Anton Blanchard 0a93ea2e89 RxRPC: Allocate tokens with kzalloc to avoid oops in rxrpc_destroy ... With slab poisoning enabled, I see the following oops: Unable to handle kernel paging request for data at address 0x6b6b6b6b6b6b6b73 ... NIP [c0000000006bc61c] .rxrpc_destroy+0x44/0x104 LR [c0000000006bc618] .rxrpc_destroy+0x40/0x104 Call Trace: [c0000000feb2bc00] [c0000000006bc618] .rxrpc_destroy+0x40/0x104 (unreliable) [c0000000feb2bc90] [c000000000349b2c] .key_cleanup+0x1a8/0x20c [c0000000feb2bd40] [c0000000000a2920] .process_one_work+0x2f4/0x4d0 [c0000000feb2be00] [c0000000000a2d50] .worker_thread+0x254/0x468 [c0000000feb2bec0] [c0000000000a868c] .kthread+0xbc/0xc8 [c0000000feb2bf90] [c000000000020e00] .kernel_thread+0x54/0x70 We aren't initialising token->next, but the code in destroy_context relies on the list being NULL terminated. Use kzalloc to zero out all the fields. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>		2011-02-25 11:12:37 -08:00
..
af_rxrpc.c	rxrpc: rxrpc_workqueue isn't used during memory reclaim	2011-01-14 09:25:11 -08:00
ar-accept.c
ar-ack.c	RxRPC: Fix a potential deadlock between the call resend_timer and state_lock	2010-08-04 21:53:16 -07:00
ar-call.c	RxRPC: Fix a potential deadlock between the call resend_timer and state_lock	2010-08-04 21:53:16 -07:00
ar-connection.c
ar-connevent.c
ar-error.c
ar-input.c
ar-internal.h	Add a dummy printk function for the maintenance of unused printks	2010-08-12 09:51:35 -07:00
ar-key.c	RxRPC: Allocate tokens with kzalloc to avoid oops in rxrpc_destroy	2011-02-25 11:12:37 -08:00
ar-local.c
ar-output.c
ar-peer.c	net: use the macros defined for the members of flowi	2010-11-17 12:27:45 -08:00
ar-proc.c
ar-recvmsg.c	net: use sk_sleep()	2010-04-26 11:18:45 -07:00
ar-security.c
ar-skbuff.c
ar-transport.c
Kconfig
Makefile	Net: rxrpc: Makefile: Remove deprecated kbuild goal definitions	2010-11-22 08:16:15 -08:00
rxkad.c