In testing a FIP patch on top of the current fcoe-fixes tree, I get this
slab corruption, just after the FCF issues a clear_link and then becomes
available again.
This only happens if I'm running tcpdump, so may have something to do with
the shared data of a cloned skb. Although the message says "probably bad RAM",
it's probably something clearing a bit in the skb data after it has been freed
by skb_release_data().
[ 1007.281467] host4: FIP selected Fibre-Channel Forwarder MAC 00:0d:ec:6d:b2:c0
[ 1010.145715] Slab corruption: size-1024 start=ffff8801372395d0, len=1024
[ 1010.152412] Redzone: 0x9f911029d74e35b/0x9f911029d74e35b.
[ 1010.157895] Last user: [<ffffffff805234ba>](skb_release_data+0xc2/0xc7)
[ 1010.164635] 080: 6a 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b
[ 1010.171006] Single bit error detected. Probably bad RAM.
[ 1010.176372] Run memtest86+ or a similar memory test tool.
[ 1010.181842] Prev obj: start=ffff8801372391b8, len=1024
[ 1010.187032] Redzone: 0xd84156c5635688c0/0xd84156c5635688c0.
[ 1010.192657] Last user:
[<ffffffff802a3e78>](__kmalloc_node_track_caller+0x71/0xc2)
[ 1010.200363] 000: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
[ 1010.206758] 010: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
[ 1010.213135] Next obj: start=ffff8801372399e8, len=1024
[ 1010.218330] Redzone: 0xd84156c5635688c0/0xd84156c5635688c0.
[ 1010.223955] Last user:
[<ffffffff802a3e78>](__kmalloc_node_track_caller+0x71/0xc2)
[ 1010.231668] 000: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
[ 1010.238074] 010: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a
I'll look into it further, but thought I should let you know it's there.
Joe
_______________________________________________
devel mailing list
[email protected]
http://www.open-fcoe.org/mailman/listinfo/devel
