* slab corruption on 2.6.10-rc4-bk7
@ 2005-02-19 22:16 Dave Jones
2005-02-19 22:41 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Dave Jones @ 2005-02-19 22:16 UTC (permalink / raw)
To: linux-kernel
(This has actually been there for a while, but I only
noticed it in dmesg this morning).
During boot on a dual em64t I see ..
scsi2 : ata_piix
isa bounce pool size: 16 pages
slab error in cache_free_debugcheck(): cache `size-2048': double free, or memory outside object was overwritten
Call Trace:<ffffffff80163448>{cache_free_debugcheck+392} <ffffffff801646aa>{kfree+234}
<ffffffff88065189>{:libata:ata_pci_init_one+937} <ffffffff801fe9ea>{pci_bus_read_config_word+122}
<ffffffff880707f2>{:ata_piix:piix_init_one+498} <ffffffff80202926>{pci_device_probe+134}
<ffffffff802691ad>{driver_probe_device+77} <ffffffff802692cb>{driver_attach+75}
<ffffffff802696c9>{bus_add_driver+169} <ffffffff802025e3>{pci_register_driver+131}
<ffffffff88074010>{:ata_piix:piix_init+16} <ffffffff80152c58>{sys_init_module+344}
<ffffffff8010e52a>{system_call+126}
ffff81011e49f4a0: redzone 1: 0x5a2cf071, redzone 2: 0x5a2cf071.
and..
SELinux: initialized (dev sysfs, type sysfs), uses genfs_contexts
Slab corruption: start=ffff81011e49f4a8, len=2048
Redzone: 0x170fc2a5/0x170fc2a5.
Last user: [<ffffffff8015290c>](load_module+0x180c/0x1a00)
000: 38 e0 35 80 ff ff ff ff b8 f9 49 1e 01 81 ff ff
010: e0 f4 49 1e 01 81 ff ff 80 e7 08 88 ff ff ff ff
020: 24 01 00 00 5a 5a 5a 5a 10 0b 15 80 ff ff ff ff
030: 00 00 00 00 00 00 00 00 2e 74 65 78 74 00 5a 5a
040: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
050: 5a 5a 5a 5a 5a 5a 5a 5a 00 e0 07 88 ff ff ff ff
Prev obj: start=ffff81011e49ec90, len=2048
Redzone: 0x170fc2a5/0x170fc2a5.
Last user: [<ffffffff88064c79>](ata_probe_ent_alloc+0x29/0xc0 [libata])
000: 90 ec 49 1e 01 81 ff ff 90 ec 49 1e 01 81 ff ff
010: 80 8e 26 08 00 81 ff ff 60 24 07 88 ff ff ff ff
slab error in cache_alloc_debugcheck_after(): cache `size-2048': double free, or memory outside object was overwritten
Call Trace:<ffffffff802d52c1>{alloc_skb+81} <ffffffff8016375a>{cache_alloc_debugcheck_after+186}
<ffffffff801638e8>{__kmalloc+216} <ffffffff802d52c1>{alloc_skb+81}
<ffffffff801f66e4>{send_uevent+84} <ffffffff801f6bb0>{kobject_hotplug+592}
<ffffffff801f635d>{kobject_add+349} <ffffffff8026a377>{class_device_add+135}
<ffffffff8026ab0d>{class_simple_device_add+317} <ffffffff80133a53>{__wake_up+67}
<ffffffff8023ab8d>{init_dev+1197} <ffffffff80243f3b>{vcs_make_devfs+43}
<ffffffff8024941e>{con_open+158} <ffffffff8023bebe>{tty_open+574}
<ffffffff801e51ea>{selinux_file_alloc_security+42}
<ffffffff80188d79>{chrdev_open+393} <ffffffff801e51ea>{selinux_file_alloc_security+42}
<ffffffff8017ef46>{dentry_open+246} <ffffffff8017f09e>{filp_open+62}
<ffffffff8017f18b>{get_unused_fd+219} <ffffffff8017f2dc>{sys_open+76}
<ffffffff8010e52a>{system_call+126}
ffff81011e49f4a0: redzone 1: 0x170fc2a5, redzone 2: 0x170fc2a5.
SELinux: initialized (dev usbfs, type usbfs), uses genfs_contexts
This box got new hardware just before xmas, and a >24hr run of memtest86
so I can probably rule out bad memory.
Dave
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: slab corruption on 2.6.10-rc4-bk7
2005-02-19 22:16 slab corruption on 2.6.10-rc4-bk7 Dave Jones
@ 2005-02-19 22:41 ` Andrew Morton
2005-02-20 2:42 ` Jeff Garzik
2005-02-21 4:28 ` [PATCH] " Jeff Garzik
0 siblings, 2 replies; 5+ messages in thread
From: Andrew Morton @ 2005-02-19 22:41 UTC (permalink / raw)
To: Dave Jones; +Cc: linux-kernel, Jeff Garzik
Dave Jones <davej@redhat.com> wrote:
>
> (This has actually been there for a while, but I only
> noticed it in dmesg this morning).
> During boot on a dual em64t I see ..
>
> scsi2 : ata_piix
> isa bounce pool size: 16 pages
> slab error in cache_free_debugcheck(): cache `size-2048': double free, or memory outside object was overwritten
>
> Call Trace:<ffffffff80163448>{cache_free_debugcheck+392} <ffffffff801646aa>{kfree+234}
> <ffffffff88065189>{:libata:ata_pci_init_one+937} <ffffffff801fe9ea>{pci_bus_read_config_word+122}
> <ffffffff880707f2>{:ata_piix:piix_init_one+498} <ffffffff80202926>{pci_device_probe+134}
> <ffffffff802691ad>{driver_probe_device+77} <ffffffff802692cb>{driver_attach+75}
> <ffffffff802696c9>{bus_add_driver+169} <ffffffff802025e3>{pci_register_driver+131}
> <ffffffff88074010>{:ata_piix:piix_init+16} <ffffffff80152c58>{sys_init_module+344}
> <ffffffff8010e52a>{system_call+126}
> ffff81011e49f4a0: redzone 1: 0x5a2cf071, redzone 2: 0x5a2cf071.
>
It's plain to see how ata_pci_init_one() will free `probe_ent' twice. Jeff
wanna fix that up please? A naive fix would be
--- 25/drivers/scsi/libata-core.c~a 2005-02-19 14:41:08.000000000 -0800
+++ 25-akpm/drivers/scsi/libata-core.c 2005-02-19 14:41:24.000000000 -0800
@@ -3743,8 +3743,6 @@ int ata_pci_init_one (struct pci_dev *pd
if (legacy_mode) {
if (legacy_mode & (1 << 0))
ata_device_add(probe_ent);
- else
- kfree(probe_ent);
if (legacy_mode & (1 << 1))
ata_device_add(probe_ent2);
else
_
>
>
> SELinux: initialized (dev sysfs, type sysfs), uses genfs_contexts
> Slab corruption: start=ffff81011e49f4a8, len=2048
> Redzone: 0x170fc2a5/0x170fc2a5.
> Last user: [<ffffffff8015290c>](load_module+0x180c/0x1a00)
> 000: 38 e0 35 80 ff ff ff ff b8 f9 49 1e 01 81 ff ff
> 010: e0 f4 49 1e 01 81 ff ff 80 e7 08 88 ff ff ff ff
> 020: 24 01 00 00 5a 5a 5a 5a 10 0b 15 80 ff ff ff ff
> 030: 00 00 00 00 00 00 00 00 2e 74 65 78 74 00 5a 5a
> 040: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
> 050: 5a 5a 5a 5a 5a 5a 5a 5a 00 e0 07 88 ff ff ff ff
> Prev obj: start=ffff81011e49ec90, len=2048
> Redzone: 0x170fc2a5/0x170fc2a5.
> Last user: [<ffffffff88064c79>](ata_probe_ent_alloc+0x29/0xc0 [libata])
> 000: 90 ec 49 1e 01 81 ff ff 90 ec 49 1e 01 81 ff ff
> 010: 80 8e 26 08 00 81 ff ff 60 24 07 88 ff ff ff ff
> slab error in cache_alloc_debugcheck_after(): cache `size-2048': double free, or memory outside object was overwritten
>
> Call Trace:<ffffffff802d52c1>{alloc_skb+81} <ffffffff8016375a>{cache_alloc_debugcheck_after+186}
> <ffffffff801638e8>{__kmalloc+216} <ffffffff802d52c1>{alloc_skb+81}
> <ffffffff801f66e4>{send_uevent+84} <ffffffff801f6bb0>{kobject_hotplug+592}
> <ffffffff801f635d>{kobject_add+349} <ffffffff8026a377>{class_device_add+135}
> <ffffffff8026ab0d>{class_simple_device_add+317} <ffffffff80133a53>{__wake_up+67}
> <ffffffff8023ab8d>{init_dev+1197} <ffffffff80243f3b>{vcs_make_devfs+43}
> <ffffffff8024941e>{con_open+158} <ffffffff8023bebe>{tty_open+574}
> <ffffffff801e51ea>{selinux_file_alloc_security+42}
> <ffffffff80188d79>{chrdev_open+393} <ffffffff801e51ea>{selinux_file_alloc_security+42}
> <ffffffff8017ef46>{dentry_open+246} <ffffffff8017f09e>{filp_open+62}
> <ffffffff8017f18b>{get_unused_fd+219} <ffffffff8017f2dc>{sys_open+76}
> <ffffffff8010e52a>{system_call+126}
> ffff81011e49f4a0: redzone 1: 0x170fc2a5, redzone 2: 0x170fc2a5.
> SELinux: initialized (dev usbfs, type usbfs), uses genfs_contexts
>
Last user: ata_probe_ent_alloc(). Hopefully that's the same bug.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: slab corruption on 2.6.10-rc4-bk7
2005-02-19 22:41 ` Andrew Morton
@ 2005-02-20 2:42 ` Jeff Garzik
2005-02-21 4:28 ` [PATCH] " Jeff Garzik
1 sibling, 0 replies; 5+ messages in thread
From: Jeff Garzik @ 2005-02-20 2:42 UTC (permalink / raw)
To: Andrew Morton; +Cc: Dave Jones, linux-kernel
On Sat, Feb 19, 2005 at 02:41:47PM -0800, Andrew Morton wrote:
> Dave Jones <davej@redhat.com> wrote:
> >
> > (This has actually been there for a while, but I only
> > noticed it in dmesg this morning).
> > During boot on a dual em64t I see ..
> >
> > scsi2 : ata_piix
> > isa bounce pool size: 16 pages
> > slab error in cache_free_debugcheck(): cache `size-2048': double free, or memory outside object was overwritten
> >
> > Call Trace:<ffffffff80163448>{cache_free_debugcheck+392} <ffffffff801646aa>{kfree+234}
> > <ffffffff88065189>{:libata:ata_pci_init_one+937} <ffffffff801fe9ea>{pci_bus_read_config_word+122}
> > <ffffffff880707f2>{:ata_piix:piix_init_one+498} <ffffffff80202926>{pci_device_probe+134}
> > <ffffffff802691ad>{driver_probe_device+77} <ffffffff802692cb>{driver_attach+75}
> > <ffffffff802696c9>{bus_add_driver+169} <ffffffff802025e3>{pci_register_driver+131}
> > <ffffffff88074010>{:ata_piix:piix_init+16} <ffffffff80152c58>{sys_init_module+344}
> > <ffffffff8010e52a>{system_call+126}
> > ffff81011e49f4a0: redzone 1: 0x5a2cf071, redzone 2: 0x5a2cf071.
> >
>
> It's plain to see how ata_pci_init_one() will free `probe_ent' twice. Jeff
> wanna fix that up please? A naive fix would be
I'll take a look at it tomorrow when I get back home; it's not that
simple, as if you look through BK history, you'll see an almost reverse
patch applied, to fix another double-kfree (author: zaitcev).
Jeff
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] Re: slab corruption on 2.6.10-rc4-bk7
2005-02-19 22:41 ` Andrew Morton
2005-02-20 2:42 ` Jeff Garzik
@ 2005-02-21 4:28 ` Jeff Garzik
2005-02-21 4:35 ` [PATCH #2] " Jeff Garzik
1 sibling, 1 reply; 5+ messages in thread
From: Jeff Garzik @ 2005-02-21 4:28 UTC (permalink / raw)
To: Andrew Morton; +Cc: Dave Jones, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 1311 bytes --]
Andrew Morton wrote:
> Dave Jones <davej@redhat.com> wrote:
>
>>(This has actually been there for a while, but I only
>> noticed it in dmesg this morning).
>>During boot on a dual em64t I see ..
>>
>>scsi2 : ata_piix
>>isa bounce pool size: 16 pages
>>slab error in cache_free_debugcheck(): cache `size-2048': double free, or memory outside object was overwritten
>>
>>Call Trace:<ffffffff80163448>{cache_free_debugcheck+392} <ffffffff801646aa>{kfree+234}
>> <ffffffff88065189>{:libata:ata_pci_init_one+937} <ffffffff801fe9ea>{pci_bus_read_config_word+122}
>> <ffffffff880707f2>{:ata_piix:piix_init_one+498} <ffffffff80202926>{pci_device_probe+134}
>> <ffffffff802691ad>{driver_probe_device+77} <ffffffff802692cb>{driver_attach+75}
>> <ffffffff802696c9>{bus_add_driver+169} <ffffffff802025e3>{pci_register_driver+131}
>> <ffffffff88074010>{:ata_piix:piix_init+16} <ffffffff80152c58>{sys_init_module+344}
>> <ffffffff8010e52a>{system_call+126}
>>ffff81011e49f4a0: redzone 1: 0x5a2cf071, redzone 2: 0x5a2cf071.
>>
>
>
> It's plain to see how ata_pci_init_one() will free `probe_ent' twice. Jeff
> wanna fix that up please? A naive fix would be
Here's the initial fix... about to test with some other fixes here.
Anybody who is seeing this wanna give it a try?
Jeff
[-- Attachment #2: libata_probe_ent_free_fix.patch --]
[-- Type: text/x-patch, Size: 331 bytes --]
===== drivers/scsi/libata-core.c 1.116 vs edited =====
--- 1.116/drivers/scsi/libata-core.c 2005-02-01 20:23:51 -05:00
+++ edited/drivers/scsi/libata-core.c 2005-02-20 23:25:52 -05:00
@@ -3751,8 +3751,8 @@
kfree(probe_ent2);
} else {
ata_device_add(probe_ent);
+ kfree(probe_ent);
}
- kfree(probe_ent);
return 0;
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH #2] Re: slab corruption on 2.6.10-rc4-bk7
2005-02-21 4:28 ` [PATCH] " Jeff Garzik
@ 2005-02-21 4:35 ` Jeff Garzik
0 siblings, 0 replies; 5+ messages in thread
From: Jeff Garzik @ 2005-02-21 4:35 UTC (permalink / raw)
To: Andrew Morton; +Cc: Dave Jones, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 41 bytes --]
Actually, here's a better fix.
Jeff
[-- Attachment #2: libata_probe_ent_free_fix.patch --]
[-- Type: text/x-patch, Size: 621 bytes --]
===== drivers/scsi/libata-core.c 1.116 vs edited =====
--- 1.116/drivers/scsi/libata-core.c 2005-02-01 20:23:51 -05:00
+++ edited/drivers/scsi/libata-core.c 2005-02-20 23:34:32 -05:00
@@ -2800,7 +2800,7 @@
return 1;
/* fall through */
-
+
default:
return 0;
}
@@ -3743,16 +3743,13 @@
if (legacy_mode) {
if (legacy_mode & (1 << 0))
ata_device_add(probe_ent);
- else
- kfree(probe_ent);
if (legacy_mode & (1 << 1))
ata_device_add(probe_ent2);
- else
- kfree(probe_ent2);
- } else {
+ } else
ata_device_add(probe_ent);
- }
+
kfree(probe_ent);
+ kfree(probe_ent2);
return 0;
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2005-02-21 4:35 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-02-19 22:16 slab corruption on 2.6.10-rc4-bk7 Dave Jones
2005-02-19 22:41 ` Andrew Morton
2005-02-20 2:42 ` Jeff Garzik
2005-02-21 4:28 ` [PATCH] " Jeff Garzik
2005-02-21 4:35 ` [PATCH #2] " Jeff Garzik
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox