linux-ext4 - Re: [PATCH 0/5] ext4: direct IO via iomap infrastructure

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20190812173150.AF04F5204F@d06av21.portsmouth.uk.ibm.com>
Date:   Mon, 12 Aug 2019 23:01:50 +0530
From:   RITESH HARJANI <riteshh@...ux.ibm.com>
To:     Matthew Bobrowski <mbobrowski@...browski.org>,
        linux-ext4@...r.kernel.org
Cc:     linux-fsdevel@...r.kernel.org, jack@...e.cz, tytso@....edu
Subject: Re: [PATCH 0/5] ext4: direct IO via iomap infrastructure

Hello Matthew,

On 8/12/19 6:22 PM, Matthew Bobrowski wrote:

> This patch series converts the ext4 direct IO code paths to make use of the
> iomap infrastructure and removes the old buffer_head direct-io based
> implementation. The result is that ext4 is converted to the newer framework
> and that it may _possibly_ gain a performance boost for O_SYNC | O_DIRECT IO.
>
> These changes have been tested using xfstests in both DAX and non-DAX modes
> using various configurations i.e. 4k, dioread_nolock, dax.

I had some minor review comments posted on Patch-4.
But the rest of the patch series looks good to me.
I will also do some basic testing of xfstests which I did for my patches 
and will revert back.

One query, could you please help answering below for my understanding :-

I was under the assumption that we need to maintain 
ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN) or 
atomic_read(&EXT4_I(inode)->i_unwritten))
in case of non-AIO directIO or AIO directIO case as well (when we may 
allocate unwritten extents),
to protect with some kind of race with other parts of code(maybe 
truncate/bufferedIO/fallocate not sure?) which may call for 
ext4_can_extents_be_merged()
to check if extents can be merged or not.

Is it not the case?
Now that directIO code has no way of specifying that this inode has 
unwritten extent, will it not race with any other path, where this info 
was necessary (like
in above func ext4_can_extents_be_merged())?

Thanks
Ritesh

>
> Matthew Bobrowski (5):
>    ext4: introduce direct IO read code path using iomap infrastructure
>    ext4: move inode extension/truncate code out from ext4_iomap_end()
>    iomap: modify ->end_io() calling convention
>    ext4: introduce direct IO write code path using iomap infrastructure
>    ext4: clean up redundant buffer_head direct IO code
>
>   fs/ext4/ext4.h        |   3 -
>   fs/ext4/extents.c     |   8 +-
>   fs/ext4/file.c        | 329 +++++++++++++++++++++++++++-------
>   fs/ext4/inode.c       | 488 +++++---------------------------------------------
>   fs/iomap/direct-io.c  |   9 +-
>   fs/xfs/xfs_file.c     |  17 +-
>   include/linux/iomap.h |   4 +-
>   7 files changed, 322 insertions(+), 536 deletions(-)
>