summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Merge pull request #962 from sgraham/pool-use-fixNico Weber2015-04-294-21/+43
|\ | | | | Fix pool use count going unbalanced
| * simplify & inlineScott Graham2015-04-272-18/+8
| |
| * avoid calling ResumeDelayedJobs insteadScott Graham2015-04-243-10/+8
| |
| * add commentScott Graham2015-04-241-0/+3
| |
| * Fix pool use count going unbalancedScott Graham2015-04-233-9/+40
| |
* | Merge pull request #743 from nicolasdespres/sigtermNico Weber2015-04-273-35/+72
|\ \ | |/ |/| Handle SIGTERM properly.
| * Forward interruption signal to child processes.Nicolas Despres2015-04-242-29/+34
| |
| * Allow SIGTERM for interruption.Nicolas Despres2015-04-243-6/+38
|/ | | | | Default signal sent by many other programs (mainly kill(1)) to gently terminates another one is SIGTERM.
* Merge pull request #958 from rui314/masterNico Weber2015-04-191-1/+1
|\ | | | | Run more than 34 processes on Win32 if we have 32+ cores.
| * Run more than 34 processes on Win32 if we have 32+ cores.Rui Ueyama2015-04-171-1/+1
|/ | | | | | | | For compatiblity reason, dwNumberOfProcessors in Win32 is capped at 32. So even if your machine has more than 32 cores, Ninja spawns at most 34 subprocesses. This patch fixes the issue by using GetNativeSystemInfo, which returns the system info from Wow64 point of view, instead of GetSystemInfo.
* Merge pull request #954 from nico/fixNico Weber2015-04-091-1/+1
|\ | | | | Fix an assert (and tests in --debug mode) after #921.
| * Fix an assert (and tests in --debug mode) after #921.Nico Weber2015-04-091-1/+1
| |
* | Merge pull request #953 from Sushisource/graphivz_fixNico Weber2015-04-081-1/+4
|\ \ | |/ |/| Fix backslashes in graphviz causing incorrect rendering on windows.
| * Fix backslashes in graphviz causing incorrect rendering on windows.Spencer Judge2015-04-081-1/+4
|/
* Merge pull request #951 from nico/cyclefix2Nico Weber2015-04-012-4/+58
|\ | | | | Don't get stuck on cyclic graphs where one edge has multiple outputs.
| * Don't get stuck on cyclic graphs where one edge has multiple outputs.Nico Weber2015-04-012-4/+58
|/ | | | | | | | | | | | | | | | | | | Fixes #934. Plan::AddSubTarget() tracks in want_ if each edge has been visited and visits every edge only once. But Plan::CheckDependencyCycle() worked on nodes however, so if a cycle was entered through an edge with multiple outputs, ninja would fail to detect that cycle. Move cycle detection to look for duplicate edges instead of nodes to fix this. The extra jump makes CheckDependencyCycle() a bit slower: for a synthetic build file with 10000 serial build steps, `ninja -n` now takes 0.32s instead of 0.26s before on my laptop. In practice, projects have a dependency change length on the order of 50, so there shouldn't be a noticeable slowdown in practice. (If this does end up being a problem: CheckDependencyCycle() currently does O(n) work and is called O(m) times from AddSubTarget(), so I think cycle checking is O(n^2) in the build depth. So instead of worrying about constant overhead, we could use a set<> instead of a stack<>. But it doesn't seem to be a problem in practice.)
* Merge pull request #950 from nico/cleanupNico Weber2015-04-012-14/+10
|\ | | | | Simplify Plan's cycle detection code. No behavior change.
| * Cleanup: Don't search stack for cycle elements twice.Nico Weber2015-04-011-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | The common case is that there is no cycle. In that case, CheckDependencyCycle() searched the stack for a dupe from the back, wouldn't find one, and return false. If there was a cycle, it would then search again from the front (probably because the push_back() that used to be here would invalidate the ri iterator). Since the push_back() is gone, just search from the front once. No intended behavior change.
| * Cleanup: Make stack a const ref now that it's no longer modified.Nico Weber2015-04-012-8/+10
| |
| * Cleanup: Don't modify during cycle checking.Nico Weber2015-04-011-6/+2
|/
* Merge pull request #949 from nico/staterrNico Weber2015-03-3112-138/+192
|\ | | | | Let Stat() have an err outparam instead of writing to stderr.
| * Let Stat() have an err outparam instead of writing to stderr.Nico Weber2015-03-3112-138/+192
|/ | | | | | | | | Also check for Stat() failure in a few more places. This way, ninja doesn't print two "ninja: error: " lines if stat() fails during a build. It also makes it easier to keep the stat tests quiet. Every caller of Stat() needs to explicitly log the error string if that's desired.
* Merge pull request #945 from nico/reldocsNico Weber2015-03-251-3/+6
|\ | | | | make notes on how to update docs more detailed
| * make notes on how to update docs more detailedNico Weber2015-03-251-3/+6
|/
* Merge pull request #944 from nico/flagsNico Weber2015-03-254-10/+66
|\ | | | | Add an opt-in flag to make duplicate edges an error (`-w dupbuild=err`).
| * Add an opt-in flag to make duplicate edges an error (`-w dupbuild=err`).Nico Weber2015-03-244-9/+65
| | | | | | | | | | This is step 1 on #931. Duplicated edges will become an error by default in the future.
| * Swap order of -k and -l in help output, to keep the list alphabetized.Nico Weber2015-03-241-1/+1
|/
* Merge pull request #943 from nico/ssshhNico Weber2015-03-244-11/+16
|\ | | | | Make tests quiet again, and prepare for making duplicate build edges an error.
| * Make tests quiet again.Nico Weber2015-03-242-5/+9
| |
| * Move warning emission on dupe edges from State to ManifestParser.Nico Weber2015-03-243-10/+11
|/ | | | | | | | This will make it easier to optionally make this an error (because ManifestParser has a way of printing errors), and it'll also make it easier to make the tests quiet again. No behavior change.
* Merge pull request #942 from nico/cyclefix2Nico Weber2015-03-212-1/+23
|\ | | | | Don't crash on cyclic references between rule bindings.
| * Preallocate edge node vectors. ~1% faster.Nico Weber2015-03-211-0/+2
| |
| * Recover slowdown for cyclic rule bindings fix.Nico Weber2015-03-211-9/+17
| |
| * Don't crash on cyclic references between rule bindings.Nico Weber2015-03-211-0/+12
|/ | | | | | | | | | | | | Fixes #902. This dynamically detects cycles. I like this approach less than detecting them statically when parsing rules [1], but it has the advantage that it doesn't break existing ninja manifest files. It has the disadvantage that it slows down manifest_parser_perftest by 3.9%. 1: https://github.com/martine/ninja/commit/cc6f54d6d436047
* Merge pull request #940 from nico/moredupesNico Weber2015-03-194-14/+38
|\ | | | | Another crash fix for duplicate edges. Fixes #939.
| * Another crash fix for duplicate edges. Fixes #939.Nico Weber2015-03-194-14/+38
|/ | | | | | | | | | | Patch #933 fixed a crash with duplicate edges by not adding edges to the graph if all the edge's outputs are already built by other edges. However, it added the edge to the out_edges of the edge's input nodes before deleting it, letting inputs refer to dead edges. To fix, move the check for deleting an edge above the code that adds inputs. Expand VerifyGraph() to check that nodes don't refer to edges that aren't present in the state.
* Merge pull request #938 from nico/aflclangNico Weber2015-03-1918-0/+51
|\ | | | | Add notes on using afl-fuzz to HACKING.
| * Add notes on using afl-fuzz to HACKING.Nico Weber2015-03-1918-0/+51
|/
* Merge pull request #937 from nico/statfixNico Weber2015-03-196-40/+88
|\ | | | | Make failing stat() calls abort the build.
| * Make failing stat() calls abort the build.Nico Weber2015-03-196-40/+88
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes #830, fixes #904. In practice, this either happens with 64-bit inodes and a 32-bit userspace when building without -D_FILE_OFFSET_BITS=64 in CFLAGS, or when a filename is longer than the system file length limit. Since DiskInterface::Stat() returns -1 on error, and Node used -1 on "stat state unknown", not aborting the build lead to ninja stat()ing the same file over and over again, until it finally ran out of stack. That's now fixed. * Change RecomputeOutputsDirty() to return success instead of dirty state (like RecomputeDirty()) and return the dirty state in a bool outparam * Node::Stat()s old return value wasn't used anywhere, change the function to return success instead and add an |err| outparam * Node::StatIfNecessary()'s old return value was used only in one place. Change that place to explicitly check status_known() and make StatIfNecessary() return success and add an |err| outparam * Plan::CleanNode() can now fail, make it return bool and add an |err| outparam
* | Merge pull request #936 from nico/cleanupNico Weber2015-03-193-10/+4
|\ \ | |/ | | Minor cleanups. No behavior change.
| * Add a missing &. (No behavior change, only used in tests.)Nico Weber2015-03-182-2/+2
| |
| * Env should only be about variables. No behavior change.Nico Weber2015-03-182-8/+2
|/
* Merge pull request #930 from nico/depcolonNico Weber2015-03-175-6/+31
|\ | | | | On unexpected output in a .d file, rebuild instead erroring.
| * On unexpected output in a .d file, rebuild instead erroring.Nico Weber2015-03-122-3/+21
| | | | | | | | Fixes #417.
| * Reject depfiles that don't contain a : after the target name.Nico Weber2015-03-124-3/+10
| | | | | | | | This is a prerequisite for fixing #417.
* | Merge pull request #933 from nico/dupeedgefixNico Weber2015-03-155-1/+46
|\ \ | | | | | | Build self-consistent graphs for dupe edges with multiple outputs.
| * | Build self-consistent graphs for dupe edges with multiple outputs.Nico Weber2015-03-155-1/+46
| |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes #867, both the crashes and "[stuck]" issues. The problem was that a duplicate edge would modify the in_edge of the outputs of the new build rule, but the edge corresponding to the old build rule would still think that the in_edge points to itself. `old_edge->outputs_[0]->in_edge()` would not return `old_edge`, which confused the scan logic. As fix, let `State::AddOut()` reject changing in_edge if it's already set. This changes behavior in a minor way: Previously, if there were multiple edges for a single output, the last edge would be kept. Now, the first edge is kept. This only had mostly-well-defined semantics if all duplicate edges are the same (which is the only case I've seen in practice), and for that case the behavior doesn't change. For testing, add a VerifyGraph() function and call that every time any test graph is parsed. That's a bit more code than just copying the test cases from the bug into build_test.cc, but it also yields better test coverage overall.
* | Merge pull request #928 from nico/novarEvan Martin2015-03-101-1/+0
|\ \ | |/ |/| Remove option no longer needed now that we don't use gtest.
| * Remove option no longer needed now that we don't use gtest.Nico Weber2015-03-101-1/+0
|/
generated by cgit v0.12 at 2025-03-24 01:41:57 (GMT)