On 9/27/05, Todd Bowman <[EMAIL PROTECTED]> wrote:
Never mind, I found the post from Roland discussing this issue:
Roland wrote:
I didn't try to fix uDAPL, because some thought probably
needs to go into how to use completion channels most efficiently.
On 9/27/05, James Lentini < [EMAIL PROTECTED]> wrote:
On Tue, 27 Sep 2005, Todd Bowman wrote:
> On 9/27/05, James Lentini <[EMAIL PROTECTED] > wrote:
>
> What is the output of cat /sys/class/net/*/ifindex?
> >
> > cat /sys/class/net/*/ifindex
> 1 #eth0
> 10 #ib0
> 11 #ib1
> 2 #lo
> 3 #tunl0
This looks like the problem I reported last week:
http://openib.org/pipermail/openib-general/2005-September/011668.html
If so, this is fixed in the current subversion sources. Could you
update your sources, specifically infiniband/core/at.c, and try again?
james
That worked. Thanks.
It is printing the addresses backwards. But I assume this is also a bug in printing.
ips_by_gid: RET 0 at_rec 0x1ffff9e5f40 -> id 1
dapli_at_event_cb()
ip_comp_handler: rec 0x1ffff9e5f40 ->id 1 id 1 num 1 b0f0add
open_hca: mthca0, port 1, AF_INET 221.10.15.11 INLINE_MAX=128
query_hca: mthca0 AF_INET 221.10.15.11
query_hca: (0.30002) ep 64512 ep_q 65535 evd 65408 evd_q 65535
query_hca: msg 2147483648 rdma 2147483648 iov 28 lmr 131056 rmr 0
Bus error
The bus error is probably due to changes in ibv_context. I didn't recompile libdapl.
The changes in ibv_context that broke udapl are:
num_comp is now num_comp_vectors
I don't know what is used in place of cq_fd[1]. Any pointers?
Todd
Never mind, I found the post from Roland discussing this issue:
Roland wrote:
I didn't try to fix uDAPL, because some thought probably
needs to go into how to use completion channels most efficiently.
_______________________________________________ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
