Skip to content

Conversation

@vehre
Copy link
Collaborator

@vehre vehre commented Nov 1, 2016

This pull-request fixes the issue the on certain systems the HAVE_INT_128T guard was available although the int128_t was not. The commit syncs the guard to be HAVE_GFC_INTEGER_16 and the type to be __int128 like it is used in previous code.

Fixed most issues. One still needs a patch to gfortran. Therefore
the pde-solver might fail with an unpatched gfortran still.
…n program

	or the scope where array descriptors are stored may be already
	destroyed when accessed by other (slower) images.
mpi_caf.c: Fixing sync_images, when a stopped image is slow to reach its
	caf_finalize and set the stopped status, while other images have read an
	ok status already and now wait on its send. Realized by using waitany
	and propagating a stopped image status.
	Small speed up in get_by_ref and smaller memory footprint.

Still failing tests:
hello_multiverse: This test fails in general with the gcc-7 fortran compiler,
	because the string is not composed correctly. This is independent
	of coarrays being present or not.
@codecov-io
Copy link

codecov-io commented Nov 1, 2016

Current coverage is 37.37% (diff: 100%)

Merging #237 into master will not change coverage

@@             master       #237   diff @@
==========================================
  Files             2          2          
  Lines           907        907          
  Methods          42         42          
  Messages          0          0          
  Branches        195        195          
==========================================
  Hits            339        339          
  Misses          504        504          
  Partials         64         64          

Sunburst

Powered by Codecov. Last update 1c9f8a6...10f4714

@zbeekman
Copy link
Collaborator

@vehre I'm confused as to why github wants to merge all of these commits into master: I thought we had already merged them, albeit perhaps onto a different branch. I think I'm going to just cherry pick 32fc2e4 onto a new branch, and open that as the pull request if you don't mind.

@zbeekman
Copy link
Collaborator

Superseded by #250

@zbeekman zbeekman closed this Nov 12, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants