summaryrefslogtreecommitdiff
path: root/testsuite/driver/testlib.py
Commit message (Collapse)AuthorAgeFilesLines
...
* Add a framework test for the package cache file being modifiedIan Lynagh2010-11-141-1/+17
| | | | Tests shouldn't touch the compiler's package database
* Add 'unless_os' testlib helperIan Lynagh2010-11-101-0/+6
|
* Add a test for #4255: can't use TH with a profiled compilerIan Lynagh2010-09-221-0/+6
|
* Add an if_compiler_profiled helperIan Lynagh2010-09-211-0/+6
| | | | and use it to skip the th, ghci and debugger tests when GHC is profiled.
* when calling ghc-pkg for reqlib(), add --no-user-package-confSimon Marlow2010-09-201-1/+1
|
* Remove some errmsg normalisation (removing paths from filenames)Ian Lynagh2010-08-081-3/+0
| | | | This wasn't needed, but was breaking "data/newtype" in an error message
* Make the dynlib tests work on WindowsIan Lynagh2010-05-031-0/+16
|
* Fix the ffi002 testIan Lynagh2010-04-281-1/+14
|
* Fix running hp2ps in a directory containing spacesIan Lynagh2010-03-041-1/+1
|
* Give more info in the testsuite outputIan Lynagh2010-01-011-1/+4
|
* Check for duplicate test names, and report them as framework failuresIan Lynagh2010-01-011-0/+5
|
* Refactor the cleaning codeIan Lynagh2009-12-191-31/+32
|
* Remove unused clean_o_hi functionIan Lynagh2009-12-191-3/+0
|
* Add clean_cmd to the testsuite, and use it in bug1465Ian Lynagh2009-12-191-0/+17
|
* Remove debugging printIan Lynagh2009-12-191-1/+0
|
* Add pre-command support to the testsuite, and fix annrun01 by using itIan Lynagh2009-12-191-0/+19
|
* Allow tests to behave differently depending on whether the compiler is in-treeIan Lynagh2009-12-181-0/+12
| | | | | And skip testwsdeque if it is not in-tree, as we rely on some headers from the build tree.
* Print out how many tests we have done and the total when we run a testIan Lynagh2009-11-281-1/+1
| | | | | | | This isn't perfect, as it doesn't account for tests that will be skipped in the total. But that's hard to work out, as we might skip a test in only some ways and we currently don't work out which ways to run it until later, so I think this is good enough for now.
* Gather all tests at once, rather than doing them directory by directoryIan Lynagh2009-11-281-19/+19
| | | | | This increases the parallelism possible, and allows us to track what progress we are making.
* grab the target architecture from GHC, and add an if_arch() testSimon Marlow2009-11-241-0/+6
|
* Normalise slashesIan Lynagh2009-10-211-0/+1
|
* Always use / to join paths; stops "make -C .\ clean" being misparsedIan Lynagh2009-10-211-1/+1
|
* add -no-user-package-conf to GHC command linesSimon Marlow2009-07-241-1/+1
|
* Follow Integer changesIan Lynagh2009-07-211-1/+1
|
* Follow Integer library changesIan Lynagh2009-06-231-0/+4
|
* Use +RTS -V0 when looking at timing informationIan Lynagh2009-06-051-2/+2
| | | | This should make things more deterministic
* Tweak the driver so that 2566 passes in teh new build systemIan Lynagh2009-04-261-0/+4
|
* Add support for compiler -t statsIan Lynagh2009-04-041-37/+63
|
* Define if_not_platformIan Lynagh2009-04-031-0/+6
|
* Define if_wordsize in the driverIan Lynagh2009-04-031-0/+6
|
* Improve the testsuite driver support for -t stats, and enhance space_leak_001Ian Lynagh2009-04-031-37/+33
|
* Rewrite the result testing logic in simple_runIan Lynagh2009-04-031-8/+11
|
* Re-indent simple_run in the driverIan Lynagh2009-04-031-54/+54
|
* Define and use opts in simple_run in the driverIan Lynagh2009-04-031-6/+8
|
* clean .eventlog filesSimon Marlow2009-03-301-1/+1
|
* Fix test 2566 in the new build systemIan Lynagh2009-01-091-0/+9
|
* check for valid .prof files when profilingSimon Marlow2008-12-151-3/+21
|
* Clean .exe.manifest filesIan Lynagh2008-12-031-1/+1
|
* small fix to the space-leak testing frameworkSimon Marlow2008-11-251-1/+1
|
* Don't show whitespace differences when diffing test resultsIan Lynagh2008-11-231-2/+10
|
* Add a space leak test, and some infrastructure for checking space usageIan Lynagh2008-11-231-0/+42
|
* Print out exceptions when cleaning testsIan Lynagh2008-10-011-3/+11
|
* Remove no_clean from the driverIan Lynagh2008-08-251-3/+0
| | | | | We don't use it any more, and it shouldn't be used as it effectively means that the test won't work in the parallel testsuite.
* Simplify the driver a bit: Break skip_{if,unless}_tag down into smaller partsIan Lynagh2008-08-171-10/+8
|
* Tweak whitespaceIan Lynagh2008-06-281-4/+4
|
* Clean <test>.std{out,err}.normalisedIan Lynagh2008-06-281-0/+1
|
* Clean empty directories as well as filesIan Lynagh2008-06-281-1/+6
|
* Use -fforce-recomp instead of -no-recompIan Lynagh2008-06-141-3/+3
|
* Break expect_fail_if_compiler_type and expect_broken_if_compiler_type downIan Lynagh2008-06-101-16/+6
| | | | | | | into their component parts if_compiler_type expect_fail expect_broken
* following the fix for #1955, the .ps file is now in foo.ps, not foo.exe.ps ↵Simon Marlow2008-05-141-4/+1
| | | | on Windows
View Lorry Controller Status