[Git][ghc/ghc][wip/angerman/cross-test-suite] 2 commits: disable flakey divbyzero test

[Moritz Angermann]

Moritz Angermann pushed to branch wip/angerman/cross-test-suite at Glasgow Haskell Compiler / GHC


Commits:
8cc21fdf by Moritz Angermann at 2020-07-14T14:00:38+00:00
disable flakey divbyzero test

- - - - -
52fa3a95 by Moritz Angermann at 2020-07-14T14:56:28+00:00
add python :(

- - - - -


2 changed files:

- testsuite/driver/id
- testsuite/tests/rts/all.T


Changes:

=====================================
testsuite/driver/id
=====================================
@@ -1,4 +1,4 @@
-#! /usr/bin/env bash
+#! /usr/bin/env python3
 #
 # This is the identity test-wrapper.  For native tests we do not need to run
 # executables through a wrapper, as we would need for cross compiled binaries.
@@ -6,4 +6,10 @@
 #
 # Therefore this backup identity test-wrapper is the default for native builds.
 #
-"$*"
\ No newline at end of file
+import sys
+import subprocess
+
+if __name__ == "__main__":
+  if len(sys.argv) < 2:
+    exit(1)
+  exit(subprocess.run(sys.argv[1:]).returncode)
\ No newline at end of file


=====================================
testsuite/tests/rts/all.T
=====================================
@@ -50,6 +50,9 @@ test('divbyzero',
       # behavior on division-by-zero (#10332).
       omit_ways(llvm_ways),
       when(not(have_ncg()), skip),
+      # This test is so flaky, and with test-wrappers, and cross compilers there
+      # is almost no way to make this test pass consistently.
+      when(cross(), skip),
       # Apparently the output can be different on different
       # Linux setups, so just ignore it. As long as we get
       # the right exit code we're OK.



View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/c4b836312a39b6416b5c8057149f2b170d24be91...52fa3a951161a72babc72d282ef8f1b18066ec33

-- 
View it on GitLab: https://gitlab.haskell.org/ghc/ghc/-/compare/c4b836312a39b6416b5c8057149f2b170d24be91...52fa3a951161a72babc72d282ef8f1b18066ec33
You're receiving this email because of your account on gitlab.haskell.org.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-commits/attachments/20200714/f1de8e5a/attachment-0001.html>