9bb3c0d71b
compiler.cfg.alias-analysis: simplify and speed up
2010-05-14 18:37:07 -04:00
db0f40e481
compiler: small fixes and cleanups
2010-05-13 02:48:20 -04:00
705b03b92b
compiler.cfg.alias-analysis: storing a reference into an object now merges the reference's alias class with the heap alias class
2010-05-03 18:20:28 -04:00
e2f164a52f
compiler.cfg: cleanup
2010-05-03 17:34:26 -04:00
c35eba5e34
compiler.cfg.alias-analysis: write unit tests and fix redundant store elimination
2010-05-03 17:34:22 -04:00
2b50b9b315
compiler.cfg.value-numbering: merge 'simplify' pass into 'rewrite'
2010-05-03 17:34:08 -04:00
6fdcd9fb02
Untagged fixnums work in progress
2010-05-03 17:34:02 -04:00
19412e4ad1
compiler: Start using tagged-rep for stuff, and split up compiler.cfg.representations into several sub-vocabularies
2010-05-03 17:34:01 -04:00
c4cf2a4085
compiler: change how 'f' is represented in low level IR to simplify some code, and fuse a ##load-constant of a word with a ##compare into a ##compare-imm on x86-32. This eliminates a spill from binary-search
2010-04-18 21:42:45 -05:00
4f585f991d
compiler: new set-special-object intrinsic; more efficient special-object intrinsic
2010-04-01 20:09:14 -04:00
84d6cf5d7b
Maing vocabs.hierarchy use sets; modifying alias analysis's use of sets
2010-02-28 22:55:22 -05:00
d797e8d29e
Making alias analysis use sets
2010-02-28 22:42:10 -05:00
0f0571e48a
Moving new-sets to sets
2010-02-26 16:01:01 -05:00
d65296b334
vm: 4 bit tags, new representation of alien objects makes unbox-any-c-ptr more efficient (work in progress)
2009-11-02 04:25:54 -06:00
10a49a4613
compiler.cfg.alias-analysis: fix ##compare where operands are copies
2009-10-01 23:54:19 -05:00
231a2d1b71
Fix alias analysis of a ##compare whose input is another ##compare
2009-10-01 23:20:05 -05:00
d3acd0a530
compiler.cfg.alias-analysis: optimize ##vm-field-ptr and ##alien-global instructions, and optimize out ##compare between values of different alias classes; this optimizes '[ [ >float ] bi@ [ + ] [ - ] 2bi eq? ]' down to an o-op and removes boxing from '[ [ >float ] bi@ [ + ] [ - ] 2bi = ]'
2009-10-01 23:03:17 -05:00
67aec74439
cleaned up vm-field-ptr compiler code
2009-09-24 08:16:57 +01:00
2a1a4ccf27
fixed up getenv compiler intrinsic to use vm struct userenv
2009-09-16 08:16:32 +01:00
775b9af2f7
compiler: eliminate boilerplate by centralizing info in declarative INSN: syntax
2009-09-02 06:22:37 -05:00
d1ce837569
Delete empty unit tests files, remove 1- and 1+, reorder IN: lines in a lot of places, minor refactoring
2009-08-13 19:21:44 -05:00
2d575d7ec9
compiler.cfg: virtual registers are integers now, and representations are stored off to the side. Fix bug in representation selection that would manifest if a value was used as a float and a fixnum in different branches; cannot globally unbox float in this case
2009-08-08 04:02:18 -05:00
4d2160799f
Split off the notion of a register representation from a register class
2009-08-07 17:44:50 -05:00
c355c21de6
Move liveness back to compiler.cfg.liveness, fix a corner-case where it didn't work on some forms of non-SSA code
2009-07-22 06:07:28 -05:00
5d9f7b0ed2
compiler.cfg: Major restructuring -- do not compute liveness before local optimization, and instead change local optimizations to be more permissive of undefined values. Now, liveness is only computed once, after phi elimination and before register allocation. This means liveness analysis does not need to take phi nodes into account and can now use the new compiler.cfg.dataflow-analysis framework
2009-07-22 03:08:28 -05:00
3de85158de
Merge branch 'master' into global_optimization
2009-06-01 03:12:32 -05:00
57d9d9f961
Split off local-optimization combinator into compiler.cfg.local, factor out CFG -> MR into compiler.cfg.mr, split off GC check insertion into a new compiler.cfg.gc-checks pass
2009-05-31 12:20:46 -05:00
968a9bb666
Various codegen improvements:
...
- new-insn word to construct instructions
- cache RPO in the CFG
- re-organize low-level optimizer so that MR is built after register allocation
- register allocation now stores instruction numbers in the instructions themselves
- split defs-vregs into defs-vregs and temp-vregs
2009-05-29 13:11:34 -05:00
40949800bf
Fixing various bugs; alias analysis wasn't handling ##phi nodes, stack analysis incorrectly handled height-changing back edges and ##fixnum-*, clean up ##dispatch generation
2009-05-29 01:39:14 -05:00
cef9ac778b
compiler.cfg.alias-analysis: ##peek needs to set alias class of output value
2009-05-28 18:19:01 -05:00
4292033da6
Fix alias analysis
2009-05-28 02:48:58 -05:00
3954c3fdc3
Add a new ##allocation union to remove some code duplication
2009-05-27 18:55:49 -05:00
1242a3a411
New local-optimization combinator removes some boilerplate
2009-05-26 19:56:56 -05:00
25805f9bfe
Refactoring low-level optimizer to support stack analysis pass
2009-05-26 19:31:19 -05:00
8da9d0f203
rename peek -> last and update all usages
2009-05-25 16:38:33 -05:00
6b25e99470
Add summary for heaps more vocabs
2009-02-16 21:05:13 -06:00
44bfff7c7b
Rename ##load-indirect to ##load-reference since this is more descriptive; value numbering doesn't assign expressions to ##load-reference nodes since this would end up folding literals which were eq? but not =
2009-01-29 01:44:58 -06:00
daf2857d7b
define-inline now takes an effect parameter
...
C: now sets the stack effect of the new word to slot names
add new basis/constants vocab; CONSTANT: foo bar is equivalent to : foo bar ; inline
(going to fold CONSTANT:, ALIAS: and qualified into core soon perhaps)
tighten the screws: inline words now need a stack declaration too
update some usages of inline words
2008-12-15 19:44:56 -06:00
4f0a9f311e
Untangling some dependencies
2008-12-08 14:58:00 -06:00
a56d480aa6
Various optimizations leading to a 10% speedup on compiling empty EBNF parser:
...
- open-code getenv primitive
- inline tuple predicates in finalization
- faster partial dispatch
- faster built-in type predicates
- faster tuple predicates
- faster lo-tag dispatch
- compile V{ } clone and H{ } clone more efficiently
- add fixnum fast-path to =; avoid indirect branch if two fixnums not eq
- faster >alist on hashtables
2008-12-06 09:16:29 -06:00
08b277e892
Fix alias anlaysis bug: result of ##box-alien and ##box-float did not get an AC, so calling hi-tag would not compile
2008-11-03 01:52:55 -06:00
492a15e345
Move insn class to compiler.cfg.instructions
2008-10-24 09:17:06 -05:00
885adc8dc9
Various improvements
2008-10-23 05:27:54 -05:00
f09813f6fd
Alias analysis
2008-10-22 18:37:47 -05:00
Slava Pestov
Daniel Ehrenberg
Phil Dawes
Doug Coleman
Slava Pestov
Doug Coleman