Actually it can also be a directory or any argument for
test/unit runner. [Fixes GH-363]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@42091 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/gcbench.rb: separate GC bench framework and process.
* tool/hashbench1.rb, tool/hashbench2.rb: add two types GC bench.
hashbench1: many temporal objects (GC by newobj)
hashbench2: hash size becomes bigger and bigger (GC by malloc)
Two benchs are executed by `gcbench-hash' rule.
* tool/rdocbench.rb: separated.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@41458 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (loadpath.o): $LOAD_PATH should not be affected by
the revision number.
* common.mk (dmyloadpath.o): depends on nothing except for the source
file.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@41352 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Postponed jobs are registered with this API. Registered jobs
are invoked at `ruby-running-safe-point' as soon as possible.
This timing is completely same as finalizer timing.
There are two APIs:
* rb_postponed_job_register(flags, func, data): register a
postponed job with data. flags are reserved.
* rb_postponed_job_register_one(flags, func, data): same as
`rb_postponed_job_register', but only one `func' job is
registered (skip if `func' is already registered).
This change is mostly written by Aman Gupta (tmm1).
https://bugs.ruby-lang.org/issues/8107#note-15
[Feature #8107]
* gc.c: use postponed job API for finalizer.
* common.mk: add dependency from vm_trace.c to debug.h.
* ext/-test-/postponed_job/extconf.rb, postponed_job.c,
test/-ext-/postponed_job/test_postponed_job.rb: add a test.
* thread.c: implement postponed API.
* vm_core.h: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40940 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
using RDoc. Generate all rdoc related files
(same as `make rdoc') in temporary directory
and remove them. Excecution time, GC::Profiler
and results of GC.stat are printed.
* tool/rdocbench.rb: added for `rdoc-bench'.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40872 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Makefile.in (loadpath): $LOAD_PATH in miniruby is empty by default
now.
* win32/Makefile.sub (loadpath): separate for quirky cmd.exe syntax.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40784 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/verconf.h.in: generate verconf.h from the template and
rbconfig.rb.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40779 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (verconf.h): depend on the timestamp file instead of
rbconfig.rb, for parallel build.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40777 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (verconf.h): no needs to update always unless rbconfig.rb
is updated.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40771 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (verconf.h): $< cannot be used in explicit rules with
nmake.
* win32/Makefile.sub (CONFIG_H): create verconf.in instead of
verconf.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40770 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.in: save configured load path values into verconf.in.
* common.mk (verconf.h): create from verconf.in with shvar_to_cpp.rb.
* tool/shvar_to_cpp.rb: turn shell variables into C macros.
[Bug #7959]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40758 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* loadpath.c: split load path staffs from version.c.
* dmyloadpath.c: miniruby has no builtin load paths, so verconf.h is
not needed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40757 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
instead of {$(VPATH)}version.h to avoid confusion by VPATH between
top level version.h and include/ruby/version.h for build in-place.
[ruby-dev:47249] [Bug #8256]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/update-deps: Use "make -p all miniruby ruby golf" to extract
deendencies in makefiles.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40246 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
other than source directory. (its commit log also says the same
thing, but such failure is not reproducible on my environment
and the commit breaks build on my environment)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40239 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (id.h, id.c): Always generated in $(srcdir).
(ext/ripper/ripper.c): Passes $(PATH_SEPARATOR) too to the sub make.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40183 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (showflags): show the compiler version.
* configure.in (CC_VERSION): compiler with version option.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@39873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (realclean-local): miniprelude.c is made by srcs, so it
should not removed by distclean but by realclean. [Bug #6807]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@39492 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (showconfig): get rid of variable expansion by shell.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@39177 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/vpath.rb (VPath#def_options): hack for msys make, which
converts a command line argument to non-msys command seems like a
path list automagically. [Bug #7710] [ruby-core:51489]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38886 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (ext/dl/callback/callback.c): include in srcs-ext for
snapshot.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38562 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
up at top-level.
* .document: Moved doc/* entries to doc/.document
* doc/.document: ditto
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38312 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
"* common.mk ($(MINIPRELUDE_C)): -I may break make dist."
This didn't solve the issue.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38308 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk, defs/id.def, template/id.c.tmpl: generate id.c as well as id.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@38192 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/generic_erb.rb, tool/id2token.rb: add --path-separator option
for mingw where make and built ruby live in different world.
* tool/vpath.rb: extract from tool/instruction.rb.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37985 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
explicit rules for same target, but not merges exlicit rules and
implicit rules -- always explict rules win. So, need to add an
explist rule for probes.h.
reported by Heesob Park at [Bug #7421] [ruby-core:49839]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37802 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Debugging/profiling features will be located.
* vm_trace.c: expose C-level TracePoint APIs.
Note that they are experimental.
* vm_trace.c, include/ruby/debug.h: rename `rb_hook_flag_t'
to `rb_event_hook_flag_t'.
Macro names `RUBY_HOOK_FLAG_*' are also renamed to
`RUBY_EVENT_HOOK_FLAG_*'.
* debug.h, vm_debug.h: rename debug.h to vm_debug.h.
* common.mk: ditto.
* debug.c, main.c, vm_core.h: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37765 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
on it.
* common.mk (vm_dump.$(OBJEXT)): remove dependency on probes.h
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37733 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (RUBYLIB): purelib option in runruby.rb is deprecated
since r28841, so set to an empty list to get rid of a bogus path in
child processes, which caused an insecure operation exception in
test/ruby/test_encoding.rb:test_unsafe.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37719 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.in: see enable_dtrace for adding libelf on FreeBSD.
* common.mk: VPATH is not needed.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37717 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
see $(srcdir).
* common.mk (probes.dmyh): now created it on $(srcdir) always.
* win32/Makefile.sub (.SUFFIXES): removed common suffix.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37715 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
RUBY_DTRACE_FUNC_RETURN_HOOK): move from vm_core.h to new file
probes_helper.h for narrowing dependency to probes.h.
* common.mk (VM_CORE_H_INCLUDES): remove dependency to probes.h.
* common.mk (vm.$(OBJEXT)): add dependency to probes_helper.h.
* vm.c, vm_insnhelper.c: include probes_helper.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37711 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
based on r26235 by Yugui. On Solaris 10, low optimization level
may also be needed, e.g. optflags="-xO1" or "-xO0" with Oracle
SolarisStudio 12.3 cc.
* configure.in (--enable-dtrace): new option to enable/disable
DTrace support. By default, trying to enable if dtrace command
is found on the system. It is disabled when cross compiling.
* configure.in (RUBY_DTRACE_POSTPROCESS): new macro. checks whether
the dtrace on the system needs postprocessing with "dtrace -G".
The postprocessing is needed on Solaris 10 and other platforms.
* configure.in (RUBY_DTRACE_BSD_BROKEN): new macro. checks whether
the dtrace supports USDT.
* configure.in (DTRACE): move after RUBY_DTRACE_POSTPROCESS.
* configure.in (LD): On Solaris, /usr/ccs/bin/ld is preferred.
* configure.in, Makefile.in, common.mk (DTRACE_OBJ): new macro for
DTrace probe object generated by postprocessing with "dtrace -G".
* Makefile.in, common.mk (probes.$(OBJEXT)): DTrace probe object
generated by the postprocessing. New file probes.stamp is for
rebuilding related objects that may be modified by "dtrace -G".
* configure.in, Makefile.in, common.mk (DTRACE_GLOMMED_OBJ): new
macro for DTrace static library hacks.
* configure.in, Makefile.in (LIBRUBY_A_OBJS): ditto.
* Makefile.in, common.mk (ruby-glommed.$(OBJEXT)): new target with
rule for DTrace static library hacks.
* common.mk (DTRACE_DEPENDENT_OBJS): objects depended on probes.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37709 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Makefile.in (incs): probes.h is a platform dependent file, so it
cannot be a part of prereq target. move it to all-incs.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37678 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.in: disable dtrace because it doesn't work on FreeBSD.
* common.mk (clean-local): rm probes.h.
* common.mk (parse.o): depend $(PROBES_H_INCLUDES).
* common.mk (.d.h): moved from Makefile.in and use BASERUBY.
* tool/gen_dummy_probes.rb: reimplemented with ruby because sed is not
available on Windows Microsoft VC++ environment.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37636 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* array.c (empty_ary_alloc, ary_new): added array create DTrace probe.
* compile.c (rb_insns_name): allowing DTrace probes to access
instruction sequence name.
* Makefile.in: translate probes.d file to appropriate header file.
* common.mk: declare dependencies on the DTrace header.
* configure.in: add a test for existence of DTrace.
* eval.c (setup_exception): add a probe for when an exception is
raised.
* gc.c: Add DTrace probes for mark begin and end, and sweep begin and
end.
* hash.c (empty_hash_alloc): Add a probe for hash allocation.
* insns.def: Add probes for function entry and return.
* internal.h: function declaration for compile.c change.
* load.c (rb_f_load): add probes for `load` entry and exit, require
entry and exit, and wrapping search_required for load path search.
* object.c (rb_obj_alloc): added a probe for general object creation.
* parse.y (yycompile0): added a probe around parse and compile phase.
* string.c (empty_str_alloc, str_new): DTrace probes for string
allocation.
* test/dtrace/*: tests for DTrace probes.
* vm.c (vm_invoke_proc): add probes for function return on exception
raise, hash create, and instruction sequence execution.
* vm_core.h: add probe declarations for function entry and exit.
* vm_dump.c: add probes header file.
* vm_eval.c (vm_call0_cfunc, vm_call0_cfunc_with_frame): add probe on
function entry and return.
* vm_exec.c: expose instruction number to instruction name function.
* vm_insnshelper.c: add function entry and exit probes for cfunc
methods.
* vm_insnhelper.h: vm usage information is always collected, so
uncomment the functions.
12 19:14:50 2012 Akinori MUSHA <knu@iDaemons.org>
* configure.in (isinf, isnan): isinf() and isnan() are macros on
DragonFly which cannot be found by AC_REPLACE_FUNCS(). This
workaround enforces the fact that they exist on DragonFly.
12 15:59:38 2012 Shugo Maeda <shugo@ruby-lang.org>
* vm_core.h (rb_call_info_t::refinements), compile.c (new_callinfo),
vm_insnhelper.c (vm_search_method): revert r37616 because it's too
slow. [ruby-dev:46477]
* test/ruby/test_refinement.rb (test_inline_method_cache): skip
the test until the bug is fixed efficiently.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37631 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
on Solaris 10. [ruby-dev:46414] [Bug #7287]
* gc.c, signal.c, vm_core.h, common.mk: reflect the rename from atomic.h
to ruby_atomic.h.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37604 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (WPROGRAM): need same dependencies as PROGRAM.
* cygwin/GNUmakefile.in (uncommon.mk): move include position
below WPROGRAM definition to be defined in uncommon.mk.
* ext/extmk.rb (all, static): fix make rubyw.exe failure with make -jN.
If make of ruby.exe and rubyw.exe run in parallel, link dll and link
exe run in parallel, which causes link failure on mingw. To fix this,
we make ruby.exe and rubyw.exe in one make process.
[ruby-core:48007] [Bug #7165]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37246 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Makefile.in (RUNRUBY_COMMAND): split from RUNRUBY.
* common.mk (gdb-ruby): use runruby.rb to set up library path.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@37093 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk: revert r36867 partially, to use files in the source
directory when out-of-place build.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36909 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk: remove VPATH from rules for generated files, because
nmake does not work. [Bug #6956]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36867 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/id.h.tmpl, tool/id2token.rb: make id.h independent from
parse.h, and make parse.c dependent on it instead.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36864 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
so need to move denendency rules under the definition of it.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36802 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (ID_H_TARGET): timestamp file makes no senses now. object
files depending on id.h are always updated after parse.y is touched.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@36751 b2dd03c8-39d4-4d8f-98ff-823fe69b080e