96cb1d5038
glue conversion intrinsics to instructions
2009-10-06 20:13:38 -05:00
f103018580
assert that vconvert's inputs are of the right type
2009-10-06 17:42:36 -05:00
0cbae3362f
math.vectors.conversion vocab with primitive words (to be mapped to intrinsics soon) and a super all-in-one "vconvert" macro wrapping the whole thing
2009-10-06 14:37:16 -05:00
3e44e0bb02
rename SIMD vmerge and kids to (vmerge), make new vmerge more generally useful
2009-10-05 17:55:39 -05:00
2d13aa8e6f
typo in vmerge-head, vmerge-tail docs
2009-10-05 11:14:14 -05:00
f8a4469c5f
go back to using random instead of uniform-random-float in math.vectors.simd tests because the software and SIMD implementations of norm and v. can sporadically diverge when their different operation orders lead to cancellation of catastrophically small inputs
2009-10-03 22:57:04 -05:00
c5993dbb39
fix v? software fallback
2009-10-03 22:37:45 -05:00
dca9d3e535
add %merge-vector-head and %merge-vector-tail instructions to back vmerge
2009-10-03 21:48:53 -05:00
398b460263
link vmerge into math.vectors docs
2009-10-03 21:48:16 -05:00
a233acf1b2
software version of vmerge word (to be backed by UNPCK instructions on x86 and VMRG instructions on ppc)
2009-10-03 20:22:37 -05:00
16cccffb71
sprinkle some NaNs onto math.vectors.simd fuzz tests to improve flavor
2009-10-03 11:47:02 -05:00
335df20713
add intrinsics for v<=, v<, v=, v>, v>=, vunordered?
2009-10-03 11:29:34 -05:00
e2e75c6b3a
add intrinsic for vnot/vbitnot
2009-10-02 20:04:28 -05:00
4b1b3ffbc7
i messed up a merge. weird
2009-10-02 16:05:00 -05:00
ff3ed1a3de
Merge branch 'master' of git://factorcode.org/git/factor into klazuka
...
Conflicts:
basis/math/vectors/vectors-docs.factor
2009-10-02 15:59:19 -05:00
c10b6d4ba7
Merge branch 'docs' of git://github.com/klazuka/factor into klazuka
...
Conflicts:
basis/math/vectors/vectors-docs.factor
2009-10-02 15:58:45 -05:00
271e6ddde5
give math.vectors shuffle words their own docs subsection in anticipation of having more ops in there
2009-10-02 15:37:49 -05:00
147a73441f
proofreading math.vectors docs
2009-10-02 15:18:42 -05:00
a4f8f87a21
add notes to math.vectors docs about simd booleans
2009-10-02 15:11:06 -05:00
e153d544eb
implement vand, vor, vandn, and vxor as bitwise intrinsics for simd types
2009-10-02 14:17:01 -05:00
8f79ea91ba
docs: change $subsection to $subsections
2009-10-02 12:15:48 -04:00
34eb7a9286
change vector logical words to reuse the input sequence types so that they work as simd fallbacks
2009-10-01 23:49:53 -05:00
0bc690e749
define simd equal? methods as v= vall?
2009-10-01 23:46:37 -05:00
9d424a1092
Merge branch 'master' of git://factorcode.org/git/factor
...
Conflicts:
basis/compiler/codegen/codegen.factor
2009-10-01 23:14:16 -05:00
c08cc860bf
create special intrinsic wrappers for 256-vector>scalar operations so that vall?, vany?, vnone? work on 256-vectors
2009-10-01 23:07:10 -05:00
aaeaa9c506
add tests for v=, vany?, vall?, vnone?
2009-10-01 21:24:14 -05:00
228ad950bb
%test-vector instruction for vany?, vall?, vnone?
2009-10-01 15:35:38 -05:00
94070c11aa
%compare-vector instruction (only does v= for now)
2009-10-01 14:31:37 -05:00
2303c0107e
add software vall?, vany?, vnone? words
2009-10-01 13:21:10 -05:00
16d08c01e3
math.vectors.simd.functor: don't open-code simd-nth and simd-with if we cannot generate double precision FP code. Should fix illegal instruction trap on CPUs with only SSE1
2009-10-01 07:36:50 -05:00
ef9abec15a
write v? and vmask in terms of bitwise ops
2009-10-01 00:09:25 -05:00
3ba79be651
Revert "add a %blend-vector intrinsic for v?"
...
This reverts commit 21e4b28b67
2009-09-30 23:40:37 -05:00
37a091a188
Merge branch 'master' of git://factorcode.org/git/factor
2009-09-30 23:04:04 -05:00
21e4b28b67
add a %blend-vector intrinsic for v?
2009-09-30 23:03:59 -05:00
65421b111b
math.vectors.simd: use fallbacks for hlshift, hrshift, vshuffle if parameter is not a literal;al; element access in int-4 on x86-64 now sign-extends the value; don't throw error at compile time if parameter for vshuffle does not have enough elements
2009-09-30 20:04:37 -05:00
36918a03d9
accept f and t as elements of literal simd vectors, storing binary all-zeroes or all-ones
2009-09-30 19:04:02 -05:00
d065203df4
break math.vectors docs into subsections
2009-09-30 13:21:25 -05:00
ddcbdae778
Merge branch 'master' of git://factorcode.org/git/factor
2009-09-30 00:07:45 -05:00
f38107c4cd
docs for new math.vectors logical ops
2009-09-30 00:07:37 -05:00
14e82bcd7a
math.vectors.simd: fix bad comparison logic in unit tests
2009-09-29 23:46:21 -05:00
1c8662ce4a
math.vectors.simd: add vbroadcast intrinsic, fix integer overflow issues
2009-09-29 22:58:20 -05:00
3c497cd2b8
add a "vmask" vector op to selectively zero out elements
2009-09-29 18:10:36 -05:00
2b13245704
math.vectors.simd: add fast intrinsic for 'nth', replace broadcast primitive with shuffles
2009-09-29 04:48:11 -05:00
a6e8277b2c
math.vectors.simd: add vshuffle intrinsic
2009-09-28 23:12:13 -05:00
db217295b0
Work in progress
2009-09-28 17:31:34 -05:00
e8243f7504
Fix some load errors in various places
2009-09-28 16:33:39 -05:00
f08521bf83
Fixing various test failures caused by C type parser change, and clarify C type docs some more
2009-09-28 08:48:39 -05:00
1109fb5725
math.vectors.simd: add intrinsic for int-4-boa, uint-4-boa, fix tests for C type parser change, fix software fallback for horizontal shifts
2009-09-28 06:34:22 -05:00
10c5fe5933
math.vectors.simd: add hlshift, hrshift (128-bit shift), vbitandn intrinsics
2009-09-28 02:17:46 -05:00
23ec1b85e8
Merge branch 'master' of git://factorcode.org/git/factor
2009-09-27 17:18:16 -05:00
31cb338fc9
math.vectors.simd: add *-cast words for converting between representations
2009-09-27 17:18:02 -05:00
4035bf88a1
software implementations of vector logical ops (vand, vor, vxor, vnot, v</=/>, vunordered?, v?)
2009-09-27 13:19:22 -05:00
be214f9142
math.vectors: remove special handling for negative shifts, now we just say the behavior of vlshift and vrshift with negative shifts is undefined
2009-09-26 02:26:31 -05:00
26c83d262c
math.vectors: fix SIMD unit tests
2009-09-25 21:44:14 -05:00
d518bed6b7
math.vectors: change vlshift and vrshift to mask the shift count by HEX: ff, to make them behave consistently with their SIMD counterparts
2009-09-25 18:50:44 -05:00
e5b94b11d7
Some fixes and cleanups in math.vectors
...
- Tighten up type inference for operations on complex float arrays
- Fix v. to have correct behavior with complex numbers
- Rename v<< and v>> to vlshift and vrshift to avoid clashing with v>> accessor
2009-09-24 06:58:33 -05:00
24039cb56a
math.vectors.simd: add v<< and v>> intrinsics for bitwise shifts on elements
2009-09-24 03:32:39 -05:00
165496d2f2
Add longlong-2, ulonglong-2, longlong-4, ulonglong-4 SIMD types, fix int-4 multiplication on SSE2
2009-09-23 20:23:25 -05:00
ebbcfd7ffc
fix using
2009-09-23 12:47:20 -05:00
abb8a221db
Merge branch 'master' of git://factorcode.org/git/factor
...
Conflicts:
basis/math/vectors/simd/simd-docs.factor
2009-09-23 10:11:49 -05:00
7cc383614c
math.vectors.simd: Improved documentation
2009-09-23 03:14:06 -05:00
8d84bcb61a
math and math.vectors: improved documentation
2009-09-23 03:08:45 -05:00
abac963882
math.vectors.simd: new operations: vabs vsqrt vbitand vbitor vbitxor
2009-09-23 02:47:14 -05:00
863ccb61d6
math.vectors.simd: fix all tests
2009-09-23 01:05:19 -05:00
66871995c9
math.vectors.simd: add saturated arithmetic operations
2009-09-20 23:16:02 -05:00
026761ed62
Merge branch 'master' of git://factorcode.org/git/factor
...
Conflicts:
basis/classes/struct/struct-tests.factor
basis/functors/functors-tests.factor
basis/specialized-arrays/specialized-arrays-tests.factor
2009-09-20 23:02:49 -05:00
47fab85a00
replace usages of <" with """
2009-09-20 22:42:40 -05:00
78c949b9b7
math.vectors: add v+- word which is accelerated by SSE3
2009-09-20 17:43:16 -05:00
dfb43bd2ca
More integer SIMD work
...
- move generated vocab support from specialized-arrays to vocabs.generated
- add fuzz testing to math.vectors.simd
- add alien type support for integer SIMD vectors
- SIMD: parsing word generates a SIMD type, instead of pre-generating them all in math.vectors.simd
2009-09-20 16:48:17 -05:00
ea2bcd69c7
math.vectors.simd: redesign to be more flexible, integer SIMD work in progress
2009-09-20 02:08:32 -05:00
4a1422e7fe
move some allocation words that don't really have much to do with c types out of alien.c-types into a new alien.data vocab
2009-09-17 22:36:05 -05:00
21a0722752
fix more ambiguities
2009-09-17 09:29:23 -05:00
18cb09a115
math.vectors.simd: don't run x86-specific tests on PPC
2009-09-11 21:38:40 -05:00
527db8995a
Specialized array overhaul
...
- Replace hand-written specialized-arrays.* subvocabularies with new system; instead of USE:ing specialized-arrays.T, do SPECIALIZED-ARRAY: T
- Ditto for specialized-vectors; use SPECIALIZED-VECTOR:
- io.mmap.functor: removed entirely, use <mapped-array> instead
- struct-arrays and struct-vectors have been removed because specialized arrays and vectors subsume them entirely
2009-09-09 22:33:34 -05:00
3e90786bc1
Fix various test failures
2009-09-08 19:18:56 -05:00
0d01371baf
math.vectors.simd: fix help rendering
2009-09-08 14:49:27 -05:00
c23e1dc1de
math.vectors.simd: remove useless dependency
2009-09-08 13:55:56 -05:00
ab64d187c9
Merge branch 'master' into simd
2009-09-08 13:38:14 -05:00
1e9f870e66
math.vectors.specialization: specialized vector words are now subwords of the generic vector word. This ensures that specializations get compiled correctly in all cases
2009-09-08 13:37:25 -05:00
ef09991500
Fixes
2009-09-08 00:13:18 -05:00
1c87486320
math.vectors.simd: allow punning SIMD vectors between types
2009-09-04 02:35:58 -05:00
136d6a42f3
math.vectors.simd: slightly faster 'sum' on 256-bit vectors: add the two components then do horizontal add, instead of doing a horizontal add on each one and adding the results
2009-09-04 02:23:25 -05:00
af14fd78dd
math.vectors.simd: docs
2009-09-04 01:22:18 -05:00
fc195f33f1
functors: support private words with DEFINES-PRIVATE; use this to make some words generated by math.vectors.simd.functor private
2009-09-04 01:21:59 -05:00
698f32c4a1
math.vectors.simd: define fallbacks for all vector constructors so that code can still work even if SIMD is not available
2009-09-03 21:37:55 -05:00
20dfbf7ac8
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
9cc705f6ba
math.vectors.simd: split off intrinsics into a sub-vocabulary, to avoid loading most of the SIMD code on bootstrap
2009-09-03 03:43:43 -05:00
f811208271
Detect SSE version and enable the correct set of SIMD intrinsics
2009-09-03 03:28:38 -05:00
52b99c050e
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
f45d80374c
alien.structs: struct-type now has a class slot; fix specialized complex-float/double arrays
2009-08-09 16:10:11 -05:00
9b4ffd65d3
math.vectors.specialization: first attempt at some call site splitting for vector ops. Specialized array types generate customized variants of all vector words, if input types are known at compile time, a call to the specialized version is inserted
2009-08-09 03:07:33 -05:00
189043eae7
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
23ae3f4ab6
homogeneous coordinates coated in nurbsauce
2009-05-21 20:55:44 -05:00
4dba6979da
vector bilerp word
2009-05-20 10:26:55 -05:00
1216ea2fd8
cut perlin-noise time in half
2009-05-10 10:41:50 -05:00
1d58b94bf7
send bilerp upstream to spawn
2009-05-09 11:43:04 -05:00
e4c84a91f6
more vector operations; perlin noise vocab
2009-05-05 22:17:04 -05:00
ca0c8937ca
lerp functions
2009-05-05 13:55:00 -05:00
ae0da1cf98
Fix conflict
2009-02-05 22:29:15 -06:00
ccdd8999e1
swizzle word to swizzle sequences
2009-02-04 19:46:19 -06:00
f79041545a
New distance word in math.vectors, replaces distance in math.points, more efficient
2008-11-29 10:19:23 -06:00
e17f519480
Minor documentation fixes
2008-11-17 22:51:57 -05:00
9c27e9d61b
Replace >r r> usage with dip in math.vectors
2008-11-17 18:17:14 -05:00
c19f2257f4
Fix permission bits
2008-10-02 08:34:49 -05:00
d5140cf248
Fix math.vectors unit tests
2008-09-12 15:52:43 -05:00
f497c7e151
Move map-reduce combinator to core, re-implement norm-sq and v. for better performance
2008-09-12 11:29:12 -05:00
d66f887736
Create basis vocab root
2008-07-28 22:03:13 -05:00
Joe Groff
Doug Coleman
Keith Lazuka
Slava Pestov
Slava Pestov
Aaron Schaefer