Hi Jerry, thank you very much for taking on the job of reviewing and sorry for my late answer. In fact, I was having a hard time figuring regressions in the OpenCoarrays library.
This also answers your first question: Yes, OpenCoarrays will make use of the new interface. Most of the changes in the interface are required by OpenCoarrays. Today I got all OpenCoarray's tests passing. The OpenCoarrays tests all run a little bit faster than with the old method. Please keep in mind, that those tests keep starting and stopping tiny apps. I.e. the overhead of this sequential part is significant. Unfortunately the speedup is tiny (about 3 seconds for the whole suite running now in 1:21.38 (m:ss.ms; Release-build, i.e. -O3; mpich and Intel's mpi). I will look for a better benchmark suite. I think to remember that in some ticket on OpenCoarrays one was mentioned. Nevertheless are all these tests run on single machine. I have no cluster to command. I will rebase, rename rewrite.cc to coarray.cc, retest and merge shortly, if no one objects. Then I unfortunately have to post a new small bugfix (about 10 lines). Thanks again, Andre On Fri, 14 Feb 2025 10:19:28 -0800 Jerry D <jvdelis...@gmail.com> wrote: > On 2/13/25 11:48 AM, Jerry D wrote: > > On 2/10/25 2:25 AM, Andre Vehreschild wrote: > >> [PATCH 7/7] Fortran: Remove deprecated coarray routines [PR107635] > >> > > > > I have applied all patches. Regression tested OK here. > > > > From patch 5 there was one reject: > > > > patching file gcc/testsuite/gfortran.dg/coarray/send_char_array_1.f90 > > Hunk #1 FAILED at 39. > > 1 out of 1 hunk FAILED -- saving rejects to file gcc/testsuite/ > > gfortran.dg/coarray/send_char_array_1.f90.rej > > > ---------------------------- > > I commented earlier about changing the name of rewrite.cc. > ---------------------------- this please. > > > > I am now going through the whole enchilada for editorial stuff. > > > > Regards, > > > > I finished going through the last nine yards and it looks good. I have a > couple of questions: > > Have you been able to test against the OpenCoarray tests? > > Have you been able to measure any performance improvements? > > I suspect that the latter question may relate only to multi-node large > systems. > > I think this is good to commit. (all 7 parts) > > Does anyone else have any comments? > > Regards, > > Jerry > > > -- Andre Vehreschild * Email: vehre ad gmx dot de