On Wed, 2005-06-01 at 19:34, Roland Dreier wrote: > Hal> ibv_asyncwatch Couldn't get context for mthca0 > > Hal> dmesg shows: Jun 1 19:01:00 hal kernel: UCM: Created struct > Hal> Jun 1 19:01:00 hal kernel: UCM: Deleted struct > > Looks like it's coming from UCM, not any of these tests...
That message is coming from a global constructor for user CM. > Hal> Any idea on what is wrong ? > > Nope. Are all your userspace and kernel versions in sync? Yes. > If so, then send the output of 'strace -ewrite=all -etrace=write > ibv_asyncwatch' Not very informative :-( strace -etrace=write -ewrite=all ./examples/ibv_asyncwatch --- SIGCHLD (Child exited) @ 0 (0) --- --- SIGCHLD (Child exited) @ 0 (0) --- --- SIGCHLD (Child exited) @ 0 (0) --- --- SIGCHLD (Child exited) @ 0 (0) --- --- SIGCHLD (Child exited) @ 0 (0) --- write(2, "Couldn\'t get context for mthca0\n", 32Couldn't get context for mthca0 ) = 32 | 00000 43 6f 75 6c 64 6e 27 74 20 67 65 74 20 63 6f 6e Couldn't get con | | 00010 74 65 78 74 20 66 6f 72 20 6d 74 68 63 61 30 0a text for mthca0. | Same for all examples except ibv_devices. -- Hal _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
