gxw
48698b2b1d
LoongArch64: Rename core
Use microarchitecture name instead of meaningless strings to name the core,
the legacy core is still retained.
1. Rename LOONGSONGENERIC to LA64_GENERIC
2. Rename LOONGSON3R5 to LA464
3. Rename LOONGSON2K1000 to LA264
1 year ago
Martin Kroeker
3ee9e9d8d0
Merge pull request #4879 from martin-frbg/issue4868-2
Ensure a memory buffer has been allocated for each thread before invoking it (take 2)
1 year ago
Martin Kroeker
a8d6b0219a
Merge pull request #4877 from XiWeiGu/fixed_undefined_blas_set_parameter
Fixed the undefined reference to blas_set_parameter
1 year ago
Martin Kroeker
d24b3cf393
properly fix buffer allocation and assignment
1 year ago
gxw
fd033467ac
Fixed the undefined reference to blas_set_parameter
Fixed the undefined reference to blas_set_parameter when
enabling USE_OPENMP and DYNAMIC_ARCH.
1 year ago
Martin Kroeker
23b5d66a86
Ensure a memory buffer has been allocated for each thread before invoking it
1 year ago
Martin Kroeker
753c7ebe17
Merge pull request #4835 from martin-frbg/revertwin4359
Temporarily revert to the coarse-grained locking in the Windows thread server
1 year ago
Martin Kroeker
50397e017a
Merge pull request #4838 from martin-frbg/fix4662-3
fix invalid ifdef syntax in HUGETLB handling
1 year ago
Martin Kroeker
5257f807a9
fix invalid ifdef syntax in HUGETLB handling
1 year ago
Martin Kroeker
2aed90171a
Add riscv sources for DYNAMIC_ARCH
1 year ago
Martin Kroeker
6468dc1142
restore the coarse locking of the pre-4359 version
1 year ago
yamazaki-mitsufumi
821ef34635
Add A64FX to the list of CPUs supported by DYNAMIC_ARCH
1 year ago
Martin Kroeker
a815594fd1
Merge pull request #4801 from markdryan/markdryan/riscv-dynamic-arch
Add autodetection for riscv64
1 year ago
Martin Kroeker
a373d0f107
Improve the error message for thread creation failure
1 year ago
Mark Ryan
3b715e6162
Add autodetection for riscv64
Implement DYNAMIC_ARCH support for riscv64. Three cpu types are
supported, riscv64_generic, riscv64_zvl256b, riscv64_zvl128b.
The two non-generic kernels require CPU support for RVV 1.0 to
function correctly. Detecting that a riscv64 device supports
RVV 1.0 is a little complicated as there are some boards on the
market that advertise support for V via hwcap but only support
RVV 0.7.1, which is not binary compatible with RVV 1.0. The
approach taken is to first try hwprobe. If hwprobe is not
available, we fall back to hwcap + an additional check to distinguish
between RVV 1.0 and RVV 0.7.1.
Tested on a VM with VLEN=256, a CanMV K230 with VLEN=128 (with only
the big core enabled), a Lichee Pi with RVV 0.7.1 and a VF2 with no
vector.
A compiler with RVV 1.0 support must be used to build OpenBLAS for
riscv64 when DYNAMIC_ARCH=1.
Signed-off-by: Mark Ryan <markdryan@rivosinc.com>
1 year ago
Martin Kroeker
d0b9948b23
Guard against invalid thread_status.queue
1 year ago
Martin Kroeker
7e9a4ba427
Merge pull request #4741 from shivammonaka/Pthread_Scalability_Improvement
Enhancing Core Utilization in BLAS Calls: A Scalable Architecture
1 year ago
Martin Kroeker
9b2a0c79cb
Add Zhaoxin KX7000
1 year ago
shivammonaka
9e22d70957
Dynamic locking in Pthread Backend to allow multiple BLAS calls to be executed parallelly
1 year ago
Martin Kroeker
db070a9223
add gemm_batch drivers
1 year ago
Martin Kroeker
d0794f88dc
add gemm_batch driver
1 year ago
Martin Kroeker
0073affe63
Merge pull request #4693 from goplanid/locks-improvement
Lock Management Improvements for Memory Allocation Efficiency
1 year ago
Martin Kroeker
6ca9ffa7f5
Merge pull request #4655 from yamazakimitsufumi/update_2d_thread_distribution
Expanding the scope of 2D thread distribution to improve multi-threaded DGEMM performance
1 year ago
Deeksha Goplani
0dc80a5c8d
locks improvement
1 year ago
Martin Kroeker
8da6f7e5f2
Merge pull request #4686 from XiWeiGu/loongarch64_dgemm_kernel_16x6
Loongarch64: Improving the Performance and Stability of dgemm
1 year ago
gxw
637c650f4f
loongarch64: Add buffer offset for target LOONGSON3R5
1 year ago
Martin Kroeker
5500b4ab26
Merge pull request #4680 from theAeon/develop
Expose whether locking is enabled in get_config
1 year ago
Martin Kroeker
f0f1ff7820
fix HUGETLB allocation for TLS mode as well
1 year ago
Andrew Robbins
edfe1aa471
Expose whether locking is enabled in get_config
1 year ago
Martin Kroeker
dc99b61380
sort unwanted interdependencies of alloc_shm and alloc_hugetlb
1 year ago
Martin Kroeker
ddcd7d6fa8
Merge branch 'develop' into Threading_Callback
1 year ago
yamazaki-mitsufumi
51ab1903e7
Expanding the scop of 2D thread distribution
1 year ago
gxw
d8c4ea8793
loongarch: Optimizing the performance of the GEMM on servers
1 year ago
shivammonaka
7102367fde
Introduced callback to Pthread, Win32 and OpenMP backend
1 year ago
Mark Seminatore
b0ad8a78ff
code to fix lost work in case of re-entrant calls to exec_blas_async()
1 year ago
Martin Kroeker
88b5330ae7
Restore outer loop of blas_buffer_inuse setup
1 year ago
shivammonaka
d49ebc54e1
Merge branch 'shivam-develop' into shivam-Locks
1 year ago
shivammonaka
bc191015e3
Using OpenMP locks with NUM_PARALLEL
1 year ago
Mark Seminatore
b29fd48998
Merge branch 'develop' into win_tidy
1 year ago
Mark Seminatore
98c56a7314
more cleanup
1 year ago
Chip Kerchner
d408ecedba
Add environment variable to display coretype for dynamic arch.
1 year ago
Chip Kerchner
ac6b4b7aa4
Make sure CPU ID works for all POWER_10 conditions
1 year ago
Chip Kerchner
08ce6b1c1c
Add missing CPU ID definitions for old versions of AIX.
1 year ago
Martin Kroeker
a4fde2c5ac
Merge pull request #4451 from martin-frbg/overflow_reset
Reset "buffer management structure overflowed" state and free auxiliary struct on blas_shutdown
1 year ago
Martin Kroeker
e61d96303d
Fix missing NO_AVX2 fallback for SapphireRapids
1 year ago
Mark Seminatore
42cb567f0f
more cleanup
1 year ago
Mark Seminatore
0d7fe5ea61
clean up whitespace
1 year ago
Martin Kroeker
d938aed7fe
reset "mem structure overflowed" state on shutdown
1 year ago
Chris Sidebottom
aaf65210cc
Add dynamic support for Arm(R) Neoverse(TM) V2 processor
Whilst I figure out how best to map the L2 parameters without
duplicating all of `ARMV8SVE`, lets just map this to `NEOVERSEV1`.
1 year ago
Martin Kroeker
152a6c43b6
Add blas_omp_threads_local
1 year ago