linux-kernel - Re: [PATCH] Percpu tag allocator

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: <20130613223028.GG13970@mtj.dyndns.org>
Date:	Thu, 13 Jun 2013 15:30:28 -0700
From:	Tejun Heo <tj@...nel.org>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	Kent Overstreet <koverstreet@...gle.com>,
	linux-kernel@...r.kernel.org, Oleg Nesterov <oleg@...hat.com>,
	Christoph Lameter <cl@...ux-foundation.org>,
	Ingo Molnar <mingo@...hat.com>,
	Andi Kleen <andi@...stfloor.org>, Jens Axboe <axboe@...nel.dk>,
	"Nicholas A. Bellinger" <nab@...ux-iscsi.org>,
	Jeff Layton <jlayton@...hat.com>,
	"J. Bruce Fields" <bfields@...ldses.org>
Subject: Re: [PATCH] Percpu tag allocator

Hello, Andrew.

On Thu, Jun 13, 2013 at 03:10:40PM -0700, Andrew Morton wrote:
> Look, we're falling into the age-old trap of trying to justify existing
> code just because it exists.
> 
> Stop.  Take a step back and pretend that the percpu tag allocator patch
> never existed.  Now, define the problem and propose solutions.
> 
> The absolute dead last and worst solution is "implement something new
> which largely duplicates existing code".  Such a step requires an
> extraordinary amount of justification and that hasn't happened.

I think you're looking at it from the complete opposite point from
mine.  That problem has always existed and the solutions implemented
many times over in the whole IO stack across aio, block layer proper,
SCSI and individual drivers, and most, if not all, suck.  If we can
consolidate those into a properly designed tag allocator, it's a win,
whether it's integrated into idr or not.  It's a well-known problem
with incomplete bad solutions implemented in multiple places.

Again, I'm not saying integrating this into idr is a bad idea.  If
that's reasonably doable, awesomer, but I think even just
consolidating the existing bad ones into a common one is beneficial
enough.

Thanks.

-- 
tejun
--
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/