linux-kernel - Re: [PATCHSET for-5.3/block] block: add blkcg bio punt mechanism

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] [day] [month] [year] [list]

Message-ID: <be63af25-6245-18f8-4c0b-4f48682a34ed@fb.com>
Date:   Wed, 10 Jul 2019 13:53:11 +0000
From:   Jens Axboe <axboe@...com>
To:     Tejun Heo <tj@...nel.org>, "axboe@...nel.dk" <axboe@...nel.dk>
CC:     "jack@...e.cz" <jack@...e.cz>,
        "josef@...icpanda.com" <josef@...icpanda.com>,
        Chris Mason <clm@...com>,
        "dsterba@...e.com" <dsterba@...e.com>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "linux-btrfs@...r.kernel.org" <linux-btrfs@...r.kernel.org>,
        Kernel Team <Kernel-team@...com>
Subject: Re: [PATCHSET for-5.3/block] block: add blkcg bio punt mechanism

On 6/27/19 2:39 PM, Tejun Heo wrote:
> Hello,
> 
> This patchset contains only the block part of the following
> 
>    [1] [PATCHSET v2 btrfs/for-next] blkcg, btrfs: fix cgroup writeback support
> 
> with the following changes
> 
>   * wbc_account_io() is renamed to wbc_account_cgroup_owner() and
>     wbc->no_account_io to wbc->no_cgroup_owner for clarity.
> 
> When writeback is executed asynchronously (e.g. for compression), bios
> are bounced to and issued by worker pool shared by all cgroups.  This
> leads to significant priority inversions when cgroup IO control is in
> use - IOs for a low priority cgroup can tie down the workers forcing
> higher priority IOs to wait behind them.
> 
> This patchset adds an bio punt mechanism to blkcg and updates btrfs to
> issue async IOs through it.  A bio tagged with REQ_CGROUP_PUNT flag is
> bounced to the asynchronous issue context of the associated blkcg on
> bio_submit().  As the bios are issued from per-blkcg work items,
> there's no concern for priority inversions and it doesn't require
> invasive changes to the filesystems.  The mechanism should be
> generally useful for IO control support across different filesystems.

Applied for 5.3, thanks Tejun.

-- 
Jens Axboe