[petsc-dev] *unwind* next branch

2017-11-12 Thread Satish Balay
Petsc 'next' branch users, I've recreated 'next' branch - and have a backup for current 'next' at next-nov-2017 [This is eliminate some binary files - that got inadvertently added. next-nov-2017 will be deleted at a later time] So (next branch users) please do the following (in all your git

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Smith, Barry F.
> On Nov 12, 2017, at 11:21 AM, Matthew Knepley wrote: > > On Sun, Nov 12, 2017 at 12:17 PM, Satish Balay wrote: > On Sun, 12 Nov 2017, Matthew Knepley wrote: > > > > > Have we tried histogramming test times? It would be nice to know how much > >

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Matthew Knepley
On Sun, Nov 12, 2017 at 12:17 PM, Satish Balay wrote: > On Sun, 12 Nov 2017, Matthew Knepley wrote: > > > > > Have we tried histogramming test times? It would be nice to know how much > > cumulative > > time it takes to run 37%, 67%, 95%, etc. > > I'm not sure what

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Satish Balay
On Sun, 12 Nov 2017, Matthew Knepley wrote: > > Have we tried histogramming test times? It would be nice to know how much > cumulative > time it takes to run 37%, 67%, 95%, etc. I'm not sure what 'histogramming test time' means. All logs record time. And Karl's script summarizes those times on

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Satish Balay
On Sun, 12 Nov 2017, Jed Brown wrote: > > If we get basic testing before merging, the only breakage in 'next' > should be the more interesting interactions. Most days should be clean. This is what I've been sayin [in all the emails yesterday - wrt improving current next testing]. Sure my

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Matthew Knepley
On Sun, Nov 12, 2017 at 11:45 AM, Jed Brown wrote: > Satish Balay writes: > > > On Sun, 12 Nov 2017, Smith, Barry F. wrote: > > > >> If the testing/fixing would be faster if we bought more machines > then decide what machines we need and we order them

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Jed Brown
Satish Balay writes: > On Sun, 12 Nov 2017, Smith, Barry F. wrote: > >> If the testing/fixing would be faster if we bought more machines then >> decide what machines we need and we order them now. Buying machines is far >> better than wasting even more people time

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Satish Balay
On Sun, 12 Nov 2017, Smith, Barry F. wrote: > If the testing/fixing would be faster if we bought more machines then > decide what machines we need and we order them now. Buying machines is far > better than wasting even more people time (which is much more expensive). Its wading through

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Smith, Barry F.
A time out would/should/does have clear indication that it is a timeout, so it shouldn't be that. Satish, If the testing/fixing would be faster if we bought more machines then decide what machines we need and we order them now. Buying machines is far better than wasting even more

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Satish Balay
On Sun, 12 Nov 2017, Matthew Knepley wrote: > > 2.not ok diff-ts_tutorials-ex45_3d_q2_r3 > > > > I have no idea what is triggering this. [It will take a little while > > for me to rebuild and reproduce - as the daybuild is now in progress] > > Is this just a timeout? Nothing in those branches

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Matthew Knepley
On Sun, Nov 12, 2017 at 10:36 AM, Satish Balay wrote: > On Sun, 12 Nov 2017, Satish Balay wrote: > > > [Starting a new thread for this] > > > > Tonight's builds will be on 'next-tmp' with: > > > > $ git fetch -p && comm -12 <(git branch -r --merged origin/next-tmp | > sort)

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Satish Balay
On Sun, 12 Nov 2017, Satish Balay wrote: > [Starting a new thread for this] > > Tonight's builds will be on 'next-tmp' with: > > $ git fetch -p && comm -12 <(git branch -r --merged origin/next-tmp | sort) > <(git branch -r --no-merged origin/master | sort) |grep -v ' origin/next-tmp' >

Re: [petsc-dev] nightlybuilds (next vs next-tmp)

2017-11-12 Thread Matthew Knepley
On Sat, Nov 11, 2017 at 9:40 PM, Satish Balay wrote: > [Starting a new thread for this] > > Tonight's builds will be on 'next-tmp' with: > How easy would it be to setup a system that throttles (or in some way intelligently selects) the branches for 'next'? And when a branch