Skip to contentSkip to conversation

BUG: unable to handle kernel NULL pointer dereference at 0000000000000040

DISCUSSION nirinA raseliarison 4 participants 10 messages Jun 14, 2013 – Jun 15, 2013 Driver core

lore
Share & tools
lore
#0 lore raw

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast
Ethernet controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1 (GCC) )
#1 SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013 x86_64
Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks, -----------------8<------------------------------8<---------------------------------------

[   57.877560] BUG: unable to handle kernel NULL pointer dereference at 0000000000000040
[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev r8169 microcode mii
[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614 #1 To be filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01
[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX: 0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI: 0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09: 00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12: ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15: ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000) knlGS:0000000000000000
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4: 00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo ffff8802119a6000, task ffff8802158fe250)
[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0 0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0 ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098 ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79 ff ff ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48 89 e5 <f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f
[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---
Signature · 1 line
nirinA
#1 lore raw

[+cc Ming, Hayes, Francois, r8169 list]

Quoted 91 lines · nirinA raseliarison

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet
controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1 (GCC) ) #1
SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013 x86_64
Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks, -----------------8<------------------------------8<---------------------------------------

[   57.877560] BUG: unable to handle kernel NULL pointer dereference at
0000000000000040
[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev r8169
microcode mii
[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614 #1 To be
filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01
[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>]
fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX:
0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI:
0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09:
00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12:
ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15:
ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000)
knlGS:0000000000000000
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4:
00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo ffff8802119a6000,
task ffff8802158fe250)
[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0
0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0
ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098
ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79 ff ff
ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48 89 e5
<f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f
[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---

--
nirinA
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#2 lore raw
Quoted 88 lines · Bjorn Helgaas

[+cc Ming, Hayes, Francois, r8169 list]

Quoted 84 lines · nirinA raseliarison

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet
controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1 (GCC) ) #1
SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013 x86_64
Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks, -----------------8<------------------------------8<---------------------------------------

[   57.877560] BUG: unable to handle kernel NULL pointer dereference at
0000000000000040
[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev r8169
microcode mii
[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614 #1 To be
filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01
[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>]
fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX:
0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI:
0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09:
00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12:
ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15:
ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000)
knlGS:0000000000000000
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4:
00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo ffff8802119a6000,
task ffff8802158fe250)
[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0
0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0
ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098
ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79 ff ff
ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48 89 e5
<f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f
[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---

Please try the following patch.

[ Bjorn, sorry I dropped you from the recipient list, but unfortunately Google still considers me to be a spammer and doesn't let me send any e-mail to you ]

Guenter

----------

Quoted 1 line

From 9feae0b1b33721573c41fbf2323db2a12c34c725 Mon Sep 17 00:00:00 2001

From: Guenter Roeck <[email protected]>
Date: Fri, 14 Jun 2013 08:39:06 -0700
Subject: [PATCH] firmware: Fix race condition in firmware_loading_store

Fix:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000040
IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
...
Call Trace:
[<ffffffff81491917>] firmware_loading_store+0x77/0x150
[<ffffffff81483063>] dev_attr_store+0x13/0x20
[<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[<ffffffff81133e8a>] vfs_write+0x9a/0x160
[<ffffffff81134164>] sys_write+0x44/0x90
[<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
---
 drivers/base/firmware_class.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/base/firmware_class.c +2 -1
index 4b1f926..f34b489 100644
@@ -570,12 +570,13 @@ static ssize_t firmware_loading_store(struct device *dev,
 				      const char *buf, size_t count)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	struct firmware_buf *fw_buf = fw_priv->buf;
 	int loading = simple_strtol(buf, NULL, 10);
+	struct firmware_buf *fw_buf;
 	int i;
 
 	mutex_lock(&fw_lock);
 
+	fw_buf = fw_priv->buf;
 	if (!fw_buf)
 		goto out;
 
Signature · 1 line
1.7.9.7
#3 lore raw
Quoted 93 lines · Guenter Roeck
Quoted 92 lines · Bjorn Helgaas

[+cc Ming, Hayes, Francois, r8169 list]

Quoted 4 lines · nirinA raseliarison

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast

Ethernet

Quoted 7 lines

controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

Quoted 8 lines

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1

(GCC) ) #1

Quoted 4 lines

SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013

x86_64

Quoted 4 lines

Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks,

-----------------8<------------------------------8<---------------------------------------

Quoted 2 lines
[   57.877560] BUG: unable to handle kernel NULL pointer dereference  

at

Quoted 5 lines

0000000000000040

[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev  

r8169

Quoted 3 lines

microcode mii

[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614  

#1 To be

Quoted 16 lines

filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01

[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>]
fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX:
0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI:
0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09:
00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12:
ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15:
ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000)

knlGS:0000000000000000

Quoted 8 lines
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4:
00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo  

ffff8802119a6000,

Quoted 16 lines

task ffff8802158fe250)

[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0
0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0
ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098
ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79  

ff ff

Quoted 1 line

ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48

89 e5

Quoted 8 lines
Quoted 6 lines
Quoted 6 lines

<f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f

[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---

Please try the following patch.

patch applied and no longer have the bug message when i reboot and wake up the ethernet controller.

thanks,

Quoted 49 lines

[ Bjorn, sorry I dropped you from the recipient list, but unfortunately Google still considers me to be a spammer and doesn't let me send any e-mail to you ]

Guenter

----------

From 9feae0b1b33721573c41fbf2323db2a12c34c725 Mon Sep 17 00:00:00 2001
From: Guenter Roeck <[email protected]>
Date: Fri, 14 Jun 2013 08:39:06 -0700
Subject: [PATCH] firmware: Fix race condition in firmware_loading_store

Fix:

BUG: unable to handle kernel NULL pointer dereference at 0000000000000040
IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
...
Call Trace:
[<ffffffff81491917>] firmware_loading_store+0x77/0x150
[<ffffffff81483063>] dev_attr_store+0x13/0x20
[<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[<ffffffff81133e8a>] vfs_write+0x9a/0x160
[<ffffffff81134164>] sys_write+0x44/0x90
[<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
---
 drivers/base/firmware_class.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/base/firmware_class.c +2 -1
index 4b1f926..f34b489 100644
@@ -570,12 +570,13 @@ static ssize_t firmware_loading_store(struct  
 device *dev,
 				      const char *buf, size_t count)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	struct firmware_buf *fw_buf = fw_priv->buf;
 	int loading = simple_strtol(buf, NULL, 10);
+	struct firmware_buf *fw_buf;
 	int i;
 	mutex_lock(&fw_lock);
+	fw_buf = fw_priv->buf;
 	if (!fw_buf)
 		goto out;
 
Signature · 1 line
nirinA
#4 lore raw
Quoted 2 lines · nirinA raseliarison

patch applied and no longer have the bug message when i reboot and wake up the ethernet controller.

I am wondering if Guenter's patch can fix the race really, but I'd like to see Guenter's explanation on his patch.

The race should be caused by below:

- request timeout triggered by internal timer

- user space aborts the requests before the line in _request_firmware_load()

             fw_priv->buf = NULL

which is run in timeout path

- then the abort() called from firmware_loading_store() may use a freed fw buf since the timeout path will free the fw buffer.

Considered clearing 'fw_priv->buf' in _request_firmware_load()() isn't protected by fw_lock now, so Guenter's patch can't avoid the race entirely.

Thanks,
--
Ming Lei

#5 lore raw
Quoted 17 lines · Ming Lei
Quoted 2 lines · nirinA raseliarison

patch applied and no longer have the bug message when i reboot and wake up the ethernet controller.

I am wondering if Guenter's patch can fix the race really, but I'd like to see Guenter's explanation on his patch.

The race should be caused by below:

- request timeout triggered by internal timer

- user space aborts the requests before the line in _request_firmware_load()

             fw_priv->buf = NULL
Quoted 8 lines

which is run in timeout path

- then the abort() called from firmware_loading_store() may use a freed fw buf since the timeout path will free the fw buffer.

Considered clearing 'fw_priv->buf' in _request_firmware_load()() isn't protected by fw_lock now, so Guenter's patch can't avoid the race entirely.

I agree; my patch only protects one specific path, and was based on the observation that access to fw_priv->buf is protected elsewhwere in the code. My suspicion was that fw_priv->buf was freed while waiting for the mutex in firmware_loading_store().

Your patch is more comprehensive.

Guenter

#6 lore raw
Quoted 26 lines · Guenter Roeck
Quoted 17 lines · Ming Lei
Quoted 2 lines · nirinA raseliarison

patch applied and no longer have the bug message when i reboot and wake up the ethernet controller.

I am wondering if Guenter's patch can fix the race really, but I'd like to see Guenter's explanation on his patch.

The race should be caused by below:

- request timeout triggered by internal timer

- user space aborts the requests before the line in _request_firmware_load()

             fw_priv->buf = NULL
Quoted 8 lines

which is run in timeout path

- then the abort() called from firmware_loading_store() may use a freed fw buf since the timeout path will free the fw buffer.

Considered clearing 'fw_priv->buf' in _request_firmware_load()() isn't protected by fw_lock now, so Guenter's patch can't avoid the race entirely.

Quoted 6 lines

I agree; my patch only protects one specific path, and was based on the observation that access to fw_priv->buf is protected elsewhwere in the code. My suspicion was that fw_priv->buf was freed while waiting for the mutex in firmware_loading_store().

Your patch is more comprehensive.

OK, thanks for your reply.

I will post out one version for merge, and this one moves the "fw_priv->buf = NULL;" into fw_load_abort() for simplifying change.

Thanks,
--
Ming Lei

#7 lore raw
Quoted 34 lines · Ming Lei
Quoted 33 lines · Guenter Roeck
Quoted 18 lines · Ming Lei
Quoted 2 lines · nirinA raseliarison

patch applied and no longer have the bug message when i reboot and wake up the ethernet controller.

I am wondering if Guenter's patch can fix the race really, but I'd like to see Guenter's explanation on his patch.

The race should be caused by below:

- request timeout triggered by internal timer

- user space aborts the requests before the line in _request_firmware_load()

             fw_priv->buf = NULL
Quoted 9 lines

which is run in timeout path

- then the abort() called from firmware_loading_store() may use a freed fw buf
since the timeout path will free the fw buffer.

Considered clearing 'fw_priv->buf' in _request_firmware_load()() isn't protected by fw_lock now, so Guenter's patch can't avoid the race entirely.

I agree; my patch only protects one specific path, and was based on the observation that access to fw_priv->buf is protected elsewhwere in the code. My suspicion was that fw_priv->buf was freed while waiting for the mutex in firmware_loading_store().

Quoted 6 lines
Quoted 1 line

Your patch is more comprehensive.

OK, thanks for your reply.

I will post out one version for merge, and this one moves the "fw_priv->buf = NULL;" into fw_load_abort() for simplifying change.

this is just to let you know that i've tested Ming Lei's latest patch. thank you very much for the fix and the explanation.

Signature · 1 line
nirinA
#8 lore raw
In reply to Bjorn Helgaas
Quoted 87 lines · Bjorn Helgaas

[+cc Ming, Hayes, Francois, r8169 list]

Quoted 83 lines · nirinA raseliarison

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet
controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1 (GCC) ) #1
SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013 x86_64
Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks, -----------------8<------------------------------8<---------------------------------------

[   57.877560] BUG: unable to handle kernel NULL pointer dereference at
0000000000000040
[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev r8169
microcode mii
[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614 #1 To be
filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01
[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>]
fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX:
0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI:
0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09:
00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12:
ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15:
ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000)
knlGS:0000000000000000
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4:
00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo ffff8802119a6000,
task ffff8802158fe250)
[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0
0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0
ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098
ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79 ff ff
ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48 89 e5
<f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f
[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---

Looks it is a double abort race, could you try below patch? (also attached for applying)

--

drivers/base/firmware_class.c +9 -3
index 6ede229..a217ba8 100644
@@ -550,7 +550,12 @@ static ssize_t firmware_loading_show(struct device *dev,
 				     struct device_attribute *attr, char *buf)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	int loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	int loading = 0;
+
+	mutex_lock(&fw_lock);
+	if (fw_priv->buf)
+		loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	mutex_unlock(&fw_lock);
 
 	return sprintf(buf, "%d\n", loading);
 }
@@ -592,12 +597,12 @@ static ssize_t firmware_loading_store(struct device *dev,
 				      const char *buf, size_t count)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	struct firmware_buf *fw_buf = fw_priv->buf;
+	struct firmware_buf *fw_buf;
 	int loading = simple_strtol(buf, NULL, 10);
 	int i;
 
 	mutex_lock(&fw_lock);
-
+	fw_buf = fw_priv->buf;
 	if (!fw_buf)
 		goto out;
 
@@ -636,6 +641,7 @@ static ssize_t firmware_loading_store(struct device *dev,
 		/* fallthrough */
 	case -1:
 		fw_load_abort(fw_buf);
+		fw_priv->buf = NULL;
 		break;
 	}
 out:
@@ -704,6 +710,7 @@ static int fw_realloc_buffer(struct firmware_priv

*fw_priv, int min_size)
GFP_KERNEL);
if (!new_pages) {
fw_load_abort(buf);
+ fw_priv->buf = NULL;
return -ENOMEM;
}
memcpy(new_pages, buf->pages,
@@ -721,6 +728,7 @@ static int fw_realloc_buffer(struct firmware_priv *fw_priv, int min_size)

if (!buf->pages[buf->nr_pages]) {
fw_load_abort(buf);
+ fw_priv->buf = NULL;
return -ENOMEM;
}
buf->nr_pages++;
@@ -805,6 +813,7 @@ static void firmware_class_timeout_work(struct work_struct *work)
return;
}
fw_load_abort(fw_priv->buf);
+ fw_priv->buf = NULL;
mutex_unlock(&fw_lock);
}

@@ -886,8 +895,6 @@ static int _request_firmware_load(struct firmware_priv *fw_priv, bool uevent,

cancel_delayed_work_sync(&fw_priv->timeout_work);

- fw_priv->buf = NULL;
-
device_remove_file(f_dev, &dev_attr_loading);
err_del_bin_attr:
device_remove_bin_file(f_dev, &firmware_attr_data);

Thanks,
--
Ming Lei

drivers/base/firmware_class.c +12 -5
index 6ede229..a217ba8 100644
@@ -550,7 +550,12 @@ static ssize_t firmware_loading_show(struct device *dev,
 				     struct device_attribute *attr, char *buf)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	int loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	int loading = 0;
+
+	mutex_lock(&fw_lock);
+	if (fw_priv->buf)
+		loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	mutex_unlock(&fw_lock);
 
 	return sprintf(buf, "%d\n", loading);
 }
@@ -592,12 +597,12 @@ static ssize_t firmware_loading_store(struct device *dev,
 				      const char *buf, size_t count)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	struct firmware_buf *fw_buf = fw_priv->buf;
+	struct firmware_buf *fw_buf;
 	int loading = simple_strtol(buf, NULL, 10);
 	int i;
 
 	mutex_lock(&fw_lock);
-
+	fw_buf = fw_priv->buf;
 	if (!fw_buf)
 		goto out;
 
@@ -636,6 +641,7 @@ static ssize_t firmware_loading_store(struct device *dev,
 		/* fallthrough */
 	case -1:
 		fw_load_abort(fw_buf);
+		fw_priv->buf = NULL;
 		break;
 	}
 out:
@@ -704,6 +710,7 @@ static int fw_realloc_buffer(struct firmware_priv *fw_priv, int min_size)
 				    GFP_KERNEL);
 		if (!new_pages) {
 			fw_load_abort(buf);
+			fw_priv->buf = NULL;
 			return -ENOMEM;
 		}
 		memcpy(new_pages, buf->pages,
@@ -721,6 +728,7 @@ static int fw_realloc_buffer(struct firmware_priv *fw_priv, int min_size)
 
 		if (!buf->pages[buf->nr_pages]) {
 			fw_load_abort(buf);
+			fw_priv->buf = NULL;
 			return -ENOMEM;
 		}
 		buf->nr_pages++;
@@ -805,6 +813,7 @@ static void firmware_class_timeout_work(struct work_struct *work)
 		return;
 	}
 	fw_load_abort(fw_priv->buf);
+	fw_priv->buf = NULL;
 	mutex_unlock(&fw_lock);
 }
 
@@ -886,8 +895,6 @@ static int _request_firmware_load(struct firmware_priv *fw_priv, bool uevent,
 
 	cancel_delayed_work_sync(&fw_priv->timeout_work);
 
-	fw_priv->buf = NULL;
-
 	device_remove_file(f_dev, &dev_attr_loading);
 err_del_bin_attr:
 	device_remove_bin_file(f_dev, &firmware_attr_data);
#9 lore raw
Quoted 92 lines · Ming Lei
Quoted 90 lines · Bjorn Helgaas

[+cc Ming, Hayes, Francois, r8169 list]

Quoted 86 lines · nirinA raseliarison

hello there,
i have this ethernet controler:

Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast
Ethernet
controller (rev 05)

that uses the r8169 module.
it works fine, but sometimes after a reboot and issueing:

ifconfig eth0 192.168.1.1 up

i got the message below. after another reboot the
message disappears. i also get the same message this 3.9.5 and 3.9.4.

it seems i catch my first oops and don't know what to do with it. currently running:

cat /proc/version
Linux version 3.9.6.20130614 (root@supernova) (gcc version 4.8.1
(GCC) ) #1
SMP Fri Jun 14 09:14:50 EAT 2013

uname -a
Linux supernova 3.9.6.20130614 #1 SMP Fri Jun 14 09:14:50 EAT 2013
x86_64
Intel(R) Celeron(R) CPU G1610 @ 2.60GHz GenuineIntel GNU/Linux

thanks, -----------------8<------------------------------8<---------------------------------------

[   57.877560] BUG: unable to handle kernel NULL pointer dereference at
0000000000000040
[   57.877603] IP: [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.877634] PGD 21330a067 PUD 211a3a067 PMD 0
[   57.877660] Oops: 0002 [#1] SMP
[   57.877681] Modules linked in: fuse coretemp kvm_intel kvm evdev r8169
microcode mii
[   57.877735] CPU 0
[   57.877746] Pid: 1950, comm: firmware Not tainted 3.9.6.20130614 #1 To be
filled by O.E.M. To be filled by O.E.M./ONDA H61V Ver:4.01
[   57.877790] RIP: 0010:[<ffffffff81491844>]  [<ffffffff81491844>]
fw_load_abort.isra.5+0x4/0x20
[   57.877824] RSP: 0018:ffff8802119a7e80  EFLAGS: 00010246
[   57.877844] RAX: ffff8802158fe250 RBX: ffff880211a03b40 RCX:
0000000000000000
[   57.877869] RDX: ffffffff81c742c8 RSI: ffff8802158fe250 RDI:
0000000000000000
[   57.877895] RBP: ffff8802119a7e80 R08: ffff8802119a6000 R09:
00000000000005aa
[   57.877920] R10: 0000000000000000 R11: 0000000000000000 R12:
ffffffffffffffff
[   57.877945] R13: ffff880213d34088 R14: 0000000000000003 R15:
ffff88020eafc230
[   57.877970] FS:  00007f3c6cb2a740(0000) GS:ffff88021f200000(0000)
knlGS:0000000000000000
[   57.877998] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   57.878019] CR2: 0000000000000040 CR3: 0000000203155000 CR4:
00000000001407f0
[   57.878044] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[   57.878069] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[   57.878094] Process firmware (pid: 1950, threadinfo ffff8802119a6000,
task ffff8802158fe250)
[   57.878124] Stack:
[   57.878133]  ffff8802119a7eb0 ffffffff81491917 ffff880211a4d5a0
0000000000000003
[   57.878168]  ffff8802119a7f50 ffffffff818765a0 ffff8802119a7ec0
ffffffff81483063
[   57.878203]  ffff8802119a7f08 ffffffff8119bc9e ffff880213d34098
ffff880211a4d5c0
[   57.878237] Call Trace:
[   57.878251]  [<ffffffff81491917>] firmware_loading_store+0x77/0x150
[   57.878275]  [<ffffffff81483063>] dev_attr_store+0x13/0x20
[   57.878297]  [<ffffffff8119bc9e>] sysfs_write_file+0xce/0x140
[   57.878320]  [<ffffffff81133e8a>] vfs_write+0x9a/0x160
[   57.878340]  [<ffffffff81134164>] sys_write+0x44/0x90
[   57.878360]  [<ffffffff817d70ed>] system_call_fastpath+0x1a/0x1f
[   57.879379] Code: 6b ff ff ff 48 89 df 31 db e8 b9 b0 c9 ff e9 79 ff ff
ff 0f 1f 40 00 48 83 c4 10 5b 41 5c 41 5d 41 5e 5d c3 0f 1f 00 55 48 89 e5
<f0> 80 4f 40 04 48 83 c7 18 e8 8e a9 bd ff 5d c3 66 66 66 2e 0f
[   57.881753] RIP  [<ffffffff81491844>] fw_load_abort.isra.5+0x4/0x20
[   57.882888]  RSP <ffff8802119a7e80>
[   57.884019] CR2: 0000000000000040
[   57.885166] ---[ end trace 6705f6d4ce6b6a12 ]---
Quoted 2 lines

Looks it is a double abort race, could you try below patch? (also attached for applying)

i've also applied this patch and up to now, after
reboot a few times all thing seems to work fine.

thanks,

Quoted 87 lines

--

drivers/base/firmware_class.c +6 -1
index 6ede229..a217ba8 100644
@@ -550,7 +550,12 @@ static ssize_t firmware_loading_show(struct device  
 *dev,
 				     struct device_attribute *attr, char *buf)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	int loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	int loading = 0;
+
+	mutex_lock(&fw_lock);
+	if (fw_priv->buf)
+		loading = test_bit(FW_STATUS_LOADING, &fw_priv->buf->status);
+	mutex_unlock(&fw_lock);
 
 	return sprintf(buf, "%d\n", loading);

}

drivers/base/firmware_class.c +3 -2
@@ -592,12 +597,12 @@ static ssize_t firmware_loading_store(struct  
 device *dev,
 				      const char *buf, size_t count)
 {
 	struct firmware_priv *fw_priv = to_firmware_priv(dev);
-	struct firmware_buf *fw_buf = fw_priv->buf;
+	struct firmware_buf *fw_buf;
 	int loading = simple_strtol(buf, NULL, 10);
 	int i;
 
 	mutex_lock(&fw_lock);
-
+	fw_buf = fw_priv->buf;
 	if (!fw_buf)
 		goto out;
@@ -636,6 +641,7 @@ static ssize_t firmware_loading_store(struct device  
 *dev,
 		/* fallthrough */
 	case -1:
 		fw_load_abort(fw_buf);
+		fw_priv->buf = NULL;
 		break;
 	}

out:

drivers/base/firmware_class.c +1 -0
@@ -704,6 +710,7 @@ static int fw_realloc_buffer(struct firmware_priv
 *fw_priv, int min_size)
 				    GFP_KERNEL);
 		if (!new_pages) {
 			fw_load_abort(buf);
+			fw_priv->buf = NULL;
 			return -ENOMEM;
 		}
drivers/base/firmware_class.c +1 -0
 		memcpy(new_pages, buf->pages,
@@ -721,6 +728,7 @@ static int fw_realloc_buffer(struct firmware_priv
 *fw_priv, int min_size)
 
 		if (!buf->pages[buf->nr_pages]) {
 			fw_load_abort(buf);
+			fw_priv->buf = NULL;
 			return -ENOMEM;
 		}
drivers/base/firmware_class.c +1 -2
 		buf->nr_pages++;
@@ -805,6 +813,7 @@ static void firmware_class_timeout_work(struct
 work_struct *work)
 		return;
 	}
 	fw_load_abort(fw_priv->buf);
+	fw_priv->buf = NULL;
 	mutex_unlock(&fw_lock);
 }
@@ -886,8 +895,6 @@ static int _request_firmware_load(struct
 firmware_priv *fw_priv, bool uevent,
 
 	cancel_delayed_work_sync(&fw_priv->timeout_work);
 
-	fw_priv->buf = NULL;
-
 	device_remove_file(f_dev, &dev_attr_loading);
 err_del_bin_attr:

device_remove_bin_file(f_dev, &firmware_attr_data);

Thanks,
--
Ming Lei

Signature · 1 line
nirinA