Commit Graph

620 Commits (e88e7f70be34325ef3ebdbef05e9ea28c78508e9)

Author SHA1 Message Date
Slava Pestov 4507bdcbc0 compiler.cfg.linear-scan: code cleanups 2009-07-08 23:28:28 -05:00
Slava Pestov c20e6c290f compiler.cfg.linear-scan: split off parallel mapping code from resolve pass, use it in assignment pass to resolve parallel copies 2009-07-08 23:07:06 -05:00
Slava Pestov e2044602aa compiler.cfg.linear-scan: debugging spilling, add more assertions 2009-07-07 13:01:27 -05:00
Slava Pestov 43c873f00a compiler.cfg.linear-scan: code cleanup 2009-07-07 03:45:27 -05:00
Slava Pestov 0bcf4ce535 compiler.cfg.linear-scan: Re-implement spilling, add some additional runtime assertions, simplify assignment pass since it doesn't have to track spill slots anymore; just assume a live value that's not in active or inactive is spilled 2009-07-07 03:28:55 -05:00
Slava Pestov 4402d8652c compiler.cfg.linear-scan: minor fixes 2009-07-05 22:51:53 -05:00
Slava Pestov a452f32e3a compiler.cfg.linear-scan: Get cycle breaking in resolve pass to work by allocating a spare spill slot for this purpose 2009-07-05 21:32:23 -05:00
Slava Pestov f14a61fac2 Fix compiler.cfg.linear-scan.resolve test failure 2009-07-04 16:30:16 -05:00
Slava Pestov c0f1b2205a compiler.cfg.value-numbering.rewrite: disable ##compare optimizations for now 2009-07-04 03:26:37 -05:00
Slava Pestov fb488025aa compiler.cfg.value-numbering.rewrite: fix ##compare-imm rewrite rule 2009-07-04 02:50:50 -05:00
Slava Pestov c00af97fa1 compiler.cfg.linear-scan.resolve: More fixes 2009-07-04 00:45:06 -05:00
Slava Pestov da13681bc8 compiler.cfg.linear-scan: redo resolve pass to fix a correctness issue 2009-07-03 23:38:52 -05:00
Slava Pestov 8d3a45dee2 compiler.cfg: clean up GC check generation to use spill-slot data type 2009-07-03 23:11:23 -05:00
Slava Pestov d07c0429fc compiler.cfg.linear-scan: assign spill slots to vregs globally, so that resolve pass never has to perform a memory->memory transfer 2009-07-03 22:51:40 -05:00
Slava Pestov 0180209f4e compiler.tree.dead-code: methods on flushable generics should be flushable 2009-07-03 21:31:26 -05:00
Doug Coleman cae8fed16c convert ##compare with immediates to ##compare-imm 2009-07-03 19:19:33 -05:00
Slava Pestov 2c3b2a0b71 compiler.cfg.predecessors: delete dead predecessors from phi nodes 2009-07-03 18:45:03 -05:00
Doug Coleman 84de04e502 Merge branch 'master' of git://factorcode.org/git/factor 2009-07-02 19:04:50 -05:00
Doug Coleman e7a227ad40 add constant folding for integer ops, refactor some rewrites 2009-07-02 19:03:21 -05:00
Slava Pestov c5e7266e17 Merge branch 'master' of git://factorcode.org/git/factor 2009-07-02 18:36:49 -05:00
Doug Coleman f11ba00667 Revert "fix a unit test so it's sure to overflow 64bit machines"
This reverts commit 88a545a215.
2009-07-02 18:18:50 -05:00
Doug Coleman 559a77be43 fix a unit test so it's sure to overflow 64bit machines 2009-07-02 18:05:55 -05:00
Doug Coleman 2aade588ee Merge branch 'master' of git://factorcode.org/git/factor 2009-07-02 18:01:58 -05:00
Doug Coleman e11e3cf6d2 unit test to make sure adds are not optimized incorrectly 2009-07-02 18:01:00 -05:00
Doug Coleman 80eb5068e2 fix bug in rewriting #add -- wasn't checking small-enough?, and change negative adds to subtractions/negative subtractions to adds 2009-07-02 17:55:35 -05:00
Slava Pestov 9404a7d0e3 half-floats: Add some more unit tests 2009-07-02 17:52:53 -05:00
Slava Pestov 6d46bf623c compiler.cfg.liveness: update for new ##phi node representation 2009-07-02 17:28:02 -05:00
Slava Pestov 90daf6d03d compiler.cfg: change representation of ##phi nodes. The input slot of ##phi nodes is now an assoc mapping predecessors to values, rather than a sequence of values. This ensures that optimization which delete predecessors can be implemented 2009-07-02 17:10:50 -05:00
Doug Coleman 86901eb01e rewrite rules for add/sub/mul/and/or/xor-imm, rewrite load-immediate/add to be add-imm 2009-07-02 11:35:31 -05:00
Doug Coleman 04cb320c55 simplify sub and sub-imm 2009-07-02 11:33:15 -05:00
Doug Coleman 826464724e add a utility word vreg>constant 2009-07-02 11:32:23 -05:00
Slava Pestov 3a4916462b compiler.cfg.dce: now performs three passes over the CFG to fix a phase ordering issue 2009-07-02 03:17:19 -05:00
Slava Pestov 513a600664 compiler.cfg.dce: minor renamings 2009-07-02 02:41:10 -05:00
Daniel Ehrenberg 392e313f05 Merge branch 'master' of git://factorcode.org/git/factor 2009-07-02 02:31:06 -05:00
Daniel Ehrenberg fcfd7f01a3 Dead code elimination eliminates some dead allocations 2009-07-02 02:30:46 -05:00
Slava Pestov 2de41f01da compiler.cfg.optimizer: enable branch folding 2009-07-02 01:05:59 -05:00
Slava Pestov af875ba836 compiler.cfg: bug fixes in GC check insertion and fixnum intrinsics 2009-07-02 00:51:06 -05:00
Slava Pestov e6b1c21590 compiler.cfg.linear-scan: fix spilling 2009-07-01 22:37:27 -05:00
Slava Pestov 41f48e3cfe compiler.tree.optimizer: Remove redundant call to compute-def-use 2009-07-01 17:41:20 -05:00
Slava Pestov a19f22ec82 compiler.cfg.linear-scan: debugging spilling and resolve pass 2009-07-01 17:41:07 -05:00
Slava Pestov 7e3fa45512 compiler.cfg.stack-analysis: fix unit tests for useless-blocks renaming and ##no-tco instruction 2009-06-30 23:17:33 -05:00
Slava Pestov a3b21292b6 Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 23:07:36 -05:00
Slava Pestov f894baf60e compiler.cfg.optimizer: disable branch folding for now 2009-06-30 23:07:27 -05:00
Slava Pestov 72b6bef19d compiler.cfg.builder: emit less crap after a #terminate node 2009-06-30 23:03:25 -05:00
Doug Coleman 8b728e0a8a Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 23:02:04 -05:00
Doug Coleman b1c7374175 optimize fixnum*fast and fixnum-shift-fast 2009-06-30 23:01:44 -05:00
Doug Coleman ded1e29d2f combine add-imm instructions into a single add 2009-06-30 22:59:53 -05:00
Slava Pestov 3385e50c43 compiler.cfg.branch-folding: fold away branches where both inputs are the same register. This results in 'push' being compiled slightly better 2009-06-30 22:43:32 -05:00
Slava Pestov fc595a7075 Fix possible infinite loop in skip-empty-blocks and remove redundant test in compiler.cfg.useless-conditionals 2009-06-30 22:43:02 -05:00
Slava Pestov f412bbc6d6 Change branch splitting around a bit, rename useless-blocks to useless-conditionals and rewrite it 2009-06-30 22:11:15 -05:00
Slava Pestov eb7f50adf4 Fix conflict 2009-06-30 21:37:49 -05:00
Slava Pestov 0402790001 compiler.cfg Remove height tracking for ##call instructions, wire in ##no-tco instruction 2009-06-30 21:21:46 -05:00
Slava Pestov 3da560130a compiler.cfg.tco: fix tail call optimization for ##fixnum-mul 2009-06-30 21:08:08 -05:00
Slava Pestov eca718b32d compiler.cfg.builder: fix stack effect 2009-06-30 21:07:55 -05:00
Slava Pestov 612e4b99b0 compiler.cfg.linear-scan.assignment: insert-copy did the wrong thing if the second interval had been split. Fixes compilation of 'trilerp' 2009-06-30 21:07:39 -05:00
Slava Pestov 1acf243cce Fix conflict 2009-06-30 20:16:09 -05:00
Slava Pestov 5c6c3ecd85 compiler.cfg.tco: Tail call optimization moved out of compiler.cfg.builder into its own pass 2009-06-30 20:13:35 -05:00
Doug Coleman 2cd202d175 remove duplicate using 2009-06-30 18:48:52 -05:00
Doug Coleman 357408df3e Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 18:48:10 -05:00
Doug Coleman 97cd0d584e make commutative operations with immediates output the same IR 2009-06-30 18:47:24 -05:00
Slava Pestov bc2a6c0ecc compiler.cfg.stack-analysis: fix tests 2009-06-30 18:22:54 -05:00
Slava Pestov d199154179 Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 18:12:00 -05:00
Slava Pestov 3355fa4809 compiler.cfg.stack-analysis: enable global optimization 2009-06-30 18:11:45 -05:00
Slava Pestov 54876d0cba compiler.cfg.linear-scan.assignment: get check-assignment? to work with coalescing 2009-06-30 18:10:53 -05:00
Doug Coleman f1b2550035 Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 16:45:57 -05:00
Doug Coleman d1f6871081 fix bug where traversal order was lost 2009-06-30 16:44:14 -05:00
Slava Pestov 578b0126fc compiler.cfg.linear-scan: compute-free-pos was broken in the case where more than one inactive interval had the same physical register assigned 2009-06-30 16:07:58 -05:00
Doug Coleman f41f84436c spill to spill-temp, not to virtual spill addresses 2009-06-30 15:40:13 -05:00
Doug Coleman ede8e1ca30 Merge branch 'master' of git://factorcode.org/git/factor 2009-06-30 15:32:11 -05:00
Doug Coleman ea072731d4 add branch splitting pass to compiler.cfg 2009-06-30 15:31:48 -05:00
Slava Pestov 32738dddde compiler.cfg.stack-analysis: fix typos in unit tests 2009-06-29 16:37:40 -05:00
Slava Pestov 82bfaca188 Fix conflict 2009-06-29 01:59:53 -05:00
Slava Pestov 0f68ee3e9a compiler.cfg.linear-scan.resolve: remove redundant vocabs from using list 2009-06-28 22:58:35 -05:00
Slava Pestov 6e72cf9711 compiler.cfg.stack-analysis: fix another corner case with ##peek insertion, and fix changed-loc merging 2009-06-28 22:52:28 -05:00
Doug Coleman 1fab8efeee Merge branch 'master' of git://factorcode.org/git/factor 2009-06-28 16:43:45 -05:00
Doug Coleman 3b0954f63e fix bug in linear-scan.resolve by rewriting entire algorithm 2009-06-28 16:43:17 -05:00
Slava Pestov d4de1d38b9 Merge branch 'master' of git://factorcode.org/git/factor 2009-06-27 17:33:00 -05:00
Slava Pestov d5179b724e compiler.cfg.stack-analysis: Fix case where both an ##inc-d/r and a ##peek get inserted 2009-06-27 17:32:37 -05:00
Doug Coleman 1cb6bc99da Merge branch 'master' of git://factorcode.org/git/factor
Conflicts:
	basis/compiler/cfg/linear-scan/resolve/resolve-tests.factor
2009-06-26 21:51:08 -05:00
Doug Coleman 90017eb248 add spill-temp to compiler.cfg.instructions, implement parallel register assignment in linear-scan.resolve 2009-06-26 21:48:21 -05:00
Slava Pestov ef32eb2e72 Fix bootstrap and unit test 2009-06-26 17:44:33 -05:00
Slava Pestov 96baede36a Redesign compiler.cfg.stack-analysis to make compiler.cfg.height redundant, and to fix some problems 2009-06-26 17:29:55 -05:00
Slava Pestov 22fa524db1 compiler.cfg.stack-analysis: split up into three vocabs 2009-06-24 13:54:43 -05:00
Slava Pestov f3cf8fad2b compiler.cfg.linear-scan: fix unit tests 2009-06-24 13:54:12 -05:00
Slava Pestov c53aca6016 compiler.cfg.linear-scan: fix coalescing to take lifetime holes into account 2009-06-23 21:32:51 -05:00
Slava Pestov 387f6df9e5 compiler.cfg.linear-scan: Debugging resolve pass 2009-06-22 00:24:51 -05:00
Slava Pestov 330bea3cc2 compiler.cfg.linear-scan.resolve: unit tests 2009-06-21 00:23:34 -05:00
Slava Pestov 5a78956956 compiler.cfg.linear-scan: Working on resolve pass 2009-06-21 00:20:01 -05:00
Slava Pestov c1839200dd compiler.cfg.linear-scan.assignment: correctly compute live registers at GC check points 2009-06-20 15:33:12 -05:00
Slava Pestov 5c912504d7 compiler.cfg.linear-scan: untangle add-active/delete-active/add-handled calls in spilling, replace 'sort-values last' with 'alist-max' in compiler.utilities 2009-06-19 18:28:30 -05:00
Slava Pestov 2f15ac3c8f compiler.cfg.linear-scan: Fix a couple of bugs 2009-06-19 03:42:42 -05:00
Slava Pestov 1f5f8393c3 compiler.cfg.linear-scan: Further progress on inactive interval handling 2009-06-17 15:27:20 -05:00
Slava Pestov 29327a787c Remove initial-quot feature 2009-06-15 13:07:15 -05:00
Slava Pestov 6fb7dca404 compiler.cfg.linear-scan.resolve: fix compile error 2009-06-14 17:00:52 -05:00
Slava Pestov 285c8cecc6 Add some failing unit tests exposing bugs in initial-quot: implementation 2009-06-13 18:34:27 -05:00
Slava Pestov 64f93e41a9 Various linear scan fixes
insert spill before reload to fix x86-32 regression
inactive splitting: if all inactive intervals' registers are in use, don't fail
fix stack analysis tests
2009-06-13 17:35:40 -05:00
Slava Pestov d0f6a7d048 Split up compiler.cfg.linear-scan.allocation into a number of sub-vocabularies; start work on compiler.cfg.linear-scan.resolve; start work on inactive interval splitting 2009-06-11 17:55:14 -05:00
Slava Pestov 2dcdfa2d8e compiler.cfg: add a declaration to fix tools.deploy.test.11 2009-06-09 03:36:16 -05:00
Slava Pestov 45a2105449 cpu.x86.assembler: IMUL2 instruction was busted for immediate operands
When given a register and an immediate, it would generate imul imm,dst,dst however the 64-bit prefix was generated wrong and if dst was an extended register only the first operand would be an extended register. To fix this, change IMUL2 to not work on immediates anymore, and added a new IMUL3 that takes a destination register, source register, and immediate. Also, change compiler.cfg.two-operand to not two-operandize %mul-imm, since this isn't needed anymore.
This fixes the sporadic benchmark.tuple-arrays crash on 64-bit machines.
2009-06-08 21:15:52 -05:00
Slava Pestov 4fbe99400f Merge branch 'master' of git://factorcode.org/git/factor 2009-06-07 21:46:41 -05:00