linux-kernel - [bug] Re: [PATCH 00/35] cpumask: Replace cpumask

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: <20081023125529.GA18327@elte.hu>
Date:	Thu, 23 Oct 2008 14:55:29 +0200
From:	Ingo Molnar <mingo@...e.hu>
To:	Mike Travis <travis@....com>
Cc:	Rusty Russell <rusty@...tcorp.com.au>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org
Subject: [bug] Re: [PATCH 00/35] cpumask: Replace cpumask_t with struct
	cpumask


ok, the new cpumask code blew up in -tip testing, with various sorts of 
slab corruptions during scheduler init:

[    0.544012]    groups: 0-1
[    0.546689] =============================================================================
[    0.548000] BUG kmalloc-8: Wrong object count. Counter is 15 but counted were 50
[    0.548000] -----------------------------------------------------------------------------
[    0.548000] 
[    0.548000] INFO: Slab 0xffffe200019d7ae0 objects=51 used=15 fp=0xffff88003f9cc4b0 flags=0x200000000000c3
[    0.548000] Pid: 1, comm: swapper Not tainted 2.6.27-tip-07104-g5cf7b67-dirty #45066
[    0.548000] Call Trace:
[    0.548000]  [<ffffffff802cbcf0>] slab_err+0xa0/0xb0
[    0.548000]  [<ffffffff8026df2d>] ? trace_hardirqs_on+0xd/0x10
[    0.548000]  [<ffffffff8026deca>] ? trace_hardirqs_on_caller+0x15a/0x1b0
[    0.548000]  [<ffffffff8023d102>] ? cpu_attach_domain+0x172/0x6b0
[    0.548000]  [<ffffffff802cbf4d>] ? check_bytes_and_report+0x3d/0xe0
[    0.548000]  [<ffffffff802cd927>] on_freelist+0x197/0x240
[    0.548000]  [<ffffffff802ce926>] __slab_free+0x1a6/0x310
[    0.548000]  [<ffffffff80415009>] ? free_cpumask_var+0x9/0x10
[    0.548000]  [<ffffffff802ceb47>] kfree+0xb7/0x140
[    0.548000]  [<ffffffff80415009>] ? free_cpumask_var+0x9/0x10
[    0.548000]  [<ffffffff80415009>] free_cpumask_var+0x9/0x10
[    0.548000]  [<ffffffff8023dadc>] __build_sched_domains+0x49c/0xd30
[    0.548000]  [<ffffffff8026df2d>] ? trace_hardirqs_on+0xd/0x10
[    0.548000]  [<ffffffff816698fa>] sched_init_smp+0xba/0x2b0
[    0.548000]  [<ffffffff8026deca>] ? trace_hardirqs_on_caller+0x15a/0x1b0
[    0.548000]  [<ffffffff802cbf4d>] ? check_bytes_and_report+0x3d/0xe0
[    0.548000]  [<ffffffff802cdc08>] ? check_object+0x238/0x270
[    0.548000]  [<ffffffff802cc044>] ? init_object+0x54/0x90
[    0.548000]  [<ffffffff8026df2d>] ? trace_hardirqs_on+0xd/0x10
[    0.548000]  [<ffffffff8026deca>] ? trace_hardirqs_on_caller+0x15a/0x1b0
[    0.548000]  [<ffffffff8026df2d>] ? trace_hardirqs_on+0xd/0x10
[    0.548000]  [<ffffffff816622e4>] ? check_nmi_watchdog+0x204/0x260
[    0.548000]  [<ffffffff816622e4>] ? check_nmi_watchdog+0x204/0x260
[    0.548000]  [<ffffffff8165fad6>] ? native_smp_cpus_done+0x1a6/0x2b0
[    0.548000]  [<ffffffff81654f86>] kernel_init+0x176/0x240
[    0.548000]  [<ffffffff8020c9f9>] child_rip+0xa/0x11
[    0.548000]  [<ffffffff8020bf14>] ? restore_args+0x0/0x30
[    0.548000]  [<ffffffff81654e10>] ? kernel_init+0x0/0x240
[    0.548000]  [<ffffffff8020c9ef>] ? child_rip+0x0/0x11
[    0.548000] FIX kmalloc-8: Object count adjusted.
[    0.548000] =============================================================================

i suspect it's due to:

01b8bd9: sched: cpumask: get rid of boutique sched.c allocations, use cpumask_va

note, CONFIG_MAXSMP is set in the .config, so this is with the dynamic 
cpumask_t.

	Ingo

View attachment "config" of type "text/plain" (64512 bytes)