we are going to do IO, we end up ending
>>>> in a deadlock. This is triggered by the test case generic/647 from
>>>> fstests.
>>>>
>>>> This fix depends on the iov_iter and iomap changes introduced in the
>>>> commit c03098d4b9ad ("Merge tag 'gfs2-v5.15-rc5-mmap-fault' of
>>>> git://git.kernel.org/pub/scm/linux/kernel/git/gfs2/linux-gfs2") and they
>>>> are part of this set for stable-5.15.y.
>>>>
>>>> Please note that patch 3/17 in this patchset changes the prototype and
>>>> renames an exported symbol as below. All its references are updated as
>>>> well.
>>>>
>>>> -EXPORT_SYMBOL(iov_iter_fault_in_readable);
>>>> +EXPORT_SYMBOL(fault_in_iov_iter_readable);
>>>>
>>>> Andreas Gruenbacher (15):
>>>>     powerpc/kvm: Fix kvm_use_magic_page
>>>>     gup: Turn fault_in_pages_{readable,writeable} into
>>>>       fault_in_{readable,writeable}
>>>>     iov_iter: Turn iov_iter_fault_in_readable into
>>>>       fault_in_iov_iter_readable
>>>>     iov_iter: Introduce fault_in_iov_iter_writeable
>>>>     gfs2: Add wrapper for iomap_file_buffered_write
>>>>     gfs2: Clean up function may_grant
>>>>     gfs2: Move the inode glock locking to gfs2_file_buffered_write
>>>>     gfs2: Eliminate ip->i_gh
>>>>     gfs2: Fix mmap + page fault deadlocks for buffered I/O
>>>>     iomap: Fix iomap_dio_rw return value for user copies
>>>>     iomap: Support partial direct I/O on user copy failures
>>>>     iomap: Add done_before argument to iomap_dio_rw
>>>>     gup: Introduce FOLL_NOFAULT flag to disable page faults
>>>>     iov_iter: Introduce nofault flag to disable page faults
>>>>     gfs2: Fix mmap + page fault deadlocks for direct I/O
>>>>
>>>> Bob Peterson (1):
>>>>     gfs2: Introduce flag for glock holder auto-demotion
>>>>
>>>> Filipe Manana (1):
>>>>     btrfs: fix deadlock due to page faults during direct IO reads and
>>>>       writes
>>>
>>> If this patchset is backported, then at least the following two commits
>>> must also be backported:
>>>
>>> commit fe673d3f5bf1fc50cdc4b754831db91a2ec10126
>>> Author: Linus Torvalds <torvalds@linux-foundation.org>
>>> Date:   Tue Mar 8 11:55:48 2022 -0800
>>>
>>>       mm: gup: make fault_in_safe_writeable() use fixup_user_fault()
>>>
>>> commit ca93e44bfb5fd7996b76f0f544999171f647f93b
>>> Author: Filipe Manana <fdmanana@suse.com>
>>> Date:   Wed Mar 2 11:48:39 2022 +0000
>>>
>>>       btrfs: fallback to blocking mode when doing async dio over multiple extents
>>
>> Thanks for pointing it out.
>>
>>> Maybe there's more that need to be backported. So cc'ing Andreas in
>>> case he's aware of any such other commits.
>>
>> I have scanned through the commits. I didn't find any further Fixes
>> for this series.
>>
>> I am sending these two patches in a new patch set as part2. Instead,
>> if it is better, I am ok to include these and send v2.
> 
> Please rebase the series and send a whole new one as I think there will
> be at least one change you can drop as it has been accepted already, and
> trying to deal with two different patch series is hard for everyone
> involved.
> 

  Sure I will do it. Thanks for pointing it out.

Thx, Anand

> thanks,
> 
> greg k-h