summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Fix stat when subdirectory is a fileFredrik Medley2017-12-192-1/+23
| | | | | | Make sure that stat on Windows, both with and without cache, returns "missing file" when running stat on notadir/foo where notadir is a file.
* Fix disk_interface_test.cc on Windows for 64-bit timestampFredrik Medley2017-12-192-1/+20
| | | | | subdir/subsubdir/.. seems to get the time of subdir/subsubdir on NTFS (Windows 7), not the time of subdir.
* Merge pull request #1219 from QuLogic/resolutionNico Weber2017-10-208-25/+60
|\ | | | | Use high-resolution timestamps
| * Fix minor typo.Elliott Sales de Andrade2017-09-161-1/+1
| |
| * Use strtoll when reading mtime from build log.Elliott Sales de Andrade2017-09-161-1/+1
| | | | | | | | This prevents overflow on Windows where 'long' is not 64-bit.
| * Fix some Windows troubles.Elliott Sales de Andrade2017-09-162-1/+8
| | | | | | | | | | Add parentheses so that constant does not overflow; include inttypes.h when using MinGW to get the proper macros.
| * Update checks for new stat fields.Elliott Sales de Andrade2017-09-161-4/+6
| | | | | | | | | | This uses the macros as defined by the man page, which, as noted in the comments, are defined correctly on as many libc's that I could check.
| * Use 64-bit-alignment-safe timestamp reading.Elliott Sales de Andrade2017-09-162-4/+11
| | | | | | | | | | Read and write the timestamp as two separate 32-bit integers in a fixed order to prevent any issues with alignment or byte order.
| * Add #define to get printf-format specifiers.Elliott Sales de Andrade2017-09-161-0/+3
| | | | | | | | | | This is needed on older compilers/stdlibs such as on Ubuntu Precise which is used on Travis.
| * Move #include for type definition to correct place.Elliott Sales de Andrade2017-09-162-1/+6
| | | | | | | | | | Not sure why the old way works in newer compilers; maybe they just pre-define these types by defaulting to a newer standard.
| * Read file timestamps in higher resolution.Elliott Sales de Andrade2017-09-161-5/+14
| | | | | | | | | | | | | | | | This uses nanoseconds on POSIX (±~292 years) and 100-ns increments on Windows (±~29247 years). The fallbacks to different structure fields is the only thing grabbed from #337, with a slight modification in implementation.
| * Make TimeStamp 64-bit.Elliott Sales de Andrade2017-09-168-19/+21
| | | | | | | | This prepares it for higher-resolution timestamps.
* | Merge pull request #1344 from loganchien/fix-buf-overrunNico Weber2017-10-201-1/+1
|\ \ | | | | | | Fix potential buffer overrun
| * | Fix potential buffer overrunLogan Chien2017-10-161-1/+1
| |/ | | | | | | | | This commit rearranges record size comparison and fread() to make sure fread() only reads the data that can fit into the buffer.
* | escape usage examplesRefael Ackermann2017-10-191-3/+3
|/
* mark this 1.8.2.gitNico Weber2017-09-111-1/+1
|
* Restore tolerance of self-referencing phony build statementsBrad King2017-09-088-3/+111
| | | | | | | | | | | | | | | | | | | | Since commit v1.8.0^2~3^2~1 (Teach RecomputeDirty to detect cycles in the build graph, 2015-11-13) we correctly reject self-referencing phony build statements like build a: phony a as cycles. Unfortunately this breaks support for CMake 2.8.12.x and 3.0.x because those versions incorrectly produce edges of this form (that we used to tolerate). In order to preserve compatibility with those CMake versions we need to restore tolerance of these edges. Add a special case to the manifest parser to filter out self-referencing inputs of phony edges of the form produced by those CMake versions. Warn by default, but add a `-w phonycycle={err,warn}` option to make it an error. Fixes: #1322
* Factor ManifestParser options into a structureBrad King2017-09-078-62/+75
| | | | | | This will allow more options to be added without updating everywhere that constructs a ManifestParser. Also extend the AssertParse function to take the options so tests can control them.
* mark this 1.8.1.gitNico Weber2017-09-061-1/+1
|
* fix normalizer test for _MAX_PATHTakuto Ikuta2017-09-051-28/+25
|
* mark this 1.8.0.gitNico Weber2017-09-041-1/+1
|
* Remove path component limit from input of CanonicalizePath in windowsTakuto Ikuta2017-08-302-59/+50
|
* Merge pull request #1111 from bradking/detect-cycles-earlyNico Weber2017-06-228-166/+175
|\ | | | | Detect build graph cycles as early as possible
| * Drop unnecessary cycle detection in Plan::AddTargetBrad King2017-06-194-100/+55
| | | | | | | | | | We now detect and reject cycles in DependencyScan::RecomputeDirty before Plan::AddTarget is called so we can assume DAG input to the Plan.
| * Teach RecomputeDirty to detect cycles in the build graphBrad King2017-06-193-20/+73
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | RecomputeDirty is the earliest traversal of the build graph complete with depfile-loaded dependencies. Teach it to detect cycles and fail immediately. This avoids the need to tolerate cycles in RecomputeDirty only to diagnose them later. It also enables future simplification of Plan and Builder logic because they will be able to assume DAG input. When RecomputeDirty detects a cycle, reject it with an error message like that previously produced by Plan::CheckDependencyCycle. Previously we used the stat state of each node to determine whether we reached it earlier in the walk. Retain this approach for leaf nodes, but add an explicit walk state mark for each Edge so that we can have a temporary mark to aid cycle detection.
| * Add infrastructure for efficient walks through the `Edge` graphBrad King2017-06-192-2/+11
| | | | | | | | Store a mark in each `Edge` to be updated as it is encountered by a walk.
| * Refactor RecomputeDirty to take a node instead of an edgeBrad King2017-06-195-47/+39
| | | | | | | | | | | | All call sites have a node on which they call `in_edge()` to call RecomputeDirty. Simplify call sites by taking the node directly and calling `in_edge()` internally.
| * Simplify BuildTest.StatFailureAbortsBuild test caseBrad King2017-06-191-2/+2
| | | | | | | | | | Remove a dependency cycle from the test case, as cycles are covered by other tests. Ensure this case covers stat failure on a valid graph.
* | Work around mtime being set to 0 sometimesPatrick Griffis2017-06-181-0/+5
|/
* Fix segfault on edge with no inputsColin Cross2017-06-162-1/+32
| | | | | | | | PR #1281 added a deference of most_recent_input without checking for NULL, which can occur if a build rule has no inputs. Check it for null before dereferencing, and add a test. Fixes #1290.
* Merge pull request #1156 from cdbennett/windows-binary-mode-outputNico Weber2017-06-141-0/+15
|\ | | | | Write subprocess output to stdout in binary mode
| * Write subprocess output to stdout in binary modeColin D Bennett2016-05-121-0/+15
| | | | | | | | | | | | | | | | Set stdout to binary mode while writing subprocess output, so that the CR in a CR LF sequence is not replaced with CR LF itself, which would result in CR CR LF. Based on patch posted by nico in issue #773 comment.
* | Always rebuild on errorsColin Cross2017-05-227-27/+45
| | | | | | | | | | | | | | | | | | | | https://groups.google.com/forum/#!msg/ninja-build/YQuGNrECI-4/ti-lAs9SPv8J discusses a case where an rule updates its output file and then fails. The next run of ninja considers the ouptut file clean and doesn't rebuild it. Always stat output files after they are built, and write the mtime into .ninja_log. Consider output files dirty if the recorded mtime is older than the most recent input file.
* | Add a test that fails if a rule updates it output file but failsColin Cross2017-05-223-2/+59
| | | | | | | | | | | | | | | | https://groups.google.com/forum/#!msg/ninja-build/YQuGNrECI-4/ti-lAs9SPv8J discusses a case where an rule updates its output file and then fails. The next run of ninja considers the ouptut file clean and doesn't rebuild it. Add a test for this case, which currently fails.
* | Move stat metric to DiskInterfaceColin Cross2017-05-222-1/+2
| | | | | | | | | | | | | | Stat is not always used through Node::Stat, it is often used directly through DiskInterface. THe next patches will cause it to be called even more often through DiskInterface, so move the metrics to DiskInterface.
* | Reduce GetFullPathName callsTakuto Ikuta2017-05-091-2/+4
| |
* | Fix for reviewTakuto Ikuta2017-05-095-20/+20
| |
* | Make clparser fastertikuta2017-05-086-79/+130
| | | | | | | | | | | | | | | | | | | | This patch improves perfromance of clparser. * Reduce the number of calling GetFullPathName. * Use StringPiece for Split and Join. * Add EqualsCaseInsensitive for StringPiece not to generate new string instance. * Add some utility member in StringPiece class.
* | Add string_piece_utilTakuto Ikuta2017-04-264-0/+251
| | | | | | | | | | | | | | | | | | | | | | | | | | Following functions are implemented for further performance optimization. * JoinStringPiece * SplitStringPiece * EqualsCaseInsensitiveASCII * ToLowerASCII To improve performance of CLParser, I will introduce above functions into include_normalize-win32.cc.
* | use usTakuto Ikuta2017-04-121-2/+2
| |
* | remove util.hTakuto Ikuta2017-04-121-1/+0
| |
* | Add clparser_perftestTakuto Ikuta2017-04-121-0/+158
| |
* | browse: Bind to localhost by defaultTej Chajed2017-03-091-2/+6
| | | | | | | | | | | | | | | | | | Previously the browse server would bind to "", which is translated to 0.0.0.0 (all interfaces), and then the hostname as retrieved by socket.gethostname() was presented to the user. The hostname is now "localhost" by default and is configurable, so the original behavior is achieved with `ninja -t browse -a ""`.
* | Merge pull request #1232 from nicolasdespres/fix-canon_perftestNico Weber2017-02-101-1/+1
|\ \ | | | | | | Fix compilation error in canon_perftest.
| * | Fix compilation error in canon_perftest.Nicolas Despres2017-02-021-1/+1
| | | | | | | | | | | | Introduced by 1cc730ddc27df52.
* | | Support restat when rebuilding manifestDan Willemsen2017-02-081-4/+13
|/ / | | | | | | | | | | | | | | As a fix for #874, we started reloading the entire manifest even if the manifest was never rebuilt due to a restat rule. But this can be slow, so call State::Reset instead, which also fixes the original crash. Fixes #987
* | Merge pull request #1181 from DanielWeber/issue-1161Nico Weber2017-01-2412-48/+57
|\ \ | | | | | | Allow more path components
| * | Use uint64_t for slash_bits consistentlyDaniel Weber2016-09-015-14/+15
| | | | | | | | | | | | The VS compiler complained about possible loss of data (and it was right!)
| * | Allow more path componentsDaniel Weber2016-08-229-36/+44
| | | | | | | | | | | | | | | | | | | | | - 60 instead of 30 path components - 64 instead of 32 backslashes in a path (windows only) Issue: 1161
* | | Close original pipe fd in subprocessesColin Cross2017-01-131-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Non-console subprocesses have the write end of a pipe connected to fds 1 and 2 for stdout and stderr, but they also have the it connected to whatever fd was assigned in the ninja process when the pipe was created. Add a call to posix_spawn_file_actions_addclose after the posix_spawn_file_actions_adddup2 calls to close the original fd once it has been dup'd to stdout and stderr. This fixes an issue seen in the Android build, where one of the subprocesses is used to start a background helper process. The background process attempts to close any inherited fds, but if ninja used a very large fd number due to a high parallelism count the background process would not close the fd and ninja would never consider the subprocess finished.
generated by cgit v0.12 at 2024-11-20 09:45:01 (GMT)