Bugzilla – Bug 151111
Slab corruption: start=e0840f38, len=2048
Last modified: 2006-04-11 09:07:57 UTC
Today i found the following in my logs. I have no idea what it is, but it looks scary. Blocker on request from Michael Matz. Feb 15 11:07:59 strolchi klogd: Slab corruption: start=e0840f38, len=2048 Feb 15 11:07:59 strolchi klogd: Redzone: 0x5a2cf071/0x5a2cf071. Feb 15 11:07:59 strolchi klogd: Last user: [<c01f0e31>](release_mem+0xe6/0x1ce) Feb 15 11:07:59 strolchi klogd: 0b0: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b ff ff ff ff Feb 15 11:07:59 strolchi klogd: 0c0: 00 00 00 00 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b Feb 15 11:07:59 strolchi klogd: Prev obj: start=e084072c, len=2048 Feb 15 11:07:59 strolchi klogd: Redzone: 0x5a2cf071/0x5a2cf071. Feb 15 11:07:59 strolchi klogd: Last user: [<c022f926>](kfree_skbmem+0x8/0x5f) Feb 15 11:07:59 strolchi klogd: 000: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b Feb 15 11:07:59 strolchi klogd: 010: 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b Feb 15 11:07:59 strolchi klogd: Next obj: start=e0841744, len=2048 Feb 15 11:07:59 strolchi klogd: Redzone: 0x170fc2a5/0x170fc2a5. Feb 15 11:07:59 strolchi klogd: Last user: [<f91983eb>](tg3_alloc_rx_skb+0xb8/0x15b [tg3]) Feb 15 11:07:59 strolchi klogd: 000: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a Feb 15 11:07:59 strolchi klogd: 010: 5a 5a 00 12 79 59 37 34 00 0a e4 53 2e be 08 00
Can you reproduce this? What were you doing at the time of the corruption?
> I was connected via Bluetooth rfcomm => ppp to the internet That makes sense. The corruption seems to happen when closing a tty device (possibly the rfcomm device). So it could be related to rfcomm. > IIRC no other network connection was active (WLAN was probably off), but > NetworkManager might have been polling on the tg3 for link. It seems the tg3 driver was receiving data in the not too distant past. So feel this may be rfcomm related. Maybe you can test this a little (connecting/disconnecting)?
I saw some tty race fixed in rc3-git4 or git3.
the bt-rfcomm-ppp connection was active and not interrupted at this time, dialin was at 10:46:19, RAS was set up 10:46:41, tg3 was probably unplugged at 08:55:55: Feb 15 08:55:55 strolchi klogd: tg3: cable: Link is down. Feb 15 08:55:55 strolchi NetworkManager: <information> SWITCH: terminating current connection 'cable' because it's no longer va What i did around the corruption time was trying to ssh / rsync into suse, but that did not work well since the network inside was kaputt.
Well, the slab chunks that were corrupted were previously used by tg3 and a tty driver (which was closed). I will change this to NEEDINFO, please leave it in this state until you have a reproduce with the most recent KOTD (which as Karsten noted has some tty race fix in it). Thanks!
Any news on this one? It's been a few weeks...
i have not seen it again. But a new one, with a usb-serial converter. Just plug it in, "screen /dev/ttyUSB0", unplug. Boom. Mar 6 13:19:33 susi kernel: usb 3-1: USB disconnect, address 3 Mar 6 13:19:33 susi kernel: cp2101 3-1:1.0: device disconnected Mar 6 13:19:33 susi kernel: cp2101 ttyUSB0: cp2101_set_config - Unable to send request, request=0x0 size=2 result=-19 Mar 6 13:19:33 susi kernel: cp2101 ttyUSB0: cp2101 converter now disconnected from ttyUSB0 Mar 6 13:19:33 susi kernel: Slab corruption: start=c4eaa448, len=1024 Mar 6 13:19:33 susi kernel: Redzone: 0x5a2cf071/0x5a2cf071. Mar 6 13:19:33 susi kernel: Last user: [<c020a43e>](device_release+0x14/0x41) Mar 6 13:19:33 susi kernel: 010: 6b 6b 6b 6b 6b 6b 6b 6b 6c 6b 6b 6b 6b 6b 6b 6b Mar 6 13:19:33 susi kernel: Prev obj: start=c4eaa03c, len=1024 Mar 6 13:19:33 susi kernel: Redzone: 0x170fc2a5/0x170fc2a5. Mar 6 13:19:33 susi kernel: Last user: [<e117c66c>](usb_alloc_dev+0x1f/0x153 [usbcore]) Mar 6 13:19:33 susi kernel: 000: 01 00 00 00 30 00 00 00 00 00 00 00 00 00 00 00 Mar 6 13:19:33 susi kernel: 010: 00 00 00 00 07 00 00 00 02 00 00 00 00 00 00 00 Mar 6 13:19:33 susi kernel: Next obj: start=c4eaa854, len=1024 Mar 6 13:19:33 susi kernel: Redzone: 0x170fc2a5/0x170fc2a5. Mar 6 13:19:33 susi kernel: Last user: [<c0166447>](expand_files+0x102/0x2e6) Mar 6 13:19:33 susi kernel: 000: d4 b1 98 db d4 b1 98 db d4 b1 98 db 44 9f 9b d7 Mar 6 13:19:33 susi kernel: 010: 80 32 b3 db 00 00 00 00 00 00 00 00 00 00 00 00 i have more of those in the logs, will attach.
Created attachment 71337 [details] more "unplug cp2101" => slab corruption messages i can also provide the hardware if necessary.
That is a different issue, most likely (your original report didn't mention anything about USB). Please open a separate bug report for that. Since you haven't seen the original problem again, I will resolve this as WORKSFORME. Please reopen when it happens again.
Seems to be the same as bug 154601; reopen.