summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Prepare for v3.3.6v3.3.63.3-maintJoel Rosdahl2018-01-282-2/+2
|
* Update NEWSJoel Rosdahl2018-01-281-0/+17
|
* test: Verify that UNCACHED_ERR_FD worksJoel Rosdahl2018-01-281-0/+32
|
* Only write hash debug output if there is any dataJoel Rosdahl2018-01-281-2/+2
|
* Fix broken UNCACHED_ERR_FDLuboš Luňák2018-01-281-1/+1
| | | | | It shouldn't be closed on exec since that's exactly the point, to leak it to whatever ccache calls.
* doc: Improve instructions on how to compile in different directoriesJoel Rosdahl2018-01-211-26/+30
|
* Split .incbin directive in ccache source code to enable cache hitsJoel Rosdahl2018-01-211-5/+5
|
* Prepare for v3.3.5v3.3.5Joel Rosdahl2018-01-134-8/+15
|
* doc: Tweak automatic cleanup descriptionJoel Rosdahl2018-01-131-3/+3
|
* Update NEWSJoel Rosdahl2018-01-111-0/+9
|
* Don't skip compiler argument following AFFECTS_CPP without TAKES_ARGJoel Rosdahl2018-01-111-1/+3
| | | | | | Bug found by Mike Gelfand. Fixes #203.
* Clean upJoel Rosdahl2018-01-113-11/+8
|
* Clean up redundant expressionorbitcowboy2018-01-111-2/+2
|
* -fdebug-prefix-map is supported in clang 3.8 and newerTom Hughes2018-01-111-1/+1
| | | | | It's not documented until newer versions of clang, but support was added in 3.8: https://reviews.llvm.org/rL250094
* Fix leaked file descriptorAndreas Huber2018-01-113-4/+16
|
* Clang emits warnings for unused linker arguments, respect that!Edward Z. Yang2018-01-111-3/+8
| | | | | | | | | | | If ccache concludes the invocation with/without linker arguments is the same as before, then it may show/fail to show a warning when it should. I know we're not supposed to rely on the is clang check, but this solves it in normal cases. Fixes #189. Signed-off-by: Edward Z. Yang <ezyang@fb.com>
* doc: Improve markupJoel Rosdahl2018-01-111-2/+2
|
* Add some fallthrough comments to work around GCC compiler warningsGeert Bosch2018-01-112-2/+5
|
* Document how automatic cleanup worksJoel Rosdahl2018-01-102-5/+55
|
* Make ccache -c/--cleanup not take limit_multiple into accountJoel Rosdahl2018-01-105-35/+66
| | | | | | | Now "ccache -c/--cleanup" will trim the cache to the configured max limits instead of 80% (limit_multiple) of the max limits. This feels like a more intuitive behavior, and it also makes it behave like documented.
* test: Fix printing of failure reason with glob charactersJoel Rosdahl2018-01-101-1/+1
|
* Fix grammar in cleanup_* function namesJoel Rosdahl2018-01-104-9/+9
|
* test: Unset all CCACHE variables in environmentJoel Rosdahl2018-01-101-41/+18
| | | | | | Gotta catch 'em all. As suggested by Radu Rendec.
* cleanup: Improve loggingJoel Rosdahl2018-01-102-5/+17
|
* test: Check more cases of path normalizationJoel Rosdahl2018-01-101-3/+7
|
* Don't overwrite cached .d file with a version with absolute pathsJoel Rosdahl2018-01-102-5/+3
| | | | | | | | The .d file has already been put in the cache by to_cache (and potentially converted to relative paths using use_relative_paths_in_depfile). This fixes what seems to be a mistake in 38301c0d.
* test: Fix verbose printing of test case name with spaceJoel Rosdahl2017-12-281-1/+1
|
* Update NEWSJoel Rosdahl2017-07-191-0/+4
|
* Don't use cache when building precompiled header with changed depsTor Arne Vestbø2017-07-194-1/+55
| | | | | | | | | | | | | | If one of the included files in a precompiled header is touched, changing only its mtime, clang will produce a fatal error when the precompiled header is then used: file 'foo.h' has been modified since the precompiled header 'foo.pch' was built We need to take this into account when producing precompiled headers, so that we don't pick out stale objects from the cache. This works fine for direct mode, but in preprocess mode we don't have enough information to decide if the object is still valid, so we skip the cache entirely in that mode.
* Always use include DEFAULT_SLOPPINESS when overriding CCACHE_SLOPPINESSTor Arne Vestbø2017-07-191-10/+10
| | | | Otherwise the tests will have problems with backdated files.
* Fix detection of missing argument to -opt/--options-fileJoel Rosdahl2017-06-212-9/+11
| | | | Closes #171.
* Update log message and docs when using SLOPPY_FILE_STAT_MATCHESTor Arne Vestbø2017-06-152-3/+3
| | | | The size is no longer part of this check.
* Tweak commentJoel Rosdahl2017-06-151-3/+3
|
* Add comment about why SLOPPY_INCLUDE_FILE_[CM]TIME compares using >=tTor Arne Vestbø2017-06-151-0/+4
|
* Update NEWSJoel Rosdahl2017-03-251-0/+3
|
* Add test for directory in ignore_headers_in_manifestJoel Rosdahl2017-03-251-6/+27
| | | | | I should have added this test in a61d0f24 – that would have prevented issue #165.
* Fix DIR_DELIM_CH for win32 and non-win32 platforms, values were mixed upOded Shimon2017-03-251-2/+2
|
* Update NEWSJoel Rosdahl2017-03-251-0/+2
|
* Tweak source code formatJoel Rosdahl2017-03-251-1/+2
|
* Support using multiple -fdebug-prefix-mapAnders F Björklund2017-03-252-5/+41
| | | | | | | Previously only the last one was actually being applied to the hash. Note that the compiler will actually only *use* one of all the paths. Closes #163.
* Fixed the check for empty object file vs diagnosticsAlexey Tourbin2017-03-251-6/+6
| | | | | | The test for diagnostic file currently interferes with the check for empty object file. Since the check for diagnostic file is much more similar to the check for dependency file, I put it there.
* Removed time_of_compilation check wrt SLOPPY_FILE_STAT_MATCHESAlexey Tourbin2017-03-251-7/+5
| | | | | | | | | | | The whole code seems to be a thinko. For a hit, neither ctime nor mtime should be greater than or equal to time_of_compilation. The code only seems to work because time_of_compilation is 0 at this stage (i.e. has not been initialized). While at it, I also introduce an optimization: when sizes do not match, it's a good chance to bail out early; there is no point in further hashing the file.
* Revert "Better parsing of g* Options, and only use last argument"Joel Rosdahl2017-03-252-40/+21
| | | | | | | | | | | This reverts commit 21e2652c62762e5c9a017ea1523f6cfb3f1bcf7d. The feature introduced in #92 changes the order of debug options, leading to bugs like #149. It would be possible to add more logic to handle special cases like -gsplit-dwarf, but for now I'll just back out the patch and keep ccache ignorant about debug options. Fixes #149
* Test that the order of debug options is keptJoel Rosdahl2017-03-251-1/+33
|
* Prepare for v3.3.4v3.3.4Joel Rosdahl2017-02-174-4/+5
|
* Update NEWSJoel Rosdahl2017-02-171-2/+5
|
* Run test for undefined behaviour with TravisAnders Björklund2017-02-121-0/+3
|
* Undefined: avoid shift warning in manifestAnders Björklund2017-02-121-4/+6
| | | | manifest.c:264:46: runtime error: left shift of 72057594037927935 by 8 places cannot be represented in type 'long int'
* Undefined: avoid shift warning in murmurhashAnders Björklund2017-02-121-6/+6
| | | | murmurhashneutral2.c:18:16: runtime error: left shift of 169 by 24 places cannot be represented in type 'int'
* Undefined: avoid null warning in tmp_unlinkAnders Björklund2017-02-121-3/+9
| | | | util.c:1468:6: runtime error: null pointer passed as argument 1, which is declared to never be null
View Lorry Controller Status