linux-kernel - Re: [PATCH-v5 1/6] idr: Percpu ida

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:	Tue, 03 Sep 2013 21:42:08 -0700
From:	"Nicholas A. Bellinger" <nab@...ux-iscsi.org>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	target-devel <target-devel@...r.kernel.org>,
	lkml <linux-kernel@...r.kernel.org>,
	"Michael S. Tsirkin" <mst@...hat.com>, Asias He <asias@...hat.com>,
	Kent Overstreet <kmo@...erainc.com>,
	Jens Axboe <axboe@...nel.dk>, Tejun Heo <tj@...nel.org>,
	Ingo Molnar <mingo@...hat.com>,
	Andi Kleen <andi@...stfloor.org>,
	Christoph Lameter <cl@...two.org>,
	Oleg Nesterov <oleg@...hat.com>,
	Christoph Lameter <cl@...ux-foundation.org>
Subject: Re: [PATCH-v5 1/6] idr: Percpu ida

On Tue, 2013-09-03 at 18:27 -0700, Andrew Morton wrote:
> On Tue, 03 Sep 2013 09:06:58 -0700 "Nicholas A. Bellinger" <nab@...ux-iscsi.org> wrote:
> 
> > Are there any other review comments to be addressed for this patch..?
> > 
> > If not, please kindly give your Reviewed-by if your OK for an initial
> > standalone merge.
> 
> Sorry, I'm largely offline for the next week...
> 
> It looks OK to me (that's as close as I get to an ack :))
> 

Thanks.  That quote is going in the commit log btw.  ;)

> Nit:
> 
> > +		if (cpu >= nr_cpu_ids)
> > +			cpu = cpumask_first(&pool->cpus_have_tags);
> > +
> > +		if (cpu >= nr_cpu_ids)
> > +			BUG();
> 
> The second `if' can be moved inside the first, but hopefully the
> compiler already did that.

Fixed.

> 
> I have dim memories that doing
> 
> 	local_irq_save()
> 	spin_lock()
> 
> is not equivalent to
> 
> 	spin_lock_irqsave()
> 
> in some circumstances.  Perhaps a lockdep thing.  Or perhaps this was fixed in
> the intervening years and this is no longer the case.

Mmmm, I had wondered the same thing on the release side with
kref_put_spinlock_irqsave() here before:

https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/include/linux/kref.h?id=ccf5ae83a6cf3d9cfe9a7038bfe7cd38ab03d5e1

but at some point was convinced it was not a issue anymore..
 
> 
> Such things probably won't be detected during 3.12-rcX because few
> people will test iscsi, I expect.  So please be sure that you've
> runtime tested it with all the lockdeppy things enabled. 
> Documentation/SubmitChecklist section 12 has a list, but it might be
> dated.

<nod>

FYI, my v3.11-rc5 kernel has been running with CONFIG_LOCKDEP_SUPPORT=y
using vhost/iscsi/iser fabrics on the -v5 patch, and not generated any
warnings (yet).

I'll double check against the other debug options in Section 12.

Thank you,

--nab

--
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/