netdev - Re: [PATCH 1/2] net/mlx4_core: clean up cq_res_start_move

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: <1389698630.20290.48.camel@x220>
Date:	Tue, 14 Jan 2014 12:23:50 +0100
From:	Paul Bolle <pebolle@...cali.nl>
To:	Jack Morgenstein <jackm@....mellanox.co.il>
Cc:	Or Gerlitz <ogerlitz@...lanox.com>,
	Rony Efraim <ronye@...lanox.com>,
	Hadar Hen Zion <hadarh@...lanox.com>,
	"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] net/mlx4_core: clean up cq_res_start_move_to()

On Tue, 2014-01-14 at 08:47 +0200, Jack Morgenstein wrote:
> On Tue, 07 Jan 2014 14:01:18 +0100
> Paul Bolle <pebolle@...cali.nl> wrote:
> 
> > +	} else {
> > +		/* state == RES_CQ_HW */
> > +		if (r->com.state != RES_CQ_ALLOCATED)
> 
> if (state != RES_CQ_HW || r->com.state != RES_CQ_ALLOCATED)
> to protect against any bad calls to this function
> (although I know that currently there are none).

So we end up with
         } else if (state != RES_CQ_HW || r->com.state != RES_CQ_ALLOCATED) {
                 err = -EINVAL;
         } else {
                 err = 0;
         }

don't we? Which is fine with me, as GCC still is then able to correctly
analyze this function.

> This also preserves the behavior of the switch statement.
> 
> >  			err = -EINVAL;
> > -		}
> > +		else
> > +			err = 0;
> > +	}
> >  
> > -		if (!err) {
> > -			r->com.from_state = r->com.state;
> > -			r->com.to_state = state;
> > -			r->com.state = RES_CQ_BUSY;
> > -			if (cq)
> > -				*cq = r;
> > -		}
> > +	if (!err) {
> > +		r->com.from_state = r->com.state;
> > +		r->com.to_state = state;
> > +		r->com.state = RES_CQ_BUSY;
> 
> Please keep the "if" here.  Protects against (future) bad calls.
> 
> > +		*cq = r;
> >  	}

There seems to be a school of thought that says it's better to trigger
an Oops if a programming error is made (in this case by passing a NULL
cq) then silently handle that (future) programming error and make
debugging harder. But, even it that school of thought really exists,
this is up to you. Besides, it's only a triviality I added to my
patches.

Thanks for the review! I hope to send in a v2 of my patches shortly.


Paul Bolle

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html