[OMPI users] 64-Bit MIPS support patch

2006-05-11 Thread Jonathan Day
g where it doesn't handle macros unless the file extension is .S, rather than .s. Enjoy! Jonathan Day __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com --- ompi-original/opal/include/opal/sy

Re: [O-MPI users] ompi spec file

2006-01-26 Thread Jonathan Day
I'm not entirely sure what the problem is with the --with/--without flags. Within the configure script, these flags enable/disable the setting of certain environment variables and/or the running of certain tests. Within the RPM spec file, these need to be given, sure, but in many cases there will

[O-MPI users] Tests with current version

2005-11-11 Thread Jonathan Day
Still working on the Broadcom MIPS64 port, so no news on that front yet except to say that I'm getting there. (Which seems to be fairly common with people using Broadcom CPUs.) On the Pentium 4 front, I can say that recent snapshots of the experimental GCC 4.1.0 compiled the C, C++ and Fortran bin

[O-MPI users] Still working on MIPS64 port.

2005-09-26 Thread Jonathan Day
I'll get a patch to the list as soon as I have one that works. In the meantime, here are a few of the things I've turned up. 1. The MIPS.asm file needs to load asm/asm.h on Linux boxes, so the config script needs to grab the OS for the MIPS box to identify what header to use. 2. MIPS.asm, in the

[O-MPI users] Changes required to support the Broadcom SB1

2005-09-14 Thread Jonathan Day
Ok, so far so good. It looks as if the majority of the Irix stuff will work with the Broadcom board - it's just that the tests to activate it (in ROMIO, for example) are too specific. It isn't just the main configure script that needs fixing - but the changes are more generalizing than anything. T

[O-MPI users] MIPS64 Trek, The Search for MIPS

2005-09-13 Thread Jonathan Day
Ok, I've established that the macros used do exist in the Linux MIPS64 header files. They are located in So, it seems safe to guess that something is getting confused. I'm trying an experiment, by copying the MIPS configuration section and having it look for mips64* or mipsisa64*. It then sets th

[O-MPI users] MIPS64 Trek - The story so far

2005-09-12 Thread Jonathan Day
Hi, Just a quick update on working with OpenMPI's MIPS code on the MIPS64 processor. It does not appear to work to simply use the atomic code for the 32-bit MIPS - the opcodes don't correspond. Over the next day or two, I will be updating the assembler on the MIPS64 board, and I'll know better the

Re: [O-MPI users] MIPS64 - The Long-Awaited Config File

2005-09-12 Thread Jonathan Day
I'll try the MIPS (aka MIPS32) solution - hey, it can't kill anything. There are some subtle differences, but most MIPS64 boards'll run MIPS32 code (I think in a similar way to the way AMD's 64-bit chips also run 32-bit code). However, for anyone planning a "native" MIPS64 solution, I've found a f

[O-MPI users] MIPS64 - The Long-Awaited Config File

2005-09-09 Thread Jonathan Day
It has taken me a while to try a few things out at this end, but in the end I've still drawn a blank. To recap, I'm compiling on a Broadcom SB1, which is a MIPS64 board, using a Gnu compiler and tools. The error message I get is: configure: error: No atomic primitives available for mips64-unknown-

[O-MPI users] MIPS64 question

2005-08-23 Thread Jonathan Day
Hi, I notice that MIPS is not listed amongst the "known" architectures in the README, and on trying to compile OpenMPI on a Linux/MIPS64 platform it blew up in the configuration code. Having said that, there does seem to be MIPS stuff in those directories listing assembly code, but I do NOT know

[O-MPI users] Further thoughts

2005-06-16 Thread Jonathan Day
Hi, Just a couple of additional thoughts. First off, when Open MPI officially goes gold, I would suggest one of the developers puts up a listing on Freshmeat, as there are a lot of academics using that to track projects these days. Secondly, it's great that Open MPI is going to use a CPAN-like ar

Re: [O-MPI users] Questions on status

2005-06-15 Thread Jonathan Day
Phew! I'd no idea my questions would create nearly that kind of level of response on the list. I should post more often! :) Anyways, here are a few suggestions that might - just might - work around some of the problems that have been mentioned: 1. Publish the stable header files early, or fragmen

[O-MPI users] Questions on status

2005-06-14 Thread Jonathan Day
Hi, The mailing list has fallen virtually silent, there's no beta yet out and (although I may have missed it) there doesn't seem to be much in the way of updates on the website. Can someone on the project just confirm that you are, indeed, all alive and well, not kidnapped by aliens, or been dela

[O-MPI users] ABI or API?

2005-03-26 Thread Jonathan Day
Hi, With this debate over a common interface, I've noticed that some have been talking about a common API (ie: code-level compatibility) and others an ABI (ie: binary-level compatibility). There's a big difference, so I think it might be helpful if it was cleared up as to what it was that was want

[O-MPI users] Several queries

2005-02-16 Thread Jonathan Day
oid the pitfalls of other projects? Jonathan Day __ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail