Slava Pestov
ff172f4132
Make specialized arrays and SIMD types final so that typed can unbox them
2010-02-18 02:39:40 +13:00
Slava Pestov
df4fb4a3ee
Removing integers-as-sequences
2010-01-15 07:15:33 +13:00
Joe Groff
6b9c4254ad
move simd operation methods onto simd-128 instead of concrete classes to save image space
2009-12-05 21:19:17 -08:00
Joe Groff
2b2be5f121
Merge branch 'master' of git://factorcode.org/git/factor
2009-12-05 17:18:49 -08:00
Joe Groff
3e40a36c50
fix simd tests
2009-12-05 17:17:16 -08:00
Slava Pestov
abe72ae7ae
math.vectors.simd: ensure that set-alien-vector is open-coded in the actual set-nth-unsafe method on SIMD specialized arrays, not just inlined instances
2009-12-05 18:20:57 -05:00
Joe Groff
104c29aabc
simd intrinsic implementation for v*high, v*hs+, vavg, and vsad
2009-12-05 14:52:18 -08:00
Joe Groff
1845915dc6
add v*high, v*hs+, vavg, and vsad operations to math.vectors
2009-12-05 11:32:31 -08:00
Joe Groff
36b0b65320
move sequence instance to individual SIMD types rather than simd-128 parent type so that "hashcode*" doesn't attempt to use the sequence implementation of hashcode on abstract simd-128 objects
2009-12-04 13:22:59 -08:00
Joe Groff
0795c60b83
"norm" doesn't need to be generic, "norm-sq sqrt" always works
2009-11-26 11:15:35 -08:00
Joe Groff
7da80f65e7
trick math.vectors.simd into making nicer quotations
2009-11-25 20:24:09 -08:00
Joe Groff
d344023b1c
update math.vectors.conversion
2009-11-24 20:53:40 -08:00
Joe Groff
c98eb84943
make math.vectors.simd tests pass again
2009-11-24 18:30:12 -08:00
Joe Groff
d94ffe6d78
scalar fallbacks for simd intrinsics
2009-11-23 21:24:55 -08:00
Joe Groff
6794d14652
typo
2009-11-19 12:17:14 -08:00
Joe Groff
152b0d2df5
break simd intrinsics into a separate vocab so they can be intrinsified before the simd methods compile
2009-11-19 11:53:46 -08:00
Joe Groff
736be27162
move methods back into simd functor so the intrinsics can compile
2009-11-19 11:29:15 -08:00
Joe Groff
c07f3212f9
fix effect of set-alien-vector
2009-11-18 22:18:42 -08:00
Joe Groff
a3e4ecfc7d
enable simd intrinsics and fix first-pass compiler errors
2009-11-18 20:32:05 -08:00
Joe Groff
243e5b43ce
fixes for stronger stack checker
2009-11-14 22:25:00 -06:00
Joe Groff
8a8699ac98
backend fixups
2009-11-14 20:59:03 -06:00
Joe Groff
e323071c44
sever lingering dependencies on simd from compiler
2009-11-11 16:08:40 -06:00
Joe Groff
4d54f27cd1
more intrinsic madness
2009-11-05 09:52:57 -06:00
Joe Groff
b98742be30
typos
2009-11-03 21:38:55 -06:00
Joe Groff
9cf3ab3da1
redo math.vectors.simd to use generics for specialization
2009-11-02 14:59:07 -06:00
Doug Coleman
bd13e018dd
memq? -> member-eq?, sorted-memq? -> sorted-member-eq?
2009-10-28 15:02:00 -05:00
Joe Groff
07aa2620c6
add SIMDS: ... ; and SPECIALIZED-ARRAYS: ... ; syntax
2009-10-09 10:43:37 -05:00
Slava Pestov
f395d83379
math.vectors.simd: add fast intrinsic for 'nth', replace broadcast primitive with shuffles
2009-09-29 04:48:11 -05:00
Slava Pestov
48d3f10c81
Fix some load errors in various places
2009-09-28 16:33:39 -05:00
Slava Pestov
9a06e6f424
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
Slava Pestov
dfc9fd071e
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
Slava Pestov
47d8763340
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
Slava Pestov
e77341b90c
math.vectors.simd: redesign to be more flexible, integer SIMD work in progress
2009-09-20 02:08:32 -05:00
Joe Groff
1f04ed01fe
fix more ambiguities
2009-09-17 09:29:23 -05:00
Slava Pestov
82a8ef7beb
math.vectors.simd: remove useless dependency
2009-09-08 13:55:56 -05:00
Slava Pestov
6494e7a53b
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
Slava Pestov
55c449c6e2
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
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
Slava Pestov
bf81cb4259
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
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