summaryrefslogtreecommitdiffhomepage
path: root/mrbgems/mruby-compiler/core/codegen.c
AgeCommit message (Collapse)Author
2019-04-01Avoid keeping pointers from `mrb_sym2name_len()`; fix #4342Yukihiro "Matz" Matsumoto
The addresses for packed inline symbols reference `mrb->symbuf` that could be overridden by the later call of `mrb_sym2name_len`. Since file names in call stack information are kept as symbols, keeping the address in the C structures could cause problems like #4342. This changes small incompatible changes in function prototypes: * `mrb_parser_get_filename`: return value changed to `mrb_sym`. * `mrb_debug_get_filename`: add `mrb_state*` as a first argument. * `mrb_debug_get_line`: ditto. I believe above functions are almost internal, and no third-party mrbgem use them.
2019-02-18Use key argument register for `OP_KEY_P` to reduce register use.Yukihiro "Matz" Matsumoto
2019-02-11Small refactoring in `codegen.c`KOBAYASHI Shuji
2019-02-07Add `OP_ENTER` to blocks without parameters; fix #4175Yukihiro "Matz" Matsumoto
So that `lambda{}.call(1)` raises `ArgumentError` as CRuby does. Also, fixed junk assignment for `lambda{|;a|p a}.call{}`.
2019-01-28`class`/`module` expression with empty body should return `nil`KOBAYASHI Shuji
Before: p(class A end) #=> A p(class << self; end) #=> #<Class:#<Object:0x7fdc3880e420>> p(module B end) #=> B After/Ruby: p(class A end) #=> nil p(class << self; end) #=> nil p(module B end) #=> nil
2019-01-06Fix 0.0 and -0.0 handling.KOBAYASHI Shuji
Fix the following issue: Good: $ bin/mruby -e 'p(-0.0)' #=> "-0" Bad: $ bin/mruby -e 'a=0.0; p(-0.0)' #=> "0"
2018-11-25Allow destructuring in formal arguments.Yukihiro "Matz" Matsumoto
e.g. ``` def m(a,(b,c),d); p [a,b,c,d]; end m(1,[2,3],4) # => [1,2,3,4] ``` mruby limitation: Destructured arguments (`b` and `c` in above example) cannot be accessed from the default expression of optional arguments and keyword arguments, since actual assignment is done after the evaluation of those default expressions. Thus: ``` def f(a,(b,c),d=b) p [a,b,c,d] end f(1,[2,3]) ``` raises `NoMethodError` for `b` in mruby.
2018-11-15Fixed a bug in continuous read of target files; ref #4138Yukihiro "Matz" Matsumoto
Line number information in a compiled file was wrong.
2018-11-15Remove `filename`&`lines` from `mrb_irep` struct.Yukihiro "Matz" Matsumoto
This patch slightly reduce memory consumption (2% for my test).
2018-11-02Remove reserved symbols for now.Yukihiro "Matz" Matsumoto
It should be done by planned embedded symbols.
2018-10-29Reduce unnecessary symbol table entrytake-cheeze
2018-10-29Reduce instruction sizetake-cheeze
2018-10-13Fixed a bug caused by too much optimization; fix #4137Yukihiro "Matz" Matsumoto
2018-10-12Fixed broken stack consistency; fix #4135Yukihiro "Matz" Matsumoto
2018-09-07Clear terminated spacedearblue
2018-09-06Need to support multiple `**` splats in argument list; fix #4106Yukihiro "Matz" Matsumoto
2018-09-01Avoid warning in MSVC compilation.Yukihiro "Matz" Matsumoto
2018-09-01Too much peephole optimization in `gen_call`; fix #4091Yukihiro "Matz" Matsumoto
It causes trouble for safe navigation operator.
2018-08-30Remove consequent `OP_RETURN` by peephole optimization.Yukihiro "Matz" Matsumoto
2018-08-29Merge branch 'node_negate_fix' of https://github.com/ukrainskiysergey/mruby ↵Yukihiro "Matz" Matsumoto
into ukrainskiysergey-node_negate_fix
2018-08-29Check iseq buffer size before code emission; fix #4090Yukihiro "Matz" Matsumoto
The type of `s->pc` is now `uint16_t` that can be overflowed easily. Need more checks.
2018-08-25Reduce integer casting warnings.Yukihiro "Matz" Matsumoto
2018-08-25fixup! New bytecode implementation of mruby VM.Yukihiro "Matz" Matsumoto
2018-08-25Hash splat `**` should not be ignored.Yukihiro "Matz" Matsumoto
Implemented by adding `OP_HASHCAT` that merges hashes.
2018-08-25New bytecode implementation of mruby VM.Yukihiro "Matz" Matsumoto
2018-08-10Added push() after OP_SENDUkrainskiy Sergey
2018-08-09Small fixUkrainskiy Sergey
2018-08-09Fix weird behavior of the NODE_NEGATEUkrainskiy Sergey
2018-07-31Bytecode support for `mrdb`.Yukihiro "Matz" Matsumoto
2018-07-30Keyword argument implemented.Yukihiro "Matz" Matsumoto
2018-07-30New bytecode implementation of mruby VM.Yukihiro "Matz" Matsumoto
2018-07-10move mrb_irep_remove_lv to codegen.c in mruby-compileryuri
2018-05-01Update `MRB_FLO_TO_STR_FMT` to "%.16g"; fix #4016Yukihiro "Matz" Matsumoto
2018-05-01Fix stack position in multiple assignment; fix #4008Yukihiro "Matz" Matsumoto
This was caused by a patch from #2684.
2018-01-16Limit size of iseq buffer; fix #3927Yukihiro "Matz" Matsumoto
2017-12-15Check if destinations are too distant; fix #3900 fix #3901Yukihiro "Matz" Matsumoto
2017-11-08Simplify `MRB_WITHOUT_FLOAT` condtionYukihiro "Matz" Matsumoto
2017-11-04Merge branch 'master' of github.com:mruby/mrubyYAMAMOTO Masaya
2017-10-28Heavily refactored how lexical scope links are implemented; fix #3821Yukihiro "Matz" Matsumoto
Instead of `irep` links, we added a `upper` link to `struct RProc`. To make a space for the `upper` link, we moved `target_class` reference. If a `Proc` does not have `env`, `target_class` is saved in an `union` shared with `env` (if a `Proc` has env, you can tell it by `MRB_PROC_ENV_P()). Otherwise `target_class` is referenced from `env->c`. We removed links in `env` as well. This change removes 2 members from `mrb_irep` struct, thus saving 2 words per method/proc/block. This also fixes potential memory leaks due to the circular references caused by a link from `mrb_irep`.
2017-10-28Add some peephole optimizationsYukihiro "Matz" Matsumoto
2017-10-11Add MRB_WITHOUT_FLOATYAMAMOTO Masaya
2017-09-12Remove temporary limitation of `OP_EPOP`.Yukihiro "Matz" Matsumoto
After f68f5f6, the operand of `OP_EPOP` should have been `1`. Now we have removed the limitation.
2017-09-05Restrict `OP_EPOP` operand to `1`; ref #3789Yukihiro "Matz" Matsumoto
2017-08-23Fixed register windows of OP_SENDs generated by NODE_{DREGX,REGX}; ref #3783Christopher Aue
2017-08-23Fixed register windows of OP_SENDs generated by NODE_OP_ASGN; ref #3783Christopher Aue
2017-08-23Refactored code around generation of OP_SEND in NODE_OP_ASGNChristopher Aue
2017-08-23Fixed register windows of OP_SENDs generated by NODE_{FOR,SYMBOLS}; ref #3783Christopher Aue
2017-08-23Tested register windows of OP_SENDs generated by NODE_{RESCUE,HASH,ALIAS}; ↵Christopher Aue
ref #3783
2017-08-23Fixed register windows of OP_SENDs generated by ↵Christopher Aue
NODE_{SCALL,CASE,YIELD,UNDEF}; ref #3783
2017-08-19Replace stack `pop()` by `push_n(2); pop_n(3)`; fix #3783Yukihiro "Matz" Matsumoto
To calculate correct register windows size. The fix was suggested by Christopher Aue.