Gene Cooperman

@karya0 . Our messages crossed, but you read my mind. :-) Thanks for the review of PR #886 ,

Hi Orion, Thanks for reporting this. We have fewer users for ARM, and so we don't test as often. Could you give us a few more details? An easy way...

Hi Dave, Thansk very much for reporting this. I don't have easy access to the aarch64 architecture. Would it be possible to arrange for a temporary guest account for me,...

Thanks, Dave. We haven't had much demand for PPC so far. But DMTCP could definitely be ported to that. Best, - Gene ________________________________________ From: Dave Love Sent: Monday, September 2,...

Hi Dave, Thanks. Kapil Arya is active and is the Fedora maintainer, but he has less time these days. I'm cc'int to him. It would be best if we can...

The story is that you (@karya0) were the original maintainer, but that expired. Paul Grosu was making an effort to become the new maintainer in Fall, 2019, but we were...

Hi @opoplawski , Sorry for my delayed reply. Yes, I very much want to get this back on track. We do have a response to the review ready for Fedora....

@rohgarg: Very nice analysis. This corresponds well to what I am seeing. I've now added a special comment in PR #743 to describe the sense in which this seems to...

@karya0, Yes! That seems like a great idea to have DMTCP-specific 'dmtcp_mutex_lock/unlock, etc. It seems similar to our jalloc strategy where DMTCP uses a spearate malloc arena from the normal...

@karya0 , The issue is how 'errno' was intended to be designed. Either: (a) a function returns -1 (error) and an errno; or (b) a function returns 0; and doesn't...