Граф коммитов

659 Коммитов

Автор SHA1 Сообщение Дата
mame 47ea999b46 ext/coverage/: add the oneshot mode
This patch introduces "oneshot_lines" mode for `Coverage.start`, which
checks "whether each line was executed at least once or not", instead of
"how many times each line was executed".  A hook for each line is fired
at most once, and after it is fired, the hook flag was removed; it runs
with zero overhead.

See [Feature #15022] in detail.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@65195 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-10-20 05:33:04 +00:00
nobu fa8b08b424 Prefer `rb_fstring_lit` over `rb_fstring_cstr`
The former states explicitly that the argument must be a literal,
and can optimize away `strlen` on all compilers.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@65059 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-10-13 09:59:22 +00:00
ko1 daf194745b check argument type.
* iseq.c (iseqw_s_compile): check argument type (T_STRING) to
  avoid SEGV.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64915 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-10-04 15:22:53 +00:00
k0kubun 402001d6c7 iseq.c: prefix rb_ to non-static iseq functions
I assume we always prefix rb_ to non-static functions to avoid conflict.
These functions are not exported and safe to be renamed.

iseq.h: ditto
compile.c: ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64736 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-09-13 13:59:25 +00:00
mame 7c9af3e196 iseq.c (rb_vm_encoded_insn_data_table_init): add a cast to build on clang
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64521 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-23 08:49:23 +00:00
mame 5b72097225 iseq.c (rb_iseq_trace_set): refactoring by using encoded_insn_data
Now it uses encoded_insn_data to identify and replace each encoded insn.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64519 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-23 08:32:31 +00:00
mame d574683c40 iseq.c: add a map from encoded insn to insn data
This enhances rb_vm_insn_addr2insn which retrieves a decoded insn number
from encoded insn.
The insn data table include not only decoded insn number, but also its
len, trace and non-trace version of encoded insn.
This table can be used to simplify trace instrumentation.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64518 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-23 08:32:30 +00:00
ko1 b85b10c11f check trace flags at loading [Bug #14702]
* iseq.c (iseq_init_trace): at ISeq loading time, we need to check
  `ruby_vm_event_enabled_flags` to turn on trace instructions.
  Seprate this checking code from `finish_iseq_build()` and make
  new function. `iseq_ibf_load()` calls this funcation after loading.

* test/ruby/test_iseq.rb: add a test for this fix.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64514 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-23 04:12:14 +00:00
mame d65f7458bc parse.y: remove coverage-related code fragments
The code fragments that initializes coverage data were scattered into
both parse.y and compile.c.  parse.y allocated a coverage data, and
compile.c initialize the data.

To remove this cross-cutting concern, this change moves the allocation
from "coverage" function of parse.y to "rb_iseq_new_top" of iseq.c.
For the sake, parse.y just counts the line number of the original source
code, and the number is passed via rb_ast_body_t.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64508 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-22 10:38:56 +00:00
nobu c5f50296b4 prototized
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64506 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-22 10:22:02 +00:00
mame 33af0429ea thread.c (rb_reset_coverages): remove coverage counters from all ISeqs
When coverage measurement is enabled, the compiler makes each iseq have
a reference to the counter array of coverage.
Even after coverage measurement is disabled, the reference is kept.
And, if coverage measurement is restarted, a coverage hook will increase
the counter.  This is completely meaningless; it brings just overhead.

To remove this meaninglessness, this change removes all the reference
when coverage measuement is stopped.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64504 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-22 05:24:50 +00:00
mame 2138f24c70 insns.def (invokesuper): remove a dummy receiever flag hack for ZSUPER
This is just a refactoring.

The receiver of "invokesuper" was a boolean to represent if it is ZSUPER
or not.  This was used in vm_search_super_method to prohibit ZSUPER call
in define_method. (It is currently prohibited because of the limitation
of the implementation.)

This change removes the hack by introducing an explicit flag,
VM_CALL_SUPER, to signal the information.  Now, the implementation of
"invokesuper" is consistent with "send" instruction.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64268 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-08-10 07:45:16 +00:00
k0kubun e1be448840 mjit.c: disable calling JIT-ed code
when TracePoint is enabled. We're cancelling JIT-ed code execution AFTER
each instruction, but there is no guard before the first insn of method.

To prevent spoiling performance, I don't want to modify the JIT-ed code
to fix this. So this commit replaces `mjit_enabled` check with `mjit_call_p`
check.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63734 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-06-23 14:11:19 +00:00
mame 75d25ede15 Make VM_INSN_INFO_TABLE_IMPL=1 work
rb_iseq_insns_info_decode_positions is used only when
VM_INSN_INFO_TABLE_IMPL=2.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63645 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-06-13 04:51:43 +00:00
normal 609939f8c6 rb_vm_insn_addr2insn: use st to perform addr2insn mapping
The current VM_INSTRUCTION_SIZE is 198, so the linear search
painful during a major GC phase.

I noticed rb_vm_insn_addr2insn2 showing up at the top of some
profiles while working on some malloc-related stuff, so I
decided to attack it.

Most notably, the benchmark/bm_vm3_gc.rb improves by over 40%:

  https://80x24.org/spew/20180602220554.GA9991@whir/raw

[ruby-core:87361] [Feature #14814]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63594 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-06-06 20:57:48 +00:00
yui-knk 548896c927 iseq.c: Fix segfault when debug is not 0
* iseq.c (get_insn_info_succinct_bitvector): If
  VM_CHECK_MODE is 0, `body->insns_info.positions` is
  freed in `rb_iseq_insns_info_encode_positions`.
  Print `position` only when VM_CHECK_MODE is set.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63468 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-19 03:43:00 +00:00
nobu a12bfb3bbd iseq body local variables
* compile.c, iseq.c: extract body and param.keyword in iseq as
  local variables.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63441 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-16 01:40:44 +00:00
nobu 610c6ba5d2 iseq.c: iseq body local variables
* iseq.c: extract body and param.keyword in iseq as local
  variables.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63404 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-12 01:24:18 +00:00
nobu 67221be89b iseq.c: refactoring rb_iseq_free
* iseq.c (rb_iseq_free): call mjit_free_iseq only if iseq->body is
  not NULL too, as the function accesses the body.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63403 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-12 01:24:16 +00:00
shyouhei 75d5cf55de RSTRING_PTR is not guaranteed to be VALUE-aligned (retry)
Don't abuse struct RString to hold arbitrary memory region.
Raw pointer should just suffice.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63368 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-09 05:42:06 +00:00
mame 18c40b16d0 tool/ruby_vm/views/_insn_name_info.erb: Auto-detect the longest insn name
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63318 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-02 00:57:50 +00:00
mame 8bfc46a117 iseq.c (rb_iseq_disasm_insn): Change the width of insn names
Currently "trace_opt_send_without_block" (28 letters) is the longest
insn.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63317 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-05-02 00:57:47 +00:00
nobu 0dd9f7cc5c iseq.c: consistent rb_bug messages
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63275 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-27 13:14:09 +00:00
nobu a3fe1034c4 insns.def: checktype
* insns.def (checktype): split branchiftype to checktype and
  branchif, to make branch condition negation possible.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63225 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-21 10:52:52 +00:00
tenderlove 5bafb6c596 Add write barrier calls for direct marking objects.
This commit adds write barriers for objects marked from `rb_iseq_mark`.
r62851 introduced direct marking from iseqs to:

* keyword arg default values
* catch table iseqs
* VALUEs embedded in encoded instructions

This patch adds missing write barrier calls to those references.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63147 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-13 21:36:51 +00:00
nobu bc6e61f149 compile.c: do not dump pointers
* compile.c (ibf_dump_iseq_each): do not dump succ_index_table
  pointer.  positions are dumped as integer arrays.  pointer
  values are meaningless outside the process.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63099 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-05 07:04:39 +00:00
nobu c2007e191b iseq.c: indent iseq dump
* iseq.c (rb_iseq_disasm): indent catch_table iseq entries.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63091 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-04 11:00:39 +00:00
nobu 5ff72d0569 iseq.c: strip trailing spaces in dump
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63084 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-04 04:28:47 +00:00
nobu c700bb1a2b iseq.c: constfied validate_get_insn_info [ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62869 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-21 02:20:37 +00:00
nobu ce848356ae compile.c: fix ibf_load_code
* compile.c (ibf_load_iseq_each): manage iseq_size to point loaded
  objects in iseq_encoded.  now marking iseq scans iseq_encoded
  directly.

* test/ruby/test_iseq.rb (test_to_binary_with_objects): skip for
  now, but fix argument order of assert_equal.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62856 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-20 08:36:42 +00:00
tenderlove 9e26858e8c Reverting r62775, this should fix i686 builds
We need to mark default values for kwarg methods.  This also fixes
Bootsnap.  IBF iseq loading needed to mark iseqs as "having markable
objects".

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62851 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-19 18:21:54 +00:00
nobu c6e50e513c iseq.c: append local var index to name
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62831 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-19 00:32:52 +00:00
nobu a65345417d iseq.c: mark kwrest arg
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62829 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-19 00:32:51 +00:00
naruse 94c40622f5 Revert "Add direct marking on iseq operands"
This reverts commit r62706.

It causes SEGV on i686-linux (debian) and armv7l-linux-eabihf:
http://www.rubyist.net/~akr/chkbuild/debian/ruby-trunk/log/20180309T204300Z.diff.html.gz
http://rubyci.s3.amazonaws.com/scw-9d6766/ruby-trunk/log/20180309T211706Z.diff.html.gz

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62775 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-16 07:59:10 +00:00
naruse b099c5e620 Revert "Fix warning: cast from pointer to integer of different size"
This reverts commit r62709.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62773 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-16 07:59:08 +00:00
tenderlove cc7bd8e5ff Unconditionally define `TRACE_INSN_P`
`TRACE_INSN_P` doesn't need to know about encoded iseqs, it just needs
to look at decoded iseqs.  We have the decoded iseqs available, so no
reason to look at encoded ones.  This change allows us to clear
`original_iseq` from the iseq struct without any segvs (previously,
clearing `original_iseq` would cause the tests to crash).

* iseq.c (rb_iseq_trace_set): Only use decoded iseq with `TRACE_INSN_P`

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62750 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-14 23:27:10 +00:00
kazu af0d504f2d Fix warning: cast from pointer to integer of different size
```
.../ruby/iseq.c: In function ‘rb_vm_insn_null_translator’:
.../ruby/iseq.c:137:12: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
     return (int)addr;
            ^
```

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62709 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-10 00:34:24 +00:00
tenderlove 8952964976 Add direct marking on iseq operands
Directly marking iseq operands allows us to eliminate the "mark array"
stored on ISEQ objects, which will reduce the amount of memory ISEQ
objects consume.  This patch changes the iseq mark function to:

* Directly marks ISEQ operands
* Iterate over and mark child ISEQs

It also introduces two flags on the ISEQ object.  In order to mark
instruction operands, we have to disassemble the instructions and find
the instruction parameters and types.  Instructions may also be
translated to jump addresses.  Instruction sequences may get marked by
the GC *while* they're mid flight (being compiled).  The
`ISEQ_TRANSLATED` flag is used to indicate whether or not the
instructions have been translated to jump addresses so that when we
decode the instructions we know whether or not we need to go from jump
location back to original instruction or not.

Not all ISEQ objects have any markable objects embedded in their
instructions.  We can detect whether or not an ISEQ has markable objects
in the instructions at compile time.  If the instructions contain
markable objects, we set a flag `ISEQ_MARKABLE_ISEQ` on the ISEQ object.
This means that during the mark phase, we can skip decompilation if the
flag is *not* set.  In other words, we can avoid decompilation of we
know in advance there is nothing to mark.

`once` instructions have an operand that contains the result of a
one-time compilation of a regex.  Before this patch, that operand was
called an "inline cache", even though the struct was actually an "inline
storage".  This patch changes the operand to be an "inline storage" so
that we can differentiate between caches that need marking (the inline
storage) and caches that don't need marking (inline cache).

[ruby-core:84909]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62706 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-09 20:11:45 +00:00
k0kubun 33949a0320 compile.c: set catch_except_p flag
to be used for MJIT's optimization. It's not used for optimization
in this commit yet.

vm_core.h: added catch_except_p field.

iseq.c: show the flag in ISeq disasm for debugging.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62654 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-04 07:04:28 +00:00
k0kubun ed935aa5be mjit_compile.c: merge initial JIT compiler
which has been developed by Takashi Kokubun <takashikkbn@gmail> as
YARV-MJIT. Many of its bugs are fixed by wanabe <s.wanabe@gmail.com>.

This JIT compiler is designed to be a safe migration path to introduce
JIT compiler to MRI. So this commit does not include any bytecode
changes or dynamic instruction modifications, which are done in original
MJIT.

This commit even strips off some aggressive optimizations from
YARV-MJIT, and thus it's slower than YARV-MJIT too. But it's still
fairly faster than Ruby 2.5 in some benchmarks (attached below).

Note that this JIT compiler passes `make test`, `make test-all`, `make
test-spec` without JIT, and even with JIT. Not only it's perfectly safe
with JIT disabled because it does not replace VM instructions unlike
MJIT, but also with JIT enabled it stably runs Ruby applications
including Rails applications.

I'm expecting this version as just "initial" JIT compiler. I have many
optimization ideas which are skipped for initial merging, and you may
easily replace this JIT compiler with a faster one by just replacing
mjit_compile.c. `mjit_compile` interface is designed for the purpose.

common.mk: update dependencies for mjit_compile.c.

internal.h: declare `rb_vm_insn_addr2insn` for MJIT.

vm.c: exclude some definitions if `-DMJIT_HEADER` is provided to
compiler. This avoids to include some functions which take a long time
to compile, e.g. vm_exec_core. Some of the purpose is achieved in
transform_mjit_header.rb (see `IGNORED_FUNCTIONS`) but others are
manually resolved for now. Load mjit_helper.h for MJIT header.
mjit_helper.h: New. This is a file used only by JIT-ed code. I'll
refactor `mjit_call_cfunc` later.
vm_eval.c: add some #ifdef switches to skip compiling some functions
like Init_vm_eval.

win32/mkexports.rb: export thread/ec functions, which are used by MJIT.

include/ruby/defines.h: add MJIT_FUNC_EXPORTED macro alis to clarify
that a function is exported only for MJIT.

array.c: export a function used by MJIT.
bignum.c: ditto.
class.c: ditto.
compile.c: ditto.
error.c: ditto.
gc.c: ditto.
hash.c: ditto.
iseq.c: ditto.
numeric.c: ditto.
object.c: ditto.
proc.c: ditto.
re.c: ditto.
st.c: ditto.
string.c: ditto.
thread.c: ditto.
variable.c: ditto.
vm_backtrace.c: ditto.
vm_insnhelper.c: ditto.
vm_method.c: ditto.

I would like to improve maintainability of function exports, but I
believe this way is acceptable as initial merging if we clarify the
new exports are for MJIT (so that we can use them as TODO list to fix)
and add unit tests to detect unresolved symbols.
I'll add unit tests of JIT compilations in succeeding commits.

Author: Takashi Kokubun <takashikkbn@gmail.com>
Contributor: wanabe <s.wanabe@gmail.com>

Part of [Feature #14235]

---

* Known issues
  * Code generated by gcc is faster than clang. The benchmark may be worse
    in macOS. Following benchmark result is provided by gcc w/ Linux.
  * Performance is decreased when Google Chrome is running
  * JIT can work on MinGW, but it doesn't improve performance at least
    in short running benchmark.
  * Currently it doesn't perform well with Rails. We'll try to fix this
    before release.

---

* Benchmark reslts

Benchmarked with:
Intel 4.0GHz i7-4790K with 16GB memory under x86-64 Ubuntu 8 Cores

- 2.0.0-p0: Ruby 2.0.0-p0
- r62186: Ruby trunk (early 2.6.0), before MJIT changes
- JIT off: On this commit, but without `--jit` option
- JIT on: On this commit, and with `--jit` option

** Optcarrot fps

Benchmark: https://github.com/mame/optcarrot

|         |2.0.0-p0 |r62186   |JIT off  |JIT on   |
|:--------|:--------|:--------|:--------|:--------|
|fps      |37.32    |51.46    |51.31    |58.88    |
|vs 2.0.0 |1.00x    |1.38x    |1.37x    |1.58x    |

** MJIT benchmarks

Benchmark: https://github.com/benchmark-driver/mjit-benchmarks
(Original: https://github.com/vnmakarov/ruby/tree/rtl_mjit_branch/MJIT-benchmarks)

|           |2.0.0-p0 |r62186   |JIT off  |JIT on   |
|:----------|:--------|:--------|:--------|:--------|
|aread      |1.00     |1.09     |1.07     |2.19     |
|aref       |1.00     |1.13     |1.11     |2.22     |
|aset       |1.00     |1.50     |1.45     |2.64     |
|awrite     |1.00     |1.17     |1.13     |2.20     |
|call       |1.00     |1.29     |1.26     |2.02     |
|const2     |1.00     |1.10     |1.10     |2.19     |
|const      |1.00     |1.11     |1.10     |2.19     |
|fannk      |1.00     |1.04     |1.02     |1.00     |
|fib        |1.00     |1.32     |1.31     |1.84     |
|ivread     |1.00     |1.13     |1.12     |2.43     |
|ivwrite    |1.00     |1.23     |1.21     |2.40     |
|mandelbrot |1.00     |1.13     |1.16     |1.28     |
|meteor     |1.00     |2.97     |2.92     |3.17     |
|nbody      |1.00     |1.17     |1.15     |1.49     |
|nest-ntimes|1.00     |1.22     |1.20     |1.39     |
|nest-while |1.00     |1.10     |1.10     |1.37     |
|norm       |1.00     |1.18     |1.16     |1.24     |
|nsvb       |1.00     |1.16     |1.16     |1.17     |
|red-black  |1.00     |1.02     |0.99     |1.12     |
|sieve      |1.00     |1.30     |1.28     |1.62     |
|trees      |1.00     |1.14     |1.13     |1.19     |
|while      |1.00     |1.12     |1.11     |2.41     |

** Discourse's script/bench.rb

Benchmark: https://github.com/discourse/discourse/blob/v1.8.7/script/bench.rb

NOTE: Rails performance was somehow a little degraded with JIT for now.
We should fix this.
(At least I know opt_aref is performing badly in JIT and I have an idea
 to fix it. Please wait for the fix.)

*** JIT off
Your Results: (note for timings- percentile is first, duration is second in millisecs)

categories_admin:
  50: 17
  75: 18
  90: 22
  99: 29
home_admin:
  50: 21
  75: 21
  90: 27
  99: 40
topic_admin:
  50: 17
  75: 18
  90: 22
  99: 32
categories:
  50: 35
  75: 41
  90: 43
  99: 77
home:
  50: 39
  75: 46
  90: 49
  99: 95
topic:
  50: 46
  75: 52
  90: 56
  99: 101

*** JIT on
Your Results: (note for timings- percentile is first, duration is second in millisecs)

categories_admin:
  50: 19
  75: 21
  90: 25
  99: 33
home_admin:
  50: 24
  75: 26
  90: 30
  99: 35
topic_admin:
  50: 19
  75: 20
  90: 25
  99: 30
categories:
  50: 40
  75: 44
  90: 48
  99: 76
home:
  50: 42
  75: 48
  90: 51
  99: 89
topic:
  50: 49
  75: 55
  90: 58
  99: 99

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 11:22:28 +00:00
k0kubun fd44a5777f mjit.c: merge MJIT infrastructure
that allows to JIT-compile Ruby methods by generating C code and
using C compiler.  See the first comment of mjit.c to know what this
file does.

mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>.
After he invented great method JIT infrastructure for MRI as MJIT,
Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW
in MJIT. In addition to merging it, I ported pthread to Windows native
threads. Now this MJIT infrastructure can be compiled on Visual Studio.

This commit simplifies mjit.c to decrease code at initial merge. For
example, this commit does not provide multiple JIT threads support.
We can resurrect them later if we really want them, but I wanted to minimize
diff to make it easier to review this patch.

`/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby
developers may not know the name "mjit" and the file name should make
sure it's from Ruby and not from some harmful programs.  TODO: it may be
better to store this to some temporary directory which Ruby is already using
by Tempfile, if it's not bad for performance.

mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is
for triggering MJIT. This drops interface for AOT compared to the original
MJIT.

Makefile.in: define macros to let MJIT know the path of MJIT header.
Probably we can refactor this to reduce the number of macros (TODO).
win32/Makefile.sub: ditto.

common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this
commit separates MJIT infrastructure and JIT compiler code as independent
object files. As initial patch is NOT going to have ultra-fast JIT compiler,
it's likely to replace JIT compiler, e.g. original MJIT's compiler or some
future JIT impelementations which are not public now.

inits.c: define MJIT module. This is added because `MJIT.enabled?` was
necessary for testing.
test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this
wouldn't work with current code when JIT is enabled.
test/ruby/test_io.rb: skip this too. This would make no sense with MJIT.

ruby.c: define MJIT CLI options. As major difference from original MJIT,
"-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support
not only gcc/clang but also cl.exe (Visual Studio) in the future. But it
takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit"
options are allowed since some Ruby committers preferred it at Ruby
developers Meeting on January, and some of options are renamed.
This file also triggers to initialize MJIT thread and variables.
eval.c: finalize MJIT worker thread and variables.
test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit.

thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for
functions which are used by other files.
thread_win32.c: ditto, for Windows.  Those pthread porting is one of major
works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235.
thread.c: follow rb_ prefix changes

vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid
SEGV by race between JIT and GC of ISeq. The improvement was provided by
wanabe <s.wanabe@gmail.com>.
In JIT compiler I created and am going to add in my next commit, I found
that having `mjit_exec` after `vm_loop_start:` is harmful because the
JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn.
Executing non-FINISH frame is unexpected for my JIT compiler and
`exception_handler` triggers executions of such ISeqs. So `mjit_exec`
here should be executed only when it directly comes from `vm_exec` call.
`RubyVM::MJIT` module and `.enabled?` method is added so that we can skip
some tests which don't expect JIT threads or compiler file descriptors.

vm_insnhelper.h: trigger MJIT on method calls during VM execution.

vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because
rb_control_frame_struct is likely to be casted to another struct. The
last position is the safest place to add the new field.
vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an
optimization which are done in both MJIT and YARV-MJIT. So this change
is added in this commit. Calculating bp from ep is a little heavy work,
so bp is kind of cache for it.

iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue
is GCed to avoid SEGV.  TODO: unload some GCed units in some safe way.

gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous
JIT and GC executions may cause SEGV and so we should synchronize them.

cont.c: save continuation information in MJIT worker. As MJIT shouldn't
unload JIT-ed code which is being used, MJIT wants to know full list of
saved execution contexts for continuation and detect ISeqs in use.

mjit_compile.c: added empty JIT compiler so that you can reuse this commit
to build your own JIT compiler. This commit tries to compile ISeqs but
all of them are considered as not supported in this commit. So you can't
use JIT compiler in this commit yet while we added --jit option now.

Patch author: Vladimir Makarov <vmakarov@redhat.com>.

Contributors:
Takashi Kokubun <takashikkbn@gmail.com>.
wanabe <s.wanabe@gmail.com>.
Lars Kanis <lars@greiz-reinsdorf.de>.

Part of Feature 12589 and 14235.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 06:58:09 +00:00
nobu 5ce70a0831 iseq.c: internal lvar name
* iseq.c (local_var_name): name internal local variables as `?N`.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62100 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-30 02:08:30 +00:00
shyouhei e92684dae3 variable succ_index_table never actually used
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61941 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-19 03:19:58 +00:00
kazu ce7ad3a84e Use already casted variable
r61883

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61935 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-18 12:54:24 +00:00
nobu 2f67833473 iseq.c: dump as non-trace instructions
* iseq.c (iseq_data_to_ary): when OPT_CALL_THREADED_CODE is used,
  iseq_encoded is overwritten by instructions with trace and the
  original_iseq is not stored.  convert these instructions to the
  original instructions as external representation.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61890 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-17 15:28:37 +00:00
nobu 640c3dbfa1 iseq.c: flatten TS_NUM operand conditions
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61883 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-17 06:23:57 +00:00
nobu 6b5e0bd98c exclude flexible array size with old compilers
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61814 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-14 11:19:18 +00:00
mame 7ad9975fe7 iseq.h (struct iseq_catch_table_entry, iseq_compile_data_storage): Use FLEX_ARY_LEN
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61808 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-13 12:44:17 +00:00
mame 7da7716159 internal.h (FLEX_ARY_LEN): Add a macro to define a flexible array
Also, use it in iseq.c.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61804 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-13 11:45:25 +00:00
shyouhei 307547f02f flexible array member is a C99ism
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61801 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-13 10:47:32 +00:00
nobu 64be4cfb5e iseq.c: child iseq encoding
* iseq.c (iseq_inspect): preserve the encoding of child iseqs.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61767 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-11 08:34:55 +00:00
mame 0542d61e27 iseq.c: Succinct bitvector now supports 32-bit machine
Second try of succinct bitvector.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61757 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-10 05:57:15 +00:00
mame ab2c205c00 iseq.c: Add explicit casts from long to int (to pass compile on clang)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61740 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-09 14:11:39 +00:00
mame 83262f2489 iseq.c: Add a succinct bitvector implementation for insn_info_table
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61739 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-09 14:05:23 +00:00
mame 6d2de83bf0 iseq.c (iseq_data_to_ary): Avoid direct use of insns_info.positions
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61738 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-09 14:05:21 +00:00
mame 4928900814 Rename code_range to code_location
Because the name "code_range" is ambiguous with encoding's.
Abbreviations ("crange", and "cr") are also renamed to "loc".

The traditional "code_location" (a pair of lineno and column) is
renamed to "code_position".  Abbreviations are also renamed
(first_loc to beg_pos, and last_loc to end_pos).

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61721 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-09 08:45:35 +00:00
ko1 8a83cd100b use `getblockparamproxy` to pass blocks.
* compile.c (setup_args): use `getblockparamproxy` (`rb_block_param_proxy`)
  to represent a block parameter passing.

* vm_args.c (vm_caller_setup_arg_block): check `rb_block_param_proxy`
  instead of using `VM_CALL_ARGS_BLOCKARG_BLOCKPARAM` call flag.

* vm_core.h (VM_CALL_ARGS_BLOCKARG_BLOCKPARAM): removed.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61694 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-08 16:06:33 +00:00
mame 52057336b7 parse.y: Make consistent with the terms about code ranges and locations
"loc" was ambiguous; it might refer both a location and a code range.
This change uses "loc" for a location, and "crange" or "cr" for a code
range.

A location (abbr. loc) is a point in a program and consists of line
number and column number.  A code range (abbr. crange and cr) is a range
within a program and consists of a pair of locations which is the first
and the last.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61690 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-08 12:56:58 +00:00
mame 38c62063c0 node.h: remove NODE_PRELUDE
NODE_PRELUDE contains a `BEGIN` node, a main node, and compile_option.
This node is assumed that it must be located immediately under the root
NODE_SCOPE, but this strange assumption is not so good, IMO.

This change removes the assumtion; it integrates the former two nodes by
block_append, and moves compile_option into rb_ast_body_t.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61610 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-05 08:59:23 +00:00
mame 92b81dc597 make rb_iseq_new* accept rb_ast_body_t instead of NODE*
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61609 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-05 08:59:22 +00:00
mame 503b858cef node.h: define rb_ast_body_t and restructure rb_ast_t
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61608 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-05 08:59:20 +00:00
mame e1428e5c92 Refactor ISEQ_TYPE_DEFINED_GUARD out
This commit removes ISEQ_TYPE_DEFINED_GUARD because it is no longer
needed.  And this introduces ISEQ_TYPE_PLAIN which means that the iseq
does nothing special but just wrap an expression.  Currently, this is
used for once execution: `/foo#{ bar }baz/o`.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61601 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-05 00:49:41 +00:00
mame e743a35314 iseq.c: Refactor out rb_iseq_new_ifunc from rb_iseq_new_with_opt
It is too error-prone to pass IMEMO_IFUNC object as NODE*.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61592 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-04 07:07:49 +00:00
mame 1e4be7a852 iseq.c (rb_iseq_new_with_opt): handle the case where node is imemo_func
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61591 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-04 04:44:16 +00:00
nobu 9edca48cff insns_info.inc.tmpl: share large tablese
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61590 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-04 01:46:01 +00:00
mame 691b05e83c vm_core.h: make the algorithm of get_insn_info selectable
Currently, VM_INSN_INFO_TABLE_IMPL == 0 means linear search, and
VM_INSN_INFO_TABLE_IMPL == 1 means binary search.  I plan to add
succinct bitvector algorithm later.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61537 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-01 13:18:55 +00:00
mame 295838e6eb iseq.h: Extract position array from iseq_insn_info_entry
This makes TracePoint a bit fast by reducing cache misses of
`get_insn_info_binary_search`.

Also, I plan to use succinct bitvector algorithm for `get_insn_info`
instead of binary search.  This change will make it easy.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61536 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-01 12:51:21 +00:00
mame e1fce44ca6 iseq.c: fix build error when VM_CHECK_MODE is enabled
Follow up of r61534.  Sorry.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61535 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-01 09:30:06 +00:00
mame fd950df795 vm_core.h: refactoring of insns_info
This factors rb_iseq_constant_body#insns_info and #insns_info_size to
struct iseq_insn_info.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61534 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-01-01 09:16:27 +00:00
ko1 d7c6776f7c Disasm output contains code range.
* iseq.c (iseq_inspect): show also code range information.
  Note that `iseq_inspect` is used only for header of disasm.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61463 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-25 04:40:42 +00:00
ko1 cbac40b3e5 Remove "trace_instruction" compile option.
* iseq.h (rb_compile_option_struct): trace instruction is removed so that
  remove the trace_instruction compile option.
  Don't show warning (just ignore) for Ruby 2.5.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61450 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-24 17:07:53 +00:00
ko1 3044138bd7 RubyVM::InstructionSequence#trace_points.
* iseq.c (iseqw_trace_points): add `RubyVM::InstructionSequence#trace_points`
  method for tools which want to manipulate ISeq (and traces).

* test/ruby/test_iseq.rb: add a test for this method.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61427 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-23 14:46:59 +00:00
ko1 eb8c2773cb RubyVM::InstructionSequence#each_child.
* iseq.c (iseqw_each_child): add RubyVM::InstructionSequence#each_child
  method for tools which want to manipulate ISeq.

* test/ruby/test_iseq.rb: add a test for this method.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61425 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-23 12:48:24 +00:00
k0kubun 1942e92547 iseq.c: show first_lineno on iseq inspect
My motivation was to improve the output of `ruby --dump=insns xxx.rb`.
When one file has many iseqs, it's hard to find the one I want to read.

So I wanted `iseq_disasm` to show first_lineno. I unified the behavior
of `iseqw_disasm` for consistency.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61423 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-23 09:10:34 +00:00
kazu a747834a69 iseq.c: Remove unused macro in r61395
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61422 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-23 01:53:48 +00:00
nobu 2a723132e9 iseq.c: local variable names
* iseq.c (rb_insn_operand_intern): simplified the condition for
  local variable names, assume TS_LINDEX and TS_NUM combinations
  are only for local variables.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61421 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-23 00:55:29 +00:00
nobu 6e0074d0ff iseq.c (rb_insn_operand_intern): needs parenthesies
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61403 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-22 01:12:02 +00:00
nobu b4dc51ce10 iseq.c: defined insn operand
* iseq.c (rb_insn_operand_intern): improve operands of defined
  instruction.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61395 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-22 00:29:38 +00:00
yui-knk 69ae0f0f42 iseq.c (rb_iseq_code_range): Remove needless `;`
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61384 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-21 07:09:03 +00:00
ko1 e9a184ed63 add experimental API.
* iseq.c (rb_iseq_code_range): added to access iseq's code range.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61382 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-21 06:40:28 +00:00
nobu e105ee2bed iseq.c: block param name
* iseq.c (rb_insn_operand_intern): show local variable name at
  getblockparam/setblockparam too.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61360 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-20 14:10:06 +00:00
mame be3439026a iseq.c (get_insn_info): use binary search instead of linear search
This change introduces get_insn_info_binary_search, which is (should be)
equivalent to the old get_insn_info.
The old get_insn_info is renamed to get_insn_info_linear_search.  When
VM_CHECK_MODE > 0, the equivalence is validated at finish_iseq_build.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61353 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-20 07:38:24 +00:00
mame c08e8886ba compile.c: add a RUBY_EVENT_COVERAGE_LINE event for line coverage
2.5's line coverage measurement was about two times slower than 2.4
because of two reasons; (1) vm_trace uses rb_iseq_event_flags (which
takes O(n) currently where n is the length of iseq) to get an event
type, and (2) RUBY_EVENT_LINE uses setjmp to call an event hook.

This change adds a special event for line coverage,
RUBY_EVENT_COVERAGE_LINE, and adds `tracecoverage` instructions where
the event occurs in iseq.
`tracecoverage` instruction calls an event hook without vm_trace.
And, RUBY_EVENT_COVERAGE_LINE is an internal event which does not
use setjmp.

This change also cancells lineno change due to the deletion of trace
instructions [Feature #14104].  So fixes [Bug #14191].

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61350 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-20 04:24:14 +00:00
mame 2e24a66b88 iseq.c (finish_iseq_build): fix coverage leakage [Bug #14191]
Before this change, coverage.so had failed to measure some multiple-line
code fragments.  This is because removing trace instructions (#14104)
changed TracePoint's lineno (new lineno), and coverage counter array was
based on old lineno.
This change initializes coverage counter array based on new lineno.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61313 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-18 02:44:36 +00:00
k0kubun f5090e2f5c iseq.c: dump type of branchiftype on disasm
This makes easier to debug scripts related to r59950.

* before

$ ./ruby --dump=insns -e '"#{a}"'
== disasm: #<ISeq:<main>@-e>============================================
0000 putobject        ""                                              (   1)[Li]
0002 putself
0003 opt_send_without_block <callinfo!mid:a, argc:0, FCALL|VCALL|ARGS_SIMPLE>, <callcache>
0006 dup
0007 branchiftype     5, 15
0010 dup
0011 opt_send_without_block <callinfo!mid:to_s, argc:0, FCALL|ARGS_SIMPLE>, <callcache>
0014 tostring
0015 concatstrings    2
0017 leave

* after

$ ./ruby --dump=insns -e '"#{a}"'
== disasm: #<ISeq:<main>@-e>============================================
0000 putobject        ""                                              (   1)[Li]
0002 putself
0003 opt_send_without_block <callinfo!mid:a, argc:0, FCALL|VCALL|ARGS_SIMPLE>, <callcache>
0006 dup
0007 branchiftype     T_STRING, 15
0010 dup
0011 opt_send_without_block <callinfo!mid:to_s, argc:0, FCALL|ARGS_SIMPLE>, <callcache>
0014 tostring
0015 concatstrings    2
0017 leave

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61217 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-13 16:07:52 +00:00
ko1 975e3a19a7 do not disable `trace_` prefix insns.
* vm.c: introduce `ruby_vm_event_enabled_flags` which represents which
  event flags are enabled before.

* vm_trace.c: do not turn off `trace_` prefix instructions because turn on
  overhead is a matter if a program repeats turn on and turn off frequently.

* iseq.c (finish_iseq_build): respect `ruby_vm_event_enabled_flags`.

* vm_insnhelper.c (vm_trace): check `ruby_vm_event_flags` and disable
  lazy trace-off technique (do not disable traces).


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61122 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-11 19:17:25 +00:00
mame ec02c4861b thread.c (update_line_coverage): Use RUBY_EVENT_LINE
This change makes coverage use the general event type RUBY_EVENT_LINE
instead of a special event type RUBY_EVENT_COVERAGE.
Just a refactoring.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61043 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-06 06:39:05 +00:00
mame e5614fbcb0 iseq.c (iseq_load, iseq_data_to_ary): Fix a type error (for clang)
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61028 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-05 09:45:03 +00:00
mame f9c614c668 Hide ISeq#load
I disclosed it incorrectly at r61025.  Sorry.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61027 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-05 08:58:57 +00:00
mame a5e02f249d vm_core.h (rb_iseq_locatoin_t): add a field `code_range`
This change makes each ISeq keep NODE's code range.  This information is
needed for method coverage.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61025 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-12-05 08:56:50 +00:00
ko1 9f1c00ce65 Relax `rb_bug()` condition.
* iseq.c (rb_iseq_trace_set): simply return immediately if
  ISeq::compile_data is available. Not sure why this state
  is allowed, but exception during compile (or `ISeq::load`)
  can make such states.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60915 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-27 00:43:23 +00:00
ko1 eb3e32480d check invariant.
* iseq.c (rb_iseq_trace_set): at this point ISEQ_USE_COMPILE_DATA
  should not be set.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60871 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-21 02:57:01 +00:00
ko1 26451ab3ba introduce `trace_events' info for iseq.
* vm_core.h (rb_iseq_t::aux): add `trace_events` which represents
  which events are enabled on this iseq. With this information,
  we can skip useless trace-on changes for ISeqs.

* vm_trace.c (RUBY_EVENTS_TRACE_BY_ISEQ): moved to iseq.h and rename it
  with ISEQ_TRACE_EVENTS.

* iseq.h: introduce ISEQ_USE_COMPILE_DATA iseq (imemo) flag to represent
  COMPILE_DATA is available. In other words, iseq->aux.trace_events is not
  available when this flag is set.
  * ISEQ_COMPILE_DATA() is changed from a macro.
  * ISEQ_COMPILE_DATA_ALLOC() is added.
  * ISEQ_COMPILE_DATA_CLEAR() is added.

* iseq.c: use them.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60838 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-18 09:39:41 +00:00
ko1 d2bb08ecf0 remove a wrong assertion.
* iseq.c (rb_iseq_trace_set): remove a wrong assertion.
  This assertion checked `insn` is `trace_` prefix instruction
  but threaded code `insn` is original code.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60820 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-17 07:29:07 +00:00
ko1 1a96057626 remove `trace_` prefix insns lazily.
* vm_trace.c (update_global_event_hook): set only when tracing is added.
  If tracing was off (event flags are decreased), then ignore them.
  Next `trace_` prefix instruction will trace off itself (lazy tracing off).

* vm_insnhelper.c (vm_trace): trace-off for when trace is not needed.

* iseq.c (rb_iseq_trace_set): fix trace-off process (it was never off tracing).


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60817 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-17 06:24:55 +00:00
kazu 06ba86c933 Remove redundant last newline [ci skip]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60777 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-15 13:53:15 +00:00
ko1 665ba24b44 remove `trace` instruction. [Feature #14104]
* tool/instruction.rb: create `trace_` prefix instructions.

* compile.c (ADD_TRACE): do not add `trace` instructions but add
  TRACE link elements. TRACE elements will be unified with a next
  instruction as instruction information.

* vm_trace.c (update_global_event_hook): modify all ISeqs when
  hooks are enabled.

* iseq.c (rb_iseq_trace_set): added to toggle `trace_` instructions.

* vm_insnhelper.c (vm_trace): added.
  This function is a body of `trace_` prefix instructions.

* vm_insnhelper.h (JUMP): save PC to a control frame.

* insns.def (trace): removed.

* vm_exec.h (INSN_ENTRY_SIG): add debug output (disabled).


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60763 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-14 12:58:36 +00:00
nobu 54b8b8e4f7 iseq.c: disasm only once for each iseq
* iseq.c (rb_iseq_disasm): do not dump repeatedly same iseq which
  has been dumped by catch tables.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60735 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-10 08:26:44 +00:00
ko1 61e4c99962 refactoring about source line.
* iseq.c (find_line_no): renamed to rb_iseq_line_no().

* vm_backtrace.c (calc_lineno): add a comment why we need to use "pos-1".


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60733 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-10 05:26:52 +00:00
nobu 324b6d306d iseq.c: operand lvar
* iseq.c (rb_insn_operand_intern): show local variable operand
  name in unified instructions.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60732 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-09 23:08:01 +00:00
ko1 1c5ae5a163 insn_info/insns_info
* iseq.h (iseq_line_info_entry): rename to iseq_insn_info_entry.

* vm_core.h (rb_iseq_constant_body): rename field name line_info_table
  to insns_info and also from line_info_size to insns_info_size.

* compile.c (INSN): add struct insn_info to contain per insn information.

* compile.c (add_insn_info): added to add new insn_info entry.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60726 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-09 06:57:24 +00:00
ko1 4049ffb58e use GET_EC() directly
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60681 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-11-07 05:08:09 +00:00
ko1 85fcaf025d * node.h (ast_t): renamed to `rb_ast_t`.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60565 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-29 15:51:23 +00:00
ktsj a184500be2 iseq.c: remove obsolete comment
You can see ARGS_SIMPLE flag as follows:

 $ ruby --dump=insns -e itself

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60528 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-28 15:37:00 +00:00
ktsj e6f133b66a iseq.c: fix a typo
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60502 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-28 09:22:45 +00:00
ktsj 0b8842ca8a iseq.c: show all call flags
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60501 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-28 09:22:44 +00:00
mame e35fe8d11b Revert "Revert "Manage AST NODEs out of GC""
This re-introduces r60485.
This reverts commit 5a176b75b1.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60488 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-27 16:44:57 +00:00
mame 5a176b75b1 Revert "Manage AST NODEs out of GC"
This reverts commit 620ba74778.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60486 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-27 16:02:34 +00:00
mame 620ba74778 Manage AST NODEs out of GC
NODEs in AST are no longer objects managed by GC.  This change will
remove the restriction imposed by the GC.  For example, a NODE can use
more than five words (this is my primary purpose; we want to store the
position data for each NODE, for coverage library), or even a NODE can
have variable length (some kinds of NODEs have unused fields).
To do this, however, we need more work, since Ripper still uses T_NODE
objects managed by the GC.

The life time of NODEs is more obvious than other kinds of objects; they
are created at parsing, and they become disused immediately after
compilation.  This change releases all NODEs by a few `xfree`s after
compilation, so performance will be improved a bit.  In extreme example,
`eval("x=1;" * 10000000)` runs much faster (40 sec. -> 7.8 sec. on my
machine).

The most important part of this change is `ast_t` struct, which has
three contents: (1) NODE buffer (malloc'ed memory), (2) a reference to
the root NODE, and (3) an array that contains objects that must be
marked during parsing (such as literal objects).  Some functions that
had received `NODE*` arguments, must now receive `ast_t*`.

* node.c, node.h: defines `ast_t` struct and related operations.
* gc.c, internal.h: defines `imemo_ast`.
* parse.y: makes `parser_params` struct have a reference to `ast_t`.
  Instead of `rb_node_newnode`, use `rb_ast_newnode` to create a NODE.
* iseq.c, load.c, ruby.c, template/prelude.c.tmpl: modifies some
  functions to handle `ast_t*` instead of `NODE*`.
* test/ruby/test_gc.rb: ad-hoc fix for a failed test.  The test assumes
  GC eden is increased at startup by NODE object creation.  However,
  this change now create no NODE object, so GC eden is not necessarily
  increased.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60485 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-27 15:59:02 +00:00
nobu fd14454688 compile.c, iseq.c: consitfied NODE pointers
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60479 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-27 12:00:38 +00:00
ko1 f56288d7a8 replace `GET_THREAD()->ec` to `GET_EC()`.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60454 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-26 14:44:09 +00:00
ko1 837fd5e494 Use rb_execution_context_t instead of rb_thread_t
to represent execution context [Feature #14038]

* vm_core.h (rb_thread_t): rb_thread_t::ec is now a pointer.
  There are many code using `th` to represent execution context
  (such as cfp, VM stack and so on). To access `ec`, they need to
  use `th->ec->...` (adding one indirection) so that we need to
  replace them by passing `ec` instead of `th`.

* vm_core.h (GET_EC()): introduced to access current ec. Also
  remove `ruby_current_thread` global variable.

* cont.c (rb_context_t): introduce rb_context_t::thread_ptr instead of
  rb_context_t::thread_value.

* cont.c (ec_set_vm_stack): added to update vm_stack explicitly.

* cont.c (ec_switch): added to switch ec explicitly.

* cont.c (rb_fiber_close): added to terminate fibers explicitly.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60440 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-26 08:32:49 +00:00
nobu 6b818dd961 common conversion functions
* array.c (rb_to_array_type): make public to share common code
  internally.

* hash.c (rb_to_hash_type): make public to share common code
  internally.

* symbol.c (rb_to_symbol_type): make public to share common code
  internally.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60438 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-10-26 07:23:23 +00:00
yui-knk 29fa7a9f91 Remove a meaningless local variable assignment
* iseq.c (rb_iseq_disasm_insn): Remove a meaningless
  local variable assignment. `insn` is never changed
  in this function, so a result of `insn_op_types(insn)`
  is also never changed.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59629 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-08-21 13:38:35 +00:00
ko1 8dd9c12c58 move fields to ec.
* vm_core.h (rb_thread.h): move errinfo and trace_arg to
  rb_execution_context_t.

* cont.c (fiber_switch, rb_cont_call): do not restore "trace_arg" here.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59199 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-06-28 14:27:49 +00:00
nobu a57d295e36 iseq.c: preserve encoding
* iseq.c (iseqw_inspect): preserve path encoding in the result.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59065 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-06-12 04:35:53 +00:00
ko1 478003f6df rename absolute_path to realpath internally and introduce pathobj.
* vm_core.h: rename absolute_path to realpath because it is expected name.
  external APIs (#absolute_path methods) are remained.

* vm_core.h: remove rb_iseq_location_struct::path and
  rb_iseq_location_struct::absolute_path and introduce pathobj.
  if given path equals to given absolute_path (and most of case
  it is true), pathobj is simply given path String. If it is not same,
  pathobj is Array and pathobj[0] is path and pathobj[1] is realpath.

  This size optimization reduce 8 bytes and
  sizeof(struct rb_iseq_constant_body) is 200 bytes -> 192 bytes
  on 64bit CPU.

  To support this change, the following functions are introduced:
    * pathobj_path() (defined in vm_core.h)
    * pathobj_realpath() (ditto)
    * rb_iseq_path() (decl. in vm_core.h)
    * rb_iseq_realpath() (ditto)
    * rb_iseq_pathobj_new() (ditto)
    * rb_iseq_pathobj_set() (ditto)

* vm_core.h (rb_binding_t): use pathobj instead of path. If binding
  is given at eval methods, realpath (absolute_path) was caller's
  realpath. However, they should use binding's realpath.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58979 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-06-01 00:05:33 +00:00
watson1978 d0015e4ac6 Improve performance of implicit type conversion
To convert the object implicitly, it has had two parts in convert_type() which are
  1. lookink up the method's id
  2. calling the method

Seems that strncmp() and strcmp() in convert_type() are slightly heavy to look up
the method's id for type conversion.

This patch will add and use internal APIs (rb_convert_type_with_id, rb_check_convert_type_with_id)
to call the method without looking up the method's id when convert the object.

Array#flatten -> 19 % up
Array#+       ->  3 % up

[ruby-dev:50024] [Bug #13341] [Fix GH-1537]

### Before
       Array#flatten    104.119k (± 1.1%) i/s -    525.690k in   5.049517s
             Array#+      1.993M (± 1.8%) i/s -     10.010M in   5.024258s

### After
       Array#flatten    124.005k (± 1.0%) i/s -    624.240k in   5.034477s
             Array#+      2.058M (± 4.8%) i/s -     10.302M in   5.019328s

### Test Code
require 'benchmark/ips'

class Foo
  def to_ary
    [1,2,3]
  end
end

Benchmark.ips do |x|

  ary = []
  100.times { |i| ary << i }
  array = [ary]

  x.report "Array#flatten" do |i|
    i.times { array.flatten }
  end

  x.report "Array#+" do |i|
    obj = Foo.new
    i.times { array + obj }
  end

end

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58978 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-05-31 12:30:57 +00:00
nobu 14a730e91e immediate message mode of compile error
* compile.c (append_compile_error): set Qtrue for erred state with
  showing the message immediately.

* iseq.c (prepare_iseq_build): make immediate message mode if main
  or top level context, not to show the failed path twice in the
  first line.

* iseq.c (cleanup_iseq_build): raise default message exception if
  immediate message mode.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57313 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2017-01-12 07:41:35 +00:00
nobu a38b2f84f4 iseq.c: syntax error in compile_file
* iseq.c (iseqw_s_compile_file): deal with syntax error as well as
  compile, and should not abort when rescued.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56268 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-09-27 08:35:31 +00:00
nobu b15f76556f iseq.c: kwarg to compile
* iseq.c (iseqw_s_compile, iseqw_s_compile_file): allow keyword
  arguments as the options.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56266 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-09-27 06:23:36 +00:00
nobu 802e5677a4 iseq.c: file name encoding
* iseq.c (iseqw_s_compile): keep the encoding of file name.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56265 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-09-27 05:34:41 +00:00
rhe 2f551849c4 Use PRIuSIZE format specifier for size_t values
Use PRIuSIZE instead of PRIdSIZE. This fixes the exception message shown
on too large xmalloc2. This commit also fixes other incorrect use of
PRIdSIZE in other functions; though most of them are debug print.

* gc.c (heap_extend_pages, get_envparam_size, ruby_malloc_size_overflow,
  gc_profile_dump_on): Use PRIuSIZE instead of PRIdSIZE as the passed
  value is size_t, not ssize_t.

* iseq.c (get_line_info, rb_iseq_disasm_insn): Ditto.

* sprintf.c (rb_str_format): Ditto.

* thread_win32.c (native_thread_create): Ditto.

* vm.c (get_param): Ditto.

* ext/objspace/objspace_dump.c (dump_append_string_content,
  dump_object): Ditto.

* ext/socket/raddrinfo.c (host_str, port_str): Ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@56156 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-09-13 12:33:13 +00:00
nobu 36b9a37a80 iseq.c: undef allocator
* iseq.c (Init_ISeq): undefine allocator of InstructionSequence,
  to get rid of segfaults at method call on uninitialized object.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55981 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-08-22 00:02:59 +00:00
ko1 03cad83593 * vm_core.h: introduce VM_FRAME_RUBYFRAME_P()
and VM_FRAME_CFRAME_P().
  Most of case, RUBY_VM_NORMAL_ISEQ_P() is no
  longer needed.

* vm_core.h: introduce rb_obj_is_iseq().

* cont.c, vm.c: VM_FRAME_MAGIC_DUMMY with
  VM_FRAME_FLAG_CFRAME.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55804 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-08-03 01:50:50 +00:00
nobu c463366dfd rb_funcallv
* *.c: rename rb_funcall2 to rb_funcallv, except for extensions
  which are/will be/may be gems.  [Fix GH-1406]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55773 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-07-29 11:57:14 +00:00
ko1 9f60791a04 * vm_core.h: revisit the structure of frame, block and env.
[Bug #12628]

  This patch introduce many changes.

  * Introduce concept of "Block Handler (BH)" to represent
    passed blocks.

  * move rb_control_frame_t::flag to ep[0] (as a special local
    variable). This flags represents not only frame type, but also
    env flags such as escaped.

  * rename `rb_block_t` to `struct rb_block`.

  * Make Proc, Binding and RubyVM::Env objects wb-protected.

  Check [Bug #12628] for more details.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55766 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-07-28 11:02:30 +00:00
ko1 d8c3672b01 * iseq.c (Init_ISeq): undef ISeq.translate and ISeq.load_iseq
to prevent calling super classes' methods.

  Without this patch, you can write workaround like:

    class << RubyVM::InstructionSequence
      def translate; end
      undef translate
    end

* test/ruby/test_iseq.rb: add a test.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55656 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-07-13 05:26:00 +00:00
naruse 3b4e68cfab * iseq.h (struct iseq_compile_data): use struct rb_id_table
instead of st_table.

* iseq.c (prepare_iseq_build): don't allocate ivar_cache_table
  until it has at least one element.

* iseq.c (compile_data_free): free ivar_cache_table only if it
  is allocated.

* compile.c (get_ivar_ic_value): allocate if the table is not
  allocated yet.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55002 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-05-14 18:43:11 +00:00
tarui 44916ec448 * compile.c (iseq_compile_each): share InlineCache during same
instance variable accesses. Reducing memory consumption,
	  rasing cache hit rate and rasing branch prediction hit rate
	  are expected. A part of [Bug #12274].

	* iseq.h (struct iseq_compile_data): introduce instance
	  variable IC table for sharing.

	* iseq.c (prepare_iseq_build, compile_data_free):
	  construct/destruct above table.


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54976 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-05-11 12:50:38 +00:00
nobu 7504be10ef remove rb_thread_t::base_block
* error.c (rb_compile_err_append): rb_thread_t::base_block is no
  longer used.
* iseq.c (rb_iseq_compile_with_option): ditto, no protection is
  needed.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54345 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-28 21:40:55 +00:00
nobu d7935475fc remove rb_thread_t::parse_in_eval
* parse.y (struct parser_params): move parse_in_eval flag from
  rb_thread_t.
* parse.y (rb_parser_set_context): set parsing context, not only
  mild error flag.
* iseq.c (rb_iseq_compile_with_option): the parser now refers no
  thread local states to be restored.
* vm_eval.c (eval_string_with_cref): ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54343 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-28 21:39:24 +00:00
nobu 3176a634ca suppress warning
* iseq.c (rb_iseq_compile_with_option): suppress
  maybe-uninitialized warning by gcc 4.8.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54217 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-21 13:53:25 +00:00
nobu bc343b851d SyntaxError message at iseq compile
* iseq.c (rb_iseq_compile_with_option): make the parser in mild
  error.
* load.c (rb_load_internal0): ditto.
* parse.y (yycompile0): return the error message within the error
  to be raised.  [Feature #11951]
* parse.y (parser_compile_error): accumulate error messages in the
  error_buffer.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-19 05:46:20 +00:00
nobu cf09c0c351 iseq.h: coverage_enabled flag
* iseq.c (prepare_iseq_build): enable coverage by coverage_enabled
  option, not by parse_in_eval flag in the thread context.
* iseq.h (rb_compile_option_struct): add coverage_enabled flag.
* parse.y (yycompile0): set coverage_enabled flag if coverage
  array is made.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54081 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-10 08:34:18 +00:00
nobu 9871d88ee4 iseq.c: set coverage at once
* iseq.c (prepare_iseq_build): set coverage at once, not
  repeatedly resetting.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54076 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-03-10 05:32:49 +00:00
normal 1282a4a895 fstring early for internal iseq
All of the strings created here eventually get converted to
fstrings when they are frozen into the iseq.  Prepare the
fstring early so we may reduce a one or two objects.

This is a very minor change, mainly for the '<main>' dedupe.

* compile.c (caller_location): use rb_fstring_cstr for "<compiled>"
  (it is converted to fstring anyways inside rb_iseq_new_with_opt)
* iseq.c (iseqw_s_compile): ditto
* iseq.c (rb_iseq_new_main): use rb_fstring_cstr for "<main>"
* vm.c (Init_VM): ditto, share with with above
* iseq.c (iseqw_s_compile_file): rb_fstring before rb_io_t->pathv
  share "<main>" with above
* vm.c (rb_binding_add_dynavars): fstring "<temp>" immediately

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53659 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-26 06:23:47 +00:00
nobu 22121544e1 iseq.c: mark parent iseq
* iseq.c (rb_iseq_mark): mark parent iseq to prevent dynamically
  generated iseq by eval from GC.  [ruby-core:72620] [Bug #11928]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53524 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-13 07:56:51 +00:00
nobu 1b39a6e539 iseq.c: mark parents of wrapped iseq
* iseq.c (iseqw_mark): as wrapped iseq is isolated from the call
  stack, it needs to take care of its parent and ancestors, so
  that they do not become orphans.  [ruby-core:72620] [Bug #11928]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53514 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-12 15:17:19 +00:00
nobu 92f33ef216 iseq.c: volatile only on gcc4.8
* iseq.c (rb_iseq_compile_with_option): no volatile on gcc other
  than 4.8.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-08 10:52:24 +00:00
nobu 0455a25541 iseq.c: make local variables volatile
* iseq.c (rb_iseq_compile_with_option): prepare arguments outside
  EXEC_TAG, and make local variables volatile not to be clobbered
  by longjmp.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53460 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-08 07:47:49 +00:00
shugo 115163421a * iseq.c (rb_iseq_compile_with_option): move variable initialization
code to avoid maybe-uninitialized warnings by gcc 4.8.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53458 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2016-01-08 00:38:40 +00:00
nobu f84ba30af5 iseq.c: suppress warnings
* iseq.c (rb_iseq_compile_with_option): suppress "clobbered"
  warnings by old gcc.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53355 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-28 06:42:18 +00:00
nobu 89b3281668 iseq.c: narrow down protected region
* iseq.c (rb_iseq_compile_with_option): narrow down protected
  region, and check/convert/prepare arguments before setting
  base_block which needs to roll back.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53282 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-24 14:15:14 +00:00
hsbt 52cd994814 * enum.c: fix a typo in documentation.
[ci skip][fix GH-1140] Patch by @jutaz
* io.c: ditto.
* iseq.c: ditto.
* numeric.c: ditto.
* process.c: ditto.
* string.c: ditto.
* vm_trace.c: ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53105 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-14 02:52:14 +00:00
nobu a2fcdb4fbe iseq.c: use rb_check_funcall
* iseq.c (rb_iseq_load_iseq): use rb_check_funcall instead of
  repeating method search.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53048 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-12 07:07:42 +00:00
ko1 4b8da4b9c8 * iseq.c: rename methods
RubyVM::InstructionSequence#to_binary_format -> #to_binary
  RubyVM::InstructionSequence.from_binary_format -> .load_from_binary
  RubyVM::InstructionSequence.from_binary_format_extra_data ->
                             .load_from_binary_extra_data

* iseq.c: fix document of iseq.to_binary.
  [Fix GH-1134]

* sample/iseq_loader.rb: catch up this change.

* test/lib/iseq_loader_checker.rb: ditto.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53004 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-09 15:10:57 +00:00
ko1 eef8beb695 add documentation for RubyVM::InstructionSequence#to_binary_format and RubyVM::InstructionSequence.from_binary_format
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52990 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-09 02:41:14 +00:00
nobu bb4d72c51f make local symbols static
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52964 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-08 15:07:41 +00:00
ko1 23124ffbe8 * iseq.c (iseq_s_load): fix mysterious bug.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52953 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-08 14:05:44 +00:00
svn c49cb7fed4 * remove trailing spaces.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52950 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-08 13:59:15 +00:00
ko1 3dbb390180 * introduce new ISeq binary format serializer/de-serializer
and a pre-compilation/runtime loader sample.
  [Feature #11788]

* iseq.c: add new methods:
  * RubyVM::InstructionSequence#to_binary_format(extra_data = nil)
  * RubyVM::InstructionSequence.from_binary_format(binary)
  * RubyVM::InstructionSequence.from_binary_format_extra_data(binary)

* compile.c: implement body of this new feature.

* load.c (rb_load_internal0), iseq.c (rb_iseq_load_iseq):
  call RubyVM::InstructionSequence.load_iseq(fname) with
  loading script name if this method is defined.

  We can return any ISeq object as a result value.
  Otherwise loading will be continue as usual.

  This interface is not matured and is not extensible.
  So that we don't guarantee the future compatibility of this method.
  Basically, you should'nt use this method.

* iseq.h: move ISEQ_MAJOR/MINOR_VERSION (and some definitions)
  from iseq.c.

* encoding.c (rb_data_is_encoding), internal.h: added.

* vm_core.h: add several supports for lazy load.
  * add USE_LAZY_LOAD macro to specify enable or disable of
    this feature.
  * add several fields to rb_iseq_t.
  * introduce new macro rb_iseq_check().

* insns.def: some check for lazy loading feature.

* vm_insnhelper.c: ditto.

* proc.c: ditto.

* vm.c: ditto.

* test/lib/iseq_loader_checker.rb: enabled iff suitable
  environment variables are provided.

* test/runner.rb: enable lib/iseq_loader_checker.rb.

* sample/iseq_loader.rb: add sample compiler and loader.

    $ ruby sample/iseq_loader.rb [dir]

  will compile all ruby scripts in [dir].
  With default setting, this compile creates *.rb.yarb files
  in same directory of target .rb scripts.

    $ ruby -r sample/iseq_loader.rb [app]

  will run with enable to load compiled binary data.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52949 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-08 13:58:50 +00:00
nobu 506b25aabf error.c: name_err_local_variables
* error.c (name_err_local_variables): new method
  NameError#local_variables for internal use only.
  [Feature #11777]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52942 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-08 05:27:10 +00:00
ko1 8dce6b15cc * iseq.c (iseq_load): disable peephole optimization option
because apply it multiple times change the sequence.
  (iseq != peephole_optimize(load(iseq.to_a)))




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52924 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-07 18:46:27 +00:00
ko1 6757d13ee1 * iseq.c (iseq_translate): at the end of constructing an iseq,
call RubyVM::InstructionSequence.translate(iseq) if this method
  is defined. If the return value is also an object of
  RubyVM::InstructionSequence, then use it instead of created one.

  For example, this method is useful to test iseq dumper/loader
  such as RubyVM::InstructionSequence#to_a and rb_iseq_load().

  Because this method is for such internal experimental usage,
  the interface is not matured. For example, this interface has
  no extensibility. Two or more translaters can not run
  simultaneously.

  So that we don't guarantee future compatibility of this method.
  Basically, do not use this method.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52921 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-07 17:23:18 +00:00
ko1 2962b6e063 * vm_core.h, iseq.h: remove rb_iseq_t::variable_body.
Fields in rb_iseq_t::variable_body are contained by
  rb_iseq_t::body::mark_ary (hidden Array object).

  Index 0 to 2 of mark_ary are reserved by these objects.

* iseq.c: catch up this fix.

* compile.c (rb_iseq_original_iseq): trivial rewrite.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52865 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-02 13:58:07 +00:00
ko1 cfd1157f11 * iseq.h: introduce ISEQ_ORIGINAL_ISEQ() and
ISEQ_ORIGINAL_ISEQ_ALLOC() macro.

* compile.c: use them to access original iseq buffer.

* iseq.c: ditto.

* vm_core.h: rename iseq field to support this fix.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52862 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-02 08:20:35 +00:00
ko1 417240b7fb * iseq.h: introduce ISEQ_COVERAGE() and ISEQ_COVERAGE_SET() macro.
* compile.c: use them.

* iseq.c: ditto.

* iseq.c (rb_iseq_coverage): added.

* thread.c (update_coverage): use rb_iseq_coverage().

* vm_core.h: rename coverage field name to support this fix.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52860 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-02 08:05:36 +00:00
ko1 7afefea564 * iseq.h: introduce ISEQ_COMPILE_DATA() macro.
* compile.c, iseq.c: use ISEQ_COMPILE_DATA().

* vm_core.h: rename compile_data field to support this fix.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52858 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-12-02 07:52:12 +00:00
nobu 012368007f iseq.h: rename member
* iseq.h (rb_compile_option_struct): rename the member
  frozen_string_literal_debug as debug_frozen_string_literal.
  [Feature #11725]

* ruby.c (proc_options): do not set $DEBUG and $VERBOSE only if no
  arguments is given.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52749 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-11-25 08:02:29 +00:00
normal 513d8b349f iseq.c (iseq_memsize): account for rb_call_cache entries
Add some comments to clarify the allocated field used for the
allocations while we're at it.

TODO: figure out a better way of testing/maintaining this...

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52419 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-11-01 02:13:54 +00:00
ko1 92d5da521d * iseq.c (make_compile_option_value): include frozen_string_literal*
in a made option value.

* vm_opts.h: forgot to add OPT_FROZEN_STRING_LITERAL_DEBUG
  at last commit.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52258 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-10-23 19:13:51 +00:00
ko1 54716fe7eb * ruby.c: introduce --enable-frozen-string-literal-debug option.
If this option is enabled, the modify error will be:
    can't modify frozen String (RuntimeError) =>
    can't modify frozen String, created at test.rb:3 (RuntimeError)

* iseq.h: add compile option frozen_string_literal_debug.

* compile.c: catch up this fix.

* error.c (rb_error_frozen): ditto.

* iseq.c (set_compile_option_from_hash): ditto.

* test/ruby/test_rubyoptions.rb: add a test for this fix.




git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52257 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-10-23 19:02:55 +00:00
nobu 4200ddd5d0 iseq.c: list of keywords [Fix GH-1056]
* iseq.c (rb_insn_operand_intern): change kw in callinfo disasm from the
  number of keyword arguments to an ordered list of the keywords used.
  [Feature #11589]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52123 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-10-14 05:11:44 +00:00
ko1 a22d6f55e8 * iseq.c (rb_iseq_free): free iseq::variable_body to avoid memory
leak.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52051 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-10-05 21:55:56 +00:00
nobu 859337b17b fronzen-string-literal pragma
* compile.c (iseq_compile_each): override compile option by option
  given by pragma.
* iseq.c (rb_iseq_make_compile_option): extract a function to
  overwrite rb_compile_option_t.
* parse.y (parser_set_compile_option_flag): introduce pragma to
  override compile options.
* parse.y (magic_comments): new pragma "fronzen-string-literal".
  [Feature #8976]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51953 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-27 06:44:02 +00:00
ko1 bffa617d36 * iseq.c (rb_iseq_free): free rb_iseq_t::body::cc_entries.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51904 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-19 18:21:35 +00:00
ko1 d5ec9ec308 * vm_core.h: split rb_call_info_t into several structs.
* rb_call_info (ci) has compiled fixed information.
  * if ci->flag & VM_CALL_KWARG, then rb_call_info is
    also rb_call_info_with_kwarg. This technique reduce one word
    for major rb_call_info data.
  * rb_calling_info has temporary data (argc, blockptr, recv).
    for each method dispatch. This data is allocated only on
    machine stack.
  * rb_call_cache is for inline method cache.
  Before this patch, only rb_call_info_t data is passed.
  After this patch, above three structs are passed.
  This patch improves:
  * data locarity (rb_call_info is now read-only data).
  * reduce memory consumption (rb_call_info_with_kwarg,
    rb_calling_info).
* compile.c: use above data.
* insns.def: ditto.
* iseq.c: ditto.
* vm_args.c: ditto.
* vm_eval.c: ditto.
* vm_insnhelper.c: ditto.
* vm_insnhelper.h: ditto.
* iseq.h: add iseq_compile_data::ci_index and
  iseq_compile_data::ci_kw_indx.
* tool/instruction.rb: introduce TS_CALLCACHE operand type.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51903 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-19 17:59:58 +00:00
ko1 22ea03bb7e * iseq.c: disable ISeq.load. It enabled accidentally at r51794.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51825 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-11 15:50:42 +00:00
ko1 eec17b718a * vm_core.h: remove rb_call_info_t::blockiseq.
* insns.def (send, invokesuper): pass blockiseq explicitly.
* compile.c: catch up this fix.
* iseq.c: ditto.
* vm_args.c: ditto.
* iseq.c (ISEQ_MINOR_VERSION): 2->3 because instruction spec was
  changed.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51794 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-08 08:21:07 +00:00
ko1 970399a9ae * iseq.c (iseq_memsize): functions for wrapper object should have
iseqw_ prefix.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51755 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-09-03 19:48:12 +00:00
ko1 7cf523c7db * vm_opts.h, iseq.c, iseq.h: add compile option to force frozen
string literals.
  [Feature #11473]
  This addition is not specification change, but to try frozen
  string literal world discussed on [Feature #11473].
  You can try frozen string literal world using this magical line:
    RubyVM::InstructionSequence.compile_option =
      {frozen_string_literal: true}
  Note that this is a global compilation option, so that you need to
  compile another script like that:
    p 'foo'.frozen? #=> false
    RubyVM::InstructionSequence.compile_option =
      {frozen_string_literal: true}
    p 'foo'.frozen? #=> false, because this line is already compiled.
    p eval("'foo'.frozen?") #=> true
  Details:
    * String literals are deduped by rb_fstring().
    * Dynamic string literals ("...#{xyz}...") is now only frozen,
      not deduped. Maybe you have other ideas.
  Now, please do not use this option on your productions :)
  Of course, current specification can be changed.
* compile.c: ditto.
* test/ruby/test_iseq.rb: add a test.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51659 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-08-21 20:47:53 +00:00
normal e7bb5d44e6 iseq.c (rb_iseq_mark): reduce NULL checks
iseq_location_setup always sets path, label and base_label fields,
and the only caller of iseq_location_setup (prepare_iseq_build) will
always pass non-nil `name' and `path' arguments.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51581 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-08-14 19:23:02 +00:00
normal 7630957eba iseq.c (iseq_memsize): reimplement for wrapper
* iseq.c (iseq_memsize): reimplement for wrapper
  (param_keyword_size): extracted from iseq_memsize
  (iseqw_mark): new mark function
  (iseqw_data_type): new data type
  (iseqw_new): wrap as iseqw_data_type
  (iseqw_check): adjust for wrapper
  (Init_ISeq): remove iseqw_iseq_key initialization
* test/objspace/test_objspace.rb: new test
  [ruby-core:70344] [Feature #11435]

v2 changes:
- added RUBY_TYPED_WB_PROTECTED and write barrier
- account for rb_call_info_kw_arg_t entries

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51547 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-08-12 12:18:51 +00:00
nobu 6e54a6df22 iseq.c: fix type
* iseq.c (rb_iseq_disasm, iseq_data_to_ary): fix loop counter type
  for array length.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51373 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-25 02:18:32 +00:00
ko1 22be6d06ab * vm_core.h: size should be unsigned.
* rb_call_info_t::index
  * rb_iseq_constant_body::stack_max
  * rb_iseq_constant_body::local_size
  * rb_iseq_constant_body::param::size
  * rb_iseq_constant_body::local_table_size
  * rb_iseq_constant_body::is_size
  * rb_iseq_constant_body::callinfo_size
* iseq.h: same for iseq_catch_table::size.
* compile.c: catch up these fix.
* iseq.c: ditto.
* proc.c: ditto.
* vm.c: ditto.
* vm_args.c: ditto.
* vm_eval.c: ditto.
* vm_insnhelper.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51369 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-24 21:44:14 +00:00
ko1 14428f09b1 * vm_core.h: constify rb_iseq_constant_body::line_info_table.
* iseq.c: catch up this fix.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51368 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-24 21:01:09 +00:00
ko1 8fbf5dd9e5 * vm_core.h: constify rb_iseq_constant_body::catch_table.
* compile.c (iseq_set_exception_table): catch up this fix.
* iseq.c: ditto.
* vm.c (vm_exec): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51365 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-24 19:49:16 +00:00
ko1 eed613f4b1 * vm_core.h: constify rb_iseq_constant_body::local_table and
rb_iseq_param_keyword::table and
  rb_iseq_param_keyword::default_values.
* compile.c: catch up this fix.
* iseq.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51357 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-23 09:53:16 +00:00
ko1 18f6978625 * vm_core.h: constify rb_iseq_constant_body::iseq_encoded and
rb_control_frame_t::pc.
* compile.c (rb_iseq_translate_threaded_code): catch up this fix.
* iseq.c: ditto.
* vm_exec.c (vm_exec_core): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-23 09:34:31 +00:00
ko1 0d775f2a3c * vm_core.h: separate rb_iseq_body into rb_iseq_constant_body and
rb_iseq_variable_body (rb_iseq_t::variable_body).
  rb_iseq_variable_body can be modified after compilation.
* compile.c: use rb_iseq_t::variable_body.
* iseq.c: ditto.
* thread.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51339 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-22 10:55:02 +00:00
ko1 e4198a73d4 * make rb_iseq_t T_IMEMO object (type is imemo_iseq).
All contents of previous rb_iseq_t is in rb_iseq_t::body.
  Remove rb_iseq_t::self because rb_iseq_t is an object.
  RubyVM::InstructionSequence is wrapper object points T_IMEMO/iseq.
  So RubyVM::ISeq.of(something) method returns different wrapper
  objects but they point the same T_IMEMO/iseq object.
  This patch is big, but most of difference is replacement of
  iseq->xxx to iseq->body->xxx.
  (previous) rb_iseq_t::compile_data is also located to
  rb_iseq_t::compile_data.
  It was moved from rb_iseq_body::compile_data.
  Now rb_iseq_t has empty two pointers.
  I will split rb_iseq_body data into static data and dynamic data.
* compile.c: rename some functions/macros.
  Now, we don't need to separate iseq and iseqval (only VALUE).
* eval.c (ruby_exec_internal): `n' is rb_iseq_t (T_IMEMO/iseq).
* ext/objspace/objspace.c (count_imemo_objects): count T_IMEMO/iseq.
* gc.c: check T_IMEMO/iseq.
* internal.h: add imemo_type::imemo_iseq.
* iseq.c: define RubyVM::InstructionSequnce as T_OBJECT.
  Methods are implemented by functions named iseqw_....
* load.c (rb_load_internal0): rb_iseq_new_top() returns
  rb_iseq_t (T_IMEMO/iesq).
* method.h (rb_add_method_iseq): accept rb_iseq_t (T_IMEMO/iseq).
* vm_core.h (GetISeqPtr): removed because it is not T_DATA now.
* vm_core.h (struct rb_iseq_body): remove padding for
  [Bug #10037][ruby-core:63721].



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51327 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-21 22:52:59 +00:00
ko1 331fb4a2b3 * vm_core.h: constify rb_call_info_t::kw_arg,
rb_control_frame_t::iseq and rb_control_frame_t::block_iseq.
* iseq.c (iseq_free): catch up this fix.
* vm.c: ditto.
* vm_dump.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51325 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-21 21:41:04 +00:00
ko1 948bd807c3 * vm_core.h: constify rb_call_info_t::blockiseq and rb_iseq_t::iseq.
* vm.c, vm_insnhelper.c: catch up this fix.
* iseq.c (iseq_data_to_ary): constify the first iseq parameter.
* vm_insnhelper.c (vm_make_proc_with_iseq): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51323 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-21 21:28:43 +00:00
ko1 c0c20ccb51 * compile.c: constify the first parameter (iseq).
* iseq_add_mark_object()
  * iseq_add_mark_object_compile_time()
* iseq.c, iseq.h (rb_iseq_add_mark_object): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51311 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-21 10:47:45 +00:00
normal 4fd2443580 avoid redundant GET_THREAD() calls
This reduces binary size slightly on my 32-bit system:

   text    data     bss     dec     hex filename
2847705   12360   30632 2890697  2c1bc9 ruby.orig
2847641   12360   30632 2890633  2c1b89 ruby

* iseq.c (rb_iseq_compile_with_option): reuse result of previous
  GET_THREAD() call
* thread.c (thread_create_core): ditto
  (rb_mutex_trylock): ditto
  (rb_mutex_lock): ditto
* process.c (rb_waitpid): avoid multiple eval from RUBY_VM_CHECK_INTS
* thread.c (rb_thread_check_ints): ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51273 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-16 20:42:43 +00:00
ko1 c3e8cca950 * vm_core.h: constify rb_iseq_t::parent_iseq.
rb_iseq_t::local_iseq is not constant data because
  local_iseq::flip_cnt can be modified (commentted).
* compile.c: catch up this fix.
* iseq.c: ditto.
* vm_insnhelper.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51269 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-16 13:13:50 +00:00
ko1 7b08338ab1 * iseq.c (rb_iseq_disasm): rename rb_iseq_t *iseqdat to iseq
and VALUE *iseq to code.
* iseq.c (rb_iseq_disasm_insn): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51267 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-16 12:50:25 +00:00
normal a0908cb413 remove redundant NULL check in mark functions
gc.c (gc_mark_children)only calls mark_func if the T_DATA ptr is
non-NULL, so avoid redundantly checking for that in each
mark function.

* iseq.c (iseq_mark): remove check for data pointer
* proc.c (binding_mark): ditto
* vm.c (rb_thread_mark): ditto
* vm_trace.c (tp_mark): ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51252 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-15 08:29:22 +00:00
normal 011347e4d7 iseq.c (iseq_data_to_ary): dump kw_arg as symbol
Fix RubyVM::InstructionSequence#to_a after r49517
Keywords were made symbols to fix [Bug #10831] [ruby-core:68031],
so we should dump symbols as-is instead of attempting to convert
them from IDs

* iseq.c (iseq_data_to_ary): dump kw_arg as symbol
* test/-ext-/iseq_load/test_iseq_load.rb: test kw_arg roundtrip
  [ruby-core:69891] [Bug #11338]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51190 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-07 19:50:49 +00:00
ko1 f395c5be87 * vm_core.h: remove rb_iseq_t::orig because rb_iseq_clone()
no longer exists.
* iseq.c: don't use rb_iseq_t::orig.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51172 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-07 02:41:52 +00:00
ko1 16a68369d2 * iseq.c, internal.h (rb_iseq_clone): removed because we don't need to
clone iseq any more.
* class.c (clone_method): share iseq between cloned methods. All of
  method dependent information are able to refer from method entry.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51171 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-07 02:27:41 +00:00
ko1 5c4d7b9a9e * vm_core.h: remove rb_iseq_t::defined_method_id because it is not
needed.
* eval.c (frame_func_id): simplify. rb_callable_method_entry_t
  has enough information.
* eval.c (frame_called_id): ditto.
* iseq.c (prepare_iseq_build): catch up this fix.
* proc.c (rb_mod_define_method): ditto.
* vm.c (vm_define_method): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51168 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-06 19:08:05 +00:00
ko1 02d5868962 * vm_core.h: remove rb_iseq_t::klass to reduce dynamic data.
* internal.h, iseq.c (rb_iseq_klass): remove it because
  rb_iseq_t::klass is removed.
* vm_insnhelper.c (vm_super_outside): do not see cfp->iseq, but
  check callable method entry on a frame.
  This fix simplify the logic to search super class.
* test/ruby/test_method.rb: support super() from Proc.
  Now, [Bug #4881] and [Bug #3136] was solved.
* proc.c (rb_mod_define_method): catch up this change.
* vm.c (vm_define_method): ditto.
* vm_backtrace.c (rb_profile_frames): now, each `frame' objects
  are rb_callable_method_entry_t data or iseq VALUEs.
  This fix introduce minor compatibility issue that
  rb_profile_frame_label() always returns
  rb_profile_frame_base_label().
* test/-ext-/debug/test_profile_frames.rb: catch up this change.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51166 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-07-06 18:44:54 +00:00
normal b49075cdf1 move RB_GC_GUARD responsibility to rb_add_method_iseq
This simplifies all the callers and makes code easier to use
and review.  I was confused about the need for RB_GC_GUARD
in define_{aset,aref}_method of struct.c without reading
rb_add_method_iseq.

Likewise, do the same for rb_iseq_clone, where the GC guard
only seems neccesary iff RGenGC is disabled.

* vm_method.c (rb_add_method_iseq): add RB_GC_GUARD
* class.c (clone_method): remove RB_GC_GUARD
* struct.c (define_aref_method): ditto
  (define_aset_method): ditto
* vm.c (vm_define_method):
* iseq.c (rb_iseq_clone): add RB_GC_GUARD

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51079 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-06-30 21:07:18 +00:00
ko1 7239111ef5 * iseq.c: constify.
* iseq.h: ditto.
* method.h: ditto.
* proc.c: ditto.
* vm_method.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50593 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-05-21 09:01:44 +00:00
ko1 0dba10a47f * proc.c (rb_method_get_iseq): rename to rb_method_iseq.
* iseq.c: catch up this fix.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50587 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-05-21 07:54:44 +00:00
ko1 08f923bdc1 * iseq.c (exception_type2symbol): show correct bug message.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50566 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-05-20 19:11:47 +00:00
nobu b813198627 iseq.c: check srouce type
* iseq.c (rb_iseq_compile_with_option): check srouce type, must be
  an IO or a String.  [ruby-core:69219] [Bug #11159]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50541 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-05-19 09:54:44 +00:00
ko1 4690e62284 * iseq.c (iseq_mark): skip some marking if iseq->orig is available.
* iseq.c (rb_iseq_clone): need WB for iseq1->klass = iseq0->klass
  (done in MEMCPY).



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@50015 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-03-18 20:31:50 +00:00
ko1 d84f9b1694 * fix namespace issue on singleton class expressions. [Bug #10943]
* vm_core.h, method.h: remove rb_iseq_t::cref_stack. CREF is stored
  to rb_method_definition_t::body.iseq_body.cref.
* vm_insnhelper.c: modify SVAR usage.
  When calling ISEQ type method, push CREF information onto method
  frame, SVAR located place. Before this fix, SVAR is simply nil.
  After this patch, CREF (or NULL == Qfalse for not iseq methods)
  is stored at the method invocation.
  When SVAR is requierd, then put NODE_IF onto SVAR location,
  and NDOE_IF::nd_reserved points CREF itself.
* vm.c (vm_cref_new, vm_cref_dump, vm_cref_new_toplevel): added.
* vm_insnhelper.c (vm_push_frame): accept CREF.
* method.h, vm_method.c (rb_add_method_iseq): added. This function
  accepts iseq and CREF.
* class.c (clone_method): use rb_add_method_iseq().
* gc.c (mark_method_entry): mark method_entry::body.iseq_body.cref.
* iseq.c: remove CREF related codes.
* insns.def (getinlinecache/setinlinecache): CREF should be cache key
  because a different CREF has a different namespace.
* node.c (rb_gc_mark_node): mark NODE_IF::nd_reserved for SVAR.
* proc.c: catch up changes.
* struct.c: ditto.
* insns.def: ditto.
* vm_args.c (raise_argument_error): ditto.
* vm_eval.c: ditto.
* test/ruby/test_class.rb: add a test.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@49874 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2015-03-06 12:24:58 +00:00
nobu 467830e867 compile.c: struct accessors
* compile.c (rb_method_for_self_aref, rb_method_for_self_aset):
  move from iseq.c to build from node instead of arrays.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48876 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-17 03:20:58 +00:00
nobu 8d84693323 iseq.c: default option
* iseq.c (rb_iseq_new_with_opt): allow NULL option as
  COMPILE_OPTION_DEFAULT.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-17 02:57:51 +00:00
nobu d327d445c8 iseq.c: absolute_path
* iseq.c (caller_location): separate absolute_path.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48872 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-17 02:52:19 +00:00
nobu e4cf2f328d iseq.c: remove useless code
* iseq.c (rb_method_for_self_aref): remove unused declaration.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48865 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-16 07:38:52 +00:00
nobu e4717eaa4c iseq.c: struct accessors
* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset): call
  accessor functions directly, not to be affected by [] and []=
  methods.  [ruby-core:66846] [Bug #10601]
* struct.c (define_aref_method, define_aset_method): ditto.
* vm_insnhelper.c (rb_vm_opt_struct_aref, rb_vm_opt_struct_aset):
  direct accessors of Struct.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48864 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-16 06:18:25 +00:00
nobu 97628bff12 compile.c: wrap funcptr
* compile.c (iseq_build_from_ary_body): enclose funcptr with
  Integer as raw pointer cannot appear in an Array.

* iseq.c (iseq_data_to_ary): extract funcptr from Integer.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48863 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-16 05:54:32 +00:00
nobu 1d5c5ea153 iseq.c: show function name if possible
* iseq.c (rb_insn_operand_intern): show the name of the nearest
  run-time symbol if possible.

* compile.c (insn_data_to_s_detail): ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48858 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-16 01:14:27 +00:00
nobu 389ea73014 compile.c: check funcptr
* compile.c (iseq_build_from_ary_body): cannot load dumped funcptr.

* iseq.c (iseq_data_to_ary): cannot dump funcptr.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48857 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-16 01:14:20 +00:00
nobu a3450f4733 iseq.c: use caller location
* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset): set
  same location as the caller.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48851 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-15 19:13:15 +00:00
nobu 20748861d4 iseq.c: intermediate arrays
* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset): hide
  and clear intermediate array objects.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48850 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-15 19:13:07 +00:00
ko1 1b6aa03eb9 * class.c (class_alloc): Strat from age == 2.
Class and Module objects can be living long life.
* iseq.c: Same for ISeq objects.
* gc.c (RVALUE_AGE_RESET): added.
* gc.c (newobj_of): allow to generate (age != 0) objects.
* gc.c (rb_copy_wb_protected_attribute): reset age for wb unprotected
  objects.
* include/ruby/ruby.h: add RUBY_TYPED_PROMOTED1 as an unrecommended
  flag.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48771 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-11 10:15:30 +00:00
nobu 8efe878d11 struct.c: use iseqval
* struct.c (define_aref_method, define_aset_method): use iseq
  VALUE instead of rb_iseq_t to prevent from GC, as RB_GC_GUARD
  makes sense only for local variables.  [Feature #10575]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48754 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-10 04:39:29 +00:00
normal 65651b34b1 struct: avoid all O(n) behavior on access
This avoids O(n) on lookups with structs over 10 members.
This also avoids O(n) behavior on all assignments on Struct members.
Members 0..9 still use existing C methods to read in O(1) time

Benchmark results:

vm2_struct_big_aref_hi*	1.305
vm2_struct_big_aref_lo*	1.157
vm2_struct_big_aset*	3.306
vm2_struct_small_aref*	1.015
vm2_struct_small_aset*	3.273

Note: I chose use loading instructions from an array instead of writing
directly to linked-lists in compile.c for ease-of-maintainability.  We
may move the method definitions to prelude.rb-like files in the future.

I have also tested this patch with the following patch to disable
the C ref_func methods and ensured the test suite and rubyspec works

	--- a/struct.c
	+++ b/struct.c
	@@ -209,7 +209,7 @@ setup_struct(VALUE nstr, VALUE members)
		ID id = SYM2ID(ptr_members[i]);
		VALUE off = LONG2NUM(i);

	-	if (i < N_REF_FUNC) {
	+	if (0 && i < N_REF_FUNC) {
		    rb_define_method_id(nstr, id, ref_func[i], 0);
		}
		else {

* iseq.c (rb_method_for_self_aref, rb_method_for_self_aset):
  new methods to generate bytecode for struct.c
  [Feature #10575]
* struct.c (rb_struct_ref, rb_struct_set): remove
  (define_aref_method, define_aset_method): new functions
  (setup_struct): use new functions
* test/ruby/test_struct.rb: add test for struct >10 members
* benchmark/bm_vm2_struct_big_aref_hi.rb: new benchmark
* benchmark/bm_vm2_struct_big_aref_lo.rb: ditto
* benchmark/bm_vm2_struct_big_aset.rb: ditto
* benchmark/bm_vm2_struct_small_aref.rb: ditto
* benchmark/bm_vm2_struct_small_aset.rb: ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48748 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-09 15:43:49 +00:00
normal d71c745b40 iseq.c: remove unused rb_iseq_new_with_bopt
* iseq.c (prepare_iseq_build): remove unused block_opt param
  (rb_iseq_new_with_bopt_and_opt): remove
  (rb_iseq_new_with_opt): inline removed function
  (rb_iseq_new_with_bopt): remove
  (iseq_load): adjust prepare_iseq_build call
  [Feature #10565]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48713 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-05 02:10:29 +00:00
normal 9581954a96 mostly fix rb_iseq_load
This allows reporters commenters of [Feature #8543] to load
instruction sequences directly.  Some test cases are still failing
but documented in test/-ext-/iseq_load/test_iseq_load.rb.

* compile.c (rb_iseq_build_from_exception): entry->sp is unsigned
  (iseq_build_callinfo_from_hash): account for kw_arg
  (iseq_build_from_ary_body): update for r35459
  (CHECK_STRING, CHECK_INTEGER): remove unused checks
  (int_param): new function for checking new `params' hash
  (iseq_build_kw): new function for loading rb_iseq_param_keyword
  (rb_iseq_build_from_ary): account for `misc' entry and general
   structure changes
  [Feature #8543]
* iseq.c (CHECK_HASH): new macro (for `misc' and `param' entries)
  (iseq_load): account for `misc' and `params' hashes
  (iseq_data_to_ary): add final opt to arg_opt_labels,
   fix kw support, account for unsigned entry->sp
* ext/-test-/iseq_load/iseq_load.c: new ext for test
* ext/-test-/iseq_load/extconf.rb: ditto
* test/-ext-/iseq_load/test_iseq_load.rb: new test

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48705 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-03 22:16:58 +00:00
normal 0c662b3431 iseq.c: avoid segfault on incomplete iseq
Compile failures will trigger iseq_free before
iseq->callinfo_entries are allocated at all.

* iseq.c (iseq_free): avoid segfault on incomplete iseq
* test/ruby/test_syntax.rb (test_invalid_next): new test
  for syntax error, not segfault

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48704 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-03 22:03:59 +00:00
nobu a81979ec55 iseq.c: use lvar index
* iseq.c (iseq_data_to_ary): use local variable indexes for hidden
  variable instead of meaningless ID values.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48679 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-02 04:54:57 +00:00
normal 7ec4604700 iseq.c (iseq_data_to_ary): keep hidden variables
We must not ignore hidden variables when rb_id2str fails.
Thanks to wanabe [ruby-core:66566]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48678 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-01 22:32:56 +00:00
nobu e580a631be use 0 for reserved
use 0 for rb_data_type_t::reserved instead of NULL, since its type
may be changed in the future and possibly not a pointer type.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48662 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-12-01 06:38:04 +00:00
normal a7361f9c64 iseq.c (iseq_data_to_ary): add missing GC guard
It seems possible for a compiler to optimize away nbody
because we only access it via RARRAY_AREF and RARRAY_LEN
macros.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48597 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-27 02:48:14 +00:00
normal ef8c2e4be5 iseq.c (iseq_s_compile_file): close IO when done
This will prevent some leaked FD warnings in future tests.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48578 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-26 02:03:06 +00:00
nobu b58802a3c1 iseq.c: preserve encoding at disassembling
* iseq.c (rb_insn_operand_intern): preserve encoding of method
  name in CALL_INFO at disassembling.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48548 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-23 17:03:32 +00:00
akr 7cd76ab0c5 * internal.h: Include ruby.h and ruby/encoding.h to be
includable without prior inclusion.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48447 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-15 11:49:06 +00:00
ko1 86e9a1c005 * iseq.c (iseq_free): resolve memory leak.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48251 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-03 05:43:31 +00:00
ko1 af97388bc4 * iseq.c (iseq_memsize): catch up recent changes.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48248 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-03 04:54:27 +00:00
ko1 ef25be7391 * vm_core.h: change semantics of opt_num and opt_table.
`opt_num' was the number of optional parameters + 1.
  `opt_table' has "opt_num" entries.
  Change them to:
  `opt_num' is the number of optional parameters.
  `opt_talbe' has "opt_num + 1" entries.
  This change simplify parameter fitting logics.
* compile.c: catch up this change.
* iseq.c: ditto.
* proc.c: ditto.
* vm_args.c: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48247 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-03 04:43:07 +00:00
ko1 831e33c780 * vm_core.h: change iseq parameter data structure.
https://bugs.ruby-lang.org/issues/10440#change-49694
  * change terminology `arg' to `param'.
  * move rb_iseq_t::arg_* to rb_iseq_t::param.
  * move rb_iseq_t::arg_size to rb_iseq_t::param::size.
  * move rb_iseq_t::argc to rb_iseq_t::param::lead_num.
  * move rb_iseq_t::arg_opts to rb_iseq_t::param::opt_num.
  * move rb_iseq_t::arg_rest to rb_iseq_t::param::rest_start.
  * move rb_iseq_t::arg_post_num to rb_iseq_t::param::post_num.
  * move rb_iseq_t::arg_post_start to rb_iseq_t::param::post_start.
  * move rb_iseq_t::arg_block to rb_iseq_t::param::block_start.
  * move rb_iseq_t::arg_keyword* to rb_iseq_t::param::keyword.
    rb_iseq_t::param::keyword is allocated only when keyword
    parameters are available.
  * introduce rb_iseq_t::param::flags to represent parameter
    availability. For example, rb_iseq_t::param:🎏:has_kw
    represents that this iseq has keyword parameters and
    rb_iseq_t::param::keyword is allocated.
    We don't need to compare with -1 to check availability.
  * remove rb_iseq_t::arg_simple.
* compile.c: catch up this change.
* iseq.c: ditto.
* proc.c: ditto.
* vm.c, vm_args.c, vm_dump.c, vm_insnhelper.c: ditto.
* iseq.c (iseq_data_to_ary): support keyword argument.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48242 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-02 23:14:21 +00:00
ko1 fbebd502f9 * rewrite method/block parameter fitting logic to optimize
keyword arguments/parameters and a splat argument.
  [Feature #10440] (Details are described in this ticket)
  Most of complex part is moved to vm_args.c.
  Now, ISeq#to_a does not catch up new instruction format.
* vm_core.h: change iseq data structures.
  * introduce rb_call_info_kw_arg_t to represent keyword arguments.
  * add rb_call_info_t::kw_arg.
  * rename rb_iseq_t::arg_post_len to rb_iseq_t::arg_post_num.
  * rename rb_iseq_t::arg_keywords to arg_keyword_num.
  * rename rb_iseq_t::arg_keyword to rb_iseq_t::arg_keyword_bits.
    to represent keyword bitmap parameter index.
    This bitmap parameter shows that which keyword parameters are given
    or not given (0 for given).
    It is refered by `checkkeyword' instruction described bellow.
  * rename rb_iseq_t::arg_keyword_check to rb_iseq_t::arg_keyword_rest
    to represent keyword rest parameter index.
  * add rb_iseq_t::arg_keyword_default_values to represent default
    keyword values.
  * rename VM_CALL_ARGS_SKIP_SETUP to VM_CALL_ARGS_SIMPLE
    to represent
      (ci->flag & (SPLAT|BLOCKARG)) &&
      ci->blockiseq == NULL &&
      ci->kw_arg == NULL.
* vm_insnhelper.c, vm_args.c: rewrite with refactoring.
  * rewrite splat argument code.
  * rewrite keyword arguments/parameters code.
  * merge method and block parameter fitting code into one code base.
* vm.c, vm_eval.c: catch up these changes.
* compile.c (new_callinfo): callinfo requires kw_arg parameter.
* compile.c (compile_array_): check the last argument Hash object or
  not. If Hash object and all keys are Symbol literals, they are
  compiled to keyword arguments.
* insns.def (checkkeyword): add new instruction.
  This instruction check the availability of corresponding keyword.
  For example, a method "def foo k1: 'v1'; end" is cimpiled to the
  following instructions.
    0000 checkkeyword     2, 0  # check k1 is given.
    0003 branchif         9     # if given, jump to address #9
    0005 putstring        "v1"
    0007 setlocal_OP__WC__0 3   # k1 = 'v1'
    0009 trace            8
    0011 putnil
    0012 trace            16
    0014 leave
* insns.def (opt_send_simple): removed and add new instruction
  "opt_send_without_block".
* parse.y (new_args_tail_gen): reorder variables.
  Before this patch, a method "def foo(k1: 1, kr1:, k2: 2, **krest, &b)"
  has parameter variables "k1, kr1, k2, &b, internal_id, krest",
  but this patch reorders to "kr1, k1, k2, internal_id, krest, &b".
  (locate a block variable at last)
* parse.y (vtable_pop): added.
  This function remove latest `n' variables from vtable.
* iseq.c: catch up iseq data changes.
* proc.c: ditto.
* class.c (keyword_error): export as rb_keyword_error().
* common.mk: depend vm_args.c for vm.o.
* hash.c (rb_hash_has_key): export.
* internal.h: ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@48239 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-11-02 18:02:55 +00:00
nobu 5617e31771 symbol.c: rename rb_str_dynamic_intern
* iseq.c, marshal.c, string.c: use rb_str_intern instead of
  rb_str_dynamic_intern.
* symbol.c (rb_str_intern): rename rb_str_dynamic_intern.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47912 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-10-14 07:23:01 +00:00
normal 57da36ebcf iseq.c (rb_iseq_line_trace_each): explicit cast
Fix https://travis-ci.org/ruby/ruby/jobs/36814282
Tested with -Wconversion since my gcc lacks -Wshorten-64-to-32

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47760 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-10-01 23:03:35 +00:00
normal 39fd4a87e8 iseq.c (rb_iseq_defined_string): trim redundant semi-colon
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47666 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-09-21 03:50:23 +00:00
normal ea4c97904e compile: translate iseq in-place
running "ruby -rpp -e 'pp GC.stat'", a reduction in
malloc usage is shown:

before:
	:malloc_increase=>118784,
	:oldmalloc_increase=>1178736,

after:
	:malloc_increase=>99832,
	:oldmalloc_increase=>1031976,

For "ruby -e exit", valgrind reports over 300K reduction in
overall allocations (and unnecessary memory copies).

before:
  total heap usage: 49,622 allocs, 20,492 frees, 8,697,493 bytes allocated
after:
  total heap usage: 48,935 allocs, 19,805 frees, 8,373,773 bytes allocated

(numbers from x86-64)

v2 changes based on ko1 recommendations [ruby-core:64883]:
- squashed in-place direct thread translation to avoid alloc+copy
- renamed rb_iseq_untranslate_threaded_code to rb_iseq_original_iseq,
  cache new iseq->iseq_original field.

* compile.c (rb_iseq_translate_threaded_code): modify in-place w/o copy
  (rb_vm_addr2insn): new function for debug
  (rb_iseq_original_iseq): ditto
  (iseq_set_sequence): assign iseq_encoded directly
  [Feature #10185]

* vm_core (rb_iseq_t): move original ->iseq to bottom

* iseq.c (iseq_free, iseq_free): adjust for new layout
  (rb_iseq_disasm): use original iseq for dump
  (iseq_data_to_ary): ditto
  (rb_iseq_line_trace_each): ditto
  (rb_iseq_build_for_ruby2cext): use iseq_encoded directly

* vm_dump.c (rb_vmdebug_debug_print_pre): use original iseq

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47508 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-09-10 06:14:07 +00:00
normal 0ae8dd168e rb_iseq_location_t: change first_lineno type to VALUE
Nearly all current uses of iseq->location.first_lineno are as a
VALUE, not a size_t.  The only exception was the experimental
(and currently unused) rb_iseq_build_for_ruby2cext function.

* vm_core.h (rb_iseq_location_t): change first_lineno type to VALUE
* iseq.c (rb_iseq_build_for_ruby2cext): update based on argument

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47269 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-08-25 02:24:10 +00:00
ko1 3e759ed78d * iseq.c (rb_iseq_clone): Should not insert write barrier from
non-RVALUE data (to non-RVALUE data, of course).
  Ruby 2.1 also has a same problem.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@47191 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-08-15 10:39:27 +00:00
nobu 95d1b61a14 symbols instead of IDs
* encoding.c (rb_enc_get_index): deal with symbols instead of IDs
  to get rid of inadvertent pin-downs.

* enum.c (chunk_ii): ditto.

* enumerator.c (append_method): ditto.

* iseq.c (iseq_load): ditto.

* marshal.c (w_symbol, r_symlink, r_symreal, r_symbol): ditto.

* signal.c (trap_handler): ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46988 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-28 08:15:42 +00:00
normal 36b476cd1e struct iseq_compile_data_storage: 16 bytes (from 32) overhead
This reduces the iseq_compile_data_storage header from 32 to 16
bytes on 64-bit systems.

pos and size fields cannot exceed 32-bit sizes due to stack size
limits.  Using a flexible array for the buffer also saves us 8
bytes of pointer overhead.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46958 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-26 07:57:44 +00:00
normal a9c7629ece rb_iseq_t: reduce to (280 bytes from 288 bytes) on 64-bit
iseq_size and line_info_size may be 32-bit.

4GB instruction sequences should be big enough for anyone.
Other existing line info counters are 32-bit, and nobody
should need all that.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46957 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-26 07:30:26 +00:00
normal 48a2b96d2b introduce ZALLOC{,_N} to replace ALLOC{,_N}+MEMZERO use
Using calloc where possible reduces code and binary sizes.

* include/ruby/ruby.h (ZALLOC, ZALLOC_N): implement
  (Data_Make_Struct, TypedData_Make_Struct):
  ZALLOC replaces ALLOC+memset
* compile.c (iseq_seq_sequence): ZALLOC_N replaces ALLOC_N+MEMZERO
* cont.c (fiber_t_alloc): ZALLOC replaces ALLOC+MEMZERO
* io.c (rb_io_reopen): ditto
* iseq.c (prepare_iseq_build): ditto
* parse.y (new_args_tail_gen, parser_new, ripper_s_allocate): ditto
* re.c (match_alloc): ditto
* variable.c (rb_const_set): ditto
* ext/socket/raddrinfo.c (get_addrinfo): ditto
* ext/strscan/strscan.c (strscan_s_allocate): ditto
* gc.c (rb_objspace_alloc): calloc replaces malloc+MEMZERO

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46952 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-25 21:34:35 +00:00
ko1 20a9bd6763 * iseq.c (rb_iseq_defined_string): use rb_gc_mark_object() instead of
marking from vm_mark().
* vm.c (rb_vm_mark): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46833 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-16 09:09:40 +00:00
normal e91ae784bc vm_core.h (struct rb_iseq_struct): reduce to 296 bytes on 64-bit
Most iseq do not have a catch_table, so avoid needlessly adding
4-8 bytes to the struct for the common case.

Changes from v2:
- iseq_catch_table_size removed, use if (...) for (;...;)

Changes from v1:
- renamed iseq->_catch_table to iseq->catch_table
- iseq_catch_table_bytes: made a static inline function
- iseq_catch_table_size: new function replaces the
  iseq_catch_table_each iterator macro

* iseq.h (struct iseq_catch_table): new flexible array struct
         (iseq_catch_table_bytes): allocated size function
* vm_core.h (struct rb_iseq_struct): uupdate catch_table member
* compile.c (iseq_set_exception_table): update for struct changes
* iseq.c (iseq_free): ditto
* iseq.c (iseq_memsize): ditto
* iseq.c (rb_iseq_disasm): ditto
* iseq.c (iseq_data_to_ary): ditto
* iseq.c (rb_iseq_build_for_ruby2cext): ditto (untested)
* vm.c (vm_exec): ditto
* vm_core.h (struct rb_iseq_struct): ditto
* vm_insnhelper.c (vm_throw): ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46811 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-07-14 07:06:26 +00:00
nobu e99ee55abc constify parameters
* include/ruby/intern.h: constify `argv` parameters.

* include/ruby/ruby.h: ditto.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@46459 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-06-18 06:16:39 +00:00
nobu 59d1856def iseq.c: linear search
* iseq.c (iseq_type_from_id): linear search instead of hash lookup for
  small fixed number keys.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44569 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-01-12 07:49:26 +00:00
nobu ff00300834 iseq.c: potential memory leak
* iseq.c (iseq_load): keep type_map to get rid of memory leak.
  based on a patch by Eric Wong at [ruby-core:59699].  [Bug #9399]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44568 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-01-12 07:41:10 +00:00
nobu 5e9b7bacc3 ruby/util.h: DECIMAL_SIZE_OF_BITS
* include/ruby/util.h (DECIMAL_SIZE_OF_BITS): a preprocessor
  constant macro to approximate decimal representation size of n-bits
  integer.
* iseq.c (register_label): use DECIMAL_SIZE_OF_BITS for better
  approximation.
* ext/bigdecimal/bigdecimal.c (BigMath_s_log): ditto.
* common.mk (iseq.o), ext/bigdecimal/depend (bigdecimal.o): add
  dependency to ruby/util.h for DECIMAL_SIZE_OF_BITS.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44562 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2014-01-12 00:21:37 +00:00
nobu 27682eefa4 compile.c: unnamed keyword rest check
* compile.c (iseq_set_arguments): set arg_keyword_check from
  nd_cflag, which is set by parser.  internal ID is used for
  unnamed keyword rest argument, which should be separated from no
  keyword check.
* iseq.c (rb_iseq_parameters): if no keyword check, keyword rest is
  present.
* parse.y (new_args_tail_gen): set keywords check to nd_cflag, which
  equals to that keyword rest is not present.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44420 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-12-25 13:44:18 +00:00
nobu 4398093755 iseq.c: fix parameter of unnamed rest keywords argument
* iseq.c (rb_iseq_parameters): push argument type symbol only for
  unnamed rest keywords argument.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44414 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-12-25 11:30:15 +00:00
ko1 c702005a7b * include/ruby/ruby.h: rename OBJ_WRITE and OBJ_WRITTEN into
RB_OBJ_WRITE and RB_OBJ_WRITTEN.
* array.c, class.c, compile.c, hash.c, internal.h, iseq.c,
  proc.c, process.c, re.c, string.c, variable.c, vm.c,
  vm_eval.c, vm_insnhelper.c, vm_insnhelper.h,
  vm_method.c: catch up this change.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44299 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-12-20 08:07:47 +00:00
tmm1 8f77cfb308 gc.c: promote long-lived NODE_CREF objects to oldgen
* include/ruby/ruby.h: add RGENGC_WB_PROTECTED_NODE_CREF setting
  In a large app, this reduces the size of
  remembered_shady_object_count by 80%. [Bug #9225] [ruby-core:58947]
* gc.c (rb_node_newnode): add FL_WB_PROTECTED flag to NODE_CREF
* class.c (rewrite_cref_stack): insert OBJ_WRITE for NODE_CREF
* iseq.c (set_relation): ditto
* iseq.c (rb_iseq_clone): ditto
* vm_eval.c (rb_yield_refine_block): ditto
* vm_insnhelper.c (vm_cref_push): ditto
* vm_insnhelper.h (COPY_CREF): ditto

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@44059 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-12-08 02:28:04 +00:00
tmm1 cbb56e30a4 * compile.c: Use rb_fstring() to de-duplicate string literals in code. [ruby-core:58599] [Bug #9159] [ruby-core:54405]
* iseq.c (prepare_iseq_build): De-duplicate iseq labels and source locations.
* re.c (rb_reg_initialize): Use rb_fstring() for regex string.
* string.c (rb_fstring): Handle non-string and already-fstr arguments.
* vm_eval.c (eval_string_with_cref): De-duplicate eval source filename.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@43866 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-11-26 23:30:25 +00:00