- Implemented multithreading framework for the DGEMV API on Zen architectures. Architecture specific AOCL-dynamic logic determines the optimal number of threads for improved performance.
- The condition check for the value of beta is optimized by utilizing masked operations. The mask value is set based on value of beta, and the masked operations are applied when the vector y is loaded or scaled with beta.
AMD-Internal: [CPUPL-6746]
- Introducted new assembly kernel that copies data from source
to destination from the front and back of the vector at the
same time. This kernel provides better performance for larger
input sizes.
- Added a wrapper function responsible for selecting the kernel
used by DCOPYV API to handle the given input for zen5
architecture.
- Updated AOCL-dynamic threshold for DCOPYV API in zen4 and
zen5 architectures.
- New unit-tests were included in the grestsuite for the new
kernel.
AMD-Internal: [CPUPL-6650]
Change-Id: Ie2af88b8e97196b6aa02c089e59247742002f568
- Added a set of AVX512 fringe kernels(using masked loads and
stores) in order to avoid rerouting to the GEMV typed API
interface(when m = 1). This ensures uniformity in performance
across the main and fringe cases, when the calls are multithreaded.
- Further tuned the thresholds to decide between ZGEMM Tiny, Small
SUP and Native paths for ZEN4 and ZEN5 architectures(in case
of parallel execution). This would account for additional
combinations of the input dimensions.
- Moved the call to Tiny-ZGEMM before the BLIS object creation,
since this code-path operates on raw buffers.
- Added the necessary test-cases for functional and memory testing
of the newly added kernels.
AMD-Internal: [CPUPL-6378][CPUPL-6661]
Change-Id: I9af73d1b6ef82b26503d4fc373111132aee3afd6
- Implemented an AVX512 rank-1 kernel that is
expected to handle column-major storage schemes
of A, B and C(without transposition) when k = 1.
- This kernel is single-threaded, and acts as a direct
call from the BLAS layer for its compatible inputs.
- Defined custom BLAS and BLIS_IMPLI layers for CGEMM
(instead of using the macro definition), in order to
integrate the call to this kernel at runtime(based on
the corresponding architecture and input constraints).
- Added unit-tests for functional and memory testing of the
kernel.
- Updated the ZEN5 context to include the AVX512 CGEMM
SUP kernels, with its cache-blocking parameters.
AMD-Internal: [CPUPL-6498]
Change-Id: I42a66c424325bd117ceb38970726a05e2896a46b
- Implemented the following AVX512 SUP
column-preferential kernels(m-variant) for CGEMM :
Main kernel : 24x4m
Fringe kernels : 24x3m, 24x2m, 24x1m,
16x4, 16x3, 16x2, 16x1,
8x4, 8x3, 8x2, 8x1,
fx4, fx3, fx2, fx1(where 0<f<8).
- Utlized the packing kernel to pack A when
handling inputs with CRC storage scheme. This
would in turn handle RRC with operation transpose
in the framework layer.
- Further adding C prefetching to the main kernel,
and updated the cache-blocking parameters for
ZEN4 and ZEN5 contexts.
- Added a set of decision logics to choose between
SUP and Native AVX512 code-paths for ZEN4 and ZEN5
architectures.
- Updated the testing interface for complex GEMMSUP
to accept the kernel dimension(MR) as a parameter, in
order to set the appropriate panel stride for functional
and memory testing. Also updated the existing instantiators
to send their kernel dimensions as a parameter.
- Added unit tests for functional and memory testing of these
newly added kernels.
AMD-Internal: [CPUPL-6498]
Change-Id: Ie79d3d0dc7eed7edf30d8d4f74b888135f31d6b4
- Included a new code section to handle input having non-unit strided y
vector for dgemv transpose case. Removed the same from the respective
kernels to avoid repeated branching caused by condition checks within
the 'for' loop.
- The condition check for beta is equal to zero in the primary kernels
are moved outside the for loop to avoid repeated branching.
- The '_mm512_reduce_pd' operations in the primary kernel is replaced by
a series of operations to reduce the number of instructions required
to reduce the 8 registers.
- Changing naming convention for DGEMV transpose kernels.
- Modified unit kernel test to avoid y increment for dgemv tranpose
kernels during the test.
AMD-Internal: [CPUPL-6565]
Change-Id: I1ac516d6b8f156ac53ac9f6eb18badd50e152e05
- Implemented the following AVX512 native
computational kernels for CGEMM :
Row-preferential : 4x24
Column-preferential : 24x4
- The implementations use a common set of macros,
defined in a separate header. This is due to the
fact that the implementations differ solely on
the matrix chosen for load/broadcast operations.
- Added the associated AVX512 based packing kernels,
packing 24xk and 4xk panels of input.
- Registered the column-preferential kernel(24x4) in
ZEN4 and ZEN5 contexts. Further updated the cache-blocking
parameters.
- Removed redundant BLIS object creation and its contingencies
in the native micro-kernel testing interface(for complex types).
Added the required unit-tests for memory and functionality
checks of the new kernels.
AMD-Interal: [CPUPL-6498]
Change-Id: I520ff17dba4c2f9bc277bf33ba9ab4384408ffe1
Optionally enable parallelism inside gtestsuite so that we can
check BLIS functions perform correctly when nested parallelism
is in operation. Enable with:
cmake ... -DOPENMP_NESTED={0,1,2,1diff}
where in gtestsuite
- 0 is the default choice with no parallelism.
- 1 and 2 are simple nested parallelism.
- 1diff has one level of parallelism setting different numbers
of threads to be used by BLIS and reference library calls
from each gtestsuite thread.
Note: OMP_NUM_THREADS must be set appropriately to enable or
disable parallelism at each level in the test programs
as desired.
OMP_NUM_THREADS will also define the parallelism used
within the BLIS library (if it is multithreaded), unless
BLIS-specific ways of specifying parallelism have been
used. If a BLIS-specific parallelism option has been set,
the same mechanism will be used in the 1diff option to
vary the number of threads in BLIS per application thread.
AMD-Internal: [CPUPL-3902]
Change-Id: I89f9edb4125c64ef03e025a9f6ccb84960ba8771
More changes to standardize copyright formatting and correct years
for some files modified in recent commits.
AMD-Internal: [CPUPL-5895]
Change-Id: Ie95d599710c1e0605f14bbf71467ca5f5352af12
- Added 32x3n n-biased kernels to directly handle the cases where n=3
which were earlier being handled by the primary n-biased, 32x8n,
kernel.
- Modified the n-biased fringe kernels to further handle the smaller
m-fringe cases. Thus, now the kernels handle the following range of m
for any value of n:
- 16x8n : m = [16, 31)
- 8x8n : m = [8, 15)
- m_leftx8n : m = [1, 7]
- Updated the function pointer map for n-biased kernels with added
granularity to invoke the smaller fringe cases directly on the basis
of m-dimension.
- Added micro-kernel unit tests for all the dgemv_n kernels.
AMD-Internal: [CPUPL-6231]
Change-Id: Ibe88848c2c1bbb65b3e79fbc90a2800dc15f5119
- Added new ZTRSM kernels for right and left variants.
- Kernel dimensions are 12x4.
- 12x4 ZGEMM SUP kernels are used internally
for solving GEMM subproblem.
- These kernels do not support conjugate transpose.
- Only column major inputs are supported.
- Tuned thresholds to pick efficent code path for ZEN5.
AMD-Internal: [CPUPL-6356]
Change-Id: I33ba3d337b0fcd972ca9cfe4668cb23d2b279b6e
- Added support for gemv kernels unit test in gtestsuite.
- Added micro-kernel tests and memory tests for DGEMV
transpose case kernels.
AMD-Internal: [CPUPL-5835]
Change-Id: I7d2d3cdbfea436f6c9b2cce9f2e85bfc5c51f201
- Restore test for old bli_dgemm_zen4_asm_8x24 kernel, so that
we can test this if linking with older AOCL versions.
- Move K_bli_dgemm_avx512_asm_8x24 definition from AOCL_42 list
to AOCL_50 list.
AMD-Internal: [CPUPL-4500]
Change-Id: Id522f4bc5b89e86f77c4e1d26c75e261736ab450
This patch introduces comprehensive optimizations to the DGEMM kernel, focusing on loop
efficiency and edge kernel performance. The following technical improvements have been implemented:
1. **IR Loop Optimization:**
- The IR loop has been re-implemented in hand-written assembly to eliminate the overhead associated
with `begin_asm` and `end_asm` calls, resulting in more efficient execution.
2. **JR Loop Integration:**
- The JR loop is now incorporated into the micro kernel. This integration avoids the repetitive overhead
of stack frame management for each JR iteration, thereby enhancing loop performance.
3. **Kernel Decomposition Strategy:**
- The m dimension is decomposed into specific sizes: 20, 18, 17, 16, 12, 11, 10, 9, 8, 4, 2, and 1.
- For remaining cases, masked variants of edge kernels are utilized to handle the decomposition efficiently.
1. **Interleaved Scaling by Alpha:**
- Scaling by the alpha factor is interleaved with load instructions to optimize the instruction pipeline
and reduce latency.
2. **Efficient Mask Preparation:**
- Masks are prepared within inline assembly code only at points where masked load-store operations are necessary,
minimizing unnecessary overhead.
3. **Broadcast Instruction Optimization:**
- In edge kernels where each FMA (Fused Multiply-Add) operation requires a broadcast without subsequent reuse,
the broadcast instruction is replaced with `mem_1to8`.
- This allows the compiler to optimize by assigning separate vector registers for broadcasting, thus avoiding
dependency chains and improving execution efficiency.
4. **C Matrix Update Optimization:**
- During the update of the C matrix in edge kernels, columns are pre-loaded into multiple vector registers.
This approach breaks dependency chains during FMA operations following the scaling by alpha, thereby mitigating
performance bottlenecks and enhancing throughput.
These optimizations collectively improve the performance of the DGEMM kernel, particularly in handling edge cases and
reducing overhead in critical loops. The changes are expected to yield significant performance gains in matrix multiplication
operations.
This patch also involves changes for tiny gemm interface. A light
interface for calling kernels and removing calls to avx2 dgemm kernels
as we use avx512 dgemm kernels for all the sizes for zen4 and zen5.
For zen4 and zen5 when A matrix transposed(CRC, RRC), tiny kernel does not have
the support to handle such inputs and thus such inputs are routed to
gemm_small path.
AMD-Internal: [CPUPL-6054]
Change-Id: I57b430f9969ca39aa111b54fa169e4225b900c4a
- Merged ZEN4 and ZEN5 DGEMM 8x24 kernel.
- Replaced 32x6 kernel with 8x24. Now same
kernel is used for ZEN4 and ZEN5.
- Blocksizes have been tuned for genoa only.
- DGEMM kernel for DTRSM native code path
is replaced with 8x24 kernel.
- Enabled alpha scaling during packing for ZEN4.
- ZEN4 8x24 kernel has been removed.
AMD-Internal: [CPUPL-5912]
Change-Id: I89a16a7e3355af037d21d453aabf53c5ecccb754
Add definitions in gtestsuite header to list available kernel
by AOCL BLIS version. Check these definitions in ukr test
programs to avoid missing symbol errors when testing with an
older version of BLIS.
Currently AOCL_41, AOCL_42, AOCL_50 and AOCL_DEV are supported,
with AOCL_DEV inferred from the version being later than the
value of AOCL_BLAS_LATEST_VERSION set in CMakeLists.txt. Thanks
to Eleni Vlachopoulou for the cmake functionality to automatically
detect the version from the library.
AMD-Internal: [CPUPL-4500]
Change-Id: I40ffd3d3789324fbb1dabfbf5e1dd4e0c94d54d9
- Correct matsize and NumericalComparison functions for
tests with first matrix dimension <= 0.
- BLAS1:
- Fix for BLAS vs CBLAS differences in amaxv IIT_ERS tests.
- Threshold adjustments in ddotxf and zaxpy.
- Break axpyv and scalv into separate executables for
each data type.
- BLAS2:
- Threshold adjustments in symv and hemv.
- Break ger into separate executables for each data type.
- UKR:
- Break gemm and trsm ukr test into separate executables
for each data type.
- Threshold adjustments in daxpyf
- Disable {z,c}trsm ukr tests when BLIS_INT_ELEMENT_TYPE
is used, as matrix generator is not currently suitable
for this.
AMD-Internal: [CPUPL-4500]
Change-Id: I1d9e7acc11025f1478b8b511c14def5517ef0ae6
Add BLAS_TEST_IMPL option for TEST_INTERFACE to test the
wrapper layer underneath BLAS and CBLAS interfaces. This is
particularly useful if building a BLIS library with these
interfaces disabled, e.g.
./configure --disable-blas amdzen
or
cmake . -DENABLE_BLAS=OFF -DBLIS_CONFIG_FAMILY=amdzen
The ?_blis_impl wrappers should have the same arguments as the
BLAS interfaces, thus we define TEST_BLAS_LIKE as an additional
definition for convenience when selecting tests and options in
the C++ files.
AMD-Internal: [CPUPL-5650]
Change-Id: I0275a387563f3efc2b40029950c8569956f2df7b
Corrections for spelling and other mistakes in code comments
and doc files.
AMD-Internal: [CPUPL-4500]
Change-Id: I33e28932b0e26bbed850c55602dee12fd002da7f
Add tests to check input arguments have not been modified by BLIS
routine. These tests add a large runtime overhead, so they are
disabled by default. To enable them, configure gtestsuite with:
cmake -DTEST_INPUT_ARGS=ON ...
and run desired tests as normal.
Also:
- Correct testinghelpers::chktrans to handle upper case values of
argument trns.
- Change testinghelpers::matsize to return size 0 if m, n or
leading dimension are 0, or if leading dimension is too small.
AMD-Internal: [CPUPL-4379]
Change-Id: I9494af800f9383195272ce99f622104a38fd0ed8
Various improvements:
- Where appropriate, test both:
- with nullptr for suitable arguments that should never
be touched.
- with all arguments correct except the one we want to
test, to check we are not returning early because
another argument is a nullptr.
- Test incorrect values for order argument in CBLAS calls.
- Test early exits with limited data changes, e.g. set
C to 0 or scale C in GEMM when alpha = 0.
- Bugfix in gemmt test when alpha is 0 and beta is 1.
- Use reference library gemmt for comparison when library
is not netlib BLAS.
AMD-Internal: [CPUPL-4500]
Change-Id: Ibde7eaba5a484a87674044ca44855c6f6ee4ff4b
Unnecessary whitespace (spaces, tabs) at the end of lines
has been removed.
AMD-Internal: [CPUPL-4500]
Change-Id: Ice5f5504232cb22460c14ac47e6a3a43309cba22
- Updated the existing code-path for ?AXPBYV to
reroute the inputs to the appropriate L1 kernel,
based on the alpha and beta value. This is done
in order to utilize sensible optimizations with
regards to the compute and memory operations.
- Updated the typed API interface for ?AXPBYV to include
an early exit condition(when n is 0, or when alpha is
0 and beta is 1). Further updated this layer to query
the right kernel from context, based on the input values
of alpha and beta.
- Added the necessary L1 vector kernels(i.e, ?SETV, ?ADDV,
?SCALV, ?SCAL2V and ?COPYV) to be used as part of special
case handling in ?AXPBYV.
- Moved the early return with negative increments from ?SCAL2V
kernels to its typed API interface.
- Updated the zen, zen2 and zen3 context to include function
pointers for all these vector kernels.
- Updated the existing ?AXPBYV vector kernels to handle only
the required computation. Additional cleanup was done to
these kernels.
- Added accuracy and memory tests for AVX2 kernels of ?SETV
?COPYV, ?ADDV, ?SCALV, ?SCAL2V, ?AXPYV and ?AXPBYV APIs
- Updated the existing thresholds in ?AXPBYV tests for complex
types. This is due to the fact that every complex multiplication
involves two mul ops and one add op. Further added test-cases
for API level accuracy check, that includes special cases of
alpha and beta.
- Decomposed the reference call to ?AXPBYV with several other
L1 BLAS APIs(in case of the reference not supporting its own
?AXPBYV API). The decomposition is done to match the exact
operations that is done in BLIS based on alpha and/or beta
values. This ensures that we test for our own compliance.
AMD-Internal: [CPUPL-4861]
Change-Id: Ia6d48f12f059f52b31c0bef6c75f47fd364952c6
- Fixed bug in ddotxf generic tests where the parameters lda_inc and
inca were being read incorrectly.
- Fixed bug in dotxf test wherein the y vector was being generated with
length m instead of b.
- Corrected function signatures to use type gtint_t instead of gint_t.
- Updated the tests to use conjugate values of type char and convert to
conj_t type only while invoking BLIS tests for both DOTXF and AXPYF.
AMD-Internal: [CPUPL-5117]
Change-Id: I0ef7af429057583a1cbf34827802e72401181caf
- Correct value of alpha in ger ERS test.
- rename ERS_IIT.cpp files to match naming convention
used for other APIs.
- Change all cases of gint_t to gtint_t except for
dotxf, which is fixed in another commit.
- Add TEST_UPPERCASE_ARGS to imatcopy and omatcopy{2}
headers.
- Corrected typo.
AMD-Internal: [CPUPL-4500]
Change-Id: I8844bb8c5941785e64daa9df5569092c19f91838
Correction to commit 8657e661fc
to allocate matrix or vector correctly when special read-only
case occurs.
Also define a set_matrix generator for symmetric matrices
to only set upper or lower triangle to the supplied value,
while setting the unused elements to a large value to help
catch incorrect access to those elements.
AMD-Internal: [CPUPL-4548]
Change-Id: I22b3a20e2ce8be70eb27179247cd47fdb2d87b9d
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
- 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
- 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
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
Previous commit introduced an infinite recursion problem in
generators for symmetric matrices. This was reported as a
compiler warning by gcc 12.2 but not by gcc 11.4.
AMD-Internal: [CPUPL-4862]
Change-Id: I8642b81a62f0643b5a9ebedb4fcc83b25542de1b
- Added test-cases to verify the functional behaviour
of the BLAS-extension API ?imatcopy_() and ?omatcopy2_().
The test-cases cover the following categories for the
supported datatypes :
- Functional and memory testing.
- Negative parameter testing with invalid inputs.
- Early return scenarios.
- Exception value testing.
- Updated functions in testinghelpers to include strides
in addition to leading-dimension, when initializing
a matrix. The default value for stride is set as 1.
- Implemented functions to load the reference symbol, based
on the choice of the reference library. The function definition
is overloaded due to different API standards being exposed by
different libraries.
- Code cleanup of files for ?OMATCOPY API.
AMD-Internal: [CPUPL-4862]
Change-Id: If63b348f517e2cde1fe48f3a195808b33a91c312
- Added overflow and underflow tests for dgemm
These tests cause floating point overflow and underflow by feeding
values close to DBL_MAX and DBL_MIN values to matrices
DBL_MAX = 1.7976931348623158e+308
DBL_MIN = 2.2250738585072014e-308
When computations result in values beyond the range [DBL_MIN, DBL_MAX],
it leads to an overflow or underflow condition
Two new arguments are added to test_gemm routine - over_under and input_range
over_under = 0 indicates overflow
over_under = 1 indicates underflow
input_range = -1 indicates values within overflow or underflow limits
input_range = 0 indicates values very close to DBL_MIN or DBL_MAX
input_range = 1 indicates values beyond DBL_MIN or DBL_MAX
- New file: dgemm_ovr_undr.cpp
Overflow and underflow tests are called from this file
dgemm_overflow and dgemm_underflow. This file uses
cfloat header file for DBL_MIN and DBL_MAX values
Signed-off-by: Nimmy Krishnan <nimmy.krishnan@amd.com>
AMD-Internal: [CPUPL-4492]
Change-Id: I4bbd519abacc56f322c73d6c0187ed6e1abbbf2b
- Added test-cases to verify the functional behaviour
of the BLAS-extension API ?omatcopy_(). The test-cases
cover the following categories for the supported datatypes :
- Functional and memory testing.
- Negative parameter testing with invalid inputs.
- Early return scenarios.
- Exception value testing.
- Implemented a function to load the reference symbol, based
on the choice of the reference library. The function definition
is overloaded due to different API standards being exposed by
different libraries.
AMD-Internal: [CPUPL-4810][SWLCSG-2706]
Change-Id: I8dcaeeaa36d392b752eb0685e32583a12ddc4220
- Handle -0.0 separately in get_value_string()
- Avoid unused variable warning when not TEST_BLIS_TYPED in
subv_evt_testing.cpp
- Remove unused variables in dgemm_ukernel.cpp
- Remove unnecessary local copies of greenzone1 in test
programs now that greenzone_1 and greenzone_2 will
not overlap.
- Protect tests of haswell kernels by ifdef on
BLIS_KERNELS_HASWELL rather than BLIS_KERNELS_ZEN.
- Added GTEST_ALLOW_UNINSTANTIATED_PARAMETERIZED_TEST
statements in TRSM kernel tests.
- Correct descriptions of trsm and trmm operations.
- Correct typos.
AMD-Internal: [CPUPL-4500]
Change-Id: If8520347e417785e6aa953a0c8a65d4f5f3c1591
- Added API tests
- Added Invalid input test cases (IIT).
- Added memory testing for SWAPV API.
- Added micro kernel testing for single and double precision
- Added reference swapv functionality in testinghelpers
- Added binary comparison method for two vectors with different
increments in check_error.h
AMD-Internal: [CPUPL-4814]
Change-Id: I32bcca51b4e998d51ede70869035da76a7f6dbca
- Added API tests for DTRSV.
- Added Extreme Value Test cases (EVT) for DTRSV.
- Tests for various combinations of INFs
and NANs for X vector and B matrix are added.
- Added Invalid input test cases (IIT).
- Added memory testing for DTRSV
kernels.
- Fixed a bug in alphax function where scaling
of a vector with a scalar was not handled
correctly when incx was negative.
AMD-Internal: [CPUPL-4715]
Change-Id: I84c873e98f845e05b11860e7ef6083d1184489b4
- Updating printing functionality for vectors and matrices.
- Adding macro definition checks so that GTestSuite builds successfully for shared libraries on zen3.
- Casting integers so that code builds for ILP64.
AMD-Internal: [CPUPL-4500]
Change-Id: I03afd08d5ad8ae50193d9559cf4ab8fc1d08753c
Modifications to testinghelpers::get_value_string() to allow
floating point values (e.g. for alpha and beta) to be used in
generating test names. Values will be generated in the form
1p3 or m2p4, or 3p0_4p5i for complex data. One decimal place
is currently enabled but this can be increased if needed. This
helps prevent duplicate test name errors when the list of values
for alpha or beta includes e.g. 1.0 and 1.3.
Also add support in testinghelpers::get_value_string() for
variables of type gtint_t.
AMD-Internal: [CPUPL-4500]
Change-Id: Icc8ca3c3cfacd7d46fffefee5a6e05452f704d4e
- dotxf is a blis specific kernel, which performs dotxv
operation but in multiple of fused factors to speed up
the operations.
- So dotxf reference function is implemented for gtestsuite,
where dotxf computation compared against computation done by
looping over dotxv function.
AMD-Internal: [CPUPL-4764]
Change-Id: I342dab066ceb1710649e54bb73afc5a23e2a8177
- axpyf is a blis specific kernel, which performs axpy
operation but in multiple of fused factors to speed up
the operations.
- So axpyf reference function is implemented for gtestsuite,
where axpyf computation compared against computation done by
looping over axpy function.
AMD-Internal: [CPUPL-4763]
Change-Id: I4713fd0b0d9e9cf688c9aaa82ac0e6ae07a05989
- Added API tests for STRSM.
- Added Extreme Value Test cases (EVT) for STRSM.
- Tests for various combinations of (+/-) INFs
and NANs in A and B matrix are added.
- Added micro kernel testing
- Added unit tests for small and native
path kernels.
- Added memory testing for STRSM
kernels.
- Edited the protected buffer in memory testing to
make sure that greenzone1 and greenzone2 do not
intersect.
AMD-Internal: [CPUPL-4640]
Change-Id: Ic48590d3b4ad12c4f2f6beaec2e1106a7aaa5213
- Updated test_scalv and ref_scalv templates for SCALV gtestsuite to
support unit-tests for mixed precision SCALV.
- Added unit-tests for the following kernels:
ZSCALV
- bli_zscalv_zen_int( ... )
ZDSCALV
- bli_zdscalv_zen_int10( ... )
- bli_zdscalv_zen_int_avx512( ... )
- Also, added API level unit-tests for the following cases:
- Unit Positive Increments
- Non-Unit Positive Increments
- Updated comments in DSCALV unit-tests with the correct kernel name.
AMD-Internal: [CPUPL-4624]
Change-Id: I96db8d3612687be07cd0e638a3119d41c3641ce8
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: Ic4540a2f1f6cd6380228b6a2884ac62850d6d8c6
- Added framework for memory testing.
- Out of bound reads and writes can be
detected in both C and assembly.
- Added memory tests for DTRSM.
- Test methodology:
- Use linux's protected pages to set some memory
before and after the required buffer as protected.
- Set the first and last page_size bytes as
read, write and execute protected (red_zones).
- If any part of code tries to read/write
in redzones, a SIGSEGV signal will be
generated, which can be used to detect a
out of bounds read and write.
- Page protection can only be set per page.
If required size for buffer is not a multiple
of pagesize we have to allocate more memory
than required in order make sure the start and
end of redzones align with page boundaries.
- Overwrite malloc(size) to allocate
'buffer_size+(2*pagesize)' where buffer_size =
minimum size such that buffer_size > 'size' and
buffer_size is multiple of pagesize.
- Use first and last page_size bytes of allocated
buffer as redzones, use first 'size' of the middle
buffer as first greenzone and last 'size' bytes as
second greenzone.
- Call test code once with first geenzone and then
with second greenzone. Greenzones are surrounded
by redzones, if test code read/writes before or after
greenzones, it will be detected.
|_____________________________________________________|
| red_zone1 | green_zone1 greenzone_2 | red_zone2|
|_____________________________________________________|
AMD-Internal: [CPUPL-4403]
Change-Id: Ic5c22a9adf8f833c77510686eee886485e894354
Add cmake option to override thresholds and set them all to zero.
In this case we don't switch to binary comparison as we want the
error to be calculated and printed. This functionality is intended for:
- Helping to determine or alter thresholds.
- To compare different max errors between different reference libraries.
- To test when we expect identical results, e.g. some comparisons of
BLIS vs BLIS.
To simplify coding, this is implemented by setting epsilon to zero
in the testinghelpers function.
AMD-Internal: [CPUPL-4400]
Change-Id: I2cf021e0cc24c62e7600ba80fd810f3aa55a6ea5
