[openib-general] [PATCH] libibat: little / big endian problems in example programs

[Hal Rosenstock]

On Thu, 2005-10-06 at 08:14, Heiko J Schick wrote:
> Hello,
> 
> during (some) test with libibat I found out that the example programs
> include a little/big endian problem.
> Below you will find the patch for ats.c and att.c which will solve
> this problem on PPC64:
> 
> Signed-off-by: Heiko Joerg Schick <schickhj at de.ibm.com>

Thanks. Applied.

> --- /home/source/trunk_3615_orig/src/userspace/libibat/examples/ats.c
> 2005-08-23 18:49:39.000000000 +0200
> +++ ats.c       2005-10-06 13:42:02.492909848 +0200
> @@ -225,7 +225,7 @@ int main(int argc, char **argv)
>         }
> 
>         for (i = 0; i < MAX_REQ; i++) {
> -               r = ib_at_route_by_ip(0x0100a8c0, 0, 0,
> +               r = ib_at_route_by_ip(htonl(0xc0a80001), 0, 0,
>                                       IB_AT_ROUTE_FORCE_ATS,
>                                       att_rt + i, att_rt_comp + i,
> &req_id);

The patch din't apply. It indicated it was malformed here. I think your
mailer line wrapped this. That needs to be turned off when submitting
patches.

> 
> --- /home/source/trunk_3615_orig/src/userspace/libibat/examples/att.c
> 2005-08-23 18:49:39.000000000 +0200
> +++ att.c       2005-10-06 13:40:26.293891760 +0200
> @@ -190,7 +190,7 @@ int main(int argc, char **argv)
>         }
> 
>         for (i = 0; i < MAX_REQ; i++) {
> -               r = ib_at_route_by_ip(0x0100a8c0, 0, 0, 0,
> +               r = ib_at_route_by_ip(htonl(0xc0a80001), 0, 0, 0,
>                                       att_rt + i, att_rt_comp + i,
> &req_id);
> 
>  #if __WORDSIZE == 64
> 
> BTW. Does the output of the uatt program looks alright?

Yes, that looks OK to me but would need to be verified with your subnet
config. It looks like your test node was not 192.168.0.1 and had a LID
of 0xab and the 192.168.0.1 node was a different node with LID 0xae. You
could also verify the GIDs which were indicated as well.

-- Hal

> uatt: att_path_comp_fn: id 21 context 0x10012ae8 completed with
> rec_num 1
> ===> slid 0xab dlid 0xae
> uatt: main: ib_at_route_by_ip: ret 1 errno 0 for request 21 id 0 0
> uatt: att_rt_comp_fn: id 0 context 0x100135f0 completed with rec_num 1
> ===> rt 0x100135f0 sgid 0xfe8000000000000067eafbe000040001 dgid
> 0xfe8000000000000067eafbe000040002
> uatt: att_rt_comp_fn: ib_at_paths_by_route: ret 0 errno 0 id 22 22
> uatt: att_path_comp_fn: id 22 context 0x10012b30 completed with
> rec_num 1
> ===> slid 0xab dlid 0xae
> uatt: main: ib_at_route_by_ip: ret 1 errno 0 for request 22 id 0 0
> uatt: att_rt_comp_fn: id 0 context 0x10013628 completed with rec_num 1
> ===> rt 0x10013628 sgid 0xfe8000000000000067eafbe000040001 dgid
> 0xfe8000000000000067eafbe000040002
> uatt: att_rt_comp_fn: ib_at_paths_by_route: ret 0 errno 0 id 23 23
> uatt: att_path_comp_fn: id 23 context 0x10012b78 completed with
> rec_num 1
> ===> slid 0xab dlid 0xae
> ...
> 
> Many thanks in advance!
> 
> Mit freundlichen Gruessen / Kind Regards
> Heiko Joerg Schick
> 
> IBM Deutschland Entwicklung GmbH
> I/Ox Microcode Development
> Linux Infiniband Device Drivers
> 
> Schoenaicher Str. 220
> 71032 Boeblingen
> E-Mail: schickhj at de.ibm.com
> External: 49-7031-16-0 x4219,   t/l: 120-4219
> 
> 
> ______________________________________________________________________
> 
> _______________________________________________
> openib-general mailing list
> openib-general at openib.org
> http://openib.org/mailman/listinfo/openib-general
> 
> To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general