linux-kernel - Re: SLUB 0:1 SLAB (OOM during massive parallel kernel builds)

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: <Pine.LNX.4.64.0710231227590.19626@schroedinger.engr.sgi.com>
Date:	Tue, 23 Oct 2007 12:35:04 -0700 (PDT)
From:	Christoph Lameter <clameter@....com>
To:	Alexey Dobriyan <adobriyan@...il.com>
cc:	penberg@...helsinki.fi, linux-kernel@...r.kernel.org,
	linux-mm@...r.kernel.org
Subject: Re: SLUB 0:1 SLAB (OOM during massive parallel kernel builds)

On Tue, 23 Oct 2007, Alexey Dobriyan wrote:

> Nasty OOM killings appear during massive parallel kernel builds with
> SLUB, but not with SLAB. By nasty I mean, cc1 processes are killed --
> object files in .ccache and tree are corrupted which makes me to
> blow up them entirely.

Hmmmm... Memory is corrupted? I guess a race destroys memory? 

> 
> With SLAB this workload never went to OOM killer.
> With SLUB and pretty much all debugging enabled, it finishes to the end
> (albeit slowly).
> With SLUB and no debugging, OOM killer kicks in.

Not sure what this is. Maybe the slowing SLUB solves the race.
Could you try and boot with

slub_debug=F

to enable limited debugging?

> cc1 invoked oom-killer: gfp_mask=0x1201d2, order=0, oomkilladj=0

Regular Order 0 alloc .... but why is there no memory available , 
reclaimed?

-
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/