Age | Commit message (Collapse) | Author | |
---|---|---|---|
2018-07-09 | [PerfTests] add makefile rules for performance testing | iain holmes | |
2018-05-10 | Always regenerate BuildVariables.cs in `make dist` | Jo Shields | |
It's not regenerated ordinarily, causing `make dist` to include a "wrong" file if switching branches. Fixes occasional accidental bad Help/About versions | |||
2018-03-20 | Added target for deploying tests | Lluis Sanchez | |
2018-02-16 | Make dist2 (#3879) | Marius Ungureanu | |
* Fix `make dist` (cherry picked from commit 7732445e89434e17ff1e0217a640d841b29d5fc9) * Don't strip assemblies from tarballs. It'll make distro packagers cry, but the tarballs will actually work. (cherry picked from commit b1820981f8acced4d4b721ce44ed5860ce8fb758) * Include case-sensitivity fix from NRefactory * Ensure BuildVariables.cs is up to date and in `make dist` tarballs | |||
2018-02-13 | [AOT] Fix wrong target name for aot-all (#3842) | Marius Ungureanu | |
2018-02-13 | [AOT] Fix Cecil not being symbolicated in instruments (#3837) | Marius Ungureanu | |
2018-02-12 | [AOT] Improve targets for AOT (#3831) | Marius Ungureanu | |
* [AOT] Make the targets use --assembly-loader=strict. This fixes exclusion path necessity and AOT's everything * [AOT] Make target relocatable. Passing APP=path/to/app will automatically AOT the .app * [AOT] Clean up rules and make a target to AOT everything needed for profiling. * Fixup * Filter azure-functions-cli rather than --assembly-loader=strict | |||
2018-02-09 | [AOT] Fix the AOT profile not working on Mac. (#3813) | Marius Ungureanu | |
Seems like AzureFunctions included a roslyn library, and it ended up being used when AOT-ing, causing mismatches with the core assembly | |||
2018-01-19 | [AOT] Apply binding redirects (#3694) | Marius Ungureanu | |
2018-01-09 | Makefile: fix all-recursive when $PATH has spaces | Sandy Armstrong | |
2017-11-27 | [A11y] Allow Accessibility to be temporarily disabled from commandline | iain holmes | |
If DISABLE_ATKCOCOA is set then accessibilty will be disabled. make run-no-accessibility will do the same. | |||
2017-05-04 | [profiling] Add the ability to load up the gobject leak tracker | Alan McGovern | |
https://github.com/alanmcgovern/gobject-tracker | |||
2017-01-13 | [Misc] Switch AOT target to use mono64 instead of mono. | therzok | |
2016-05-09 | [build] Bail out early if `git` is not installed | Andrés G. Aragoneses | |
Otherwise typing `make` would fail with this error: ``` $ make if test -d ".git"; then \ git submodule sync; \ git submodule update --init --recursive || exit 1; \ fi /bin/sh: 2: git: not found /bin/sh: 3: git: not found Makefile:12: recipe for target 'update_submodules' failed make: *** [update_submodules] Error 1 ``` | |||
2016-01-24 | Update Makefile | ArsenShnurkov | |
2015-12-04 | Merge branch 'master' into roslyn | Matt Ward | |
Conflicts: main/src/addins/NUnit/Gui/AbstractUnitTestEditorExtension.cs version-checks | |||
2015-12-03 | Add code of conduct | Miguel de Icaza | |
2015-10-30 | [Mac Platform] Enable 64bit with an environment variable | iain holmes | |
2015-10-06 | [UITest] Allow running a UITest by test case, test fixture or a namespace | Manish Sinha | |
2015-07-10 | Missing SEMICOLON | Marius Ungureanu | |
2015-07-10 | [Build] Add `git submodule sync` as a pre-build command. | Marius Ungureanu | |
2015-06-23 | [build] Let fsharp compile in El Capitan | Alan McGovern | |
With El Capitan we will no longer be able to symlink our binaries in /usr/bin. As such we need to either use full paths to our binaries like 'xbuild' and 'nuget' or we need to ensure they are available in PATH by adding in the path to mono's bin dir. I opted for the latter. This fix is not the best way, so if someone else wants to implement this in a better way feel free. | |||
2015-04-16 | Create a new test target named uitest to run UITests | Manish Sinha | |
2015-03-10 | [Coverage] Add a make coverage option to run coverage tests | iain holmes | |
2014-09-16 | Forcibly remove all assemblies from generated tarball, for source-only build. | Jo Shields | |
2014-09-16 | Always regenerate buildinfo when making a tarball | Jo Shields | |
(buildinfo is cached otherwise) | |||
2014-06-04 | explicitly generate buildinfo during make dist; | Timotheus Pokorra | |
fixing problems with comments in Makefile, caused by my previous commit | |||
2014-06-03 | put the buildinfo file into the top srcdir of the tarball. | Timotheus Pokorra | |
the buildinfo target should check if that file exists in the top srcdir, and copy it, or generate with the configure.exe from scripts directory, if the file does not exist yet | |||
2014-05-30 | improvement: the valid buildinfo file is now copied to the tarball | Timotheus Pokorra | |
2014-05-28 | fix previous commit, need to escape dollar signs in makefile | Timotheus Pokorra | |
2014-05-28 | address the problems with version.config missing in the tarball, and the ↵ | Timotheus Pokorra | |
scripts directory. do not call scripts/configure.cs in the tarball, because it tries to resolve the version from the git history | |||
2013-07-09 | Nuke some unnecessary local copies | Alan McGovern | |
2013-07-09 | Hide the verbose output in make aot | Alan McGovern | |
2013-07-09 | More better the 'make aot' target | Alan McGovern | |
It can now find all the required assemblies by autogenerating MONO_PATH to include all the paths we place our binaries in. | |||
2013-06-25 | [build] AOT with debug symbols | Alan McGovern | |
2013-06-25 | [build] Ensure the addins can be AOT'ed when they reference core libs | Alan McGovern | |
2013-06-25 | [build] better messages | Alan McGovern | |
2013-06-25 | [build] Fix up make aot a bit ;) | Alan McGovern | |
2013-06-25 | [build] Hello make aot | Alan McGovern | |
If someone wants to AOT their binaries after compilation they can now just run 'make aot' from the toplevel makefile. This is useful for getting nice information in the macos profiler, or in gdb when backtracing. | |||
2012-12-03 | [Build] if something goes wrong updating the submodules, bail out | Alan McGovern | |
2012-10-22 | [build] Only update the submodules if inside a git checkout | Alan McGovern | |
2012-09-28 | [Makefiles] Add run-sgen targets that run MD with SGen enabled. | Bojan Rajkovic | |
2012-09-07 | Revert "Remove the other call to update the submodules" | Michael Hutchinson | |
This reverts commit 71f54037d767fc4d37ec7a7728006c6ce870db95. | |||
2012-09-07 | Revert "[build] Don't reset the submodule here, it breaks normal developer ↵ | Michael Hutchinson | |
workflow" This reverts commit 2b9a4c1a1373828114a9af1b5c9cee90d7058fba. | |||
2012-09-04 | Remove the other call to update the submodules | Alan McGovern | |
2012-09-04 | [build] Don't reset the submodule here, it breaks normal developer workflow | Alan McGovern | |
If we constantly reset submodules here it makes it impossible to do in-tree changes and use them. | |||
2012-06-05 | [Test] Propagate arguments when invoking 'make test' | Alan McGovern | |
2012-06-05 | [Tests] Import a new test runner using a git submodule and enable the tests | Alan McGovern | |
When MonoDevelop is compiled using --enable-tests we now build MonoMac, MacDev and the UnitTests assemblies. These can then be run on the bots as part of the build process. | |||
2012-05-17 | Fix #5113 - Improve building portability | Romain Tartière | |
2012-05-17 | [Build] Use the correct version string, it should be PACKAGE_VERSION | Alan McGovern | |