linux-kernel - Re: [PATCH 2/2] blk-plug: don't flush nested plug lists

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: <x49384cro4z.fsf@segfault.boston.devel.redhat.com>
Date:	Tue, 07 Apr 2015 10:47:40 -0400
From:	Jeff Moyer <jmoyer@...hat.com>
To:	Ming Lei <tom.leiming@...il.com>
Cc:	Jens Axboe <axboe@...nel.dk>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] blk-plug: don't flush nested plug lists

Ming Lei <tom.leiming@...il.com> writes:

> Hi Jeff,

Hi, Ming.  Thanks for reviewing!

>> Unpatched kernel:
>> Read B/W:    283,638 KB/s
>> Read Merges: 0
>>
>> Patched kernel:
>> Read B/W:    873,224 KB/s
>> Read Merges: 2,046K
>
> The data is amazing, but maybe better to provide some latency
> data.

Yes, good point.  I'll include the full fio output in the next posting.

>> Much of the patch involves modifying call sites to blk_start_plug,
>> since its signature is changed.  The meat of the patch is actually
>
> I am wondering if the type of blk_start_plug has to be changed
> since the active plug is always the top plug with your patch, and
> blk_finish_plug() can find the active plug from current->plug.

That is a much simpler idea, thank you.  I think blk_finish_plug
shouldn't take the plug argument, though, since it won't actually use
it.

Cheers,
Jeff
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/