* Configuration prefers using the AR in the compiler specific bin dir.
* Fix pre-delete of archives so it uses it's own action, and hence allows the archive action to be a single command.
* That fixes the problem of command line limits on the CMD shell, for archives that have many source files.
[SVN r33475]
rules (run/unit-test) to set LD_LIBRARY_PATH variables when running test.
This is to accomotate cases where gcc is installed in non-standard location,
and the path to that location is not in LD_LIBRARY_PATH, but the binaries
still want to link to libstdc++.
[SVN r33307]
after all flags added by V2 from features. This was, explicit
warnings disabling options will after -Wall on the command line and take
effect.
[SVN r33251]
in favour of the latter (toolset.prebuilt). This cuts down the number
of generator invocations for prebuilt target and is more clear.
Thanks to Mark Evans for pointing this out.
[SVN r32952]
1. If when generating something, we find more that one suitable generators,
run them and more then one return something, immediately report ambiguity.
Don't care if the produced targets are the same. This is better that
running several generators all the time, performance wise.
2. Remove the notion of 'intermediate' virtual-targets. IIRC, they were used
to prevent staging of RSP files, and we don't stage them anyway now.
[SVN r29491]
are configured.
* tools/gcc.jam (init-link-flags): New parameter 'toolset'.
* tools/darwin.jam (init): Move flags inheriting out of here
* tools/intel-linux.jam (init): Likewise.
[SVN r25817]
* tools/gcc.jam
(init): passing $(condition) to init-link-flags
(init-link-flags): now takes new condition parameter
moved common flags out of it
* tools/intel-linux.jam
(init): moved call to inherit-flags into it because we need to call
gcc.init-link-flags with the 'condition' parameter.
* tools/darwin.jam
(init): moved call to inherit-flags into it because we need to call
gcc.init-link-flags with the 'condition' parameter.
Fixes related to init-link-flags
* tools/gcc.jam
(init): passing $(condition) to init-link-flags
(init-link-flags): now takes new condition parameter
moved common flags out of it
* tools/intel-linux.jam
(init): moved call to inherit-flags into it because we need to call
gcc.init-link-flags with the 'condition' parameter.
* tools/darwin.jam
(init): moved call to inherit-flags into it because we need to call
gcc.init-link-flags with the 'condition' parameter.
Patch from Andre Hentz.
[SVN r25816]
write
using gcc : ... : ... : <cxxflags>foo <linkflags>bar ;
* tools/common.jam (handle-options): The login for setting the common
options.
[SVN r25767]
* v2/tools/gcc.jam
(init): parse the new options and call the new rule.
(init-link-flags): new rule. Initialize flags based on linker type.
put back -minpure-text
* v2/test/conditionals.py
small modification to avoid empty binaries.
Patch from Andre Hentz.
[SVN r25764]
lib png : z : <name>png ;
lib z : : <name>z ;
now works: if you link to 'png' you'll also link to 'z'.
lib png : z : <file>png.a ;
lib z : : <file>z.a ;
now works too. The 'prebuilt.jam' modules which used to handle <file> for
all target kinds is now removed.
[SVN r25703]
* build/toolset.jam (inherit-generator): New parameter 'generators-to-ignore'.
* tools/gcc.jam: Ignore *link* generators when inheriting. Declare the
gcc specific generator.
* test/gcc_runtime.py: Test that <link-runtime>static is correctly handled.
The gcc specific generator is necessary exactly for that reason.
[SVN r25700]
In particular, this allows to explicitly link statically to some
libs, using <linkflags> feature. In static libs are specified at
the beginning of the command line, we'll just get link errors.
[SVN r25548]
of virtual targets, not the deep one with the 'str' method. The latter is
very slow
* tools/gcc.jam: Don't declare gcc.link and gcc.link.dll generators, as
they are already inherited from unix.jam. If we declare them, then for
each targets, the whole generation process is run twice. As result,
select-alternative is called for each target, and due to sloweness of
'str' the performance is horrible.
The net result is that running time on one testcase is down from 1580 secs
to less than 10.
[SVN r25512]
