[openib-general] IPoIB oops on path record completion

[Hal Rosenstock]

On Wed, 2004-12-15 at 23:18, Roland Dreier wrote:
>     Hal> Can you shorten your timeout down to 1 msec and see what happens ?
> 
> OK, that let me reproduce the oops, and I pushed a fix out (skqueue
> was used unitialized).

That fixed the oops :-) Thanks.

I am still seeing the continual retransmission of SA Get(PathRecords)
even after I terminate the ping -b. The status on the callback is 0.

When I remove ipoib module, I get:

Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib_sa_path_rec_callback: sa_query
0xc26052e8 status 0x0 mad 0xce5e1a4c
Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib_sa_path_rec_callback: sa_query
0xc26052a8 status 0x0 mad 0xce5e18ec
Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib_sa_path_rec_callback: sa_query
0xc2605268 status 0x0 mad 0xce5e178c
Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib_sa_path_rec_callback: sa_query
0xc2605228 status 0x0 mad 0xce5e162c
Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib_sa_path_rec_callback: sa_query
0xc26051e8 status 0x0 mad 0xce5e14cc
Dec 16 08:21:31 localhost kernel: ib0: dev_queue_xmit failed to requeue
packet
Dec 16 08:21:31 localhost kernel: ib0: ib_dealloc_pd failed

When I then remove mthca, I get:
Dec 16 08:23:22 localhost kernel: ool_destroy mthca_av, c9f32000 busy
Dec 16 08:23:22 localhost kernel: ib_mthca 0000:02:00.0:
dma_pool_destroy mthca_av, c5bc5000 busy

The latter is repeated over and over for different AVs.

So there appear to me to be 2 problems:
1. In this error case, why can't resources be freed ?
2. Why does the Get(PR) keep being retried ?

-- Hal