linux-kernel - Re: [BUG] commit 3c517a61, slab: better fallback allocation behavior

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] [day] [month] [year] [list]

Date:	Tue, 12 Dec 2006 11:22:11 -0800 (PST)
From:	Christoph Lameter <clameter@....com>
To:	Paul Jackson <pj@....com>
cc:	jacliburn@...lsouth.net, akpm@...l.org,
	linux-kernel@...r.kernel.org
Subject: Re: [BUG] commit 3c517a61, slab: better fallback allocation behavior

On Tue, 12 Dec 2006, Paul Jackson wrote:

> Christoph wrote:
> > +		if (local_flags & __GFP_WAIT)
> > +			local_irq_enable();
> > +		kmem_flagcheck(cache, flags);
> >  		obj = kmem_getpages(cache, flags, -1);
> > +		if (local_flags & __GFP_WAIT)
> > +			local_irq_disable();
> 
> This seems strange to me.  I am surprised that it is ok for a routine
> that is called with irq's disabled, to enable them momentarilly.

The slab itself disables the irq. In this case the slab function is 
was called irqs enabled and we forgot to reenable interrupts before 
calling kmem_getpages().

> I'd have thought the caller of this routine, who called it with irq's
> disabled, would expect irq's to remain disabled across the entire call.

Irqs only stay disabled if GFP_ATOMIC is passed to the slab function.
-
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/