Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-19 Thread Jeff Squyres (jsquyres)
On Dec 19, 2014, at 10:44 AM, George Bosilca wrote: > Regarding your second point, while I do tend to agree that such issue is > better addressed in the MPI Forum, the last attempt to fix this was certainly > not a resounding success. Yeah, fair enough -- but it wasn't a

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-19 Thread George Bosilca
On Fri, Dec 19, 2014 at 8:58 AM, Jeff Squyres (jsquyres) wrote: > George: > > (I'm not a member of petsc-maint; I have no idea whether my mail will > actually go through to that list) > > TL;DR: I do not think that George's change was correct. PETSC is relying > on undefined

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-19 Thread Jeff Squyres (jsquyres)
On Dec 19, 2014, at 8:58 AM, Jeff Squyres (jsquyres) wrote: > More specifically, George's change can lead to inconsistency/incorrectness in > the presence of multiple threads simultaneously executing attribute actions > on a single entity. Actually -- it's worse than I

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-19 Thread Jeff Squyres (jsquyres)
George: (I'm not a member of petsc-maint; I have no idea whether my mail will actually go through to that list) TL;DR: I do not think that George's change was correct. PETSC is relying on undefined behavior in the MPI standard and should probably update to use a different scheme. More

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-18 Thread Jeff Squyres (jsquyres)
On Dec 17, 2014, at 10:11 PM, Howard Pritchard wrote: > (hmm.. no MPI_Comm_attr_get man page, > needs to be fixed) FWIW: The function is actually MPI_COMM_GET_ATTR; its man page is MPI_Comm_get_attr.3. -- Jeff Squyres jsquy...@cisco.com For corporate legal information go

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-17 Thread George Bosilca
Ben, I can't find anything in the MPI standard suggesting that a recursive behavior of the attribute deletion is enforced/supported by the MPI standard. Thus, the current behavior of Open MPI (a single lock for all attributes), while maybe a little strict, is standard compliant (and thus

Re: [OMPI users] Deadlock in OpenMPI 1.8.3 and PETSc 3.4.5

2014-12-17 Thread Howard Pritchard
Hi Ben, Would you mind checking if you still observe this deadlock condition if you use the 1.8.4 rc4 candidate? openmpi-1.8.4rc4.tar.gz I realize the behavior will likely be the same, but this is just to double