linux-kernel - Re: [PATCH v4 0/3] fs/dcache: Track # of negative dentries

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 for Android: free password hash cracker in your pocket

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

Date:   Fri, 5 Oct 2018 10:55:47 -0400
From:   Waiman Long <longman@...hat.com>
To:     Alexander Viro <viro@...iv.linux.org.uk>,
        Jonathan Corbet <corbet@....net>,
        Andrew Morton <akpm@...ux-foundation.org>
Cc:     linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        linux-mm@...ck.org, linux-doc@...r.kernel.org,
        "Luis R. Rodriguez" <mcgrof@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Jan Kara <jack@...e.cz>,
        "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
        Ingo Molnar <mingo@...nel.org>,
        Miklos Szeredi <mszeredi@...hat.com>,
        Matthew Wilcox <willy@...radead.org>,
        Larry Woodman <lwoodman@...hat.com>,
        James Bottomley <James.Bottomley@...senPartnership.com>,
        "Wangkai (Kevin C)" <wangkai86@...wei.com>,
        Michal Hocko <mhocko@...nel.org>
Subject: Re: [PATCH v4 0/3] fs/dcache: Track # of negative dentries

On 09/12/2018 01:35 PM, Waiman Long wrote:
>  v3->v4:
>   - Drop patch 4 as it is just a minor optimization.
>   - Add a cc:stable tag to patch 1.
>   - Clean up some comments in patch 3.
>
>  v2->v3:
>   - With confirmation that the dummy array in dentry_stat structure
>     was never a replacement of a previously used field, patch 3 is now
>     reverted back to use one of dummy field as the negative dentry count
>     instead of adding a new field.
>
>  v1->v2:
>   - Clarify what the new nr_dentry_negative per-cpu counter is tracking
>     and open-code the increment and decrement as suggested by Dave Chinner.
>   - Append the new nr_dentry_negative count as the 7th element of dentry-state
>     instead of replacing one of the dummy entries.
>   - Remove patch "fs/dcache: Make negative dentries easier to be
>     reclaimed" for now as I need more time to think about what
>     to do with it.
>   - Add 2 more patches to address issues found while reviewing the
>     dentry code.
>   - Add another patch to change the conditional branch of
>     nr_dentry_negative accounting to conditional move so as to reduce
>     the performance impact of the accounting code.
>
> This patchset addresses 2 issues found in the dentry code and adds a
> new nr_dentry_negative per-cpu counter to track the total number of
> negative dentries in all the LRU lists.
>
> Patch 1 fixes a bug in the accounting of nr_dentry_unused in
> shrink_dcache_sb().
>
> Patch 2 removes the ____cacheline_aligned_in_smp tag from super_block
> LRU lists.
>
> Patch 3 adds the new nr_dentry_negative per-cpu counter.
>
> Various filesystem related tests were run and no statistically
> significant changes in performance outside of the possible noise range
> was observed.
>
> Waiman Long (3):
>   fs/dcache: Fix incorrect nr_dentry_unused accounting in
>     shrink_dcache_sb()
>   fs: Don't need to put list_lru into its own cacheline
>   fs/dcache: Track & report number of negative dentries
>
>  Documentation/sysctl/fs.txt | 26 ++++++++++++++++----------
>  fs/dcache.c                 | 38 +++++++++++++++++++++++++++++++++-----
>  include/linux/dcache.h      |  7 ++++---
>  include/linux/fs.h          |  9 +++++----
>  4 files changed, 58 insertions(+), 22 deletions(-)
>
Any comments on these patches. The first one actually is a bug fix.

Cheers,
Longman