[commit: testsuite] master: Deal with failures for T367, T367_letnoescape under ghci (bb4e4d5)
git at git.haskell.org
git at git.haskell.org
Mon Sep 30 13:01:07 CEST 2013
Repository : ssh://git@git.haskell.org/testsuite
On branch : master
Link : http://ghc.haskell.org/trac/ghc/changeset/bb4e4d5c3f8019a7e8caac3a3938110fa748e36b/testsuite
>---------------------------------------------------------------
commit bb4e4d5c3f8019a7e8caac3a3938110fa748e36b
Author: Reid Barton <rwbarton at gmail.com>
Date: Sun Sep 29 16:21:01 2013 -0400
Deal with failures for T367, T367_letnoescape under ghci
These tests had a very short timeout (0.3 s). With WAY=ghci,
the time ghci takes to start up and compile the test modules
is counted in this timeout, and that causes the tests to fail.
T367 really needs the very short timeout, so this commit disables
the ghci way for T367. T367_letnoescape can handle any timeout,
so I bumped up the timeout to 6 s to give ghci time to start up.
>---------------------------------------------------------------
bb4e4d5c3f8019a7e8caac3a3938110fa748e36b
tests/concurrent/should_run/all.T | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/concurrent/should_run/all.T b/tests/concurrent/should_run/all.T
index 0eec846..15457fd 100644
--- a/tests/concurrent/should_run/all.T
+++ b/tests/concurrent/should_run/all.T
@@ -22,8 +22,8 @@ test('conc072', only_ways(['threaded2']), compile_and_run, [''])
test('conc073', normal, compile_and_run, [''])
# vector code must get inlined to become non-allocating
-test('T367', [reqlib('vector'), timeout_multiplier(0.001)], compile_and_run, ['-O2 -fno-omit-yields'])
-test('T367_letnoescape', [timeout_multiplier(0.001)], compile_and_run, ['-fno-omit-yields'])
+test('T367', [reqlib('vector'), omit_ways(['ghci']), timeout_multiplier(0.001)], compile_and_run, ['-O2 -fno-omit-yields'])
+test('T367_letnoescape', [timeout_multiplier(0.02)], compile_and_run, ['-fno-omit-yields'])
test('T1980', normal, compile_and_run, [''])
test('T2910', normal, compile_and_run, [''])
More information about the ghc-commits
mailing list