Improve consistency in test names across different APIs.
In this commit, standardize storage, side, uplo, trans
diag and conj in test names.
AMD-Internal: [CPUPL-4500]
Change-Id: Ifcdb6e9f684b134841d86087218d7aefd9cabe63
- Reduced number of jump operations in AVX512
assembly kernel for SCOPYV, DCOPYV and ZCOPYV.
- Fixed memory test failure for bli_zcopyv_zen_int_avx512
kernel.
- Replaced existing AVX2 COPYV intrinsic kernels in
bli_cntx_init_zen5.c with AVX512 assembly kernels.
Change-Id: Idc11601b526d6d82cfbdf63af2fd331918b31159
Some BLAS routines do not require matrices or vectors to be
initialized in certain use cases. For example, in GEMM when
beta=zero, C is set rather than updated, thus input values of
C should not be used. In these cases set the inital values of
such matrices or vectors to an extreme value, to help detect
if these are incorrectly being read.
The extreme value can be NaN or Inf. The default is Inf,
change it by running
cmake ... -DEXT_VALUE=NaN
AMD-Internal: [CPUPL-4548]
Change-Id: I4a665363779d2496b8247f6357e970b7f23cd1eb
- Utilized the memory testing feature in GTestsuite
to update the testing interfaces for micro-kernel
testing of SCOPY, DCOPY and ZCOPY APIs.
Change-Id: I3d6905f33b000b8d5e60727aa896bd869f4f441f
- Existing vectorizes code was disabled because
of the failures observed in matlab tests.
- The issue is caused by underflow during division when diagonal
elements of A matrix are very small.
- When diagonal is very small (4E-324 in case of matlab), sqauring the
diagonal during divison causes the square to be rounded off to zero.
- Fix is to normalise (ar) and (ai) by dividing (ar) and (ai) by
max(ar, ai), this will make either (ar) or (ai) 1, and hence
reduce the likelihood of underflow.
AMD-Internal: [CPUPL-5052]
Change-Id: Iff7893fdcb92907a12e6af8e102a92637a13ce4f
- Added accuracy and memory tests for AVX2 and AVX512 ?SETV kernels,
AVX512 ZAXPYV kernel and AVX512 ZAXPYF kernels, with fuse-factors
2, 4 and 8.
- Cleanup of the code-section that declares and defines the reference
compute for AXPYF operation. Corrected the type mismatch with the
arguments that reference AXPYV would expect(this is used to decompose
AXPYF as part of reference). Ensured usage of GTestSuite's internal
alias for integer types.
- Updated the API level testsuite and testing interface for AXPYF,
based on the cleaup done to the reference code.
AMD-Internal: [CPUPL-4974]
Change-Id: I71de6c09d3877cd3dd1eaa20ab4f90e7c33eb1e1
Test programs for key APIs like GEMM take a long time to run,
and even to generate the list of test cases. Break into
separate test programs for different data types to enable
these to run in parallel (at gtest level). In this patch
we break up GEMM, TRSM, GEMV and TRSV.
AMD-Internal: [CPUPL-4500]
Change-Id: I21363b050d30e0402d5a1e8cbeaed2ebcc87aaeb
AOCL libFLAME optimizations directly call some internal
BLIS symbols. Export them to enable this to work with
the BLIS shared library.
AMD-Internal: [CPUPL-5044]
Change-Id: Icb62dcb51e12d72dde8434593ab17de3c227c93d
- Updated DOTV Gtestsuite interface to invoke C/ZDOTC when conjx='c'
and testing interface is either BLAS or CBLAS.
- Added ukr tests for bli_zdotv_zen4_asm_avx512( ... ) and
bli_zdotv_zen_int_avx512( ... ) kernels.
AMD-Internal: [CPUPL-5011]
Change-Id: I32fb69027a35d9ea92f997a095d412c8242a4b68
* Functional tests are covered for saxpyv and zaxpyv.
* As part of functional large size of m, stride greater than m, scalar
combinations(including special cases), Zero increment tests are
added for saxpyv and zaxpyv.
Signed-off-by: eseswari <sangadala.eswari@amd.com>
AMD-Internal: CPUPL-4413
Change-Id: I61473357680cb0f394e6e653796ec31110895fa4
- Added AVX512 kernel for ZDOTV.
- Multithreaded both ZDOTC and ZDOTU with AOCL_DYNAMIC support.
AMD-Internal: [CPUPL-5011]
Change-Id: I56df9c07ab3b8df06267a99835b088dcada81bd8
* As part of functional test cases, large size of m, stride greater than
m,scalar combinations, Zero increment tests are added for ?copyv.
Signed-off-by: eseswari <sangadala.eswari@amd.com>
AMD-Internal: CPUPL-4412
Change-Id: I9fa74c147975bbe21263aaf48190170c6ed0a8fd
- Before the system was assuming 3 levels in the directory structure and
was creating corresponding targets.
- Now the system looks into the subdirectories of testsuite and creates
a target for each subdirectory that has at least one cpp file.
- Also deleted a directory that seems duplicate and was breaking builds.
AMD-Internal: [CPUPL-4500]
Change-Id: I03ca362b09783f1c7c5f37ab420d8ca2c2b45e2e
Existing Design:
- GEMM AVX2 kernel performs computation and updates temporary C buffer
- Portion of temporary C buffer is copied to output C buffer
based on UPLO parameter
- For diagonal blocks, using GEMM kernels is not efficient
New Design: Implemented in current patch when UPLO='L'
- GEMMT kernel used for computation, temporary buffer is not required.
- Only required elements are computed using mask load store for all
fringe cases
- Exception: AVX2 code path is used when storage format is RRC, CRR, CRC
- AOCL-Dynamic is added based on dimension
- Check for AVX platform is added in SUP interface, It returns to
native implementation if hardware doesnot support AVX platform
- SUP ref_var2m is expanded for dcomplex datatype to avoid condition
check which exists for double datatype
AMD_Internal: [CPUPL-5006]
Change-Id: I3e21404b732b8f2df9cbdba394303752fdf36286
1. The 5 LOOP LPGEMM path is in-efficient when A or B is a vector
(i.e, m == 1 or n == 1).
2. An efficient implementation is developed considering the b matrix
reorder in case of m=1 and post-ops fusion.
3. When m = 1 the algorithm divide the GEMM workload in n dimension
intelligently at a granularity of NR. Each thread work on A:1xk
B:kx(>=NR) and produce C=1x(>NR). K is unrolled by 4 along with
remainder loop.
4. When n = 1 the algorithm divide the GEMM workload in m dimension
intelligently at a granularity of MR. Each thread work on A:(>=MR)xk
B:kx1 and produce C = (>=MR)x1. When n=1 reordering of B is avoided
to efficiently process in n one kernel.
AMD-Internal: [SWLCSG-2355]
Change-Id: I7497dad4c293587cbc171a5998b9f2817a4db880
Thanks to Zhenyu Zhu ajz34 for pointing out this bug.
When trans="t" or "conjugate transpose" in the case of complex data-types
the ldb should be greater than equal to cols.
In the bug it was checked against "rows". Fixed this bug.
Some minor code format is done.
[CPUPL-4810][SWLCSG-2706]
Change-Id: Ie796d25a361b2ba72eda80e8c5867d6352af901f
- In AVX512 ZTRSM kernel, vertorizes division code
is causing failures in matlab.
- The logic is identical in reference C code and intrinsics code,
but intrinsics code is causing failure
- Replaced optimized intrinsics code with C code.
AMD-Internal: [CPUPL-5052]
Change-Id: Iea184330b22c46d979867b870486066ef980eb84
- Modified all structs that are passed to JIT-generated code to use
integer of type uint64_t rather than dim_t so that functionality
is not affected when size of BLIS-internal integer is modified
during configure time.
Change-Id: Ib81c088072badf13da4ca73be2d4af4551b713d8
- Added a {} around zen4 switch case to avoid AOCC error.
- Error is caused because in C declarations are not a statement, therefore
they cannot be labled hence compiler is not able to create a lable
for jump.
AMD-Internal: [CPUPL-4880]
Change-Id: Icfeedafd80bf9a955e430ca967b6a93dcbbf075e
- Updated the AVX512 DOTXF kernels to use MASKZ loads
instead of MASK loads when loading X vector in fringe
case. This avoids compiler warnings of uninitialized
vector as input to the intrinsic.
- The functionality will not change when using either MASK
or MASKZ loads on X, since A matrix is loaded using MASKZ
loads.
AMD-Internal: [CPUPL-4974]
Change-Id: I1ef98a1292352d0e905cc09cd5667acd883df827
Check internal value of INFO for BLAS2 and BLAS3 routines
using the bli_info_get_info_value() function added in AOCL 4.2.
If testing a BLIS library that does not have this, use
cmake ... -DCAN_TEST_INFO_VALUE=OFF
AMD-Internal: [CPUPL-4993]
Change-Id: Ida5d252b0f6727793ebfb74bb160e8cb96b61b74
- In DGEMMT SUP AVX2 code path, traingular kernels
are added in order to avoid temporary C buffer.
- Since these kernels did not exist for AVX512,
AVX2 kernels were being used in GEMMT.
- AVX512 triangular GEMM kernel has been added
to make sure that AVX512 kernels can be used without
creating a temporary buffer.
- This kernel is added only for Lower variant of GEMMT,
for upper variant of DGEMMT, temporary C buffer is
created, full GEMM kernel is called on temporary C and
traingular region from temporary C is copied to C
buffer.
AMD-Internal: [CPUPL-4881]
Change-Id: Id70645f79ae078ab9a7006e83d328505f1fae8a9
- Kernel dimensions are 4x4.
- Two kernels are implemented, Right Upper and
Right lower.
- In case of Left variants of TRSM, transpose is
induced so that Right variant kernels can be used.
- No packing is performed in these kernels.
- Changes are made in the threshold to pick ZTRSM small
code path.
- BLIS_INLINE is removed from signature of
"TRSMSMALL_KER_PROT".
- These kernels do not support "ENABLE_TRSM_PREINVERSION".
- Newly added kernels do not support conjugate
transpose.
- Added multithreading to ZTRSM small code path.
AMD-Internal: [CPUPL-4324]
Change-Id: I683b1d5239593e54f433e7f27497d72dfbd9141c
- Added DAXPYF and DDOTXF AVX512 kernels.
- Fuse factor for ddotxf kernel is 8.
- 2 DAXPYF kernels are added, with fuse
factor 8 and 32.
- Multithreading is also added to the DAXPYf
kernel with fuse factor 32.
- These kernels are internally used by TRSM.
- Added changes in TRSV to call these kernels
in ZEN4
AMD-Internal: [CPUPL-4880]
Change-Id: I12850de974b437bbca07677b68bc3d6a35858770
- Implemented AVX512 kernels for handling the calls to ZGEMV
with transpose to A matrix.
- This includes the set of ZDOTXF and ZDOTXV kernels. ZDOTXF
kernels include those with fuse-factor 8 (main kernel), 4
and 2(fringe kernels).
- Updated the bli_zgemv_unf_var1( ... ) function to update
the function pointers to these kernels, based on the
configuration.
AMD-Internal: [CPUPL-4974]
Change-Id: I313ae0abe9dc119de849da42f9825b71f11b1fda
- Implemented AVX512 kernels for handling the calls to ZGEMV
with no-transpose to A matrix.
- This includes the ZAXPYF, ZAXPYV and ZSETV kernels.
The set of ZAXPYF kernels include those with fuse-factor 8
(main kernel), 4 and 2(fringe kernels).
- Updated the bli_zgemv_unf_var2( ... ) function to set
the function pointers to these kernels, based on the
configuration. Further added the call to ZSETV at this
layer in case beta is 0.
AMD-Internal: [CPUPL-4974]
Change-Id: Iee4b724719e49023138bb16479765be44d677cd9
- Using a template class for the printing operator that depends
on the type.
- USe a macro to denote which interface is being tested.
AMD-Internal: [CPUPL-4500]
Change-Id: I453c4ef4842c354064f49ff32ec4bf42920cc17c
Following a recent change to the data generators to allow a stride
to be specified (60cc23f3d3), seg
faults can occur if m<=0 for column storage or n<=0 for row storage.
Prevent this by having separarate code paths to handle these
scenarios.
AMD-Internal: [CPUPL-4500]
Change-Id: I23ed8b2dccaaca140e2ddfda45bcdb4c888d5708
Improve consistency in test names across different APIs.
In this commit, standardize m, n, k and b in test names.
AMD-Internal: [CPUPL-4500]
Change-Id: I53e7dd83cbf426ab1ebe8aa4af1da01594f4af23
- Implemented AVX512 kernels for scopyv_, dcopyv_ and zcopyv_
using respective AVX512 intrinsics including masked
load and store operations.
- Implemented AVX512 kernels for scopy_, dcopy_ and
zcopy_ using assembly language to prevent loss of
performance during the translation of intrinsics.
- Updated the dcopy_blis_impl( ... ) and
zcopy_blis_impl( ... ) function to support
multithreaded calls to the respective computational
kernels, if and when the OpenMP support is enabled.
- Implemented OpenMP parallelization for dcopyv_ and
zcopyv_ APIs, while scopyv_ and ccopyv_ only support
single thread.
AMD-Internal: [CPUPL-4854]
Change-Id: I5fbd0bcca4e59001fbe2b1168b624d0c33242b3e
- Updated the IIT_ERS tests for SUBV to avoid using undefined
variables. These tests are enabled only when GTestSuite is
configured for BLIS_TYPED interface testing.
- Updated an instantiator in DAXPBY accuracy tests, to avoid
parsing error(extra comma). These tests are enabled only when
GTestSuite is configured for BLIS_TYPED interface.
AMD-Internal: [CPUPL-4500]
Change-Id: If6894daadbbc353dd66968649642ff07fa663782
Return type of xerbla and xerbla_array APIs are defined as int in BLIS, but according to netlib it should be void. Updated the defination and declaration accordingly.
Signed-off-by: Sridhar Govindaswamy <Sridhar.Govindaswamy@amd.com>
Change-Id: I3072ba76111189de5c5cf08df83ea154163dd34d
First in a series of commits to improve consistency in test names
across different APIs. This will help with gtest filtering.
In this commit, standardize alpha, beta, incx and incy.
AMD-Internal: [CPUPL-4500]
Change-Id: I0cde85f9a4cf969c0b12ac589b232786ad011f09
Details:
- variable m0 is being loaded into a register without typecasting
it to uint64_t. This resulted in seg-fault when int size is set
to be 32 bits during configure time.
- Any variable that is loaded using mov in assembly needs to be
typecasted to uint64_t before begin_asm, so that change in size
of integer doesn't affect the functionality.
- Modified all instances using variable m0 to use variable 'm' where
m = (uint64_t)m0;
AMD-Internal: [CPUPL-4971]
Change-Id: I49b66d2cacf19ace40ab44c9f85904644e8921f4
- Updated test_gemv.h to pass the right boolean
to computediff( ... ), based on whether we run
it for exception value tests or not.
AMD-Internal: [CPUPL-4500]
Change-Id: I1ad2cde4f9b4bb1dadc32d1f7d02a90a457e218f
*covered large sizes, scalar combinations and strides greater than the
size for cger, dger, sger and zger.
Signed-off-by: Sangadala Eswari <Sangadala.Eswari@amd.com>
AMD-Internal: CPUPL-4414
Change-Id: I6fba26a35903d1f6dbd713f19eac6bb537b3d8d2
- Changed the macro guard for accuracy tests of SIMATCOPY,
to ensure that tests are enabled/disabled based on the reference.
- Updated test_gemv.h to make sure the contents of y vector is copied
to y_ref post inducing exception values.
AMD-Internal: [CPUPL-4500]
Change-Id: I7249e643677e7e493eba5d072567615bc913a532
Add name of variable being tested in error output from
computediff functions. First step to adding (optional)
tests on input arguments.
AMD-Internal: [CPUPL-4379]
Change-Id: I9553b660bcf5ecf1dd675cb837655078933455ac
Modify thresholds to reflect number of operations that
accumulate results into each output element. Different
limits are set for early return and special cases.
Constants are still subject to experimentation and change.
AMD-Internal: [CPUPL-4378]
Change-Id: I81f63a36c161ff1866f2d404b9e3cbb9a2948d3a
Modify thresholds to reflect number of operations that
accumulate results into each output element. Different
limits are set for early return and special cases.
Constants are still subject to experimentation and change.
AMD-Internal: [CPUPL-4378]
Change-Id: I03cd8901e574f2e44e85ce8b0bc234e36edb4819
Correct the order of tests in bli_cpuid.c to test all known zen
AVX512 platforms before considering fallback tests on AVX512
support. This avoids builds with "configure auto" or
"cmake -DBLIS_CONFIG_FAMILY=auto" incorrectly selecting zen5
sub-configuration on zen4 systems.
AMD-Internal: [CPUPL-4966]
Change-Id: I8706382e2df7c9ae4bb456e3a7f465053e15beea
1. Fixed issue related to linking reference library.
2. Clean-up of how reference library variables are set.
2. Compilation error related to std::max() and std::min().
AMD-Internal: [CPUPL-4879]
Change-Id: I427a4a4c0ea56a340a8bbd1a6649252e9680b937
- Added Memory Access Test support for GEMV.
- Added Extreme Value Tests for various combinations of NaN, Inf and
-Inf for ?GEMV.
- Also fixed some invalid IIT_ERS tests.
AMD-Internal: [CPUPL-4825]
Change-Id: Iee77b305f6c6b9427153fbbc5191176dae9fbfea
- In 2x1 fringe case in [RUN/RLT] kernel, 3 scomplex
precision numbers are being read instead of 1 scomplex.
- Fixed the code to read only one scomplex.
AMD-Internal: [CPUPL-4403]
Change-Id: If3ac03ed864618382d3a382a8cdff7ff8a94eb7d
- Scaling vector X is skipped when alpha is 1 in ZTRSV.
- Scaling matrix A is skipped when alpha is 1 in ZTRSM.
AMD-Internal: [CPUPL-4324]
Change-Id: I03c5a454ed1f5be36dac0f121408749bfc9cfc81
- Comparision using bli_eqsc is slower than direct comparison.
- Changed comparision logic for 1x1 matrix
from bli_sqsc to direct comparision.
AMD-Internal: [CPUPL-4324]
Change-Id: Ifb2d0ad7a97c8bf33b66d624a7ecc53e38c1c803
Correction to commit 2450a1813b
to add -DBLIS_CONFIG_FAMILY=zen5 support in cmake.
AMD-Internal: [CPUPL-3518]
Change-Id: Iecff2b64d5d95960cecbbf98d5269133747b122e
