Age | Commit message (Collapse) | Author | Files | Lines | |
---|---|---|---|---|---|
2017-11-06 | Revert back to using -fmodule-file for GCC and Clang | Boris Kolpackov | 1 | -35/+119 | |
Seeing that it's unclear if/when our patches for -fmodule-file-map will be accepted. | |||||
2017-10-17 | Fortify header dependency extraction against inconsistent behavior | Boris Kolpackov | 1 | -6/+19 | |
2017-10-03 | Extract system header search paths from GCC or compatible | Boris Kolpackov | 1 | -8/+16 | |
2017-09-01 | Add support for try_match(), use to handle deleted headers | Boris Kolpackov | 1 | -8/+17 | |
2017-08-30 | Switch to -fmodule-file=<name>=<file>, requires Clang 6.0/trunk | Boris Kolpackov | 1 | -7/+7 | |
2017-08-16 | Allow any directory separators when building src-out map | Boris Kolpackov | 1 | -3/+4 | |
2017-08-14 | Skip invalid paths when trying to remap generated headers | Boris Kolpackov | 1 | -45/+57 | |
2017-08-06 | Fix preprocessing/header generation race | Boris Kolpackov | 1 | -11/+16 | |
2017-08-06 | Filter out warning as error options during preprocessing | Boris Kolpackov | 1 | -13/+17 | |
2017-08-06 | Fix GCC -M/-MD logic some more | Boris Kolpackov | 1 | -9/+2 | |
2017-08-06 | And make sure warning are not treated as errors for GCC -MD run | Boris Kolpackov | 1 | -2/+9 | |
2017-08-06 | Make sure warnings are disables for GCC -M run | Boris Kolpackov | 1 | -8/+15 | |
2017-08-06 | Create output directory for module sidebuilds | Boris Kolpackov | 1 | -4/+13 | |
The fsdir{} injection doesn't do it if out == src. | |||||
2017-08-06 | Handle preprocessor errors caused by outdated generated headers | Boris Kolpackov | 1 | -9/+46 | |
This works with Clang and VC but not (yet) with GCC. | |||||
2017-08-05 | Minor changes | Boris Kolpackov | 1 | -2/+2 | |
2017-08-05 | Implement src-to-out re-mapping for generated headers | Boris Kolpackov | 1 | -58/+218 | |
2017-08-04 | Implement sidebuilding of installed modules | Boris Kolpackov | 1 | -36/+229 | |
2017-08-04 | Check for preprocessed value on source, not output target | Boris Kolpackov | 1 | -2/+4 | |
2017-08-02 | Change cc.preprocessed to {c,cxx}.preprocessed | Boris Kolpackov | 1 | -2/+2 | |
2017-08-02 | Save module map to pkg-config files | Boris Kolpackov | 1 | -2/+4 | |
2017-07-29 | Create libu{} and obj{} groups even if they don't exist | Boris Kolpackov | 1 | -4/+3 | |
Failed that target-type/pattern-specific variables for groups may not have effect. | |||||
2017-07-29 | Refine utility library semantics with "see through" theme | Boris Kolpackov | 1 | -6/+20 | |
2017-07-28 | Tweak generated header prefix mapping heuristics | Boris Kolpackov | 1 | -4/+17 | |
2017-07-28 | Implement support for linking whole archive | Boris Kolpackov | 1 | -3/+3 | |
2017-07-27 | Initial infrastructure for utility libraries | Boris Kolpackov | 1 | -56/+66 | |
2017-07-27 | Fix to print to diagnostics stream instead of std::cerr | Karen Arutyunov | 1 | -8/+6 | |
2017-07-26 | Fix bug in precise change detection logic | Boris Kolpackov | 1 | -4/+10 | |
2017-07-26 | Implement module map storage in .d, use -fmodule-file-map in GCC | Boris Kolpackov | 1 | -9/+48 | |
2017-07-25 | Fix bug in switching to preprocessed source | Boris Kolpackov | 1 | -7/+9 | |
2017-07-25 | Use absolute path to translation unit | Boris Kolpackov | 1 | -40/+46 | |
This seems to be the only sane way to obtain the same hash regardless of the working directory. | |||||
2017-07-24 | Issue diagnostics when we ignore ignorable changes | Boris Kolpackov | 1 | -2/+10 | |
2017-07-22 | Increase target data pad size | Boris Kolpackov | 1 | -2/+2 | |
2017-07-22 | Implement detection of ignorable changes (whitespaces, comments) | Boris Kolpackov | 1 | -81/+139 | |
2017-07-21 | Use binary mode when pre-parsing translation unit | Boris Kolpackov | 1 | -2/+3 | |
Failed that, the module export position we obtain cannot be reliably seeked to. | |||||
2017-07-21 | Adjust GCC module option name (-fmodule-file) | Boris Kolpackov | 1 | -1/+1 | |
2017-07-21 | Unset VC's IFCPATH if we are using custom std.* modules | Boris Kolpackov | 1 | -9/+46 | |
2017-07-21 | Keep partially preprocessed output around if using modules | Boris Kolpackov | 1 | -27/+52 | |
Since with modules we may have to recompile and otherwise unchanged translation unit because of a BMI change. | |||||
2017-07-19 | Add support for custom std.* modules in VC | Boris Kolpackov | 1 | -16/+48 | |
2017-07-05 | Keep track of translation unit type, make own BMI for impl unit first | Boris Kolpackov | 1 | -35/+54 | |
2017-06-29 | Pass entire BMI graph, not only direct imports and re-exports | Boris Kolpackov | 1 | -25/+56 | |
All current implementations require access to the entire graph (though Clang embeds the file references so we don't need to pass the options). | |||||
2017-06-24 | Make VC compatible with 'export module M;' by fixing up preprocessed output | Boris Kolpackov | 1 | -1/+31 | |
2017-06-22 | Add extra support for symbol exporting in modularized projects | Boris Kolpackov | 1 | -2/+24 | |
Since modules don't see each other's macros, we can use a single, keyword- like macro for dll-exporting that is managed by the build system (so no need for an "export" header). For example: cxx.features.symexport = true export __symexport void f (); | |||||
2017-06-20 | Adjust search logic for std.* modules | Boris Kolpackov | 1 | -14/+45 | |
1. Don't do fuzzy match for std.* modules (more precisely, for 'std' and std.*). This means if we want to build a std.* module, we have to specify its name explicitly: mxx{std-core}@./: cc.module_name = std.core The idea is that we don't want, say, std.core to be fuzzy-matched to some unrelated core.mxx while #2 below could have done the right thing. 2. We allow such modules to go unresolved with the assumption that they are pre-built and will be found by other means (e.g., VC's IFCPATH). | |||||
2017-06-20 | Force Clang not to reference source files from BMIs, add trace | Boris Kolpackov | 1 | -1/+12 | |
2017-06-16 | Add support for explicitly specifying module name on mxx{} target | Boris Kolpackov | 1 | -20/+31 | |
2017-06-16 | Clean up module-related diagnostics | Boris Kolpackov | 1 | -6/+24 | |
2017-06-16 | Store module information in depdb | Boris Kolpackov | 1 | -84/+231 | |
2017-06-15 | Implement module search in prerequisite libraries | Boris Kolpackov | 1 | -78/+164 | |
2017-06-14 | Get rid of warnings | Boris Kolpackov | 1 | -2/+2 | |
2017-06-14 | Next installment in C++ modules saga: module search, re-export support | Boris Kolpackov | 1 | -35/+341 | |