linux-kernel - Re: [ANNOUNCE] Status of unlocked

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: <4CCD5DA4.9090806@panasas.com>
Date:	Sun, 31 Oct 2010 14:14:28 +0200
From:	Boaz Harrosh <bharrosh@...asas.com>
To:	Andi Kleen <ak@...ux.intel.com>
CC:	James Bottomley <James.Bottomley@...e.de>,
	"Nicholas A. Bellinger" <nab@...ux-iscsi.org>,
	Mike Anderson <andmike@...ux.vnet.ibm.com>,
	linux-kernel <linux-kernel@...r.kernel.org>,
	linux-scsi <linux-scsi@...r.kernel.org>,
	Vasu Dev <vasu.dev@...ux.intel.com>,
	Tim Chen <tim.c.chen@...ux.intel.com>,
	Matthew Wilcox <willy@...ux.intel.com>,
	Mike Christie <michaelc@...wisc.edu>,
	Jens Axboe <jaxboe@...ionio.com>,
	James Smart <james.smart@...lex.com>,
	Andrew Vasquez <andrew.vasquez@...gic.com>,
	FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>,
	Hannes Reinecke <hare@...e.de>,
	Joe Eykholt <jeykholt@...co.com>,
	Christoph Hellwig <hch@....de>,
	Jon Hawley <warthog9@...nel.org>,
	Brian King <brking@...ux.vnet.ibm.com>,
	Christof Schmitt <christof.schmitt@...ibm.com>,
	Tejun Heo <tj@...nel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	"H. Peter Anvin" <hpa@...or.com>, julia@...u.dk
Subject: Re: [ANNOUNCE] Status of unlocked_qcmds=1 operation for .37

On 10/28/2010 08:26 PM, Andi Kleen wrote:
>> I disagree with your approach this introduces a spin_unlock_irqrestore
>> call site at every return, in the usually huge queuecommand.
> 
> I converted the full tree and in practice it turns out there 
> are very few returns in nearly all queuecommands. So it won't
> make much difference.
> 

"few returns" is too much. Any thing bigger than 1 is a total wast.

And the mess?!? Where to add the flags? where the returns? Need a 
"{...}" or not. Lots of manual intervention, and possible errors.
I bet with my approach you wouldn't need to manually fix a single
file.

> Longer term they will be all hopefully gone again anyways.
> 

That one I'm most afraid of. These that did not get fixed in this
round, will not be fixed for a long time (if ever). I'd even go
anal and not open code the lock but actually call the original
__queue_command through a MACRO, that can be change in one place.
(And will solve your patchset bisect-ability)

- XXX_queue_command(...
+ XXX_queue_command_unlocked(...


+ XXX_queue_command(...
+ {
+ 	return SCSI_LOCKED_QUEUECOMMAND(XXX_queue_command_unlocked, ...);
+ }
> -Andi

But since I'm only blabing, the one that "do", gets to decide ;-) .
Perhaps next time.

Boaz
--
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/