develooper Front page | perl.perl5.porters | Postings from September 2011

Re: A summary of Chip patches and potential impact

Thread Previous | Thread Next
From:
Jesse Vincent
Date:
September 4, 2011 14:03
Subject:
Re: A summary of Chip patches and potential impact
Message ID:
5B3E7232-EC47-452C-9F47-611EBBC9F19C@fsck.com

On Sep 3, 2011, at 6:11 AM, Steffen Mueller wrote:

> This leaves the fact that the CPAN smoking is just too damn slow to be throwing it at a lot of branches in the future. I do think it's viable for deep changes such as yours, but we need to be very careful with adopting a "if in doubt, let's CPAN smoke it" policy both for conservation of volunteer time, CPU time, and latency (this remark is for Jesse primarily).

xdg  pointed out to me that my response to the "what about just smoking a subset like Task::Kensho" misinterpreted some of the recent context on this thread. Sorry about that.

Indeed, I never intended that we'd be smoking the majority of topic branches, but I do think that having infrastructure to do such a smoke with minimal pain is incredibly important as we move forward. I know it's currently really painful to do and appreciate the effort being put in to making it less painful.

It may be that once we get somewhat further on the automation, we try to do something like smoke cpan for each blead release and dig further when that goes bad, saving the individual branch cpansmokes for "important" cases.  

Once the data is gettable, I'd really love to know how much CPU time we're actually talking about to do a full A/B cpansmoke.  Being able to peg a time/pain/cpu cost to it helps make it a lot more obvious what's not gonna work.

thanks.
Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About