aboutsummaryrefslogtreecommitdiffstats
path: root/tool/mk_call_iseq_optimized.rb
Commit message (Collapse)AuthorAgeFilesLines
* delete mk_call_iseq_optimized.rb卜部昌平2020-05-101-73/+0
| | | | To generate what is necessary via generic_erb.rb instead.
* Introduce disposable call-cache.Koichi Sasada2020-02-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | 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.
* VALUE size packed callinfo (ci).Koichi Sasada2020-02-221-5/+4
| | | | | | | | | | | | | | | | | | | | Now, rb_call_info contains how to call the method with tuple of (mid, orig_argc, flags, kwarg). Most of cases, kwarg == NULL and mid+argc+flags only requires 64bits. So this patch packed rb_call_info to VALUE (1 word) on such cases. If we can not represent it in VALUE, then use imemo_callinfo which contains conventional callinfo (rb_callinfo, renamed from rb_call_info). iseq->body->ci_kw_size is removed because all of callinfo is VALUE size (packed ci or a pointer to imemo_callinfo). To access ci information, we need to use these functions: vm_ci_mid(ci), _flag(ci), _argc(ci), _kwarg(ci). struct rb_call_info_kw_arg is renamed to rb_callinfo_kwarg. rb_funcallv_with_cc() and rb_method_basic_definition_p_with_cc() is temporary removed because cd->ci should be marked.
* more on struct rb_call_data卜部昌平2019-10-251-3/+3
| | | | | Replacing adjacent struct rb_call_info and struct rb_call_cache into a struct rb_call_data.
* add debug counters to count call cache fastpath.ko12019-03-221-0/+1
| | | | | | | | | Add counters to count ccf (call cache fastpath) usage. These counters will help which kind of method dispatch is important to optimize. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67336 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* _mjit_compile_send.erb: refactor code to setup iseqk0kubun2018-09-241-1/+1
| | | | | | | | | | | | | by sharing vm_call_iseq_setup_normal. This is a retry of r64280. vm_insnhelper.c: Remove unused argument `ci` and pass `me` instead of `cc` to share this with JIT. Declare this with ALWAYS_INLINE to make sure this function is inlined in JIT. tool/mk_call_iseq_optimized.rb: deal with the interface change of vm_call_iseq_setup_normal. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64820 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* mk_call_iseq_optimized.rb: add mode string [ci skip]nobu2017-11-091-0/+1
| | | | git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60731 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* `vm_call_handler` and related functions accept `ec` instead of `th`.ko12017-10-271-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * vm_core.h (vm_call_handler): fix to accept `ec` instead of `th`. * vm_args.c: the following functions accept `ec` instead of `th`. * raise_argument_error * argument_arity_error * argument_kw_error * setup_parameters_complex * vm_eval.c: ditto. * vm_call0 * vm_call0_cfunc_with_frame * vm_call0_cfunc * vm_call0_body * vm_insnhelper.c: ditto * vm_call_iseq_setup_tailcall_0start * vm_call_iseq_setup_normal_0start * vm_callee_setup_arg * vm_call_iseq_setup * vm_call_iseq_setup_2 * vm_call_iseq_setup_normal * vm_call_iseq_setup_tailcall * vm_cfp_consistent_p * vm_call_cfunc_with_frame * vm_call_cfunc * vm_call_ivar * vm_call_attrset * vm_call_bmethod_body * vm_call_bmethod * vm_call_opt_send * vm_call_opt_call * vm_call_method_missing * vm_call_zsuper * current_method_entry * vm_call_method_each_type * vm_call_method_nome * vm_call_method * vm_call_general * vm_call_super_method * tool/mk_call_iseq_optimized.rb: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60468 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/mk_call_iseq_optimized.rb: fix generated commentnormal2017-07-041-1/+1
| | | | | | | | "tool/mkcall_iseq.rb" never existed in our repo. * tool/mk_call_iseq_optimized.rb: fix generated comment git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59264 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* * vm_core.h: revisit the structure of frame, block and env.ko12016-07-281-3/+3
| | | | | | | | | | | | | | | | | | | | | | [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
* tool: add descriptions and fix typosnormal2016-07-021-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * tool/asm_parse.rb: add description * tool/change_maker.rb: ditto * tool/downloader.rb: ditto * tool/eval.rb: ditto * tool/expand-config.rb: ditto * tool/extlibs.rb: ditto * tool/fake.rb: ditto * tool/file2lastrev.rb: ditto * tool/gem-unpack.rb: ditto * tool/gen_dummy_probes.rb: ditto * tool/gen_ruby_tapset.rb: ditto * tool/generic_erb.rb: ditto * tool/id2token.rb: ditto * tool/ifchange: ditto * tool/insns2vm.rb: ditto * tool/instruction.rb: ditto * tool/jisx0208.rb: ditto * tool/merger.rb: ditto * tool/mkrunnable.rb: ditto * tool/node_name.rb: ditto * tool/parse.rb: ditto * tool/rbinstall.rb: ditto * tool/rbuninstall.rb: ditto * tool/rmdirs: ditto * tool/runruby.rb: ditto * tool/strip-rdoc.rb: ditto * tool/vcs.rb: ditto * tool/vtlh.rb: ditto * tool/ytab.sed: ditto * tool/enc-unicode.rb: fix typo * tool/mk_call_iseq_optimized.rb: ditto * tool/update-deps: ditto [ruby-core:76215] [Bug #12539] by Noah Gibbs <the.codefolio.guy@gmail.com> git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55564 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* compile.c: explicit addressnobu2016-03-261-5/+5
| | | | | | | * compile.c (prepare_compile_error): add & to explicit address of functions to suppress warnings by old VC. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54294 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk: vm_call_iseq_optimized.incnobu2015-10-231-1/+1
| | | | | | * common.mk (incs): generate vm_call_iseq_optimized.inc too. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* * properties.svn2015-10-231-72/+72
| | | | git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52256 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* * vm_insnhelper.c: introduce new call handler for simple ISeqs.ko12015-10-231-0/+72
vm_call_iseq_setup_normal_0start() is simple, however it has some loops/conditions depends on ISeq::param.size and ISeq::local_size (in vm_push_frame(), inlined into this function). There are many simple methods which has a few parameters and local variables. So that this patch introduces several special functions generated in vm_call_iseq_optimized.inc by tool/mk_call_iseq_optimized.rb. This script makes vm_call_iseq_setup_normal_0start_Xparams_Ylocals() where X is 0 to 3 and Y is 1 to 6 (as current setting). In this case, X * Y = 24 functions are created. These functions creates fast method dispatch by inlining vm_push_frame() with immediate params/locals sizes. On my laptop, we can have the following results. vm2_method* 1.083 (8.3% faster) vm2_poly_method* 0.961 (3.4% slower) It shows 8.3% faster for inner loop method dispatch (hit inline cache), but 3.4% slower when inline cache miss because we need to find a suitable call handler. * common.mk: add a rule for vm_call_iseq_optimized.inc. * tool/mk_call_iseq_optimized.rb: added. * vm.c: include vm_call_iseq_optimized.inc. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52254 b2dd03c8-39d4-4d8f-98ff-823fe69b080e