aboutsummaryrefslogtreecommitdiffstats
path: root/mjit.h
Commit message (Collapse)AuthorAgeFilesLines
* Add ISEQ_BODY macroPeter Zhu2022-03-241-1/+1
| | | | | | Use ISEQ_BODY macro to get the rb_iseq_constant_body of the ISeq. Using this macro will make it easier for us to change the allocation strategy of rb_iseq_constant_body when using Variable Width Allocation.
* Cleanup diff against upstream. Add commentsAlan Wu2021-10-201-0/+1
| | | | | I did a `git diff --stat` against upstream and looked at all the files that are outside of YJIT to come up with these minor changes.
* Tweak mjit_exec() to remove YJIT symbol exportsAlan Wu2021-10-201-7/+14
| | | | | | | | We were exporting a couple of symbols in yjit.h because they could be used by code generated by MJIT. We don't want MJIT calling into YJIT code anyways so let's stop exporting them to libruby.so. Also adjust indentation and comments in mjit_exec().
* Always use `ret` to return to the interpreterAaron Patterson2021-10-201-3/+15
| | | | | | | | | | | | | | | | | | | | | Always using `ret` to return to the interpreter means that we never have to check the VM_FRAME_FLAG_FINISH flag. In the case that we return `Qundef`, the interpreter will execute the cfp. We can take advantage of this by setting the PC to the instruction we can't handle, and let the interpreter pick up the ball from there. If we return a value other than Qundef, the interpreter will take that value as the "return value" from the JIT and push that to the SP of the caller The leave instruction puts the return value on the top of the calling frame's stack. YJIT does the same thing for leave instructions. However, when we're returning back to the interpreter, the leave instruction _should not_ put the return value on the top of the stack, but put it in RAX and use RET. This commit pops the last value from the stack pointer and puts it in RAX so that the interpreter is happy with SP.
* Fix GCC warningsAlan Wu2021-10-201-4/+2
| | | | Mostly unused and uninitialized warnings here and there
* Yet Another Ruby JIT!Jose Narvaez2021-10-201-4/+5
| | | | Renaming uJIT to YJIT. AKA s/ujit/yjit/g.
* Thread EC through jit state. Add codegen status enum.Maxime Chevalier-Boisvert2021-10-201-1/+1
|
* Implement --ujit-call-thresholdMaxime Chevalier-Boisvert2021-10-201-1/+1
|
* Refactor uJIT code into more files for readabilityMaxime Chevalier-Boisvert2021-10-201-1/+1
|
* Move MicroJIT compilation threshold into MicroJIT's headerAlan Wu2021-10-201-2/+1
|
* Fix MicroJIT logic in MJIT execAlan Wu2021-10-201-8/+12
|
* Never compile with MicroJIT when MJIT is enabledAlan Wu2021-10-201-1/+1
| | | | There is a crash in there that we don't want to worry about for now.
* Don't call MicroJIT from MJIT outputAlan Wu2021-10-201-0/+2
|
* MicroJIT: compile after ten callsAlan Wu2021-10-201-1/+7
|
* Print JIT cancel when all JIT-ed code is cancelledTakashi Kokubun2021-08-121-0/+2
|
* Avoid enqueueing the same ISeq twiceTakashi Kokubun2021-06-101-1/+1
| | | | | | | | | | by a race condition by multiple Ractors. Atmically incrementing body->total_calls may have its own cost, so for now we intentionally leave the unreliable total_calls. So we allow an ISeq to be never pushed when you use multiple Ractors. However, if you enqueue a single ccan node twice, get_from_list loops infinitely. Thus this patch takes care of such a situation.
* Drop JIT_ISEQ_SIZE_THRESHOLDTakashi Kokubun2021-05-311-4/+0
| | | | | | | | | Compiling everything seems to contributed to improving the final performance in general. MJIT's compilation is slow anyway, especially when you need to wait for JIT compaction. This might make sense for short-time benchmarks like Optcarrot with default parameters, but it didn't give benefits in my local environment.
* enable constant cache on ractorsKoichi Sasada2021-01-051-4/+0
| | | | | | | | | | | | | | | | constant cache `IC` is accessed by non-atomic manner and there are thread-safety issues, so Ruby 3.0 disables to use const cache on non-main ractors. This patch enables it by introducing `imemo_constcache` and allocates it by every re-fill of const cache like `imemo_callcache`. [Bug #17510] Now `IC` only has one entry `IC::entry` and it points to `iseq_inline_constant_cache_entry`, managed by T_IMEMO object. `IC` is atomic data structure so `rb_mjit_before_vm_ic_update()` and `rb_mjit_after_vm_ic_update()` is not needed.
* Skip mjit_wait if iseq is not a targetTakashi Kokubun2021-01-041-10/+1
|
* Stop managing valid class serialsTakashi Kokubun2020-12-291-4/+0
| | | | `mjit_valid_class_serial_p` has no longer been used since b9007b6c548.
* MJIT is enabled whenever an MJIT header is usedTakashi Kokubun2020-12-211-1/+5
| | | | | | | | | | | | | | | ``` $ benchmark-driver -v --rbenv 'before --jit;after --jit' --repeat-count=4 --alternate --output=all benchmark.yml before --jit: ruby 3.0.0dev (2020-12-22T05:16:39Z master 692af8e8f8) +JIT [x86_64-linux] after --jit: ruby 3.0.0dev (2020-12-22T06:05:59Z master d2ca23fdbc) +JIT [x86_64-linux] last_commit=MJIT is enabled whenever an MJIT header is used Calculating ------------------------------------- before --jit after --jit Optcarrot 180 frames 86.61115328062763 91.49245232931852 fps 90.18653732453176 92.31139517827663 91.29817884736318 93.84437713805814 92.76118222412521 93.91929147901106 ```
* Mark active_unitsTakashi Kokubun2020-12-201-0/+2
| | | | | | | | | | | | | | | | | | to avoid SEGV on mjit_recompile and compact_all_jit_code. For some reason, ISeqs on stack are sometimes GC-ed (why?) and therefore it may run mjit_recompile on a GC-ed ISeq, which I expected d07183ec85d to fix but apparently it may refer to random things if already GC-ed. Marking active_units would workaround the situation. http://ci.rvm.jp/results/trunk-mjit-wait@phosphorus-docker/3292740 Also, while compact_all_jit_code was executed, we saw some SEGVs where CCs seemed to be already GC-ed, meaning their owner ISeq was not marked properly. Even if units are still in active_units, it's not guaranteed that their ISeqs are in use. So in this case we need to mark active_units for a legitimate reason. http://ci.rvm.jp/results/trunk-mjit-wait@phosphorus-docker/3293277 http://ci.rvm.jp/results/trunk-mjit-wait@phosphorus-docker/3293090
* Stop marking unit_queueTakashi Kokubun2020-12-201-2/+0
| | | | | | | | | | The original motivation of this marking was https://github.com/k0kubun/yarv-mjit/issues/20. As wanabe said, there are multiple options to mitigate the issue, and Eric Wong introduced another fix at 143776f6fe by checking unit->iseq inside the lock. Therefore this particular condition has been covered in two ways, and the script given by wanabe no longer crashes without mjit_mark().
* Inline getconstant on JIT (#3906)Takashi Kokubun2020-12-161-0/+7
| | | | | * Inline getconstant on JIT * Support USE_MJIT=0
* ruby/internal/config.h needs to be included firstTakashi Kokubun2020-11-221-1/+1
| | | | to define USE_MJIT.
* Make --disable-jit-support compileTakashi Kokubun2020-11-221-1/+1
| | | | | | vm_core.h needs to be included to know rb_execution_context_t, etc. I also added a trivial refactoring in mjit.c and missing dependency for process.c.
* Combine mjit.h and internal/mjit.hTakashi Kokubun2020-11-221-12/+24
| | | | It's very hard to remember which mjit.h has what.
* Do not JIT inline builtin methodsTakashi Kokubun2020-06-241-0/+1
| | | | | It's probably not worth it because there's nothing we can optimize in such builtin methods. It's worth JIT only when inlined.
* Carve out mjit_exec_slowpathTakashi Kokubun2020-05-271-31/+41
| | | | and mark it as COLDFUN on JIT, assuming enqueue usually happens on VM.
* Eliminate a call instruction on JIT cancel pathTakashi Kokubun2020-05-261-1/+4
| | | | | | | by calling combined functions specialized for each cancel type. I'm hoping to improve locality of hot code, but this patch's impact should be insignificant.
* sed -i 's|ruby/impl|ruby/internal|'卜部昌平2020-05-111-1/+1
| | | | To fix build failures.
* sed -i s|ruby/3|ruby/impl|g卜部昌平2020-05-111-1/+1
| | | | This shall fix compile errors.
* Include unit id in a function name of an inlined methodTakashi Kokubun2020-04-301-1/+1
| | | | | I'm trying to make it possible to include all JIT-ed code in a single C file. This is needed to guarantee uniqueness of all function names
* add #include guard hack卜部昌平2020-04-131-3/+2
| | | | | | | | | | | | | | | | | | | | | | According to MSVC manual (*1), cl.exe can skip including a header file when that: - contains #pragma once, or - starts with #ifndef, or - starts with #if ! defined. GCC has a similar trick (*2), but it acts more stricter (e. g. there must be _no tokens_ outside of #ifndef...#endif). Sun C lacked #pragma once for a looong time. Oracle Developer Studio 12.5 finally implemented it, but we cannot assume such recent version. This changeset modifies header files so that each of them include strictly one #ifndef...#endif. I believe this is the most portable way to trigger compiler optimizations. [Bug #16770] *1: https://docs.microsoft.com/en-us/cpp/preprocessor/once *2: https://gcc.gnu.org/onlinedocs/cppinternals/Guard-Macros.html
* Merge pull request #2991 from shyouhei/ruby.h卜部昌平2020-04-081-1/+1
| | | Split ruby.h
* Optimize exivar access on JIT-ed getivarTakashi Kokubun2020-03-301-1/+3
| | | | | | | | | | | | | | | | | | JIT support of dd723771c11. $ benchmark-driver -v --rbenv 'before;before --jit;after --jit' benchmark/mjit_exivar.yml --repeat-count=4 before: ruby 2.8.0dev (2020-03-30T12:32:26Z master e5db3da9d3) [x86_64-linux] before --jit: ruby 2.8.0dev (2020-03-30T12:32:26Z master e5db3da9d3) +JIT [x86_64-linux] after --jit: ruby 2.8.0dev (2020-03-31T05:57:24Z mjit-exivar 128625baec) +JIT [x86_64-linux] Calculating ------------------------------------- before before --jit after --jit mjit_exivar 57.944M 53.579M 54.471M i/s - 200.000M times in 3.451588s 3.732772s 3.671687s Comparison: mjit_exivar before: 57944345.1 i/s after --jit: 54470876.7 i/s - 1.06x slower before --jit: 53579483.4 i/s - 1.08x slower
* Add debug counter for unload_unitsTakashi Kokubun2020-03-151-1/+0
| | | | changing add_iseq_to_process's debug counter name as well for comparison
* Move code to mark jit_unit's cc_entries to mjit.cTakashi Kokubun2020-03-121-1/+1
|
* Internalize rb_mjit_unit definition againTakashi Kokubun2020-02-261-29/+1
| | | | Fixed a TODO in b9007b6c548f91e88fd3f2ffa23de740431fa969
* Introduce disposable call-cache.Koichi Sasada2020-02-221-0/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | This patch contains several ideas: (1) Disposable inline method cache (IMC) for race-free inline method cache * Making call-cache (CC) as a RVALUE (GC target object) and allocate new CC on cache miss. * This technique allows race-free access from parallel processing elements like RCU. (2) Introduce per-Class method cache (pCMC) * Instead of fixed-size global method cache (GMC), pCMC allows flexible cache size. * Caching CCs reduces CC allocation and allow sharing CC's fast-path between same call-info (CI) call-sites. (3) Invalidate an inline method cache by invalidating corresponding method entries (MEs) * Instead of using class serials, we set "invalidated" flag for method entry itself to represent cache invalidation. * Compare with using class serials, the impact of method modification (add/overwrite/delete) is small. * Updating class serials invalidate all method caches of the class and sub-classes. * Proposed approach only invalidate the method cache of only one ME. See [Feature #16614] for more details.
* suppress clang warnings卜部昌平2020-02-201-2/+4
| | | | | | Starting clang 11, casts between pointer and (narrower-than-pointer) int are now warned. However all such thing in our repository are guaranteed safe. Let's suppress the warnings.
* decouple internal.h headers卜部昌平2019-12-261-2/+5
| | | | | | | | | | | | | | | | | | Saves comitters' daily life by avoid #include-ing everything from internal.h to make each file do so instead. This would significantly speed up incremental builds. We take the following inclusion order in this changeset: 1. "ruby/config.h", where _GNU_SOURCE is defined (must be the very first thing among everything). 2. RUBY_EXTCONF_H if any. 3. Standard C headers, sorted alphabetically. 4. Other system headers, maybe guarded by #ifdef 5. Everything else, sorted alphabetically. Exceptions are those win32-related headers, which tend not be self- containing (headers have inclusion order dependencies).
* Constified mjit_initNobuyoshi Nakada2019-12-011-1/+1
|
* Allow specifying arbitrary MJIT flags by --jit-debugTakashi Kokubun2019-12-011-0/+2
| | | | | | | | This is a secret feature for me. It's only for testing and any behavior with this flag override is unsupported. I needed this because I sometimes want to add debug options but do not want to disable optimizations, for using Linux perf.
* delete unused functions卜部昌平2019-11-141-4/+0
| | | | | | | | | | | | Looking at the list of symbols inside of libruby-static.a, I found hundreds of functions that are defined, but used from nowhere. There can be reasons for each of them (e.g. some functions are specific to some platform, some are useful when debugging, etc). However it seems the functions deleted here exist for no reason. This changeset reduces the size of ruby binary from 26,671,456 bytes to 26,592,864 bytes on my machine.
* Rename mjit_gc_finish_hook to mjit_gc_exit_hookTakashi Kokubun2019-05-151-2/+2
| | | | | because @ko1 said "gc_finish" is confusing like a finish of entire GC process
* Prefix rb_ to MJIT wait call since it's exportedk0kubun2019-04-201-2/+2
| | | | git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67643 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Update MJIT referencestenderlove2019-04-201-0/+1
| | | | | | ISeq can move, so we need to tell MJIT where the new location is. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67624 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Recompile without method inliningk0kubun2019-04-161-0/+2
| | | | | | if cancel happens in an inlined method. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67575 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* Unify comment styles across MJIT sourcesk0kubun2019-04-141-32/+32
| | | | | | | | | | | | I'm writing `//` comments in newer MJIT code after C99 enablement (because I write 1-line comments more often than multi-line comments and `//` requires fewer chars on 1-line) and then they are mixed with `/* */` now. For consistency and to avoid the conversion in future changes, let me finish the rewrite in MJIT-related code. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67533 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-24 17:27:31 +0000