Commit Graph

305 Commits (b03eaf3c32885517c23c691353c3bbb2c7a245a1)

Author SHA1 Message Date
Slava Pestov 64cddef3e1 cpu.ppc: fix %single>double-float and %double>single-float 2009-09-10 13:04:58 -05:00
Joe Groff 4d1e832f79 Merge branch 'master' of git://factorcode.org/git/factor 2009-09-09 17:14:48 -05:00
Joe Groff c50a35c062 altivec instructions for powerpc assembler 2009-09-09 17:14:36 -05:00
Slava Pestov ed09dfe3ae Merge branch 'master' of git://factorcode.org/git/factor 2009-09-09 13:56:20 -05:00
Slava Pestov daff63071f Fix the build 2009-09-09 13:44:54 -05:00
Slava Pestov a01f693335 cpu.ppc: fix bootstrap 2009-09-08 23:53:51 -05:00
Slava Pestov 0e2a3e4292 Merge branch 'master' of git://factorcode.org/git/factor 2009-09-08 22:34:17 -05:00
Slava Pestov cd5133b233 cpu.x86: tweak SIMD intrinsics 2009-09-08 22:34:01 -05:00
Joe Groff f166797dd4 no really, update ppc for argument order changes 2009-09-08 22:21:00 -05:00
Joe Groff 3003b9e5d0 Merge branch 'master' of git://factorcode.org/git/factor 2009-09-08 21:58:25 -05:00
Joe Groff c95a7febb0 update cpu.ppc for argument order changes 2009-09-08 21:58:11 -05:00
Slava Pestov 8916fb7a3c Merge branch 'master' of git://factorcode.org/git/factor 2009-09-08 21:51:21 -05:00
Slava Pestov 7e2f0e5dbf compiler: separate ##save-context instruction from ##alien-invoke, generate a ##save-context for libm calls, and add a pass to combine multiple context saves within a basic block. Fixes crashes with FP traps thrown by libm functions on x86-32 2009-09-08 21:50:55 -05:00
Joe Groff 9be1f42fe9 typos in cpu.ppc 2009-09-08 21:44:11 -05:00
Slava Pestov 410d2bf0d0 Merge branch 'master' of git://factorcode.org/git/factor 2009-09-08 19:35:14 -05:00
Slava Pestov 74fa73aeaf Fix various test failures 2009-09-08 19:18:56 -05:00
Doug Coleman d02d95ea34 Merge branch 'master' of git://factorcode.org/git/factor 2009-09-08 17:05:58 -05:00
Joe Groff 0ea9949e51 split unordered and ordered float comparison intrinsics in compiler; generate only unordered comparisons for now 2009-09-08 17:04:26 -05:00
Doug Coleman 740a175e56 duplicate using 2009-09-08 17:02:31 -05:00
Slava Pestov 8b6ec8e030 cpu.x86.features: better wording 2009-09-08 14:17:05 -05:00
Slava Pestov e5e7c1f725 cpu.x86.features: add -sse-version command-line switch to override SSE detection 2009-09-08 13:56:37 -05:00
Slava Pestov 1e5bc3b376 inline alien-vector and set-alien-vector if SIMD is not available for a small speedup 2009-09-08 13:56:17 -05:00
Slava Pestov 38c961cf6f Fixes 2009-09-08 00:13:18 -05:00
Slava Pestov bbca00e2ae Fix conflicts 2009-09-07 23:51:25 -05:00
Joe Groff d47d747ebc i had comisd/ucomisd backwards on x86 2009-09-04 12:30:30 -05:00
Slava Pestov 382e1d5b0e fix some typos in cpu.ppc 2009-09-04 11:18:41 -05:00
Slava Pestov 92d5d8f0c5 Merge branch 'master' of git://factorcode.org/git/factor into ppc-float-compare 2009-09-04 10:58:50 -05:00
Joe Groff 638e351131 update powerpc compiler to generate correct float comparisons 2009-09-04 10:51:12 -05:00
Slava Pestov fc1bf07a1f cpu.ppc: fix typo 2009-09-04 06:41:33 -05:00
Slava Pestov e36a0d7ef4 compiler: clean up code generation for alien boxing/unboxing a bit 2009-09-03 21:22:43 -05:00
Joe Groff c480bec303 convert comparison branch code in compiler to use locals 2009-09-03 21:19:39 -05:00
Slava Pestov 4d5a4222b6 More SIMD work
- Rename SIMD types and register representations: <type>-<count> rather than <count><type>-array
- Make a functor to define 256-bit vector types, use it to define float-8 type
- Make SIMD instructions pure-insns so that they participate in value numbering
2009-09-03 20:58:56 -05:00
Joe Groff 036ff77306 add compiler comparison codes for floating-point unordered comparisons; update x86 backend to generate proper code for all floating-point comparisons 2009-09-03 20:32:05 -05:00
Slava Pestov 3d4c04302a Merge branch 'master' into simd 2009-09-03 03:45:58 -05:00
Slava Pestov 906a0d212a Detect SSE version and enable the correct set of SIMD intrinsics 2009-09-03 03:28:38 -05:00
Slava Pestov ff8c70dbe0 Initial implementation of SSE vector intrinsics:
- cpu.architecture: add SSE vector representations
- compiler.cfg.intrinsics.alien: remove an attempt at optimization that value numbering handles now
- compiler.cfg.representations: support instructions where the representation is set in the 'rep' slot, and support conversions between single and double floats
- alien-float, set-alien-float now use the single float representation, and the conversion is implicit; this fixes a long-standing bug where a register could get clobbered because of how %set-alien-float was defined on x86
- math.vectors.specialization: add support for SIMD specialization (where the vector word's body is replaced by another quotation), also specialize the 'sum' word
- math.vectors.simd: 4float-array, 2double-array, 4double-array types, and specializers for the math.vectors words
2009-09-03 02:33:07 -05:00
Joe Groff 102df64ec7 i suck at reading tech docs--those were m64 instructions, not mm instructions 2009-09-02 12:58:35 -05:00
Joe Groff 962d560c10 get rid of useless mm->xmm instructions in cpu.x86.assembler, add MOVHLPS and MOVLHPS 2009-09-02 11:06:08 -05:00
Slava Pestov 85a2bfab6c compiler: eliminate boilerplate by centralizing info in declarative INSN: syntax 2009-09-02 06:22:37 -05:00
Slava Pestov f91b539c31 cpu.ppc: implement fast float function calls; 3x speedup on benchmark.struct-arrays on PowerPC 2009-09-01 15:19:26 -05:00
Slava Pestov 6f1a7c731c cpu.ppc: fix %box-displaced-alien 2009-08-30 20:56:04 -05:00
Slava Pestov 9595be4bf9 %box-displaced-alien: fix clobberage found by Doug 2009-08-30 05:11:08 -05:00
Slava Pestov 0db01f6d5f compiler.cfg.linear-scan now supports partial sync-points where all registers are spilled; taking advantage of this, there are new trigonometric intrinsics which yield a 2x performance boost on benchmark.struct-arrays and a 25% boost on benchmark.partial-sums 2009-08-30 04:52:01 -05:00
Slava Pestov 2bb6293217 compiler: add fixnum-min/max intrinsics; ~10% speedup on benchmark.yuv-to-rgb 2009-08-28 19:02:59 -05:00
Slava Pestov d957ae4e44 Performance improvements to make struct-arrays benchmark faster
- improved optimization of ##unbox-any-c-ptr on ##box-displaced-alien; convert it to ##unbox-c-ptr where possible using class info stored in the ##bda instruction
- make fcos, fsin, etc inline again; everything in math.libm inline again, except for fsqrt which is an intrinsic
- convert min and max on floats to float-min and float-max
- make min and max not inline, so that the above can work
- struct-arrays: rice a bit so that more fixnums come up
2009-08-28 05:21:16 -05:00
sheeple 98f93f799b cpu.ppc: fix ##box-displaced-alien 2009-08-27 04:43:45 -05:00
Slava Pestov f662e6403a compiler: new inline intrinsic for <displaced-alien> where the inputs have known types; value numbering now eliminates unnecessary allocation of displaced aliens if the result is immediately unboxed again 2009-08-27 00:06:19 -05:00
Slava Pestov 0df8aadce2 cpu.x86: use SQRTSD instruction for math.libm:fsqrt word 2009-08-25 23:22:15 -05:00
Slava Pestov b7e29ca8dd cpu.ppc.assembler: LOAD32 assembler macro was busted 2009-08-25 22:37:10 -05:00
Slava Pestov 1afd001393 basis/cpu: eliminate some usages of rot 2009-08-25 19:38:48 -05:00