Re: [OMPI devel] Invalid initializers in pml_base_request and bml_base_open

2009-08-03 Thread Jeff Squyres
Should they each mail you about MPI_COMPLEX first, too? ;-) On Aug 3, 2009, at 1:35 PM, George Bosilca wrote: Not before I get an email from every developer in the Open MPI project !!! george. On Aug 3, 2009, at 12:31 , Jeff Squyres wrote: This looks icky. George -- can you please

Re: [OMPI devel] [PATCH] Better error reporting when failing to load a component

2009-08-03 Thread Kenneth Lloyd
Hi all, I've just recently joined because I, too, am working on a (possible) OpenMPI component. The basics are: A Topology and Weight Evolving Artificial Neural Network (TWEANN) that learns to configure point-to-point comm. on ob1 fabrics in various compute clusters given a context of existing

Re: [OMPI devel] Invalid initializers in pml_base_request and bml_base_open

2009-08-03 Thread George Bosilca
Not before I get an email from every developer in the Open MPI project !!! george. On Aug 3, 2009, at 12:31 , Jeff Squyres wrote: This looks icky. George -- can you please fix? On Aug 3, 2009, at 12:07 AM, Ralph Castain wrote: Hi folks The following lines were entered into two

Re: [OMPI devel] Invalid initializers in pml_base_request and bml_base_open

2009-08-03 Thread Jeff Squyres
This looks icky. George -- can you please fix? On Aug 3, 2009, at 12:07 AM, Ralph Castain wrote: Hi folks The following lines were entered into two files by r21739: ompi/mca/pml/base/pml_base_request.c: ompi_free_list_t mca_pml_base_send_requests = {0}; ompi_free_list_t

Re: [OMPI devel] [PATCH] Better error reporting when failing to load a component

2009-08-03 Thread Jeff Squyres
On Aug 3, 2009, at 8:23 AM, Arthur Huillet wrote: I have recently started working on OpenMPI, and part of my job consists in adding a new module to OpenMPI. Cool. What are you adding? I would like to take this opportunity to thank the people who have been involved in writing

Re: [OMPI devel] Device failover on ob1

2009-08-03 Thread Rolf Vandevaart
I have not, but there should be no difference. The failover code only gets triggered when an error happens. Otherwise, there are no differences in the code paths while everything is functioning normally. Rolf On 08/03/09 11:14, Pavel Shamis (Pasha) wrote: Rolf, Did you compare latency/bw

Re: [OMPI devel] Device failover on ob1

2009-08-03 Thread Brian W. Barrett
On Sun, 2 Aug 2009, Ralph Castain wrote: Perhaps a bigger question needs to be addressed - namely, does the ob1 code need to be refactored? Having been involved a little in the early discussion with bull when we debated over where to put this, I know the primary concern was that the code

Re: [OMPI devel] Device failover on ob1

2009-08-03 Thread Pavel Shamis (Pasha)
Rolf, Did you compare latency/bw for failover-enabled code VS trunk ? Pasha. Rolf Vandevaart wrote: Hi folks: As some of you know, I have also been looking into implementing failover as well. I took a different approach as I am solving the problem within the openib BTL itself. This of

Re: [OMPI devel] Device failover on ob1

2009-08-03 Thread Rolf Vandevaart
Hi folks: As some of you know, I have also been looking into implementing failover as well. I took a different approach as I am solving the problem within the openib BTL itself. This of course means that this only works for failing from one openib BTL to another but that was our area of

[OMPI devel] Fwd: Upgrading subversion on sourcehaven

2009-08-03 Thread Jeff Squyres
FYI. Begin forwarded message: From: DongInn Kim Date: August 2, 2009 11:17:17 PM EDT Subject: Upgrading subversion on sourcehaven Hi, I have tried to upgrade subversion-1.4.2 to 1.5.5 on sourcehaven and this time, there is no problem with the SourceHaven web application.

[OMPI devel] btl_openib_component.c -- min_distance

2009-08-03 Thread guillaume
Hello, I'm currently looking at the carto framework and the way it is used by the openib btl. Looking at the code, I found this : 1963if(min_distance < 0 || min_distance < distance) 1964min_distance = distance; at ompi/mca/btl/openib/btl_openib_component.c in the function:

[OMPI devel] [PATCH] Better error reporting when failing to load a component

2009-08-03 Thread Arthur Huillet
Hello, I have recently started working on OpenMPI, and part of my job consists in adding a new module to OpenMPI. I would like to take this opportunity to thank the people who have been involved in writing https://svn.open-mpi.org/trac/ompi/wiki/devel/CreateComponent as it was very

[OMPI devel] Invalid initializers in pml_base_request and bml_base_open

2009-08-03 Thread Ralph Castain
Hi folks The following lines were entered into two files by r21739: ompi/mca/pml/base/pml_base_request.c: ompi_free_list_t mca_pml_base_send_requests = {0}; ompi_free_list_t mca_pml_base_recv_requests = {0}; ompi/mca/bml/bml_base_open.c: opal_list_t mca_bml_base_components_available = {0};