Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] Open MPI nightly tarballs suspended / 1.5.5rc3
From: Jeffrey Squyres (jsquyres_at_[hidden])
Date: 2012-02-29 13:35:32

On Feb 28, 2012, at 6:40 PM, Paul H. Hargrove wrote:

> Testing 1.5.5rc3 on a "representative sampling" of my many platforms looks good.
> In particular, I've retested various platforms that showed any significant problems previously and found them to be fixed.
> Though minor, I do see that the following patches I've posted are not applied
> + Add a Mellanox PCI vendor ID to the device params file
> Posted 13 hours ago and not yet on trunk

Just pinged Mellanox for confirmation. I see no reason to do this, but it is their hardware, so I want them in the loop.

> + Fix show_help_lex.l to avoid undefined behavior (and silence associated warning from flex)
> Was applied to trunk as r25983

Pushed this to v1.6 (because we know that the trailing context is ok in all OMPI cases): CMR #3025.

> + Reorder includes to avoid "'struct in_addr' declared inside parameter list" warnings
> Was applied to trunk as r25984

Pushed this to v1.6: CMR #3026.

> I am assuming there is no interest in the MIPS atomics fixes, or the PPC64 atomics work-around for an XLC bug.
> MIPS 1of2:
> MIPS 2of2:
> PPC64/XLC:
> If there *is* interest in these, let me know if there is any assistance I can lend.

There actually is, but I'm ok pushing all of these to v1.6. I'll file tickets for v1.6.

So far as I see it, there's just the 1 issue of the Mellanox device ID. Which, honestly, I'm ok pushing to v1.6 because it's for fairly old Mellanox hardware -- that ID is already in the ConnectX entry.

Jeff Squyres
For corporate legal information go to: