4de545aa7d
address minor warnings from gcc7
2019-09-07 10:21:08 +03:00
bf1430f7d7
Merge pull request #2208 from martin-frbg/munmap-debug
...
Provide more information on mmap/munmap failure
2019-08-09 07:55:35 +02:00
1776ad82c0
Add files via upload
2019-08-09 00:08:11 +02:00
4e2f81cfa1
Provide more information on mmap/munmap failure
...
for #2207
2019-08-08 23:15:35 +02:00
3d36c45116
Add CPUID identification of Intel Ice Lake
2019-08-01 22:52:35 +02:00
21d05a4835
Merge pull request #2140 from martin-frbg/pgi19
...
Do not try ancient PGI hacks with recent versions of that compiler
2019-05-26 12:39:20 +02:00
1778fd4219
Do not try ancient PGI hacks with recent versions of that compiler
...
should fix #2139
2019-05-22 13:48:27 +02:00
86dda5c2fa
Add option USE_LOCKING for SMP-like locking in USE_THREAD=0 builds
2019-05-15 23:21:20 +02:00
5cabda79d0
Merge pull request #2117 from martin-frbg/issue2114
...
Fix errors in cpu affinity setup with glibc 2.6
2019-05-07 18:18:16 +02:00
a6a8cc2b7f
Fix errors in cpu enumeration with glibc 2.6
...
for #2114
2019-05-07 13:34:52 +02:00
a387a23518
Merge pull request #2101 from luzpaz/misc-typos
...
Misc. typo fixes in comments and documentation
2019-05-04 22:28:29 +02:00
b43c8382c8
Correct argument of CPU_ISSET for glibc <2.5
...
fixes #2104
2019-05-01 10:46:46 +02:00
daf2fec12d
Misc. typo fixes
...
Found via `codespell -q 3 -w -L ith,als,dum,nd,amin,nto,wis,ba -S ./relapack,./kernel,./lapack-netlib`
2019-04-29 17:03:56 -04:00
40e53e52d6
snprintf define consolidated to common.h
2019-04-22 17:01:34 -07:00
bcdf1d4917
Add in runtime CPU detection for POWER.
2019-04-09 14:20:16 +10:00
8ba9e2a61a
Also call CloseHandle on each thread, as well as on the event so as to not leak thread handles.
2019-03-19 11:21:44 +01:00
4ad694eda1
Fix for #2063 : The DllMain used in Cygwin did not run the thread memory
...
pool cleanup upon THREAD_DETACH which is needed when compiled with
USE_TLS=1.
2019-03-19 09:26:50 +01:00
3ce28fb81a
Merge pull request #2055 from martin-frbg/atomid
...
Add CPUID data for Intel Denverton (as Nehalem)
2019-03-12 22:57:07 +01:00
04f2226ea6
Add Intel Denverton
2019-03-12 16:09:55 +01:00
4741ce803b
Merge pull request #2045 from martin-frbg/2033-3
...
Do not compile in AVX512 check if AVX support is disabled
2019-03-06 22:40:26 +01:00
11cfd0bd75
Do not compile in AVX512 check if AVX support is disabled
...
xgetbv is function depends on NO_AVX being undefined - we could change that too, but that combo is unlikely to work anyway
2019-03-05 16:04:25 +01:00
d7b2c53c0b
Merge pull request #2039 from brada4/meminit
...
Address warning in memory.c
2019-03-05 12:11:15 +01:00
10d841d8b9
Merge pull request #2026 from martin-frbg/trmv_threads
...
Correct range limiting in trmv_thread and re-enable TRMV multithreading
2019-03-04 15:08:31 +01:00
6c83b878f6
Merge pull request #2040 from martin-frbg/locks2002
...
Restore locking optimizations for OpenMP case
2019-03-04 15:07:14 +01:00
af480b02a4
Restore locking optimizations for OpenMP case
...
restore another accidentally dropped part of #1468 that was missed in #2004 to address performance regression reported in #1461
2019-03-03 14:17:07 +01:00
e4a79be6bb
address warning introed with #1814 et al
2019-03-03 09:05:11 +02:00
45333d5793
Fix error introduced during cleanup
2019-02-19 22:16:33 +01:00
78d9910236
Correct range_n limiting
...
same bug as seen in #1388 , somehow missed in corresponding PR #1389
2019-02-19 20:59:48 +01:00
03a2bf2602
Fix potential memory leak in cpu enumeration on Linux ( #2008 )
...
* Fix potential memory leak in cpu enumeration with glibc
An early return after a failed call to sched_getaffinity would leak the previously allocated cpu_set_t. Wrong calculation of the size argument in that call increased the likelyhood of that failure. Fixes #2003
2019-02-10 23:24:45 +01:00
69edc5bbe7
Restore dropped patches in the non-TLS branch of memory.c ( #2004 )
...
* Restore dropped patches in the non-TLS branch of memory.c
As discovered in #2002 , the reintroduction of the "original" non-TLS version of memory.c as an alternate branch had inadvertently used ba1f91fa8002e2#1450 , #1468 , #1501 , #1504 and #1520 .
2019-02-07 20:06:13 +01:00
29dc72889f
Add support for Hygon Dhyana
2019-01-16 14:25:19 +08:00
dbc9a060ef
Fix missing braces in support_av() call
2019-01-14 22:41:31 +01:00
21c0f2af7b
Merge pull request #1957 from martin-frbg/issue1954
...
Move TLS key deletion to openblas_quit
2019-01-10 12:04:08 +01:00
ad2c386d6a
Move TLS key deletion to openblas_quit
...
fixes #1954 (as suggested by thrasibule in that issue)
2019-01-10 00:32:50 +01:00
31ed19e8b9
Add message for SkylakeX and KNL fallbacks to Haswell
2019-01-05 19:41:13 +01:00
e1574fa2b4
Add xcr0 (os support) check
2019-01-05 18:08:02 +01:00
ae1d1f74f7
Query AVX2 and AVX512 capability for runtime cpu selection
2019-01-05 16:55:33 +01:00
8643521127
Merge pull request #1943 from martin-frbg/issue1748
...
Re-enable loop unrolling in trmv and remove the scary warning
2018-12-30 20:07:01 +01:00
5a720cf9ca
Re-enable loop unrolling in trmv and remove the scary warning
...
fixes #1748 as that half of the fix for #1332 appears to have been an overreaction on my part.
2018-12-30 15:22:37 +01:00
ccd5945d38
Merge pull request #1942 from martin-frbg/issue1720
...
Delete the pthread key on cleanup in TLS mode
2018-12-30 14:47:05 +01:00
bba1e67269
Delete the pthread key on cleanup in TLS mode
...
to avoid a crash when OpenBLAS was loaded via dlopen and libc tries to clean up the leaked TLS after dlclose
Fixes #1720
2018-12-29 21:59:31 +01:00
f343ed65b5
Avoid taking the root of a negative number
...
Fixes #1924 where numpy 1.17+ would report the (transient) FE_INVALID exception raised for the domain error.
2018-12-22 22:30:29 +01:00
0bf6d74e5f
Fix typo in previous commit for arm dynamic arch
2018-12-07 19:37:33 +01:00
2b355592e3
Make sure to use the arm version of dynamic.c in ARM64 DYNAMIC_ARCH
...
cf. #1908
2018-12-07 16:25:55 +01:00
2601cd58ab
remove surplus locking code , only enabled w x86, disabled or never enabled on all others
2018-11-30 11:38:19 +01:00
97d7298973
call it OpenBLAS not just version
2018-11-29 11:52:08 +01:00
de0d0ed52f
Improve formatting of config output
2018-11-29 11:28:19 +01:00
816775e309
Add version information to openblas_get_config output
2018-11-29 00:06:44 +01:00
f72fdf525c
Merge pull request #1875 from martin-frbg/issue1851
...
Serialize accesses to parallelized level3 functions from multiple cal…
2018-11-25 20:53:46 +01:00
113cb00b95
fix missing parenthesis
2018-11-19 21:01:36 +01:00
5192651706
Add CriticalSection handling instead of mutexes for Windows
2018-11-19 17:58:22 +01:00
2e6fae2aad
Serialize accesses to parallelized level3 functions from multiple callers
...
for #1851
2018-11-19 14:02:50 +01:00
368d14f8c8
Fix harmless typo
...
fixes #1872
2018-11-16 14:58:28 +01:00
0427277cef
Allow optimization for small m, large n only if it can be made threadsafe
...
otherwise the introduction of a static array in 8e5a108#532 breaks concurrent calls from multiple threads as seen in #1844
2018-11-10 15:45:54 +01:00
5b708e5eb1
sgemm/dgemm: add a way for an arch kernel to specify prefered sizes
...
The current gemm threading code can make very unfortunate choices, for
example on my 10 core system a 1024x1024x1024 matrix multiply ends up
chunking into blocks of 102... which is not a vector friendly size
and performance ends up horrible.
this patch adds a helper define where an architecture can specify
a preference for size multiples.
This is different from existing defines that are minimum sizes and such.
The performance increase with this patch for the 1024x1024x1024 sgemm
is 2.3x (!!)
2018-11-01 01:43:20 +00:00
f5595d0262
Merge pull request #1843 from martin-frbg/aix_numprocs
...
Add get_num_procs implementation for AIX
2018-10-31 21:25:15 +01:00
326d394a0f
Add get_num_procs implementation for AIX
...
(and copy HAIKU implementation to the non-TLS version of the code as well)
2018-10-31 18:38:22 +01:00
38cf5d9364
ensure that threading has been initialized in the first place before calling openblas_set_num_threads
2018-10-28 21:16:52 +00:00
d5aeff636f
ARM64: Enable DYNAMIC_ARCH
...
Enable DYNAMIC_ARCH feature on ARM64. This patch uses the cpuid
feature in linux kernel to detect the core type at runtime
(https://www.kernel.org/doc/Documentation/arm64/cpu-feature-registers.txt ).
If this feature is missing in kernel, then the user should use the
OPENBLAS_CORETYPE env variable to select the desired core type.
2018-10-22 01:49:35 -07:00
d50abc8903
ARM64: Move parameters from parameter.c to param.h
...
Remove the runtime setting of P, Q, R parameters for
targets ARMV8, THUNDERX2T99. Instead set them as constants
in param.h at compile time.
2018-10-22 01:45:51 -07:00
21f46a1cf2
ARM64: Use THUNDERX2T99 Neon Kernels for ARMV8
...
Currently the generic ARMV8 target uses C implementations
for many routines. Replace these with the neon implementations
written for THUNDERX2T99 target which are upto 6x faster for
certain routines.
2018-10-17 10:44:37 -07:00
3439158dea
address #1782 2nd loop
2018-10-03 21:20:50 +02:00
28aa94bf4b
Include thread numbers in failure message from blas_thread_init
...
to aid in debugging cases like #1767
2018-09-22 14:00:15 +02:00
1ad1e79062
Catch inadvertent USE_TLS=0 declaration
...
for #1766
2018-09-19 18:03:43 +02:00
b402626509
Do not use the new TLS code for non-threaded builds even if USE_TLS is set
...
Workaround for #1761 as that exposed a problem in the new code (which was intended to speed up multithreaded code only anyway).
2018-09-16 12:43:36 +02:00
b55690a659
typo fix
2018-08-26 11:31:07 +02:00
b902a40986
Rewrite glibc version check
2018-08-26 11:18:02 +02:00
5991d1a6cd
Update memory.c
2018-08-25 22:12:40 +02:00
b1b743f434
Merge branch 'develop' into interim033
2018-08-25 19:45:19 +02:00
fd42ca462d
Combo of default pre-0.3.1 memory.c and band-aided version of PR1739
2018-08-25 19:35:16 +02:00
6463bffd59
Haiku supporting patches
2018-08-02 20:49:14 +02:00
8ef7d4fb54
Merge pull request #1706 from oon3m0oo/develop
...
Fix #1705 where we incorrectly calculate page locations.
2018-08-02 18:53:34 +02:00
6400868e55
Fix #1705 where we incorrectly calculate page locations.
...
Since we now use an allocation size that isn't a multiple of PAGESIZE, finding
the pages for run_bench wasn't terminating properly. Now we detect if we've
found enough pages for the allocation and terminate the loop.
2018-08-02 16:21:19 +01:00
66fcdd5be8
Merge pull request #1695 from martin-frbg/issue1692
...
Unset memory table entry, not just the local pointer to it on shutdown
2018-07-22 16:34:09 +02:00
43ac839c16
Unset memory table entry, not just the temporary pointer to it on shutdown
...
to fix crash with multiple instances of OpenBLAS, #1692
2018-07-22 09:19:19 +02:00
7ba5936ecd
Merge pull request #1688 from martin-frbg/issue1673
...
Temporarily disable special handling of OPENMP thread memory allocation
2018-07-19 19:03:45 +02:00
b14f44d2ad
Temporarily disable special handling of OPENMP thread memory allocation
...
for issue #1673
2018-07-19 08:57:56 +02:00
36aea5ce2d
Merge pull request #1680 from martin-frbg/snprint
...
Fix wrong redefinitions of snprintf for older MSVC
2018-07-12 14:05:13 +02:00
571e9de2ac
Fix definition of snprintf for MSVC
...
MS _snprintf_s takes an additional argument for the size of the buffer, so is not a direct replacement (utest/ctest.h from which I copied was wrong)
2018-07-12 11:42:25 +02:00
448ed15115
Merge pull request #1678 from martin-frbg/issue1677
...
Define snprintf for older versions of MSVC
2018-07-12 09:21:34 +02:00
045fb5ea2c
Define snprintf for older versions of MSVC
...
for #1677
2018-07-12 07:30:58 +02:00
4dd70d98d7
Merge pull request #1667 from xianyi/revert-1642-develop
...
Revert "Rewrite &= -> = and simplify the initial blocking phase."
2018-07-04 08:27:21 +02:00
504310eeb9
Merge pull request #1665 from martin-frbg/cpuid-ryzen2
...
Add cpuid for AMD Ryzen 2
2018-07-04 08:19:40 +02:00
ea1f39518f
Merge pull request #1663 from martin-frbg/issue1641
...
Double MAX_ALLOCATING_THREADS to fix segfaults with Go and Octave
2018-07-04 08:19:11 +02:00
5f2a3c05cd
Revert "Rewrite &= -> = and simplify the initial blocking phase."
2018-07-03 21:42:28 +02:00
d0ec4325cf
Add cpuid for AMD Ryzen 2
2018-07-03 21:03:24 +02:00
a49203b48c
Double MAX_ALLOCATING_THREADS to fix segfaults with Go and Octave
...
for #1641
2018-07-03 17:35:54 +02:00
9d15a3bd16
Fix typo that broke compilation with DYNAMIC_ARCH and NO_AVX2
...
fixes 1659
2018-07-02 14:40:41 +02:00
3d3c19717c
Merge pull request #1655 from martin-frbg/issue1641
...
Fix apparent off-by-one error in calculation of MAX_ALLOCATING_THREADS
2018-07-01 08:41:22 +02:00
4e9c34018e
Fix apparent off-by-one error in calculation of MAX_ALLOCATING_THREADS
...
fixes #1641
2018-06-30 23:57:50 +02:00
750162a05f
Try gradual fallback for cores not in the dynamic core list
2018-06-25 21:02:31 +02:00
e6d93f20f1
Merge pull request #2 from martin-frbg/develop
...
merge develop
2018-06-25 20:48:10 +02:00
0144068537
Rewrite &= -> = and simplify the initial blocking phase.
2018-06-25 15:08:55 +01:00
1833a67071
Add support for a user-defined list of dynamic targets
2018-06-23 19:42:15 +02:00
28c28ed275
Fix data races reported by TSAN.
2018-06-21 16:41:02 +01:00
a399d00425
Further improvements to memory.c. ( #1625 )
...
- Compiler TLS is now used only used when the compiler supports it
- If compiler TLS is unsupported, we use platform-specific TLS
- Only one variable (an index) is now in TLS
- We only access TLS once per alloc, and never when freeing
- Allocation / release info is now stored within the allocation itself, by
over-allocating; this saves having external structures do the bookkeeping, and
reduces some of the redundant data that was being stored (such as addresses)
- We never hit the alloc lock when not using SMP or when using OpenMP (that was
my fault)
- Now that there are fewer tracking structures I think this is a bit easier to
read than before
2018-06-20 22:04:03 +02:00
5a6a2bed9a
Merge pull request #1623 from fenrus75/fast-thread
...
Initialize only the required subset of the jobs array, fix barriers and improve switch ratio on SkylakeX and Haswell. For issue #1622
2018-06-18 09:02:40 +02:00
2d8cc7193a
Support upcoming Intel Cannon Lake CPUs as Skylake X ( #1621 )
...
* Support upcoming Cannon Lake as Skylake X
2018-06-17 23:38:14 +02:00
73de17664d
Add missing barriers in gemm scheduler
...
a few places in the gemm scheduler code were missing barriers;
the code likely worked OK due to heavy use of volatile / _Atomic
but there's no reason to get this incorrect
2018-06-17 17:50:43 +00:00
d148ec4ea1
Don't use _Atomic for jobs sometimes...
...
The use of _Atomic leads to really bad code generation in the compiler
(on x86, you get 2 "mfence" memory barriers around each access with gcc8, despite
x86 being ordered and cache coherent). But there's a fallback in the code that
just uses volatile which is more than plenty in practice.
If we're nervous about cross thread synchronization for these variables, we should
make the YIELD function be a compiler/memory barrier instead.
performance before (after last commit)
Matrix SGEMM cycles MPC DGEMM cycles MPC
48 x 48 10630.0 10.6 0.7% 18112.8 6.2 -0.7%
64 x 64 20374.8 13.0 1.9% 40487.0 6.5 0.4%
65 x 65 141955.2 1.9 -428.3% 146708.8 1.9 -179.2%
80 x 80 178921.1 2.9 -369.6% 186032.7 2.8 -156.6%
96 x 96 205436.2 4.3 -233.4% 224513.1 3.9 -97.0%
112 x 112 244408.2 5.8 -162.7% 262158.7 5.4 -47.1%
128 x 128 321334.5 6.5 -141.3% 333829.0 6.3 -29.2%
Performance with this patch (roughly a 2x improvement):
Matrix SGEMM cycles MPC DGEMM cycles MPC
48 x 48 10756.0 10.5 -0.5% 18296.7 6.1 -1.7%
64 x 64 20490.0 12.9 1.4% 40615.0 6.5 0.0%
65 x 65 83528.3 3.3 -210.9% 96319.0 2.9 -83.3%
80 x 80 101453.5 5.1 -166.3% 128021.7 4.0 -76.6%
96 x 96 149795.1 5.9 -143.1% 168059.4 5.3 -47.4%
112 x 112 191481.2 7.3 -105.8% 204165.0 6.9 -14.6%
128 x 128 265019.2 7.9 -99.0% 272006.4 7.7 -5.3%
2018-06-17 15:39:15 +00:00
9e162146a9
Only initialize the part of the jobs array that will get used
...
The jobs array is getting initialized in O(compiled cpus^2) complexity.
Distros and people with bigger systems will use pretty high values
(128 or 256 or more) for this value, leading to interesting bubbles
in performance.
Baseline (single threaded performance) gets roughly 13 - 15 multiplications per cycle
in the interesting range (threading kicks in at 65x65 mult by 65x65).
The hardware is capable of 32 multiplications per cycle theoretically.
Matrix SGEMM cycles MPC DGEMM cycles MPC
48 x 48 10703.9 10.6 0.0% 17990.6 6.3 0.0%
64 x 64 20778.4 12.8 0.0% 40629.2 6.5 0.0%
65 x 65 26869.9 10.3 0.0% 52545.7 5.3 0.0%
80 x 80 38104.5 13.5 0.0% 72492.7 7.1 0.0%
96 x 96 61626.4 14.4 0.0% 113983.8 7.8 0.0%
112 x 112 91803.8 15.3 0.0% 180987.3 7.8 0.0%
128 x 128 133161.4 15.8 0.0% 258374.3 8.1 0.0%
When threading is turned on
TARGET=SKYLAKEX F_COMPILER=GFORTRAN SHARED=1 DYNAMIC_THREADS=1 USE_OPENMP=0 NUM_THREADS=128
Matrix SGEMM cycles MPC DGEMM cycles MPC
48 x 48 10725.9 10.5 -0.2% 18134.9 6.2 -0.8%
64 x 64 20500.6 12.9 1.3% 40929.1 6.5 -0.7%
65 x 65 2040832.1 0.1 -7495.2% 2097633.6 0.1 -3892.0%
80 x 80 2063129.1 0.2 -5314.4% 2119925.2 0.2 -2824.3%
96 x 96 2070374.5 0.4 -3259.6% 2173604.4 0.4 -1806.9%
112 x 112 2111721.5 0.7 -2169.6% 2263330.8 0.6 -1170.0%
128 x 128 2276181.5 0.9 -1609.3% 2377228.9 0.9 -820.1%
There is a deep deep cliff once you hit 65x65
With this patch
Matrix SGEMM cycles MPC DGEMM cycles MPC
48 x 48 10630.0 10.6 0.7% 18112.8 6.2 -0.7%
64 x 64 20374.8 13.0 1.9% 40487.0 6.5 0.4%
65 x 65 141955.2 1.9 -428.3% 146708.8 1.9 -179.2%
80 x 80 178921.1 2.9 -369.6% 186032.7 2.8 -156.6%
96 x 96 205436.2 4.3 -233.4% 224513.1 3.9 -97.0%
112 x 112 244408.2 5.8 -162.7% 262158.7 5.4 -47.1%
128 x 128 321334.5 6.5 -141.3% 333829.0 6.3 -29.2%
The cliff is very significantly reduced.
(more to follow)
2018-06-17 15:32:03 +00:00
47bf0dba8f
Add build-time option for OMP scheduler; document MULTITHREAD_THRESHOLD range ( #1620 )
...
* Allow choosing the OpenMP scheduler and add range hint for GEMM_MULTITHREAD_THRESHOLD
* Amended description of GEMM_MULTITHREAD_THRESHOLD
to reflect #742 making it track floating point operations rather than matrix size
2018-06-15 11:25:05 +02:00
bf40f806ef
Remove the need for most locking in memory.c.
...
Using thread local storage for tracking memory allocations means that threads
no longer have to lock at all when doing memory allocations / frees. This
particularly helps the gemm driver since it does an allocation per invocation.
Even without threading at all, this helps, since even calling a lock with
no contention has a cost:
Before this change, no threading:
```
----------------------------------------------------
Benchmark Time CPU Iterations
----------------------------------------------------
BM_SGEMM/4 102 ns 102 ns 13504412
BM_SGEMM/6 175 ns 175 ns 7997580
BM_SGEMM/8 205 ns 205 ns 6842073
BM_SGEMM/10 266 ns 266 ns 5294919
BM_SGEMM/16 478 ns 478 ns 2963441
BM_SGEMM/20 690 ns 690 ns 2144755
BM_SGEMM/32 1906 ns 1906 ns 716981
BM_SGEMM/40 2983 ns 2983 ns 473218
BM_SGEMM/64 9421 ns 9422 ns 148450
BM_SGEMM/72 12630 ns 12631 ns 112105
BM_SGEMM/80 15845 ns 15846 ns 89118
BM_SGEMM/90 25675 ns 25676 ns 54332
BM_SGEMM/100 29864 ns 29865 ns 47120
BM_SGEMM/112 37841 ns 37842 ns 36717
BM_SGEMM/128 56531 ns 56532 ns 25361
BM_SGEMM/140 75886 ns 75888 ns 18143
BM_SGEMM/150 98493 ns 98496 ns 14299
BM_SGEMM/160 102620 ns 102622 ns 13381
BM_SGEMM/170 135169 ns 135173 ns 10231
BM_SGEMM/180 146170 ns 146172 ns 9535
BM_SGEMM/189 190226 ns 190231 ns 7397
BM_SGEMM/200 194513 ns 194519 ns 7210
BM_SGEMM/256 396561 ns 396573 ns 3531
```
with this change:
```
----------------------------------------------------
Benchmark Time CPU Iterations
----------------------------------------------------
BM_SGEMM/4 95 ns 95 ns 14500387
BM_SGEMM/6 166 ns 166 ns 8381763
BM_SGEMM/8 196 ns 196 ns 7277044
BM_SGEMM/10 256 ns 256 ns 5515721
BM_SGEMM/16 463 ns 463 ns 3025197
BM_SGEMM/20 636 ns 636 ns 2070213
BM_SGEMM/32 1885 ns 1885 ns 739444
BM_SGEMM/40 2969 ns 2969 ns 472152
BM_SGEMM/64 9371 ns 9372 ns 148932
BM_SGEMM/72 12431 ns 12431 ns 112919
BM_SGEMM/80 15615 ns 15616 ns 89978
BM_SGEMM/90 25397 ns 25398 ns 55041
BM_SGEMM/100 29445 ns 29446 ns 47540
BM_SGEMM/112 37530 ns 37531 ns 37286
BM_SGEMM/128 55373 ns 55375 ns 25277
BM_SGEMM/140 76241 ns 76241 ns 18259
BM_SGEMM/150 102196 ns 102200 ns 13736
BM_SGEMM/160 101521 ns 101525 ns 13556
BM_SGEMM/170 136182 ns 136184 ns 10567
BM_SGEMM/180 146861 ns 146864 ns 9035
BM_SGEMM/189 192632 ns 192632 ns 7231
BM_SGEMM/200 198547 ns 198555 ns 6995
BM_SGEMM/256 392316 ns 392330 ns 3539
```
Before, when built with USE_THREAD=1, GEMM_MULTITHREAD_THRESHOLD = 4, the cost
of small matrix operations was overshadowed by thread locking (look smaller than
32) even when not explicitly spawning threads:
```
----------------------------------------------------
Benchmark Time CPU Iterations
----------------------------------------------------
BM_SGEMM/4 328 ns 328 ns 4170562
BM_SGEMM/6 396 ns 396 ns 3536400
BM_SGEMM/8 418 ns 418 ns 3330102
BM_SGEMM/10 491 ns 491 ns 2863047
BM_SGEMM/16 710 ns 710 ns 2028314
BM_SGEMM/20 871 ns 871 ns 1581546
BM_SGEMM/32 2132 ns 2132 ns 657089
BM_SGEMM/40 3197 ns 3196 ns 437969
BM_SGEMM/64 9645 ns 9645 ns 144987
BM_SGEMM/72 35064 ns 32881 ns 50264
BM_SGEMM/80 37661 ns 35787 ns 42080
BM_SGEMM/90 36507 ns 36077 ns 40091
BM_SGEMM/100 32513 ns 31850 ns 48607
BM_SGEMM/112 41742 ns 41207 ns 37273
BM_SGEMM/128 67211 ns 65095 ns 21933
BM_SGEMM/140 68263 ns 67943 ns 19245
BM_SGEMM/150 121854 ns 115439 ns 10660
BM_SGEMM/160 116826 ns 115539 ns 10000
BM_SGEMM/170 126566 ns 122798 ns 11960
BM_SGEMM/180 130088 ns 127292 ns 11503
BM_SGEMM/189 120309 ns 116634 ns 13162
BM_SGEMM/200 114559 ns 110993 ns 10000
BM_SGEMM/256 217063 ns 207806 ns 6417
```
and after, it's gone (note this includes my other change which reduces calls
to num_cpu_avail):
```
----------------------------------------------------
Benchmark Time CPU Iterations
----------------------------------------------------
BM_SGEMM/4 95 ns 95 ns 12347650
BM_SGEMM/6 166 ns 166 ns 8259683
BM_SGEMM/8 193 ns 193 ns 7162210
BM_SGEMM/10 258 ns 258 ns 5415657
BM_SGEMM/16 471 ns 471 ns 2981009
BM_SGEMM/20 666 ns 666 ns 2148002
BM_SGEMM/32 1903 ns 1903 ns 738245
BM_SGEMM/40 2969 ns 2969 ns 473239
BM_SGEMM/64 9440 ns 9440 ns 148442
BM_SGEMM/72 37239 ns 33330 ns 46813
BM_SGEMM/80 57350 ns 55949 ns 32251
BM_SGEMM/90 36275 ns 36249 ns 42259
BM_SGEMM/100 31111 ns 31008 ns 45270
BM_SGEMM/112 43782 ns 40912 ns 34749
BM_SGEMM/128 67375 ns 64406 ns 22443
BM_SGEMM/140 76389 ns 67003 ns 21430
BM_SGEMM/150 72952 ns 71830 ns 19793
BM_SGEMM/160 97039 ns 96858 ns 11498
BM_SGEMM/170 123272 ns 122007 ns 11855
BM_SGEMM/180 126828 ns 126505 ns 11567
BM_SGEMM/189 115179 ns 114665 ns 11044
BM_SGEMM/200 89289 ns 87259 ns 16147
BM_SGEMM/256 226252 ns 222677 ns 7375
```
I've also tested this with ThreadSanitizer and found no data races during
execution. I'm not sure why 200 is always faster than it's neighbors, we must
be hitting some optimal cache size or something.
2018-06-14 16:54:58 +01:00
63f7395fb4
Move some DYNAMIC_ARCH targets to new DYNAMIC_OLDER option
2018-06-09 16:31:38 +02:00
38ad05bd04
Extend loop range to find SkylakeX in force_coretype
2018-06-05 10:26:49 +02:00
8be027e4c6
Update dynamic.c
2018-06-04 14:36:39 +02:00
ac7b6e3e9a
Fix misplaced endif
2018-06-04 08:23:40 +02:00
ef626c6824
typo fix
2018-06-04 00:13:19 +02:00
5a51cf4576
Separate Skylake X from Skylake
2018-06-03 23:41:33 +02:00
99c7bba8e4
Initial support for SkylakeX / AVX512
...
This patch adds the basic infrastructure for adding the SkylakeX (Intel Skylake server)
target. The SkylakeX target will use the AVX512 (AVX512VL level) instruction set,
which brings 2 basic things:
1) 512 bit wide SIMD (2x width of AVX2)
2) 32 SIMD registers (2x the number on AVX2)
This initial patch only contains a trivial transofrmation of the Haswell SGEMM kernel
to AVX512VL; more will follow later but this patch aims to get the infrastructure
in place for this "later".
Full performance tuning has not been done yet; with more registers and wider SIMD
it's in theory possible to retune the kernels but even without that there's an
interesting enough performance increase (30-40% range) with just this change.
2018-06-03 07:58:52 +00:00
a8002e283a
Revert "take out unused variables"
...
This reverts commit e5752ff9b3
2018-06-01 23:20:00 +01:00
a91f1587b9
Work around name clash with Windows10's winnt.h
...
fixes #1503
2018-05-31 13:26:00 +02:00
191746c493
Merge pull request #1557 from martin-frbg/getconfig
...
Add threading and OpenMP information to output
2018-05-14 17:37:55 +02:00
41ae8e8d67
Add threading and OpenMP information to output
...
For #1416 and #1529 , more information about the options OpenBLAS was built with is needed. Additionally we may want to add this data to the openblas.pc file (but not all projects use pkgconfig, and as far as I am aware the cmake module for accessing it does not make such "private" declarations available)
2018-05-12 12:11:38 +02:00
53457f222f
move _Atomic define to common.h
2018-05-11 00:13:16 -07:00
3716267124
Change _STDC_VERSION__ to __STDC_VERSION__
...
Change-Id: Id3fa4e8d9eedd4ef7230df69b611e7f397301a42
2018-05-11 12:15:08 +08:00
50acc40613
Merge pull request #1536 from WestAlgo/develop
...
Fix race condition in blas_server_omp.c
2018-05-11 10:09:14 +08:00
802cf6b22d
Merge pull request #1486 from martin-frbg/atomic
...
Use _Atomic instead of volatile for thread safety where C11 is supported
2018-04-27 23:09:57 +02:00
1b83341d19
Fix race condition in blas_server_omp.c
...
Change-Id: Ic896276cd073d6b41930c7c5a29d66348cd1725d
2018-04-27 17:00:42 +08:00
f29389c7ac
Merge pull request #1520 from martin-frbg/cpucounts
...
Catch invalid cpu count returned by CPU_COUNT_S
2018-04-14 22:24:34 +02:00
7c861605b2
Catch invalid cpu count returned by CPU_COUNT_S
...
mips32 was seen to return zero here, driving nthreads to zero with subsequent fpe in blas_quickdivide
2018-04-14 18:29:10 +02:00
20c6c38e51
Merge branch 'develop' into atomic
2018-04-07 12:09:39 +02:00
d636b418af
Merge pull request #1504 from ararslan/aa/openbsd
...
Allow building on OpenBSD
2018-04-04 15:26:46 +02:00
a41d241a0e
Add support for DragonFly BSD
2018-04-03 16:39:29 -07:00
8da6b6ae52
Allow building on OpenBSD
...
With this change, OpenBLAS builds and all tests pass on OpenBSD 6.2
using Clang. Tested on x86-64 only, with and without DYNAMIC_ARCH=1.
2018-04-02 10:48:22 -07:00
01c4b82f04
Update memory.c
2018-03-31 22:32:06 +02:00
93db123f7e
Update memory.c
2018-03-29 13:13:49 +02:00
752fdb5dd8
Add workaround for old gcc and clang versions
...
Old gcc and clang do not handle constructor arguments, finally fix #875 as discussed there, using the fedora patch
2018-03-29 11:56:56 +02:00
6a99fcce94
Use _Atomic instead of volatile for thread safety where C11 is supported
...
Suggested by dodomorandi in #660
2018-03-10 00:03:49 +01:00
7646974227
Limit the additional locking from PRs 1052,1299 to non-OpenMP multithreading
2018-02-21 11:45:33 +01:00
8866e393a2
Revert "Add locks only for non-OPENMP multithreading"
2018-02-20 17:17:12 +01:00
3119b2ab4c
Add locks only for non-OPENMP multithreading
...
to migitate performance problems caused by #1052 and #1299 as seen in #1461
2018-02-20 12:17:18 +01:00
8f5f614615
On Cygwin use mmap instead of Windows native allocation functions, which are not fork-safe.
2018-02-06 12:23:27 +01:00
f5fc109fbd
Perform blas_thread_shutdown with pthread_atfork() on Cygwin
...
Even if we're directly using the win32 threading driver and not pthreads,
pthread_atfork still works fine to register a pre-fork handler, and is
necessary to restore the threading server to a pre-initialized state.
2018-02-06 12:23:27 +01:00
e388459a27
Merge pull request #1419 from brada4/develop
...
Initialize unitialized values for repeated calls
2018-01-31 23:48:34 +01:00
e5752ff9b3
take out unused variables
2018-01-20 11:42:31 +01:00
8a0b086b28
add missing bracket for old glibc (cppcheck)
2018-01-12 22:35:48 +01:00
42285d8e70
Merge pull request #1410 from brada4/develop
...
Address warnings #1357
2018-01-06 20:02:46 +01:00
8aafa0473c
address last warnings as seen by gcc7
2018-01-01 20:57:12 +01:00
11a627c54e
remove surplus parentheses to silence clang5
2018-01-01 20:56:26 +01:00
cc9500db41
Merge pull request #1403 from brada4/develop
...
Address few more warnings
2017-12-30 14:51:34 +01:00
bfc2a88594
remove unused buffer
2017-12-22 00:55:40 +01:00
177b78c8b4
Issue1388 ( #1389 )
...
* Calculation of chunk range limits was ignoring num_cpu
bug introduced by me in #1262 - should fix #1388
* Calculation of range limits was ignoring num_cpu
bug introduced by me in #1262
* Calculation of chunk range limits was ignoring num_cpu
bug introduced by me in #1262
* Calculation of chunk range limits was ignoring num_cpu
bug introduced by me in #1262
* Calculation of chunk range limits was ignoring num_cpu
bug introduced by me in #1262
* Calculation of chunk range limits was ignoring num_cpu
bug introduced by me in #1262
2017-12-09 22:29:03 +01:00
281a2b952f
warning cleanup ( #1380 )
...
* dead increments in driver/level2
* dead increments in kernel/generic
* part dead increments in kernel/x86_64
2017-12-05 19:54:10 +01:00
c49c6b237d
Merge pull request #1382 from martin-frbg/dtrmv-1332
...
Work around errors in multithreaded dtrmv
2017-12-05 19:53:23 +01:00
28ae3ca76f
Limit MAX_CPU to 1024 for now
...
Some Linux distributions (notably SuSE) have raised CPU_SETSIZE to 4096, apparently disregarding API limitations.
From #1348 , the highest value to survive array initialization (on a desktop system) is 3232, and 1024 - which is the
more usual CPU_SETSIZE limit, was demonstrated to work fine on an actual bignuma system.
2017-12-05 12:54:15 +01:00
b414283f48
Disable gemv unrolling
...
as a (hopefully temporary) workaround for #1332
2017-12-03 22:41:54 +01:00
ef95cd471f
elminate unread variable, after reiteration 3 of them (clang4)
2017-11-25 02:54:37 +01:00
e14d50d86e
eliminate Wunused-const gcc7 warning
2017-11-24 19:13:24 +01:00
07e7c36dac
Handle shmem init failures in cpu affinity setup code
...
Failures to obtain or attach shared memory segments would lead to an exit without explanation of the exact cause.
This change introduces a more verbose error message and tries to make the code continue without setting cpu affinity.
Fixes #1351
2017-11-18 23:57:44 +01:00
2a6fef9a55
Try to handle shmget or shmat failing
...
also replaces one verbatim sched_yield with the YIELDING macro for consistency as suggested in #1351
2017-11-09 23:16:13 +01:00
db72ad8f6a
Merge pull request #1320 from timmoon10/develop
...
2D thread distribution for multi-threaded GEMMs
2017-10-08 23:31:33 +02:00
514d237257
Merge pull request #1279 from xsacha/develop
...
CMake improvements
2017-10-06 21:13:45 +02:00
30486a356c
Reduce number of data partitions in n.
2017-10-04 12:37:49 -07:00
9de52b489a
Cleaning up and documenting multi-threaded GEMM code.
2017-10-03 16:32:08 -07:00
860dcfc703
Use 2D thread distribution for small GEMMs.
...
Allows maximum use of available cores if one of M and N is small and the other is large.
2017-10-03 13:43:39 -07:00
6aaa107865
Reducing threads for multi-threaded GEMMs on small matrices.
2017-09-27 19:25:33 -07:00
ba1f91f17b
Convert another caller of "allocation" to LOCK_COMMAND
...
... as the "allocation" code jumped to now does UNLOCK_COMMAND instead of blas_unlock
2017-09-09 20:30:33 +02:00
f460776f0f
Fix thread data races
2017-09-09 19:07:06 +02:00
e882f3d6f3
Fix thread data race in memory.c
2017-09-09 18:58:38 +02:00
37858d1146
Fix threading usage in CMake: s/SMP/USE_THREAD/
2017-08-19 15:07:42 +10:00
2f12ea017b
No strncasecmp with MSVC
2017-08-08 00:07:25 +05:30
719fcc56b0
Merge pull request #1262 from martin-frbg/xmv_thread-splitting
...
Make sure that range limit of last thread never exceeds data size
2017-08-06 14:11:44 +02:00
ebb04e3265
Merge pull request #1259 from isuruf/cmake
...
CMake Improvements
2017-08-02 15:31:05 +02:00
0ba64cee60
Update trmv_thread.c
2017-08-02 12:03:54 +02:00
c4e5ba1bfe
Make sure that range_n of last thread never exceeds the actual data size when splitting the workload
2017-08-02 00:37:58 +02:00
a6f533b248
Revert "Fix calculated range limit exceeding actual data size for last thread"
2017-08-01 19:28:08 +02:00
d245caa49a
Support out-of-source build
2017-08-01 15:16:14 +05:30
e70a6b92bf
Merge pull request #1257 from martin-frbg/cgroups-prereq
...
Rework __GLIBC_PREREQ checks to avoid breaking non-glibc builds
2017-08-01 11:23:03 +02:00
63cfa32691
Rework __GLIBC_PREREQ checks to avoid breaking non-glibc builds
2017-07-31 21:02:43 +02:00
585c0010a5
Fix range limit exceeding actual data size in last step
2017-07-28 00:27:02 +02:00
857f61bc5d
Fix range limit exceeding data size in last step
2017-07-28 00:21:53 +02:00
9332042d5f
Fix range exceeding actual data size in quick_divide
2017-07-28 00:13:24 +02:00
c4af196a2d
Honor cgroup/cpuset limits when enumerating cpus
2017-07-25 22:47:34 +02:00
480e697681
Revert "Honor cgroup/cpuset limits when enumerating cpus" ( #1246 )
2017-07-24 16:17:50 +02:00
80373ea039
More fixes for silly misedits
2017-07-15 12:48:42 +02:00
d12b75a6c4
Fixup braces lost in previous edit
2017-07-15 11:53:28 +02:00
7294fb1d9d
Merge branch 'develop' into cgroups
2017-07-15 10:40:42 +02:00
2a7c6930ac
Merge pull request #1234 from brada4/develop
...
Fix write past fixed size buffer
2017-07-13 20:27:37 +08:00
529bfc36ec
Fix write past fixed size buffer
2017-07-12 00:59:30 +02:00
731c518cff
Add files via upload
2017-07-11 18:42:39 +02:00
29fc429d9a
Honor cgroup/cpuset constraints when enumerating cpus
2017-07-11 18:27:33 +02:00
3db2adf872
Merge pull request #1230 from martin-frbg/rhel5
...
Add sched_getcpu implementation for pre-2.6 glibc
2017-07-09 13:16:16 +02:00
c1cf62d2c0
Add sched_getcpu implementation for pre-2.6 glibc
...
Fixes #1210 , compilation on RHEL5 with affinity enabled
2017-07-09 09:45:38 +02:00
bfe1656b8b
Merge pull request #1225 from martin-frbg/stolen_from_wernsaar_fork
...
fixed syrk_thread.c taken from wernsaar
2017-07-07 15:43:33 +08:00
49e62c0e77
fixed syrk_thread.c taken from wernsaar
...
Stride calculation fix copied from https://github.com/wernsaar/OpenBLAS/commit/88900e1
2017-07-06 17:30:12 +02:00
34513be726
Add Microsoft Windows 10 UWP build support
2017-06-23 13:07:34 -07:00
65e56cb29d
Add 64bit support for Microsoft Visual Studio
2017-06-21 13:38:22 -07:00
59c97cfee4
memory: Fix buffer overflow when position == NUM_BUFFERS
2017-05-05 17:47:03 +01:00
5fecfe0f42
memory: switch loop condition around in blas_memory_free
...
Before this commit, the "position < NUM_BUFFERS" loop condition from
blas_memory_free will be completely optimized away by GCC. This is
because the condition can only be false after undefined behavior has
already been invoked (reading past the end of an array). As a
consequence of this bug, GCC also removes the subsequent if statement
and all the code after the error label because all of it is dead.
This commit switches the loop condition around so it works as intended.
2017-05-05 16:01:58 +01:00
9c884986ad
Add an extra familiy/model combination used by AMD Steamrolller (Godavari).
2017-04-19 19:15:47 +02:00
0cbd2d34e4
Recognize ZEN when passed as OPENBLAS_CORETYPE.
2017-04-10 20:05:16 +02:00
62979fd104
Fix dynamic detection for ZEN CPUs.
2017-04-10 19:08:37 +02:00
c9ff735da6
Add ZEN support (tested for auto-detected static backend)
2017-03-19 15:32:50 +01:00
ffc1d6c468
Merge pull request #1108 from ashwinyes/develop_20170203_thunderx2t99
...
Optimized Implementations for ThunderX2T99
2017-02-28 16:02:19 +01:00
a86474c6f7
THUNDERX2T99: Performance fix for ZGEMM
2017-02-28 06:05:00 -08:00
19ba133383
THUNDERX2T99: Add Optimized ZGEMM Implementation
2017-02-28 05:31:41 +00:00
5088523786
detect apollo lake for real
2017-02-20 23:54:59 +01:00
1d8ab99e09
Add `exfamily == 9` case (Kaby Lake) to dynamic arch detection
2017-02-10 15:23:55 -08:00
76c6e33e54
Enable EXCAVATOR kernels for A12-9800
2017-02-07 21:38:28 +01:00
2757b49767
THUNDERX2T99: Add Optimized CGEMM Implementation
2017-01-30 17:44:26 +05:30
f279ff4789
THUNDERX2T99: Add Optimized SGEMM Implementation
2017-01-16 21:44:33 +05:30
0863a0d4b4
Merge pull request #1061 from ashwinyes/develop_aarch64_vulcan_thunderx_patch
...
Add new targets for ARM64
2017-01-16 13:20:10 +08:00
c1c5a63d3c
prepared parameter.c for UNROLL values, that are not a power of two
2017-01-11 09:50:28 +01:00
4b55fae337
ARM64: Add Cavium THUNDERX2T99 Target
2017-01-11 11:18:40 +05:30
0b8e876d89
VULCAN: Add optimized DGEMM implementation
2017-01-10 15:01:37 +05:30
4713e7c47f
ARM64: Add the VULCAN Target
2017-01-10 15:01:17 +05:30
1aa1e6cb54
modify the blas_l1_thread.c for support multi-threded for L1 fuction with return value
2017-01-10 11:47:06 +08:00
b9bb009236
Merge pull request #1053 from wernsaar/develop
...
prepared driver/level3 functions for UNROLL values, that are not a po…
2017-01-09 11:17:38 +01:00
a2672d5589
prepared driver/level3 functions for UNROLL values, that are not a power of two
2017-01-09 10:38:15 +01:00
51aa157e64
Relocate declaration of alloc_lock outside ifdef block
2017-01-09 01:10:43 +01:00
87c7d10b34
Fix thread data races detected by helgrind 3.12
...
Ref. #995 , may possibly help solve issues seen in 660,883
2017-01-08 23:33:51 +01:00
0ef7841473
Update xerbla.c
2017-01-04 23:16:48 +01:00
104ad066af
Use appropriate int32/int64 format for error number in message string
2016-12-30 00:45:59 +01:00
a16ace68f5
Include system headers on FreeBSD
2016-11-16 21:58:20 -08:00
596ead0f8d
Add files via upload
2016-11-06 23:26:39 +01:00
66c9a9b33d
Merge pull request #981 from howard0su/develop
...
USE NPROCESSOR_CONF instaed of NPORCESSOR_ONLN
2016-10-17 11:32:57 +08:00
8a8f3932eb
Update dynamic.c
...
Add Bay Trail "Pentium N3520" atom
2016-10-16 22:40:00 +02:00
ff1da01476
USE NPROCESSOR_CONF instaed of NPORCESSOR_ONLN
...
to determine the number of CPU. In ARM platform,
online CPU will increasing when there is more workload.
while configure cpu is the max number of CPU.
2016-10-13 12:37:50 +00:00
ef52a9266b
Fixed #979 . Patch for NetBSD.
2016-10-13 10:17:07 +08:00
7de829f713
Update dynamic.c
...
Add Braswell (extended model 4, model 12) N3150 as Nehalem
2016-07-14 12:22:55 +02:00
053044ae4d
Replace CMAKE_SOURCE_DIR/CMAKE_BINARY_DIR with PROJECT_SOURCE_DIR/PROJECT_BINARY_DIR
...
If OpenBLAS is built using add_subdirectory(OpenBlas) as part of another project
then the paths set by CMAKE_XXX_DIR are relative to the parent project
and not the OpenBLAS project.
2016-05-25 09:13:28 +02:00
0fb380c966
Update NUMA CPU binding
...
When the number of process can all be
accommodated within the current node,
then use cores from the current node only.
2016-04-29 11:58:15 +05:30
78b05f6476
bugfix for EXCAVATOR and DYNAMIC_ARCH
2016-04-25 10:13:30 +02:00
2b967590a0
bugfix in dynamic.c
2016-04-25 09:08:38 +02:00
aa744dfa59
Update memory.c
2016-03-22 20:02:37 +08:00
61cf8f74d9
Fix access violation on Windows while static linking
2016-03-22 19:14:54 +08:00
68eb4fa329
Add missing openblas_env makefile.
2016-03-09 14:52:47 -05:00
05196a8497
Refs #716 . Only call getenv at init function.
2016-03-09 12:50:07 -05:00
53ba1a77c8
ztrmv_L.c: no longer need a 4kB buffer
...
Fix #786
2016-03-05 19:07:03 +01:00
1edf30b790
Change Opteron(SSE3) to Opteron_SSE3 at dyanmaic core name.
2016-03-01 20:13:08 +08:00
6b85dbb6dc
Refs #696 . Turn off stack limit setting on Linux.
...
I cannot reproduce SEGFAULT of lapack-test with default stack size
on ARM Linux.
2016-02-24 14:21:42 -05:00
d06b92906a
Add gemm3m building for CMake.
2016-02-12 05:02:51 +08:00
78dcf5c3d5
Improve performances of ztrmv on small matrices
...
* Use stack allocation
* Disable multi-threading
* Ref #727
2016-02-08 11:25:02 +01:00
935356c34f
Update dynamic.c and cpuid_x86.c for Intel Avoton.
...
Second part of "support Intel Avoton via Nehalem kernel"
2016-02-02 13:42:55 -05:00
f5df444ceb
Merge pull request #762 from jeromerobert/bug760
...
Let openblas_get_num_threads return the number of active threads
2016-01-26 08:45:16 -06:00
aaa8551c57
Merge pull request #749 from lotheac/illumos_fixes
...
illumos fixes
2016-01-26 08:42:20 -06:00
0d87c1ffb6
Let openblas_get_num_threads return the number of active threads
...
... not the number of allocated threads.
Close #760
2016-01-26 13:04:16 +01:00
e737e32fd1
RLIMIT_NPROC doesn't exist on illumos
2016-01-22 18:55:51 +02:00
97cd4b8aee
illumos fixes to memory.c
2016-01-22 18:55:43 +02:00
b07d733a71
added updates for syrk and syr2k
2016-01-21 13:16:44 +01:00
055b481386
Fixed CMake bug for single core.
2016-01-15 06:42:54 +08:00
0d22551a6b
increase the stack size limit in the constructor
2015-11-20 09:23:01 +01:00
fbc21266e6
Minor C code fixes in driver/
2015-11-09 14:15:49 +05:30
839395fc25
Detect AMD Trinity and Richland.
2015-10-29 02:53:29 +08:00
6040858b22
Fix #673
...
Add lacking headers declarations when compiling for Android ARM7
2015-10-27 13:55:24 +01:00
70642fe4ed
Refs #668 . Raise the signal when pthread_create fails.
...
Thank James K. Lowden for the patch.
2015-10-26 19:02:51 -05:00
2feef49fa8
Merge branch 'develop' into cmake
...
Conflicts:
driver/others/memory.c
2015-10-26 14:54:34 -05:00
1ce054fcb3
Refs #669 . Fixed the build bug with gcc on Mac OS X.
2015-10-22 11:07:35 -05:00
d8392c1245
Fixe cmake config bugs.
2015-10-20 04:30:55 +08:00
94b125255f
Merge branch 'develop' into cmake
...
Conflicts:
driver/others/memory.c
2015-10-13 04:46:08 +08:00
11ac4665c8
Fixed #654 . Make sure the gotoblas_init function is run before all other static initializations.
2015-10-05 14:14:32 -05:00
cc7cab8a45
Detect other Intel Skylake cores.
...
http://users.atw.hu/instlatx64/
2015-09-09 10:47:17 -05:00
61ae47eb99
Ref #632 . Support Intel Skylake by Haswell kernels.
2015-09-09 11:07:33 -04:00
d3e2f0a1af
add missing barriers
...
should fix issue #597
2015-08-16 15:37:02 +02:00
f8eba3d548
Fixed cmake build bugs on Linux.
2015-08-11 16:25:16 -05:00
f874465bb8
Use cmake to build OpenBLAS GENERIC Target on MSVC x86 64-bit.
...
Disable CBLAS and LAPACK.
2015-08-10 14:10:44 -05:00
dcd5ba4443
Merge branch 'cmake' of https://github.com/hpanderson/OpenBLAS into hpanderson_cmake
2015-07-22 04:06:39 +08:00
a11555c715
Support Android NDK armeabi-v7a-hard ABI. (-mfloat-abi=hard)
...
e.g.
make HOSTCC=gcc CC=arm-linux-androideabi-gcc NO_LAPACK=1 TARGET=ARMV7
In Android NDK, it uses armeabi-v7a-hard ABI.
TARGET_CFLAGS += -mhard-float -D_NDK_MATH_NO_SOFTFP=1
TARGET_LDFLAGS += -Wl,--no-warn-mismatch -lm_hard
For more information, please check hard-float example at
android_ndk/tests/device/hard-float/jni/.
2015-05-20 21:57:27 -05:00
51ff17d46e
Add AMD Excavator target.
2015-05-13 16:16:30 -05:00
ebb9eba987
Fix build with ALLOC_SHM=0 (Android NDK)
...
Refactor such that you can build with ALLOC_SHM=0. HughTLB
implicity depends on ALLOC_SHM=1. This patch allows
building for Android NDK r10d.
2015-05-10 00:10:26 -07:00
8e5a1083bb
Refs #532 . Improve gemv paralel with small m and large n case.
...
Splite the matrix and reduction.
2015-05-08 05:33:17 +08:00
9798481979
Refs #478 , #482 . Fix segfault bug for gemv_t with MAX_ALLOC_STACK flag.
...
For gemv_t, directly use malloc to create the buffer.
2015-04-13 19:45:27 -05:00
8977b3f235
Refs #529 . Support Intel Broadwell by Haswell kernels.
2015-04-02 11:08:03 -05:00
e95d64333a
Refs #519 . Avoid calling strncpy.
2015-03-19 15:57:22 -05:00
b6438dedea
Fix issue #508
...
Fix race condition during shutdown causing a crash in
gotoblas_set_affinity().
2015-03-18 13:22:43 +01:00
5ae8993752
Added intrinsics for MSVC.
2015-02-25 11:52:51 -06:00
84d90d6ed8
Fixed some compiler errors/warnings for clang.
2015-02-25 11:52:25 -06:00
9eaea02f33
Added additional gemm defines for complex types.
2015-02-25 09:39:11 -06:00
ab7043373f
Fixed bug generating trmv complex source names.
2015-02-24 15:18:41 -06:00
0553476fba
Added TRANS defines for complex sources in lapack.
2015-02-24 14:30:35 -06:00
2416d9dbac
Fixed TRANSA defines for complex sources in driver/level2.
2015-02-24 13:18:07 -06:00
0d8e227ea7
Changed strategy for setting preprocessor definitions.
...
Instead of generating separate object files for each permutation of
defines for a source file, GenerateNamedObjects now writes an entirely
new source file and inserts the defines as #define c statements.
This solves a problem I ran into with ar.exe where it was refusing to
link objects that had the same filename despite having different paths.
2015-02-24 12:26:33 -06:00
1b7f427401
Added conj gemv objects for complex build.
2015-02-23 10:24:31 -06:00
fb5d5bb971
Added defines for complex trmv.
2015-02-21 12:39:03 -06:00
371071d461
Added CONJ defines for trmm/trsm.
2015-02-21 10:59:02 -06:00
8a143516e3
Added alternate_name to a couple of the name mangling schemes.
...
Added zherk_k sources to driver/level3.
2015-02-20 17:03:33 -06:00
e5897ecb9b
Added zherk_kernel.c objects to driver/level3.
2015-02-19 16:19:56 -06:00
33c5e8db7f
Added a helper function for setting the L1 kernel defaults.
...
Added loop to build objects with different KERNEL defines.
2015-02-17 21:36:23 -06:00
4662a0b13a
Changed generate functions to iterate through a list of float types.
...
This will generate obj files for SINGLE/DOUBLE/COMPLEX/DOUBLE COMPLEX.
2015-02-15 17:44:37 -06:00
e74462a3f5
Moved declarations to start of functions to satisfy MSVC C89 implementation.
2015-02-11 11:16:57 -06:00
056ba26755
Changed a number of inline calls to use __inline.
...
MSVC doesn't inmplement C99, so can't use the inline keyword. __inline
appears to work in MSVC and GCC.
2015-02-11 11:13:17 -06:00
c94fe71278
Removed incoming-stack-boundary for MSVC.
...
Made float type optional for GenerateNamedObjects.
Called GenerateNamedObjects for a couple of driver/others files that
needed NAME/CNAME set.
2015-02-11 10:54:14 -06:00
e8c39138c6
Removed return value from GenerateNamedObjects.
...
It sets DBLAS_OBJS directly to save a bunch of list appending in the
CMakeLists.txt files.
2015-02-09 12:28:09 -06:00
7fa5c4e2fd
Fixed some case issues with ARCH.
...
Added some kernel and driver/others objects.
2015-02-08 15:29:18 -06:00
cfa9392ffa
Fix openblas_get_num_threads and openblas_get_num_procs bug with single thread.
2015-02-08 01:30:23 -06:00
2f59135eb6
Added gemv to level2 CMakeLists.txt.
2015-02-07 21:15:21 -06:00
6b5d26e07b
Added SMP sources to level2 CMakeLists.txt.
2015-02-06 16:52:19 -06:00
627d5e7401
Added SMP objects to driver/level3.
2015-02-05 12:22:48 -06:00
943fa2fb58
Fixed object names in level2.
2015-02-05 10:49:11 -06:00
461e691127
Codes when define is absent are now a parameter to AllCombinations.
...
The level3 object names should now be correct.
2015-02-05 09:23:47 -06:00
cfaf1c678f
Added option to append define codes with an underscore.
...
Fixed the code array not getting reset on subsequent AllCombinations
calls.
2015-02-05 09:17:18 -06:00
0d7bad1f35
Changed GenerateObjects to append combination codes (e.g. dtrmm_TU).
2015-02-05 09:02:54 -06:00
2828f6630c
Added SMP sources to COMMONOBJS.
2015-02-04 14:01:36 -06:00
65a847cd36
Introduce openblas_get_num_threads and openblas_get_num_procs
2015-02-03 12:23:41 -05:00
7194424fef
Added missing common objects to the library.
2015-02-02 15:21:29 -06:00
d11bde60d0
DOUBLE define for DBLAS objects is now set in main CMakeLists.txt.
...
Since the objects are the same, could generate SINGLE/COMPLEX/etc here
without having to rewrite all the object enumeration code again.
2015-02-02 15:00:44 -06:00
5057a4b4df
Added openblas add_library call that uses DBLAS_OBJS ojbects.
2015-01-30 15:21:21 -06:00
3e8ea7a351
Added COMMONOBJS to driver/others CMakeLists.txt.
2015-01-30 14:06:14 -06:00
d3dcdddf75
Moved functions into util cmake file.
2015-01-30 13:47:40 -06:00
Andrew
Martin Kroeker
luz.paz
Jeff Baylor
Rashmica Gupta
Erik M. Bray
caiyu
Arjan van de Ven
Ashwin Sekhar T K
Zoltán Mizsei
Craig Donner
oon3m0oo
Matthew Brett
zhiyong.dang
Zhang Xianyi
Alex Arslan
Andrew
Tim Moon
Sacha Refshauge
Isuru Fernando
Neil Shipp
James Cowgill
Gian-Carlo Pascutto
Gian-Carlo Pascutto
Denis Steckelmacher
Elliot Saba
Martin Koehler
Werner Saar
jiahaipeng
Howard Su
John Biddiscombe
Ashwin Sekhar T K
Theoractice
theoractice
Jerome Robert
Lauri Tirkkonen
Ralph Campbell
j-bo
Yichao Yu
Grazvydas Ignotas
powderluv
Ton van den Heuvel
Hank Anderson
Erik Schnetter