Recep Aslantas
44cd0ae4fd
avx: optimize avx mat4 scale and mat4 mul
2024-04-13 00:33:57 +03:00
Recep Aslantas
d75467f93f
avx: implement transpose with AVX
2024-04-13 00:12:14 +03:00
Recep Aslantas
45c1beff51
simd: fix glmm_set1, glmm_splat
2024-04-12 21:53:20 +03:00
Recep Aslantas
14c567d9d9
sse: drop unused macros: glmm_shuff1x, glmm_shuff2
2024-04-11 21:57:46 +03:00
Recep Aslantas
480e1de048
sse: make use of int domain as default behavior if possible ( compiler may ignore it )
...
also use AVX's `_mm_permute_ps`for shuffling single vector
2024-04-11 21:57:16 +03:00
Recep Aslantas
de66f0a67f
glmm, avx: optimize splat macros
2024-04-10 23:49:18 +03:00
Recep Aslantas
68bdec4510
simd: use glmm_set1() to optimize broadcasting single float
2024-04-10 22:52:53 +03:00
Recep Aslantas
62c0448e25
simd, msvc: ensure required definitions are exist on msvc
2024-04-07 22:48:11 +03:00
Recep Aslantas
4f00ce0e52
sse: reduce some instructions in mat4 inv
2024-04-07 22:33:37 +03:00
Recep Aslantas
a7845ffc44
msvc, simd: fix simd headers for _M_ARM64EC
2024-04-07 00:54:29 +03:00
Recep Aslantas
bd941ed7fb
arm, neon: fix neon support on GCC ARM
2024-04-06 14:23:36 +03:00
Recep Aslantas
87350f809b
msvc bug: dont align types due to "ARM32 = C2719: formal parameter with requested alignment of 16 won't be aligned." on ARM32/MSVC
...
until a good solution.
2024-04-06 14:11:46 +03:00
Recep Aslantas
f50a7a7d00
arm, neon: improve glm_mat4_inv_neon
2024-04-06 14:10:13 +03:00
Recep Aslantas
b3308af146
arm: fix glmm_vhadd on ARM32
2024-04-06 14:09:52 +03:00
Recep Aslantas
a94f839d6d
Merge branch 'master' into optimize-inv
2024-04-04 00:31:08 +03:00
Recep Aslantas
0ff0e8948f
Update mat4.h
2024-04-03 00:05:42 +03:00
Recep Aslantas
5b772d0eb4
neon: mat4_inv, reduce 1mul for two extra 2xor
2024-04-03 00:03:55 +03:00
Recep Aslantas
c528ca1095
neon: mat4_inv remastered
2024-04-02 13:21:49 +03:00
Recep Aslantas
f0e09776d7
arm, neon: optimize glmm_vhadd and add glmm_vdot
2024-04-02 02:36:16 +03:00
Recep Aslantas
da4224ba32
now working on v0.9.5
2024-04-02 00:52:54 +03:00
Recep Aslantas
1796cc5ce2
Merge pull request #412 from recp/sse_only
...
separate SSE and SSE2
2024-04-01 17:52:06 +03:00
Recep Aslantas
568634a79e
tests: dont test isinf == true on fast math
2024-04-01 16:48:58 +03:00
Recep Aslantas
c9c7941a72
tests: dont test isnan on fast math
2024-04-01 14:10:38 +03:00
Recep Aslantas
4e929a81c2
Update x86.h
2024-04-01 13:26:08 +03:00
Recep Aslantas
1d09c41e18
make xor enable in SSEonly + fast math
2024-04-01 13:23:52 +03:00
Recep Aslantas
d8e933b5b1
Update test_project.h
2024-04-01 13:23:22 +03:00
Recep Aslantas
8b15fd51ba
common way to identify CGLM_FAST_MATH
2024-04-01 13:23:17 +03:00
Recep Aslantas
9ce0a3b625
tests: disable isnan check for min/max which may not work correctly especially in fast-math environment
2024-04-01 13:22:51 +03:00
Recep Aslantas
00d2e8a4cf
suppress warnings
2024-04-01 01:46:25 +03:00
Recep Aslantas
35a12ed033
make SSE2-only features not available in SSE
2024-04-01 00:52:59 +03:00
Recep Aslantas
9ad7dd3fbc
Merge pull request #409 from EasyIP2023/feature/expand-content-width
...
docs: expand wy-nav-content width to edge of screen
2024-03-31 23:44:40 +03:00
Recep Aslantas
28142b5912
Merge pull request #411 from EasyIP2023/bugfix/mat4x3-multiplication
...
mat4x3: fix multiplication functions
2024-03-31 23:44:00 +03:00
Recep Aslantas
f07d75c680
Merge pull request #410 from EasyIP2023/bugfix/mat4x2-multiplication
...
mat4x2: fix multiplication functions
2024-03-31 23:43:11 +03:00
Recep Aslantas
17d8b83a38
Merge pull request #408 from EasyIP2023/bugfix/mat3x4-multiplication
...
mat3x4: fix multiplication functions
2024-03-31 23:41:40 +03:00
Recep Aslantas
32a477ef07
separate SSE and SSE2
2024-03-31 23:35:15 +03:00
Recep Aslantas
e3ed9834a1
Update mat2x3.h
2024-03-31 23:06:50 +03:00
Recep Aslantas
8396bbf0b3
coding style
2024-03-31 23:06:30 +03:00
Vincent Davis Jr
013ac5dd07
docs: mat4x3 account for latest mulitplication changes
...
This also includes tables to explain how
mat4x3, column vectors, and row vectors are
represented. Also includes how resulting
matrix or vector is formed.
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 14:43:31 -04:00
Vincent Davis Jr
a0e3d3766f
mat4x3: fix multiplication functions
...
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 14:43:30 -04:00
Vincent Davis Jr
fc7f0e13fd
docs: mat3x4 account for latest mulitplication changes
...
This also includes tables to explain how
mat3x4, column vectors, and row vectors are
represented. Also includes how resulting
matrix or vector is formed.
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 13:43:41 -04:00
Vincent Davis Jr
1340b5d512
mat3x4: fix multiplication functions
...
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 13:43:39 -04:00
Vincent Davis Jr
85165dd3e3
docs: mat4x2 account for latest mulitplication changes
...
This also includes tables to explain how
mat4x2, column vectors, and row vectors are
represented. Also includes how resulting
matrix or vector is formed.
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 13:40:43 -04:00
Vincent Davis Jr
3445f93fbc
mat4x2: fix multiplication functions
...
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 13:40:41 -04:00
Vincent Davis Jr
d2642eb206
docs: expand wy-nav-content width to edge of screen
...
RTD theme's default is 800px as max width for the content,
but we have tables with tons of columns, which need the
full width of the view-port.
Comment from yocto project theme_overrides.css
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 12:22:13 -04:00
Recep Aslantas
929963c6eb
Merge pull request #407 from EasyIP2023/bugfix/mat3x2-multiplication
...
mat3x2: fix multiplication functions
2024-03-31 13:40:58 +03:00
Recep Aslantas
25b33fab6b
Merge pull request #405 from EasyIP2023/bugfix/mat2x4-mutli
...
mat2x4: fix multiplication functions
2024-03-31 13:40:45 +03:00
Recep Aslantas
c9adbaabd7
Merge pull request #403 from EasyIP2023/bugfix/mat2x3-multiplication
...
mat2x3: fix multiplication functions
2024-03-31 13:36:39 +03:00
Recep Aslantas
b22e8230d0
Merge pull request #406 from recp/I_macro
...
dont use I macro defined in standard
2024-03-31 13:30:42 +03:00
Recep Aslantas
bf4c5b4e26
dont use I macro defined in standard
2024-03-31 13:24:50 +03:00
Vincent Davis Jr
54dfbc5a28
docs: mat2x4 account for latest mulitplication changes
...
This also includes tables to explain how
mat2x4, column vectors, and row vectors are
represented. Also includes how resulting
matrix or vector is formed.
Signed-off-by: Vincent Davis Jr <vince@underview.tech >
2024-03-31 00:44:58 -04:00
