f9daebba0a
checked in bugfixes for ARM
2014-02-16 11:45:47 +01:00
9a557e90da
Refs #340 . Fixed SEGFAULT bug of dgemv_n on OSX.
2014-02-15 23:23:15 +08:00
2d557eb1e0
Fixed computational error of dgemv_n.
2014-02-04 21:47:51 +08:00
05bb391c3a
Refs #330 . Fixed the compatible issue with clang on Mac OSX.
2013-12-16 20:31:17 +08:00
9b5be29886
Refs #310 . Fixed Segfault bug on nehalem when Julia calling dgeqrt3 on OSX.
...
Please also check JuliaLang/julia#4099
Julia test script:
A=rand(256, 256)
qrfact(A)
I found this was a bug in kernel/x86_64/dgemm_ncopy_8.S.
However, I cannot use gdb with julia. Thus, this is a walkaround fix.
2013-12-12 23:23:04 +08:00
53eaf41901
added support for HASWELL
2013-12-02 13:17:51 +01:00
9423f980f6
modified trsm kernel
2013-12-02 10:08:14 +01:00
c6156b2ef2
added trsm kernels from origin
2013-12-01 22:39:39 +01:00
034a5b2083
modified zsymv
2013-12-01 21:07:49 +01:00
27d4234d4d
merged symv
2013-12-01 20:56:02 +01:00
402d6e91db
Merge remote branch 'origin/develop' into armv7
2013-12-01 18:18:40 +01:00
b3254eecaf
Merge remote branch 'origin/haswell' into develop
2013-12-01 18:09:12 +01:00
d910404f00
Merge remote branch 'origin/piledriver' into develop
2013-12-01 18:06:51 +01:00
ffe70b1fdc
modified Makefile.L3
2013-12-01 17:58:46 +01:00
0b6e13b689
Merge remote branch 'origin/develop' into haswell
2013-12-01 13:38:11 +01:00
e09dc279a2
Merge remote branch 'origin/develop' into piledriver
2013-12-01 13:33:18 +01:00
4be4db590c
Merge remote branch 'origin/develop' into armv7
2013-12-01 13:16:41 +01:00
5c648a8984
Merge remote branch 'origin/develop' into haswell
2013-12-01 11:25:33 +01:00
c44dc4dd3c
Merge remote branch 'origin/develop' into piledriver
2013-12-01 11:06:36 +01:00
9d3fae15a8
Merge branch 'develop' into armv7
2013-12-01 10:12:07 +01:00
2d3c884294
added complex gemv kernels for ARMV6 and ARMV7
2013-11-29 17:06:33 +01:00
d54a061713
optimized gemv_n_vfp.S
2013-11-28 17:40:21 +01:00
86afb47e83
added optimized ctrmm kernel for ARMV6
2013-11-28 14:35:07 +01:00
42a4dff056
added optimized ztrmm kernel for ARMV6
2013-11-28 13:41:06 +01:00
5bc322a66c
optimized strmm kernel for ARMV6
2013-11-28 12:45:38 +01:00
dec7ad0dfd
optimized dtrmm kernel for ARMV7
2013-11-28 12:32:12 +01:00
274304bd03
add optimized cgemm kernel for ARMV6
2013-11-28 11:54:38 +01:00
5007a534c4
optimized zgemm kernel for ARMV6
2013-11-28 10:04:43 +01:00
a537d7d8d7
optimized zgemm_kernel_2x2_vfp.S
2013-11-28 08:33:44 +01:00
b42145834f
optimized sgemm kernel for ARMV6
2013-11-28 08:08:08 +01:00
3d5e792c72
optimized sgemm kernel for ARMV6
2013-11-27 18:38:32 +01:00
a9bd12da2c
optimized dgemm kernel for ARMV6
2013-11-27 17:37:38 +01:00
697e198e8a
added zgemm_kernel for ARMV6
2013-11-27 16:15:06 +01:00
36b0f7fe1d
added optimized gemv_t kernel for ARMV6
2013-11-25 19:31:27 +01:00
d2b20c5c51
add optimized axpy kernel
2013-11-25 12:25:58 +01:00
fe5f46c330
added experimental support for ARMV8
2013-11-24 15:47:00 +01:00
25c6050593
add single and double precision gemv_n kernel for ARMV6
2013-11-24 12:03:28 +01:00
12e02a00e0
added ncopy kernels for ARMV6
2013-11-24 08:46:47 +01:00
29a3196f56
added optimized sgemm and strmm kernel for ARMV6
2013-11-23 18:09:41 +01:00
8776a73773
added optimized dgemm and dtrmm kernel for ARMV6
2013-11-23 16:24:52 +01:00
7e84acd3e8
fixed bug in SAVE macros, that are not found by any test routine
2013-11-23 14:35:19 +01:00
33d3ab6e09
small optimizations for zgemv kernels
2013-11-23 12:35:31 +01:00
9a0f978929
added nrm2 kernel for ARMV6
2013-11-22 17:21:10 +01:00
7f210587f0
renamed some ncopy and tcopy files
2013-11-22 00:20:25 +01:00
9f0a3a35b3
removed obsolete file sdot_vfpv3.S
2013-11-21 23:42:54 +01:00
dbae93110b
added sdot_vfp.S
2013-11-21 23:34:51 +01:00
19cd5c64a2
renamed swap_vfpv3.S to swap_vfp.S
2013-11-21 23:19:32 +01:00
9adf87495e
renamed some dot kernels
2013-11-21 23:07:51 +01:00
440db4cdda
delete rot_vfpv3.S
2013-11-21 22:52:24 +01:00
cd93cae5a7
renamed rot_vfpv3.S to rot_vfp.S
2013-11-21 22:49:28 +01:00
8565afb3c2
renamed asum_vfpv3.S to asum_vfp.S
2013-11-21 22:26:27 +01:00
5bf7cf8d67
renamed scal_vfpv3.S to scal_vfp.S
2013-11-21 22:03:36 +01:00
29a005c635
renamed iamax assembler kernel
2013-11-21 21:12:33 +01:00
f1be3a168a
renamed some BLAS kernels, which are compatible to ARMV6
2013-11-21 20:48:57 +01:00
410afda9b4
added cpu detection and target ARMV6, used in raspberry pi
2013-11-21 20:18:51 +01:00
bf04544902
added gemv_n kernel for single and double precision
2013-11-19 15:07:20 +01:00
86283c0be1
added gemv_t kernel for single and double precision
2013-11-19 09:55:54 +01:00
f27cabfd08
added nrm2 kernel for all precisions
2013-11-16 16:17:17 +01:00
23dd474cd0
added rot kernel for all precisions
2013-11-15 14:08:57 +01:00
f1b452e160
added scal kernel for all precisions
2013-11-15 11:56:43 +01:00
3dabd7e6e6
added swap-kernel for all precisions
2013-11-14 19:06:19 +01:00
6f4a0ebe38
added max- und min-kernels for all precisions
2013-11-14 13:52:47 +01:00
f1db386211
changes for compatibility with Pathscale compiler
2013-11-13 17:59:11 +01:00
6da558d2ab
changes for compatibility with Pathscale compiler
2013-11-13 17:39:13 +01:00
f750103336
small optimizations on dot-kernels
2013-11-11 15:47:56 +01:00
00f33c0134
added asum_kernel for all precisions and complex
2013-11-11 14:20:59 +01:00
5b36cc0f47
added blas level1 dot kernels for complex and double complex
2013-11-08 09:08:11 +01:00
c8f1aeb154
added optimized blas level1 dot kernels for single and double precision
2013-11-07 17:22:03 +01:00
8fa93be06e
added optimized blas level1 copy kernels
2013-11-07 17:18:56 +01:00
1e8128f41c
added cgemm_tcopy_2_vfpv3.S and zgemm_tcopy_2_vfpv3.S
2013-11-07 17:15:50 +01:00
2f5fdd2000
Refs #314 . Fixed clang compiling bug on OSX.
2013-11-07 08:12:03 +08:00
80a2e901b1
added dgemm_tcopy_4_vfpv3.S and sgemm_tcopy_4_vfpv3.S
2013-11-06 20:01:18 +01:00
ac50bccbd2
added cgemm_ncopy_2_vfpv3.S and made assembler labels unique
2013-11-05 20:21:35 +01:00
82015beaef
added zgemm_ncopy_2_vfpv3.S and made assembler labels unique
2013-11-05 19:31:22 +01:00
6216ab8a7e
removed obsolete gemm_kernels from haswell branch
2013-11-04 08:33:04 +01:00
370e3834a9
added missing file kernel/arm/Makefile
2013-11-03 11:54:39 +01:00
e31186efd4
deleted obsolete dgemm_kernel and dtrmm_kernel
2013-11-02 13:12:21 +01:00
2b801a00a5
small optimizations on sgemm_kernel for ARMV7
2013-11-02 13:06:11 +01:00
b3eab8fcb7
minor optimizations on zgemm_kernel for ARMV7
2013-11-02 09:43:53 +01:00
02bc36ac79
added sgemm_ncopy routine and made some improvements on cgemm_kernel for ARMV7
2013-11-01 18:22:27 +01:00
5118a7f4d1
small optimizations on dgemm_kernel for Piledriver
2013-10-31 11:53:26 +01:00
e172b70ea2
added cgemm_kernel for Piledriver
2013-10-31 08:38:17 +01:00
1cf4b974b2
added zgemm_kernel for Piledriver
2013-10-30 09:12:17 +01:00
7bccff1512
added sgemm_kernel for PILEDRIVER
2013-10-29 22:53:04 +01:00
afe44b0241
tests and code cleanup of gemm_kernels for HASWELL
2013-10-28 14:23:48 +01:00
a77c71eaf5
added highly optimized dgemm_kernel for HASWELL
2013-10-28 10:23:47 +01:00
fe8c5666f9
optimized dgemm_kernel for HASWELL
2013-10-20 16:52:26 +02:00
f6b50057e2
corrected and testet FMA3 Code
2013-10-19 10:52:20 +02:00
2840d56aeb
added dgemm_kernel for Piledriver
2013-10-19 09:47:15 +02:00
85484a42df
added kernels for cgemm, ctrmm, zgemm and ztrmm
2013-10-16 18:00:41 +02:00
3983011f0b
added sgemm- and strmm_kernel
2013-10-14 08:22:27 +02:00
2a1515c9dd
added dgemm_ncopy_4_vfpv3.S
2013-10-12 16:48:29 +02:00
31f51e78bc
minor optimizations on dgemm_kernel
2013-10-12 09:42:18 +02:00
beffee7d91
Fixed buffer overflow bug in kernel/x86_64/dgemv_t.S file.
2013-10-11 03:20:20 +08:00
e0b968c3a7
Changed kernels for dgemm and dtrmm
2013-10-05 12:59:44 +02:00
1c63180bb6
updated dgemm_kernel_8x2_vfpv3.S
2013-09-30 17:31:23 +02:00
4a474ea7dc
changed dgemm_kernel to use fused multiply add
2013-09-29 17:46:23 +02:00
69ce737cc5
modified Makefile.L3 for ARM
2013-09-28 19:13:47 +02:00
70411af888
initial checkin of kernel/arm
2013-09-28 19:02:25 +02:00
6c4a7d0828
Import AMD Piledriver DGEMM kernel generated by AUGEM.
...
So far, this kernel doesn't deal with edge.
AUGEM: Automatically Generate High Performance Dense Linear Algebra
Kernels on x86 CPUs.
Qian Wang, Xianyi Zhang, Yunquan Zhang, and Qing Yi. In the
International Conference for High Performance Computing, Networking,
Storage and Analysis (SC'13). Denver, CO. Nov, 2013.
2013-08-25 10:16:01 -03:00
wernsaar
Zhang Xianyi
wangqian