linux-kernel - Re: Query on shrink list

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: <b638d027-023a-d455-f77e-d66c10b8d9e4@codeaurora.org>
Date:   Fri, 17 Aug 2018 20:07:59 +0530
From:   Mukesh Ojha <mojha@...eaurora.org>
To:     Al Viro <viro@...IV.linux.org.uk>
Cc:     linux-fsdevel@...r.kernel.org, lkml <linux-kernel@...r.kernel.org>
Subject: Re: Query on shrink list



On 8/17/2018 6:28 PM, Al Viro wrote:
> On Fri, Aug 17, 2018 at 03:39:22PM +0530, Mukesh Ojha wrote:
>> Hi Al Viro,
>>
>> Is there is reason we have kept data->found++, if the dentry already there
>> in shrink list ?
>>
>> static enum d_walk_ret select_collect(
>> ...
>>          if (dentry->d_flags & DCACHE_SHRINK_LIST) {
>>                  data->found++;
>>          } else {
>>    ..
>>
>> If the dentry is already there on shrink list, does it not mean that
>> data->found is already non-zero ?
> Nope.  It can be on *another* shrink list - if two processes are doing
> that...

Ok, if we go out simply,  letting others to do the job will break 
`shrink_dcache_parent()`
and if someone touched that dentry made the refcount > 0 while it is on 
shrink list
  then owner will keep on looping in shrink_dentry_list() until refcount 
becomes 0 .

Am i making sense here ?

Thanks.
Mukesh

>
>> Can't we just go out from here directly?