next ior release?

[roblatham00]

Is there a time frame for a new IOR release? I find myself telling folks around here to use my fork but i'd rather see them use livermore's version.

[morrone]

Soonish? I was hoping to get to it today on your prodding, but we just had a production issue take off around here, so I won't get to it. Soon. Ish. I hope.

[roblatham00]

OK. I don't know if it's helpful but here's the one-line summary of the outstanding pull requests (80% of which are mine..):

• Add gpfs libraries if needed #23 fixes a compile error (Not linking due to undefined reference to gpfs_fcntl #15)
• A more MPI-IO friendly remove file #21 and more tolerant of whitepace in hintfile #22 cost a colleague of mine some serious time to figure out what was going on. (and will close When IOR deletes an MPI-IO file, it might not delete it #17 and ior hintfile is too sensitive to whitespace #16)
• clean up MPI resources #19 is minor but means mpich configured with resource leak logging won't find anything
• I don't have enough lustre experience to speak to Patch 1: Interoperation delay #18 or Add a functionnality for Lustre to choose specific OSTs or all available OSTs as starting OST for testfile in POSIX mode with separate files #11

That leaves #10, dealing with the HDF5 driver. I've been meaning to take a closer look at that one for...oh a year now. crud.

[roblatham00]

Reviewing papers for PDSW that use IOR: reminded me I should bug you about a release.

[adilger]

Any plans to revive this project? Arguably this is the canonical IOR repository, but nothing has been landed here in years and forks like https://github.com/IOR-LANL/ior (which is sadly not an actual github fork of this project) have sprung up because there is no progress here. Should LLNL/ior be abandoned and efforts moved to IOR-LANL or is it possible to add non-LLNL maintainers to this project (e.g. Rob) that can land patches here?

[roblatham00]

All my patches went to IOR-LANL/ior , which I now consider the official IOR
repository.

On Tue, Nov 22, 2016 at 1:50 PM, adilger [email protected] wrote:

Any plans to revive this project? Arguably this is the canonical IOR
repository, but nothing has been landed here in years and forks like
https://github.com/IOR-LANL/ior (which is sadly not an actual github fork
of this project) have sprung up because there is no progress here. Should
LLNL/ior be abandoned and efforts moved to IOR-LANL or is it possible to
add non-LLNL maintainers to this project (e.g. Rob) that can land patches
here?

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#24 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/ACLEEavOLA2oJI5465-WcCDk-Mom9zRkks5rA0edgaJpZM4CaMC_
.

[johnbent]

Someone please kill this repo! It pops up first in google for IOR github. I, like RobL, also consider the LANL one to be "official".