OpenBLAS/benchmark
Bart Oldeman 9959a60873 Benchmarks: align malloc'ed buffers.
Benchmarks should allocate with cacheline (often 64 bytes) alignment
to avoid unreliable timings. This technique, storing the offset in the
byte before the pointer, doesn't require C11's aligned_alloc for
compatibility with older compilers.

For example, Glibc's x86_64 malloc returns 16-byte aligned buffers, which is
not sufficient for AVX/AVX2 (32-byte preferred) or AVX512 (64-byte).
2022-10-20 13:28:20 -04:00
..
scripts disable NaN checks before BLAS calls dgemm.R 2019-01-16 11:54:22 +02:00
Make_exe.sh bugfixes, to build benchmarks with mingw on Windows OS 2015-05-29 12:56:22 +02:00
Makefile Change "HALF" and "sh" to "BFLOAT16" and "sb" 2020-10-12 00:08:29 +02:00
amax.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
amin.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
asum.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
axpby.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
axpy.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
bench.h Benchmarks: align malloc'ed buffers. 2022-10-20 13:28:20 -04:00
cholesky.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
copy.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
cula_wrapper.c Import GotoBLAS2 1.13 BSD version codes. 2011-01-24 14:54:24 +00:00
dot.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
geev.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
gemm.c Refactor the performance measurement system 2020-10-23 10:32:03 +08:00
gemm3m.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
gemv.c Fix flipped indices in benchmark for gemv 2021-11-03 12:45:09 +01:00
ger.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
gesv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
getri.c Handle OPENBLAS_LOOPS and OPENBLAS_TEST options 2021-07-01 17:38:45 +02:00
hbmv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
hemm.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
hemv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
her.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
her2.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
her2k.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
herk.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
hpmv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
iamax.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
iamin.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
imax.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
imin.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
linpack.c Handle OPENBLAS_LOOPS for more stable results 2021-07-01 17:39:37 +02:00
max.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
min.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
nrm2.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
plot-filter.sh added blas level1 copy benchmark 2015-05-20 11:05:00 +02:00
plot-header added a sample plot-filter scripts and a header file for gnuplot 2014-07-21 14:50:24 +02:00
potrf.c Add OPENBLAS_LOOPS support to potrf/potrs/potri benchmark 2021-06-26 23:46:00 +02:00
rot.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
rotm.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
scal.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
smallscaling.c added bugfixes for some make files and smallscaling.c 2016-04-21 12:54:32 +02:00
spmv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
spr.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
spr2.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
swap.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
symm.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
symv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
syr.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
syr2.c Handle OPENBLAS_LOOPS in SYR2 benchmark 2021-07-10 21:27:53 +02:00
syr2k.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
syrk.c Handle OPENBLAS_LOOP 2021-07-04 16:59:43 +02:00
tplot-header added plot-header to compare multithreading 2014-09-02 14:11:42 +02:00
tpmv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
tpsv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
trmm.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
trmv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
trsm.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
trsv.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
zdot-intel.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00
zdot.c Refractoring remaining benchmark cases. 2020-10-26 10:25:05 +08:00