Evan Martin [Tue, 17 Nov 2015 22:02:43 +0000 (14:02 -0800)]
provide a better title for the manual
In particular, this helps web search engines index it.
Nico Weber [Thu, 12 Nov 2015 05:46:33 +0000 (21:46 -0800)]
Merge pull request #1047 from pykello/master
Speed-up bash target auto-complete.
Nico Weber [Thu, 12 Nov 2015 05:46:19 +0000 (21:46 -0800)]
Merge pull request #1046 from zed0/master
Speed up zsh target completion
Evan Martin [Thu, 12 Nov 2015 01:19:00 +0000 (17:19 -0800)]
Merge pull request #1007 from mikesep/aix
Support for AIX
Nico Weber [Thu, 12 Nov 2015 01:13:30 +0000 (17:13 -0800)]
Merge pull request #999 from colincross/dumbstatus
Print status when edge finishes on dumb terminals
Nico Weber [Thu, 12 Nov 2015 00:51:29 +0000 (16:51 -0800)]
Merge pull request #964 from nicolasdespres/sighup
Cleanup build on SIGHUP.
Nico Weber [Thu, 12 Nov 2015 00:34:38 +0000 (16:34 -0800)]
Merge pull request #763 from drbo/master
Allow paths with international characters in depfiles
Nico Weber [Wed, 11 Nov 2015 21:54:14 +0000 (13:54 -0800)]
Try to fix #685
Evan Martin [Wed, 11 Nov 2015 21:34:13 +0000 (13:34 -0800)]
drop more references to github/martine
Nico Weber [Wed, 11 Nov 2015 21:27:16 +0000 (13:27 -0800)]
Update link to point to ninja-build.org
Nico Weber [Wed, 11 Nov 2015 19:54:07 +0000 (11:54 -0800)]
Merge pull request #1049 from nico/update-links
Make links point to org page
Nico Weber [Wed, 11 Nov 2015 19:53:15 +0000 (11:53 -0800)]
Make links point to org page
Nico Weber [Wed, 11 Nov 2015 19:36:16 +0000 (11:36 -0800)]
test commit
Hadi Moshayedi [Sat, 31 Oct 2015 10:13:33 +0000 (06:13 -0400)]
Speed-up bash target auto-complete.
These speed-ups include:
* Let compgen do the command substitution. Similar to
https://lists.gnu.org/archive/html/bug-bash/2012-03/msg00115.html
* Use "cut" instead of "awk" for separating fields.
Ben Falconer [Fri, 30 Oct 2015 14:43:32 +0000 (14:43 +0000)]
Speed up zsh target completion
Mike Seplowitz [Wed, 21 Oct 2015 22:37:50 +0000 (18:37 -0400)]
Separate bootstrapped build from final build
AIX does not support rebuilding ninja in-place from the bootstrapped ninja.
Nico Weber [Thu, 22 Oct 2015 03:07:02 +0000 (20:07 -0700)]
Merge pull request #1015 from moroten/docs-relpath
Describe why to use relative paths
Nico Weber [Thu, 22 Oct 2015 01:49:56 +0000 (18:49 -0700)]
Merge pull request #1040 from peter1000/patch-1
Fixes typo
Nico Weber [Wed, 21 Oct 2015 23:35:03 +0000 (16:35 -0700)]
Merge pull request #1029 from nicolasdespres/fix-doc-typo
Fix typo.
peter1000 [Mon, 19 Oct 2015 20:44:30 +0000 (18:44 -0200)]
Fixes typo
Nicolas Despres [Thu, 24 Sep 2015 18:02:43 +0000 (20:02 +0200)]
Fix typo.
Nicolas Despres [Wed, 29 Apr 2015 07:33:37 +0000 (09:33 +0200)]
Cleanup build on SIGHUP.
SIGHUP is sent when the connection hang up (i.e. when the terminal
window is closed or the ssh connection is closed).
Nico Weber [Mon, 7 Sep 2015 23:52:25 +0000 (16:52 -0700)]
Merge pull request #1014 from moroten/optimize-stringpiece
Optimize StringPiece hash map
Nico Weber [Mon, 7 Sep 2015 21:35:21 +0000 (14:35 -0700)]
Merge pull request #1018 from nico/spell
Merge LLVM r242069, make spell checking allocate less memory.
Nico Weber [Mon, 7 Sep 2015 21:26:19 +0000 (14:26 -0700)]
Merge LLVM r242069, make spell checking allocate less memory.
Fredrik Medley [Fri, 12 Jun 2015 21:59:12 +0000 (23:59 +0200)]
Optimize StringPiece hash map
Replace strncmp with memcmp to improve performance.
Signed-off-by: Fredrik Medley <fredrik.medley@gmail.com>
Fredrik Medley [Thu, 3 Sep 2015 19:27:19 +0000 (21:27 +0200)]
Describe why to use relative paths
Ninja does resolve relative paths and file system links in paths.
Therefore, such paths pointing to the same file will not match and may
lead to an invalid dependency graph.
Signed-off-by: Fredrik Medley <fredrik.medley@gmail.com>
Nico Weber [Thu, 27 Aug 2015 06:34:48 +0000 (23:34 -0700)]
Merge pull request #1009 from sgraham/fix-null.o
Don't output null.o during configure
Scott Graham [Thu, 27 Aug 2015 02:51:15 +0000 (19:51 -0700)]
Don't output null.o during configure
Mike Seplowitz [Tue, 13 Jan 2015 01:53:14 +0000 (20:53 -0500)]
Fix getopt for AIX
AIX supplies getopt but not getopt_long.
We can't use the embedded getopt implementation, since the constness of its
arguments doesn't match the AIX system routine.
Mike Seplowitz [Tue, 13 Jan 2015 01:48:07 +0000 (20:48 -0500)]
Implement GetLoadAverage on AIX using libperfstat
Mike Seplowitz [Tue, 13 Jan 2015 00:13:39 +0000 (19:13 -0500)]
Fix AIX compilation error related to printf macros
On AIX, inttypes.h gets indirectly included by build_log.h.
It's easiest just to ask for the printf format macros right away.
Mike Seplowitz [Tue, 13 Jan 2015 01:47:55 +0000 (20:47 -0500)]
Start AIX port
Mike Seplowitz [Mon, 12 Jan 2015 22:36:37 +0000 (17:36 -0500)]
Strip tabs from od's output
/bin/od on Solaris and AIX both generate tabs.
Nico Weber [Mon, 10 Aug 2015 15:35:33 +0000 (08:35 -0700)]
Merge pull request #1003 from sgraham/2015-compile
Fix compile on VS2015
Scott Graham [Mon, 10 Aug 2015 04:27:39 +0000 (21:27 -0700)]
Remove accidental changes
Scott Graham [Mon, 10 Aug 2015 04:25:09 +0000 (21:25 -0700)]
Some mucking with std:: for set_terminate and terminate_handler
Scott Graham [Mon, 10 Aug 2015 04:02:23 +0000 (21:02 -0700)]
Set _HAS_EXCEPTIONS=0 on MSVC
Nico Weber [Mon, 3 Aug 2015 00:30:44 +0000 (17:30 -0700)]
Merge pull request #998 from LindleyF/stdc-format-macros
Make sure not to re-define __STDC_FORMAT_MACROS.
Lindley French [Fri, 31 Jul 2015 00:33:11 +0000 (17:33 -0700)]
Make sure not to re-define __STDC_FORMAT_MACROS.
Colin Cross [Fri, 31 Jul 2015 00:30:52 +0000 (17:30 -0700)]
Print status when edge finishes on dumb terminals
On smart terminals ninja prints the status line both before
and after running a command, reusing the same line if possible.
On a dumb terminal that doesn't support reusing the line, it
only prints the status before starting the command, but prints
the output of the command when the command finishes, by which
point other commands may have started and printed their status
line. This makes it impossible to determine what command
produced a line of output.
Modify BuildEdgeStarted to only print the status line if the
command is going to lock the console, or if ninja is running
on a smart terminal. Modify BuildEdgeFinished to always
print the status line unless the command locked the console,
in which case the status was already printed and no other
command can have printed any lines.
The end result will be dumb terminal output that much more
closely matches smart terminal output. One disadvantage is
that dumb terminals won't show anything when starting a
command, making it harder to tell what commands are currently
running, but I expect most interactive uses of ninja will use
a smart terminal.
Nico Weber [Thu, 30 Jul 2015 18:21:47 +0000 (11:21 -0700)]
Merge pull request #997 from mikesep/assert-arg-order
Correct order of ASSERT_EQ arguments
Mike Seplowitz [Tue, 21 Jul 2015 22:13:26 +0000 (18:13 -0400)]
Correct order of ASSERT_EQ arguments
The correct order is ASSERT_EQ(expected, actual).
Nico Weber [Fri, 10 Jul 2015 19:24:23 +0000 (12:24 -0700)]
Merge pull request #988 from nico/slashes
win: print right slashes in 'unknown target' message
Nico Weber [Fri, 10 Jul 2015 19:15:00 +0000 (12:15 -0700)]
win: print right slashes in 'unknown target' message
Nico Weber [Sun, 5 Jul 2015 22:53:46 +0000 (15:53 -0700)]
Merge pull request #973 from sgraham/canonicalize-fix
Fix crash in attempting to canonicalize paths longer than _MAX_PATH
Nico Weber [Sun, 5 Jul 2015 22:53:28 +0000 (15:53 -0700)]
Merge pull request #979 from stinb/bootstrap-out-of-source
Bootstrap out of source
Nico Weber [Thu, 2 Jul 2015 00:52:16 +0000 (17:52 -0700)]
fix typos, punctuation, redundant words in HACKING.md
Nico Weber [Wed, 1 Jul 2015 00:00:42 +0000 (17:00 -0700)]
Merge pull request #984 from martine/vcvars
point people at more detailed build docs if they're on Windows
Evan Martin [Tue, 30 Jun 2015 22:26:20 +0000 (15:26 -0700)]
point people at more detailed build docs if they're on Windows
It's a hard balance between "most people want to run configure --bootstrap" and
"here's a massive wall of text of every detail about building". I think the
balance in this change is better than it was before -- most people on Windows
don't want to build their own binary.
Fixes #983.
Nico Weber [Mon, 29 Jun 2015 17:16:35 +0000 (10:16 -0700)]
mark this 1.6.0.git
Jason Haslam [Mon, 22 Jun 2015 22:28:04 +0000 (16:28 -0600)]
Fix bootstrap from a source path containing spaces.
Jason Haslam [Tue, 26 May 2015 22:47:18 +0000 (16:47 -0600)]
Search for generated headers relative to build dir.
Jason Haslam [Tue, 26 May 2015 21:19:28 +0000 (15:19 -0600)]
Allow configure script to bootstrap out of source.
Nico Weber [Mon, 22 Jun 2015 20:44:24 +0000 (13:44 -0700)]
Merge pull request #978 from martine/revert-971-bootstrap-out-of-source
Revert "Bootstrap out of source"
Nico Weber [Mon, 22 Jun 2015 20:35:37 +0000 (13:35 -0700)]
Revert "Bootstrap out of source"
Nico Weber [Fri, 19 Jun 2015 00:34:37 +0000 (17:34 -0700)]
Merge pull request #975 from nico/docs
Document the change from #921 in the manual.
Nico Weber [Fri, 19 Jun 2015 00:22:48 +0000 (17:22 -0700)]
Document the change from #921 in the manual.
Nico Weber [Thu, 18 Jun 2015 23:39:30 +0000 (16:39 -0700)]
Merge pull request #971 from stinb/bootstrap-out-of-source
Bootstrap out of source
Nico Weber [Thu, 18 Jun 2015 23:38:48 +0000 (16:38 -0700)]
Merge pull request #970 from nico/explain
Add a missing EXPLAIN() call.
Scott Graham [Thu, 18 Jun 2015 22:52:25 +0000 (15:52 -0700)]
don't alias input/output in ExtractDeps (i.e. actually works now)
Scott Graham [Thu, 18 Jun 2015 22:41:47 +0000 (15:41 -0700)]
propagate include normalization failure to caller instead
Scott Graham [Thu, 18 Jun 2015 22:06:42 +0000 (15:06 -0700)]
Error+exit -> Fatal
Nico Weber [Thu, 18 Jun 2015 21:47:41 +0000 (14:47 -0700)]
Merge pull request #974 from bradking/add-missing-initializer
Add missing member initializer in Edge constructor
Brad King [Tue, 16 Jun 2015 17:38:55 +0000 (13:38 -0400)]
Add missing member initializer in Edge constructor
In commit v1.1.0^2~15^2~20 (stub out an api and de-constify Pool,
2012-10-03) the Edge "pool_" member was added as a raw pointer but the
initializer was accidentally left out of the constructor. Add it now.
Scott Graham [Tue, 9 Jun 2015 17:15:11 +0000 (10:15 -0700)]
don't abort()
Scott Graham [Tue, 9 Jun 2015 00:17:55 +0000 (17:17 -0700)]
push error to caller and abort on failure
Scott Graham [Mon, 8 Jun 2015 20:21:55 +0000 (13:21 -0700)]
Fix crash in attempting to canonicalize paths longer than _MAX_PATH
Jason Haslam [Tue, 26 May 2015 22:47:18 +0000 (16:47 -0600)]
Search for generated headers relative to build dir.
Jason Haslam [Tue, 26 May 2015 21:19:28 +0000 (15:19 -0600)]
Allow configure script to bootstrap out of source.
Nico Weber [Fri, 15 May 2015 00:37:50 +0000 (17:37 -0700)]
Add a missing EXPLAIN() call.
Nico Weber [Sat, 2 May 2015 21:38:43 +0000 (14:38 -0700)]
Fix typo in comment.
Nico Weber [Wed, 29 Apr 2015 21:26:54 +0000 (14:26 -0700)]
Merge pull request #962 from sgraham/pool-use-fix
Fix pool use count going unbalanced
Nico Weber [Mon, 27 Apr 2015 23:42:39 +0000 (16:42 -0700)]
Merge pull request #743 from nicolasdespres/sigterm
Handle SIGTERM properly.
Scott Graham [Mon, 27 Apr 2015 18:12:43 +0000 (11:12 -0700)]
simplify & inline
Scott Graham [Fri, 24 Apr 2015 23:06:55 +0000 (16:06 -0700)]
avoid calling ResumeDelayedJobs instead
Scott Graham [Fri, 24 Apr 2015 21:25:17 +0000 (14:25 -0700)]
add comment
Nicolas Despres [Wed, 16 Apr 2014 08:08:42 +0000 (10:08 +0200)]
Forward interruption signal to child processes.
Nicolas Despres [Wed, 16 Apr 2014 08:05:25 +0000 (10:05 +0200)]
Allow SIGTERM for interruption.
Default signal sent by many other programs (mainly kill(1)) to gently
terminates another one is SIGTERM.
Scott Graham [Thu, 23 Apr 2015 22:54:28 +0000 (15:54 -0700)]
Fix pool use count going unbalanced
Nico Weber [Sun, 19 Apr 2015 03:43:24 +0000 (20:43 -0700)]
Merge pull request #958 from rui314/master
Run more than 34 processes on Win32 if we have 32+ cores.
Rui Ueyama [Fri, 17 Apr 2015 21:14:44 +0000 (14:14 -0700)]
Run more than 34 processes on Win32 if we have 32+ cores.
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.
Nico Weber [Thu, 9 Apr 2015 18:47:28 +0000 (11:47 -0700)]
Merge pull request #954 from nico/fix
Fix an assert (and tests in --debug mode) after #921.
Nico Weber [Thu, 9 Apr 2015 18:41:40 +0000 (11:41 -0700)]
Fix an assert (and tests in --debug mode) after #921.
Nico Weber [Wed, 8 Apr 2015 19:46:19 +0000 (12:46 -0700)]
Merge pull request #953 from Sushisource/graphivz_fix
Fix backslashes in graphviz causing incorrect rendering on windows.
Spencer Judge [Tue, 7 Apr 2015 21:18:43 +0000 (14:18 -0700)]
Fix backslashes in graphviz causing incorrect rendering on windows.
Nico Weber [Wed, 1 Apr 2015 21:18:39 +0000 (14:18 -0700)]
Merge pull request #951 from nico/cyclefix2
Don't get stuck on cyclic graphs where one edge has multiple outputs.
Nico Weber [Wed, 1 Apr 2015 15:09:28 +0000 (08:09 -0700)]
Don't get stuck on cyclic graphs where one edge has multiple outputs.
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.)
Nico Weber [Wed, 1 Apr 2015 15:18:49 +0000 (08:18 -0700)]
Merge pull request #950 from nico/cleanup
Simplify Plan's cycle detection code. No behavior change.
Nico Weber [Wed, 1 Apr 2015 15:00:27 +0000 (08:00 -0700)]
Cleanup: Don't search stack for cycle elements twice.
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.
Nico Weber [Wed, 1 Apr 2015 14:59:38 +0000 (07:59 -0700)]
Cleanup: Make stack a const ref now that it's no longer modified.
Nico Weber [Wed, 1 Apr 2015 14:56:14 +0000 (07:56 -0700)]
Cleanup: Don't modify during cycle checking.
Nico Weber [Tue, 31 Mar 2015 19:59:49 +0000 (12:59 -0700)]
Merge pull request #949 from nico/staterr
Let Stat() have an err outparam instead of writing to stderr.
Nico Weber [Tue, 31 Mar 2015 15:12:12 +0000 (08:12 -0700)]
Let Stat() have an err outparam instead of writing to stderr.
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.
Nico Weber [Wed, 25 Mar 2015 21:13:49 +0000 (14:13 -0700)]
Merge pull request #945 from nico/reldocs
make notes on how to update docs more detailed
Nico Weber [Wed, 25 Mar 2015 19:19:14 +0000 (12:19 -0700)]
make notes on how to update docs more detailed
Nico Weber [Wed, 25 Mar 2015 03:53:06 +0000 (20:53 -0700)]
Merge pull request #944 from nico/flags
Add an opt-in flag to make duplicate edges an error (`-w dupbuild=err`).
Nico Weber [Tue, 24 Mar 2015 21:16:07 +0000 (14:16 -0700)]
Add an opt-in flag to make duplicate edges an error (`-w dupbuild=err`).
This is step 1 on #931. Duplicated edges will become an error by default in
the future.
Nico Weber [Tue, 24 Mar 2015 20:37:11 +0000 (13:37 -0700)]
Swap order of -k and -l in help output, to keep the list alphabetized.
Nico Weber [Tue, 24 Mar 2015 02:51:35 +0000 (19:51 -0700)]
Merge pull request #943 from nico/ssshh
Make tests quiet again, and prepare for making duplicate build edges an error.