summaryrefslogtreecommitdiffstats
path: root/tests/auto/testlib/selftests/expected_benchlibcallgrind.tap
diff options
context:
space:
mode:
authorEdward Welbourne <edward.welbourne@qt.io>2021-11-30 09:54:03 +0100
committerEdward Welbourne <edward.welbourne@qt.io>2021-12-09 20:54:25 +0100
commite42fe5c775f982e89772eab5083b6dc641a870e3 (patch)
tree9a6b677748292663d6ef31cef154586f996c9bdf /tests/auto/testlib/selftests/expected_benchlibcallgrind.tap
parent4eb2dcd4b96cbca099d8ac0db12ced5066e1e90e (diff)
Fix argv hacking in tst_benchlibcallgrind.cpp
It unconditionally added -callgrind to its own command-line options, but the way testlib handles this argument is, in QTest::qRun(), to re-run the program under the control of valgrind --tool=callgrind, removing the -callgrind command-line option from the test and adding -callgrindchild to its command-line options. So we shouldn't re-add the -callgrind option in the resulting recursive call. The test now runs quickly, producing sensible output, where previously it took a very long time. Revised the drivers to reflect this speed-up, but continue skipping the non-.txt formats to save the need for variant-output files for many formats. To match that, removed the unused non-.txt results files. Change-Id: Iaa99c1b5964d50bccfc6076a21896791b6bbf289 Reviewed-by: MÃ¥rten Nordheim <marten.nordheim@qt.io>
Diffstat (limited to 'tests/auto/testlib/selftests/expected_benchlibcallgrind.tap')
-rw-r--r--tests/auto/testlib/selftests/expected_benchlibcallgrind.tap9
1 files changed, 0 insertions, 9 deletions
diff --git a/tests/auto/testlib/selftests/expected_benchlibcallgrind.tap b/tests/auto/testlib/selftests/expected_benchlibcallgrind.tap
deleted file mode 100644
index da30082eaf..0000000000
--- a/tests/auto/testlib/selftests/expected_benchlibcallgrind.tap
+++ /dev/null
@@ -1,9 +0,0 @@
-TAP version 13
-# tst_BenchlibCallgrind
-ok 1 - initTestCase()
-ok 2 - twoHundredMillionInstructions() # SKIP This test is only defined for gcc and x86.
-ok 3 - cleanupTestCase()
-1..3
-# tests 3
-# pass 2
-# fail 0