[petsc-dev] PETSc developers who use weird MPI

[Barry Smith]

On Nov 1, 2013, at 9:17 PM, Satish Balay <balay at mcs.anl.gov> wrote:

> On Fri, 1 Nov 2013, Barry Smith wrote:
> 
>> 
>> On Nov 1, 2013, at 8:33 PM, Satish Balay <balay at mcs.anl.gov> wrote:
>> 
>>> I thought I already fixed this. Which branch are these errors in?
>>> 
>>> http://lists.mcs.anl.gov/pipermail/petsc-dev/2013-October/013510.html
>>> 
>>> $ git branch -r --contains 13d1d6fd6eea58bc05ab56819e591ed4a281466f
>>> origin/next
>>> origin/prbrune/sf-sfbasicops
>> 
>>   Damn next/master model!  How the hey am I suppose to know about this when I am working in another branch and I can’t merge in next. We have too much crap in next that is not in master 
> 
> The commit that gives these errors is:
> 
> 53deab3 Added several additional MPI_Ops to PetscSFBasic.
> 
> And I think the commit was in 'prbrune/sf-sfbasicops' initially when I fixed this.
> 
> But now I see this:
> 
> $ git branch -r --contains 53deab3
>  origin/barry/saws
>  origin/madams/sr-driver
>  origin/next
>  origin/prbrune/fas-gscolorsecant
>  origin/prbrune/mat-matcolor
>  origin/prbrune/sf-sfbasicops
> 
> 
> Somehow the snapshot of 'prbrune/sf-sfbasicops' [prior to my fix] got merged into 'barry/saws'
> [I'll have to check these branches to see the path.]
> 
> The fix is to merge latest 'prbrune/sf-sfbasicops' in the same pathway it got into 'barry/saws’

   Thanks. That will likely fix the problem.

   I made the mistake of believing “it is ok to have some branch hang around unmerged with master etc for several months” myth


   Barry

> 
> Satish