The blis.h header file includes a lot of BLIS internal definitions. Some of these caused problems
when using a BLIS library compiled with clang on Windows from an applications compiled with
the Intel icc and icx compilers. Workaround is to use "#ifdef BLIS_IS_BUILDING_LIBRARY" to
guard these definitions from being exposed to applications including blis.h. (The BLIS configure
and cmake builds systems automatically define BLIS_IS_BUILDING_LIBRARY only for compiling
the BLIS library.)
This patch implements the minimum changes to resolve the issue. Longer term, similar changes
may need to be added around all BLIS internal definitions in blis.h.
AMD-Internal: [CPUPL-6953]
Details:
- Removed support for all induced methods except for 1m. This included
removing code related to 3mh, 3m1, 4mh, 4m1a, and 4m1b as well as any
code that existed only to support those implementations. These
implementations were rarely used and posed code maintenance challenges
for BLIS's maintainers going forward.
- Removed reference kernels for packm that pack 3m and 4m micropanels,
and removed 3m/4m-related code from bli_cntx_ref.c.
- Removed support for 3m/4m from the code in frame/ind, then reorganized
and streamlined the remaining code in that directory. The *ind(),
*nat(), and *1m() APIs were all removed. (These additional API layers
no longer made as much sense with only one induced method (1m) being
supported.) The bli_ind.c file (and header) were moved to frame/base
and bli_l3_ind.c (and header) and bli_l3_ind_tapi.h were moved to
frame/3.
- Removed 3m/4m support from the code in frame/1m/packm.
- Removed 3m/4m support from trmm/trsm macrokernels and simplified some
pointer arithmetic that was previously expressed in terms of the
bli_ptr_inc_by_frac() static inline function (whose definition was
also removed).
- Removed the following subdirectories of level-0 macro headers from
frame/include/level0: ri3, rih, ri, ro, rpi. The level-0 scalar macros
defined in these directories were used exclusively for 3m and 4m
method codes.
- Simplified bli_cntx_set_blkszs() and bli_cntx_set_ind_blkszs() in
light of 1m being the only induced method left within BLIS.
- Removed dt_on_output field within auxinfo_t and its associated
accessor functions.
- Re-indexed the 1e/1r pack schemas after removing those associated with
variants of the 3m and 4m methods. This leaves two bits unused within
the pack format portion of the schema bitfield. (See bli_type_defs.h
for more info.)
- Spun off the basic and expert interfaces to the object and typed APIs
into separate files: bli_l3_oapi.c and bli_l3_oapi_ex.c; bli_l3_tapi.c
and bli_l3_tapi_ex.c.
- Moved the level-3 operation-specific _check function calls from the
operations' _front() functions to the corresponding _ex() function of
the object API. (This change roughly maintains where the _check()
functions are called in the call stack but lays the groundwork for
future changes that may come to the level-3 object APIs.) Minor
modifications to bli_l3_check.c to allow the check() functions to be
called from the expert interface APIs.
- Removed support within the testsuite for testing the aforementioned
induced methods, and updated the standalone test drivers in the 'test'
directory so reflect the retirement of those induced methods.
- Modified the sandbox contract so that the user is obliged to define
bli_gemm_ex() instead of bli_gemmnat(). (This change was made in light
of the *nat() functions no longer existing.) Also updated the existing
'power10' and 'gemmlike' sandboxes to come into compliance with the
new sandbox rules.
- Updated BLISObjectAPI.md, BLISTypedAPI.md, Testsuite.md documentation
to reflect the retirement of 3m/4m, and also modified Sandboxes.md to
bring the document into alignment with new conventions.
- Updated various comments; removed segments of commented-out code.
Details:
- Updated all static function definitions to use the cpp macro
BLIS_INLINE instead of the static keyword. This allows blis.h to
use a different keyword (inline) to define these functions when
compiling with C++, which might otherwise trigger "defined but
not used" warning messages. Thanks to Giorgos Margaritis for
reporting this issue and Devin Matthews for suggesting the fix.
- Updated the following files, which are used by configure's
hardware auto-detection facility, to unconditionally #define
BLIS_INLINE to the static keyword (since we know BLIS will be
compiled with C, not C++):
build/detect/config/config_detect.c
frame/base/bli_arch.c
frame/base/bli_cpuid.c
- CREDITS file update.
Details:
- Fixed an innocuous bug that manifested when running the testsuite on
extremely small matrices with randomization via the "powers of 2 in
narrow precision range" option enabled. When the randomization
function emits a perfect 0.0 to fill a 1x1 matrix, the testsuite will
then compute 0.0/0.0 during the normalization process, which leads to
NaN residuals. The solution entails smarter implementaions of randv,
randnv, randm, and randnm, each of which will compute the 1-norm of
the vector or matrix in question. If the object has a 1-norm of 0.0,
the object is re-randomized until the 1-norm is not 0.0. Thanks to
Kiran Varaganti for reporting this issue (#413).
- Updated the implementation of randm_unb_var1() so that it loops over
a call to the randv_unb_var1() implementation directly rather than
calling it indirectly via randv(). This was done to avoid the overhead
of multiple calls to norm1v() when randomizing the rows/columns of a
matrix.
- Updated comments.
Details:
- Fixed a typo in the loop counter for the 'd' (duplication) dimension
in the complex macros of frame/include/level0/bb/bli_scal2bbs_mxn.h.
They shouldn't be used by anyone yet, but thankfully clang via
AppVeyor spit out warnings that alerted me to the issue.
Details:
- Added support for being able to duplicate (broadcast) elements in
memory when packing matrix B (ie: the left-hand operand) in level-3
operations. This turns out advantageous for some architectures that
can afford the cost of the extra bandwidth and somehow benefit from
the pre-broadcast elements (and thus being able to avoid using
broadcast-style load instructions on micro-rows of B in the gemm
microkernel).
- Support optionally disabling right-side hemm and symm. If this occurs,
hemm_r is implemented in terms of hemm_l (and symm_r in terms of
symm_l). This is needed when broadcasting during packing because the
alternative--supporting the broadcast of B while also allowing matrix
B to be Hermitian/symmetric--would be an absolute mess.
- Support alignment factors for packed blocks of A, B, and C separately
(as well as for general-purpose buffers). In addition, we support
byte offsets from those alignment values (which is different from
aligning by align+offset bytes to begin with). The default alignment
values are BLIS_PAGE_SIZE in all four cases, with the offset values
defaulting to zero.
- Pass pack_t schema into bli_?packm_cxk() so that it can be then passed
into the packm kernel, where it will be needed by packm kernels that
perform broadcasts of B, since the idea is that we *only* want to
broadcast when packing micropanels of B and not A.
- Added definition for variadic bli_cntx_set_l3_vir_ukrs(), which can be
used to set custom virtual level-3 microkernels in the cntx_t, which
would typically be done in the bli_cntx_init_*() function defined in
the subconfiguration of interest.
- Added a "broadcast B" kernel function for use with NP/NR = 12/6,
defined in in ref_kernels/1m/bli_packm_cxk_bb_ref.c.
- Added a gemm, gemmtrsm, and trsm "broadcast B" reference kernels
defined in ref_kernels/3/bb. (These kernels have been tested with
double real with NP/NR = 12/6.)
- Added #ifndef ... #endif guards around several macro constants defined
in frame/include/bli_kernel_macro_defs.h.
- Defined a few "broadcast B" static functions in
frame/include/level0/bb for use by "broadcast B"-style packm reference
kernels. For now, only the real domain kernels are tested and fully
defined.
- Output the alignment and offset values for packed blocks of A and B
in the testsuite's "BLIS configuration info" section.
- Comment updates to various files.
- Bumped so_version to 3.0.0.
Details:
- Updated the API and semantics of packm kernels such that they must now
handle edge cases, meaning that a c-by-k packm kernel must be able to
pack edge cases that are fewer than c rows/columns and be able to
zero-fill the remaining elements. They must also be able to zero-fill
the equivalent region when copying fewer than k columns/rows (which is
needed by trsm). The new packm kernel API is generally:
void packm_kernel
(
conj_t conja,
dim_t cdim,
dim_t n,
dim_t n_max,
ctype* restrict kappa,
ctype* restrict a, inc_t inca, inc_t lda,
ctype* restrict p, inc_t ldp,
cntx_t* restrict cntx
);
where cdim and n are the dimensions (short and long, respectively) of
the submatrix being copied from the source matrix A, and n_max is the
"full" long dimension (corresponding to the k dimension in gemm) of
the micropanel. The "full" short dimension (corresponding to the
register blocksize MR or NR) is not part of the API because it is
known intrinsically by the packm kernel implementation. Thanks to
Devin Matthews for prompting us to make this change (#282).
- Updated all reference packm kernels in ref_kernels/1m according to
above changes, as well as all optimized packm kernels (which only
consisted of those for knl).
- Bumped the major soname version number in 'so_version' to 2. At first
I was considering leaving it unchanged, but I couldn't escape the
reality that the packm kernel API is much closer to an expert API
than it is some obscure helper function interface within the framework
that nobody would ever notice.
- Removed reference packm kernels for mr/nr = 30. The only sub-config
that would have been using those kernels is knc, which is likely no
longer being used by very many people (if any). (This also mostly
offset the larger object code footprint incurred by moving the edge-
case handling into the individual packm kernels.)
- Fixed an obscure race condition for 3mh and 4mh induced methods in
which those implementations were modifying the contexts stored in the
gks rather than a local copy.
- Fixed a minor bug in the testsuite that prevented non-1m-based induced
method implementations of trsm from executing.
Details:
- Removed explicit reference to The University of Texas at Austin in the
third clause of the license comment blocks of all relevant files and
replaced it with a more all-encompassing "copyright holder(s)".
- Removed duplicate words ("derived") from a few kernels' license
comment blocks.
- Homogenized license comment block in kernels/zen/3/bli_gemm_small.c
with format of all other comment blocks.
Details:
- Lifted the constraint that 1m only be used when all operands' storage
datatypes (along with the computation datatype) are equal. Now, 1m may
be used as long as all operands are stored in the complex domain. This
change largely consisted of adding the ability to pack to 1e and 1r
formats from one precision to another. It also required adding logic
for handling complex values of alpha to bli_packm_blk_var1_md()
(similar to the logic in bli_packm_blk_var1()).
- Fixed a bug in several virtual microkernels (bli_gemm_md_c2r_ref.c,
bli_gemm1m_ref.c, and bli_gemmtrsm1m_ref.c) that resulted in the wrong
ukernel output preference field being read. Previously, the preference
for the native complex ukernel was being read instead of the pref for
the native real domain ukernel. This bug would not manifest if the
preference for the native complex ukernel happened to be equal to that
of the native real ukernel.
- Added support for testing mixed-precision 1m execution via the gemm
module of the testsuite.
- Tweaked/simplified bli_gemm_front() and bli_gemm_md.c so that pack
schemas are always read from the context, rather than trying to
sometimes embed them directly to the A and B objects. (They are still
embedded, but now uniformly only after reading the schemas from the
context.)
- Redefined cpp macro bli_l3_ind_recast_1m_params() as a static function
and renamed to bli_gemm_ind_recast_1m_params() (since gemm is the only
consumer).
- Added 1m optimization logic (via bli_gemm_ind_recast_1m_params()) to
bli_gemm_ker_var2_md().
- Added explicit handling for beta == 1 and beta == 0 in the reference
gemm1m virtual microkernel in ref_kernels/ind/bli_gemm1m_ref.c.
- Rewrote various level-0 macro defs, including axpyris, axpbyris,
scal2ris, and xpbyris (and their conjugating counterparts) to
explicitly support three operand types and updated invocations to
xpbyris in bli_gemmtrsm1m_ref.c.
- Query and use the storage datatype of the packed object instead of the
storage datatype of the source object in bli_packm_blk_var1().
- Relocated and renamed frame/ind/misc/bli_l3_ind_opt.h to
frame/3/gemm/ind/bli_gemm_ind_opt.h.
- Various whitespace/comment updates.
Details:
- Implemented support for gemm where A, B, and C may have different
storage datatypes, as well as a computational precision (and implied
computation domain) that may be different from the storage precision
of either A or B. This results in 128 different combinations, all
which are implemented within this commit. (For now, the mixed-datatype
functionality is only supported via the object API.) If desired, the
mixed-datatype support may be disabled at configure-time.
- Added a memory-intensive optimization to certain mixed-datatype cases
that requires a single m-by-n matrix be allocated (temporarily) per
call to gemm. This optimization aims to avoid the overhead involved in
repeatedly updating C with general stride, or updating C after a
typecast from the computation precision. This memory optimization may
be disabled at configure-time (provided that the mixed-datatype
support is enabled in the first place).
- Added support for testing mixed-datatype combinations to testsuite.
The user may test gemm with mixed domains, precisions, both, or
neither.
- Added a standalone test driver directory for building and running
mixed-datatype performance experiments.
- Defined a new variation of castm, castnzm, which operates like castm
except that imaginary values are not touched when casting a real
operand to a complex operand. (By contrast, in these situations castm
sets the imaginary components of the destination matrix to zero.)
- Defined bli_obj_imag_is_zero() and substituted calls in lieu of all
usages of bli_obj_imag_equals() that tested against BLIS_ZERO, and
also simplified the implementation of bli_obj_imag_equals().
- Fixed bad behavior from bli_obj_is_real() and bli_obj_is_complex()
when given BLIS_CONSTANT objects.
- Disabled dt_on_output field in auxinfo_t structure as well as all
accessor functions. Also commented out all usage of accessor
functions within macrokernels. (Typecasting in the microkernel is
still feasible, though probably unrealistic for now given the
additional complexity required.)
- Use void function pointer type (instead of void*) for storing function
pointers in bli_l0_fpa.c.
- Added documentation for using gemm with mixed datatypes in
docs/MixedDatatypes.md and example code in examples/oapi/11gemm_md.c.
- Defined level-1d operation xpbyd and level-1m operation xpbym.
- Added xpbym test module to testsuite.
- Updated frame/include/bli_x86_asm_macros.h with additional macros
(courtsey of Devin Matthews).
Details:
- Removed four trailing spaces after "BLIS" that occurs in most files'
commented-out license headers.
- Added UT copyright lines to some files. (These files previously had
only AMD copyright lines but were contributed to by both UT and AMD.)
- In some files' copyright lines, expanded 'The University of Texas' to
'The University of Texas at Austin'.
- Fixed various typos/misspellings in some license headers.
Details:
- Converted most C preprocessor macros in bli_param_macro_defs.h and
bli_obj_macro_defs.h to static functions.
- Reshuffled some functions/macros to bli_misc_macro_defs.h and also
between bli_param_macro_defs.h and bli_obj_macro_defs.h.
- Changed obj_t-initializing macros in bli_type_defs.h to static
functions.
- Removed some old references to BLIS_TWO and BLIS_MINUS_TWO from
bli_constants.h.
- Whitespace changes in select files (four spaces to single tab).
Details:
- Fixed a bug that manifested as improperly-computed 1-norm for vectors
and matrices. This is one of the few operations in BLIS that does not
have its own test module within the testsuite, hence why it went
undetected for so long. The bad 1-norms were being used to normalize
matrices in the testsuite after initialization, which led to some
matrices containing a combination of "large" and "small" values. This
tended to push the residuals computed after each test away from zero.
In some cases, they were off *just* enough to the testsuite to label
it a "failure". Many thanks to Jeff Hammond for reporting this bug.
(Wonky details: the bug was due to improperly-defined level-0 scalar
macros for abval2, an operation that computes the absolute square,
or complex magnitude/modulus. Certain complex domain instances of
abval2 were being incorrectly defined in terms of real-only solutions,
leading to bad results. This level-0 operation forms the basis of
norm1v/norm1m. absq2 was also affected, but almost nothing uses
this operation.)
Details:
- Implemented the 1m method for inducing complex domain matrix
multiplication. 1m support has been added to all level-3 operations,
including trsm, and is now the default induced method when native
complex domain gemm microkernels are omitted from the configuration.
- Updated _cntx_init() operations to take a datatype parameter. This was
needed for the corresponding function for 1m (because 1m requires us
to choose between column-oriented or row-oriented execution, which
requires us to query the context for the storage preference of the
gemm microkernel, which requires knowing the datatype) but I decided
that it made sense for consistency to add the parameter to all other
cntx initialization functions as well, even though those functions
don't use the parameter.
- Updated bli_cntx_set_blkszs() and bli_gks_cntx_set_blkszs() to take
a second scalar for each blocksize entry. The semantic meaning of the
two scalars now is that the first will scale the default blocksize
while the second will scale the maximum blocksize. This allows scaling
the two independently, and was needed to support 1m, which requires
scaling for a register blocksize but not the register storage
blocksize (ie: "packdim") analogue.
- Deprecated bli_blksz_reduce_dt_to() and defined two new functions,
bli_blksz_reduce_def_to() and bli_blksz_reduce_max_to(), for reducing
default and maximum blocksizes to some desired blocksize multiple.
These functions are needed in the updated definitions of
bli_cntx_set_blkszs() and bli_gks_cntx_set_blkszs().
- Added support for the 1e and 1r packing schemas to packm, including
1e/1r packing kernels.
- Added a minor optimization to bli_gemm_ker_var2() that allows, under
certain circumstances (specifically, real domain beta and row- or
column-stored matrix C), the real domain macrokernel and microkernel
to be called directly, rather than using the virtual microkernel
via the complex domain macrokernel, which carries a slight additional
amount of overhead.
- Added 1m support to the testsuite.
- Added 1m support to Makefile and runme.sh in test/3m4m. Also simplified
some code in test_gemm.c driver.
Details:
- Moved amaxv from being a utility operation to being a level-1v operation.
This includes the establishment of a new amaxv kernel to live beside all
of the other level-1v kernels.
- Added two new functions to bli_part.c:
bli_acquire_mij()
bli_acquire_vi()
The first acquires a scalar object for the (i,j) element of a matrix,
and the second acquires a scalar object for the ith element of a vector.
- Added integer support to bli_getsc level-0 operation. This involved
adding integer support to the bli_*gets level-0 scalar macros.
- Added a new test module to test amaxv as a level-1v operation. The test
module works by comparing the value identified by bli_amaxv() to the
the value found from a reference-like code local to the test module
source file. In other words, it (intentionally) does not guarantee the
same index is found; only the same value. This allows for different
implementations in the case where a vector contains two or more elements
containing exactly the same floating point value (or values, in the case
of the complex domain).
- Removed the directory frame/include/old/.
Details:
- Defined a new randomization operation, randn, on vectors and matrices.
The randnv and randnm operations randomize each element of the target
object with values from a narrow range of values. Presently, those
values are all integer powers of two, but they do not need to be powers
of two in order to achieve the primary goal, which is to initialize
objects that can be operated on with plenty of precision "slack"
available to allow computations that avoid roundoff. Using this method
of randomization makes it much more likely that testsuite residuals of
properly-functioning operations are close to zero, if not exactly zero.
- Updated existing randomization operations randv and randm to skip
special diagonal handling and normalization for matrices with structure.
This is now handled by the testsuite modules by explicitly calling a
testsuite function that loads the diagonal (and scales off-diagonal
elements).
- Added support for randnv and randnm in the testsuite with a new switch
in input.general that universally toggles between use of the classic
randv/randm, which use real values on the interval [-1,1], and
randnv/randnm, which use only values from a narrow range. Currently,
the narrow range is: +/-{2^0, 2^-1, 2^-2, 2^-3, 2^-4, 2^-5, 2^-6}, as
well as 0.0.
- Updated testsuite modules so that a testsutie wrapper function is called
instead of directly calling the randomization operations (such as
bli_randv() and bli_randm()). This wrapper also takes a bool_t that
indicates whether the object's elements should be normalized. (NOTE: As
alluded to above, in the test modules of triangular solve operations such
as trsv and trsm, we perform the extra step of loading the diagonal.)
- Defined a new level-0 operation, invertsc, which inverts a scalar.
- Updated the abval2ris and sqrt2ris level-0 macros to avoid an unlikely
but possible divide-by-zero.
- Updated function signature and prototype formatting in testsuite.
- Add missing axpby and xpby operations (plus test cases).
- Add special case for scal2v with alpha=1.
- Add restrict qualifiers.
- Add special-case algorithms for incx=incy=1.
Details:
- Retrofitted a new data structure, known as a context, into virtually
all internal APIs for computational operations in BLIS. The structure
is now present within the type-aware APIs, as well as many supporting
utility functions that require information stored in the context. User-
level object APIs were unaffected and continue to be "context-free,"
however, these APIs were duplicated/mirrored so that "context-aware"
APIs now also exist, differentiated with an "_ex" suffix (for "expert").
These new context-aware object APIs (along with the lower-level, type-
aware, BLAS-like APIs) contain the the address of a context as a last
parameter, after all other operands. Contexts, or specifically, cntx_t
object pointers, are passed all the way down the function stack into
the kernels and allow the code at any level to query information about
the runtime, such as kernel addresses and blocksizes, in a thread-
friendly manner--that is, one that allows thread-safety, even if the
original source of the information stored in the context changes at
run-time; see next bullet for more on this "original source" of info).
(Special thanks go to Lee Killough for suggesting the use of this kind
of data structure in discussions that transpired during the early
planning stages of BLIS, and also for suggesting such a perfectly
appropriate name.)
- Added a new API, in frame/base/bli_gks.c, to define a "global kernel
structure" (gks). This data structure and API will allow the caller to
initialize a context with the kernel addresses, blocksizes, and other
information associated with the currently active kernel configuration.
The currently active kernel configuration within the gks cannot be
changed (for now), and is initialized with the traditional cpp macros
that define kernel function names, blocksizes, and the like. However,
in the future, the gks API will be expanded to allow runtime management
of kernels and runtime parameters. The most obvious application of this
new infrastructure is the runtime detection of hardware (and the
implied selection of appropriate kernels). With contexts in place,
kernels may even be "hot swapped" at runtime within the gks. Once
execution enters a level-3 _front() function, the memory allocator will
be reinitialized on-the-fly, if necessary, to accommodate the new
kernels' blocksizes. If another application thread is executing with
another (previously loaded) kernel, it will finish in a deterministic
fashion because its kernel information was loaded into its context
before computation began, and also because the blocks it checked out
from the internal memory pools will be unaffected by the newer threads'
reinitialization of the allocator.
- Reorganized and streamlined the 'ind' directory, which contains much of
the code enabling use of induced methods for complex domain matrix
multiplication; deprecated bli_bsv_query.c and bli_ukr_query.c, as
those APIs' functionality is now mostly subsumed within the global
kernel structure.
- Updated bli_pool.c to define a new function, bli_pool_reinit_if(),
that will reinitialize a memory pool if the necessary pool block size
has increased.
- Updated bli_mem.c to use bli_pool_reinit_if() instead of
bli_pool_reinit() in the definition of bli_mem_pool_init(), and placed
usage of contexts where appropriate to communicate cache and register
blocksizes to bli_mem_compute_pool_block_sizes().
- Simplified control trees now that much of the information resides in
the context and/or the global kernel structure:
- Removed blocksize object pointers (blksz_t*) fields from all control
tree node definitions and replaced them with blocksize id (bszid_t)
values instead, which may be passed into a context query routine in
order to extract the corresponding blocksize from the given context.
- Removed micro-kernel function pointers (func_t*) fields from all
control tree node definitions. Now, any code that needs these function
pointers can query them from the local context, as identified by a
level-3 micro-kernel id (l3ukr_t), level-1f kernel id, (l1fkr_t), or
level-1v kernel id (l1vkr_t).
- Removed blksz_t object creation and initialization, as well as kernel
function object creation and initialization, from all operation-
specific control tree initialization files (bli_*_cntl.c), since this
information will now live in the gks and, secondarily, in the context.
- Removed blocksize multiples from blksz_t objects. Now, we track
blocksize multiples for each blocksize id (bszid_t) in the context
object.
- Removed the bool_t's that were required when a func_t was initialized.
These bools are meant to allow one to track the micro-kernel's storage
preferences (by rows or columns). This preference is now tracked
separately within the gks and contexts.
- Merged and reorganized many separate-but-related functions into single
files. This reorganization affects frame/0, 1, 1d, 1m, 1f, 2, 3, and
util directories, but has the most obvious effect of allowing BLIS
to compile noticeably faster.
- Reorganized execution paths for level-1v, -1d, -1m, and -2 operations
in an attempt to reduce overhead for memory-bound operations. This
includes removal of default use of object-based variants for level-2
operations. Now, by default, level-2 operations will directly call a
low-level (non-object based) loop over a level-1v or -1f kernel.
- Converted many common query functions in blk_blksz.c (renamed from
bli_blocksize.c) and bli_func.c into cpp macros, now defined in their
respective header files.
- Defined bli_mbool.c API to create and query "multi-bools", or
heterogeneous bool_t's (one for each floating-point datatype), in the
same spirit as blksz_t and func_t.
- Introduced two key parameters of the hardware: BLIS_SIMD_NUM_REGISTERS
and BLIS_SIMD_SIZE. These values are needed in order to compute a third
new parameter, which may be set indirectly via the aforementioned
macros or directly: BLIS_STACK_BUF_MAX_SIZE. This value is used to
statically allocate memory in macro-kernels and the induced methods'
virtual kernels to be used as temporary space to hold a single
micro-tile. These values are now output by the testsuite. The default
value of BLIS_STACK_BUF_MAX_SIZE is computed as
"2 * BLIS_SIMD_NUM_REGISTERS * BLIS_SIMD_SIZE".
- Cleaned up top-level 'kernels' directory (for example, renaming the
embarrassingly misleading "avx" and "avx2" directories to "sandybridge"
and "haswell," respectively, and gave more consistent and meaningful
names to many kernel files (as well as updating their interfaces to
conform to the new context-aware kernel APIs).
- Updated the testsuite to query blocksizes from a locally-initialized
context for test modules that need those values: axpyf, dotxf,
dotxaxpyf, gemm_ukr, gemmtrsm_ukr, and trsm_ukr.
- Reformatted many function signatures into a standard format that will
more easily facilitate future API-wide changes.
- Updated many "mxn" level-0 macros (ie: those used to inline double loops
for level-1m-like operations on small matrices) in frame/include/level0
to use more obscure local variable names in an effort to avoid variable
shaddowing. (Thanks to Devin Matthews for pointing these gcc warnings,
which are only output using -Wshadow.)
- Added a conj argument to setm, so that its interface now mirrors that
of scalm. The semantic meaning of the conj argument is to optionally
allow implicit conjugation of the scalar prior to being populated into
the object.
- Deprecated all type-aware mixed domain and mixed precision APIs. Note
that this does not preclude supporting mixed types via the object APIs,
where it produces absolutely zero API code bloat.
Details:
- Added "4mh" and "3mh" APIs, which implement the 4m and 3m methods at
high levels, respectively. APIs for trmm and trsm were NOT added due
to the fact that these approaches are inherently incompatible with
implementing 4m or 3m at high levels (because the input right-hand
side matrix is overwritten).
- Added 4mh, 3mh virtual micro-kernels, and updated the existing 4m and
3m so that all are stylistically consistent.
- Added new "rih" packing kernels (both low-level and structure-aware)
to support both 4mh and 3mh.
- Defined new pack_t schemas to support real-only, imaginary-only, and
real+imaginary packing formats.
- Added various level0 scalar macros to support the rih packm kernels.
- Minor tweaks to trmm macro-kernels to facilitate 4mh and 3mh.
- Added the ability to enable/disable 4mh, 3m, and 3mh, and adjusted
level-3 front-ends to check enabledness of 3mh, 3m, 4mh, and 4m (in
that order) and execute the first one that is enabled, or the native
implementation if none are enabled.
- Added implementation query functions for each level-3 operation so
that the user can query a string that describes the implementation
that is currently enabled.
- Updated test suite to output implementation types for reach level-3
operation, as well as micro-kernel types for each of the five micro-
kernels.
- Renamed BLIS_ENABLE_?COMPLEX_VIA_4M macros to _ENABLE_VIRTUAL_?COMPLEX.
- Fixed an obscure bug when packing Hermitian matrices (regular packing
type) whereby the diagonal elements of the packed micro-panels could
get tainted if the source matrix's imaginary diagonal part contained
garbage.
Details:
- Reordered the #include statements in bli_scalar_macro_defs.h so that
conventional, ri-, and ri3-based macros are grouped together.
- Renamed bli_eqri.h (and macros within) to end with 'ris' suffix.
Details:
- Updated copyright headers to include "at Austin" in the name of the
University of Texas.
- Updated the copyright years of a few headers to 2014 (from 2011 and
2012).
Details:
- Redefined xpbys_mxn and xpbys_mxn_u/_l macros to employ a copy
(instead of scaling by beta) when beta is zero. This will stamp out
any possible infs or NaNs in the output matrix, if it happens to be
uninitialized. Thanks to Tony Kelman for isolating this bug.
Details:
- Fixed various bugs in packm_*_cxk(), the 4m/3m micro-kernels, and
elsewhere in the framework that were not yet set up to work properly
when BLIS_ENABLE_C99_COMPLEX is defined in bli_config.h
- Extensive changes to f2c-derived files in frame/compat/f2c to allow
C99 complex storage. Most of these changes center around accessing
real and imaginary components via bli_?real()/bli_?imag() accessor
macros, and setting of values via bli_?sets() assignment macros.
(Thanks to Vladimir Sukarev for pointing out that _ENABLE_C99_COMPLEX
was broken.)
Details:
- Added the ability to induce complex domain level-3 operations via new
virtual complex micro-kernels which are implemented via only real
domain micro-kernels. Two new implementations are provided: 4m and 3m.
4m implements complex matrix multiplication in terms of four real
matrix multiplications, where as 3m uses only three and thus is
capable of even higher (than peak) performance. However, the 3m method
has somewhat weaker numerical properties, making it less desirable
in general.
- Further refined packing routines, which were recently revamped, and
added packing functionality for 4m and 3m.
- Some modifications to trmm and trsm macro-kernels to facilitate indexing
into micro-panels which were packed for 4m/3m virtual kernels.
- Added 4m and 3m interfaces for each level-3 operation.
- Various other minor changes to facilitate 4m/3m methods.
Details:
- Added set of basic scalar macros that take arguments' real and
imaginary components separately, named like the previous set except
with the "ris" (instead of "s") suffix.
- Redefined the previous set of scalar macros (those that take arguments
"whole") in terms of the new "ri" set.
- Renamed setris and getris macros to sets and gets.
- Renamed setimag0 macros to seti0s.
- Use bli_?1 macro instead of a local constant in bla_trmv.c, bla_trsv.c.
Details:
- Re-defined abval2s and sqrt2s macros to use scaling to avoid underflow
and overflow from squaring the real and imaginary components. (This is
the same technique used to fix recent bugs in invscals/invscaljs and
inverts.)
Details:
- Fixed complex inversion in invscals and invscaljs whereby the
imaginary component was being computed incorrectly.
- Use bli_fmaxabs() instead of bli_fabs() when choosing the scalar
in inverts, invscals, and invscaljs.
- Changed bli_abs() and bli_fabs() macro definitions to use "<="
operator instead of "<".
Details:
- Fixed the complex bli_?inverts() macros, which were inverting elements
in an "unsafe" manner, such that very large and very small values were
unnecessarily over/under-flowing. Thanks for Vladimir Sukharev for
reporting this bug.
- Comment update to bli_sumsqv_unb_var1.c.
- Removed redundant bli_min() macro in bli_scalar_macro_defs.h.
- Changed 1.0F to 1.0 for bli_drands() macro.
Details:
- Added support for C99 complex types to bli_type_defs.h and overloaded
complex arithmetic to the scalar-level macros in include/level0. This
includes a somewhat substantial reorganization and re-layering of much
of the existing machinery present in the level0 macros.
- Added new #define for BLIS_ENABLE_C99_COMPLEX to bli_config.h files,
commented-out by default, which optionally enables the use of built-in
C99 complex types and arithmetic.
- Minor changes to clarksville and reference configs' make_defs.mk files.
- Removed macro definitions from bli_param_macro_defs.h which was not being
used (bli_proj_dt_to_real_if_imag_eq0).
Details:
- Removed copynzv and copynzm operation directories. These operations
implemented a variation of copyv/m that, in the case of real source
and complex destination operands, leaves the imaginary component
untouched (rather than setting it to zero). I realize now that the
special case(s) (e.g. gemm with real A and B but complex C) that I
thought required this operation actually can be handled more simply.
- Removed level0 scalar macros implementing copynzs, copynzjs.
Details:
- Changed the way bli_type_defs.h defines integer types so that dim_t,
inc_t, doff_t, etc. are all defined in terms of gint_t (general signed
integer) or guint_t (general unsigned integer).
- Renamed Fortran types fchar and fint to f77_char and f77_int.
- Define f77_int as int64_t if a new configuration variable,
BLIS_ENABLE_BLIS2BLAS_INT64, is defined, and int32_t otherwise.
These types are defined in stdint.h, which is now included in blis.h.
- Renamed "complex" type in f2c files to "singlecomplex" and typedef'ed
in terms of scomplex.
- Renamed "char" type in f2c files to "character" and typedef'ed in terms
of char.
- Updated bla_amax() wrappers so that the return type is defined directly
as f77_int, rather than letting the prototype-generating macro decide
the type. This was the only use of GENTFUNC2I/GENTPROT2I-related macros,
so I removed them. Also, changed the body of the wrapper so that a
gint_t is passed into abmaxv, which is THEN typecast to an f77_int
before returning the value.
- Updated f2c code that accessed .r and .i fields of complex and
doublecomplex types so that they use .real and .imag instead (now that
we are using scomplex and dcomplex).
Details:
- Replaced multi-type invocations of copys_mxn, xpbys_mxn, etc. (PASTEMAC2
and PASTEMAC3) with those that only use a single type (PASTEMAC).
- Added extra macros to bli_adds_mxn_uplo.h and bli_xpbys_mxn_uplo.h to
accommodate above change.
- Fixed comment typo in bli_config.h files.
- Added .nfs* pattern to .gitignore.
Details:
- Replaced scalar constant macro definitions in bli_const_defs.h with a single,
simplier macro in bli_obj_macro_defs.h.
- Updated invocations of old macros accordingly.
- Removed bli_const_defs.h.
Details:
- Added a new set of reference gemm, gemmtrsm, and trsm micro-kernels that
contain explicit loops over MR and NR, thus allowing them to be used
unmodified by developers who want to build a reference library with
custom register blocksizes.
- Changed config/reference/bli_kernel.h to use above ukernels by default.
- Changed interfaces of new and existing gemm, gemmtrsm, and trsm micro-kernels
to use 'restrict' keyword.
- Added -funroll-loops option to config/reference/make_defs.mk.
- Updated comments in bli_kernel.h describing constraints on register and
cache blocksizes.
- Updated _adds_mxn.h, _copys_mxn.h, and _xpbys_mxn.h macros files so that
single-char macros are also defined.
Details:
- Implemented amax operation in BLIS.
- Activated BLAS2BLIS routine mapping for new amax BLIS implementation.
- Added integer support to [f]printv, [f]printm.
- Added integer support to level-0 copys macros.
- Updated printing of configuration information in test suite driver.
- Comment changes to _config.h files.
- Added comments to bla_dot.c to reminder reader what sdsdot()/dsdot() are
used for.
Details:
- Changed all filename and function prefixes from 'bl2' to 'bli'.
- Changed the "blis2.h" header filename to "blis.h" and changed all
corresponding #include statements accordingly.
- Fixed incorrect association for Fran in CREDITS file.
Details:
- Added the blas2blis compatibility layer, located in frame/compat. This
includes virtually all of the BLAS, including banded and packed level-2
operations.
- Defined bl2_init_safe(), bl2_finalize_safe(). The former allows a conditional
initialization, which stores the "exit status" in an err_t, which is then
read by the latter function to determine whether finalization should actually
take place.
- Added calls to bl2_init_safe(), bl2_finalize_safe() to all level-2 and
level-3 BLAS-like wrappers.
- Added configuration option to instruct BLIS to remain initialized whenever
it automatically initializes itself (via bl2_init_safe()), until/unless the
application code explicitly calls bl2_finalize().
- Added INSERT_GENTFUNC* and INSERT_GENTPROT* macros to facilitate type
templatization of blas2blis wrappers.
- Defined level-0 scalar macro bl2_??swaps().
- Defined level-1v operation bl2_swapv().
- Defined some "Fortran" types to bl2_type_defs.h for use with BLAS
wrappers.
Details:
- Changed name of set0 and set0_mxn macros to set0s and set0s_mxn,
respectively.
- Added code to the following operations that sets the output operand to
zero if the corresponding scalar is zero (rather than performing the
floating-point multiply, or in the case of setv, copying the value).
This will prevent nan's and inf's from creeping into results from
uninitialized memory.
- axpy
- dotxv
- scalv
- scal2v
- setv
- gemv
- ger
- hemv
- her
- her2
- gemm reference ukernels
Details:
- Added a highly configurable, unified test suite.
- Removed DUPB configuration constant from bl2_kernel.h and macro-kernel
header files. Now, instead, DUPB is computed as (NDUP != 1) within each
macro-kernel. This fixes a bug in trmm/trsm whereby bp was indexed into
incorrectly when DUPB was set to FALSE but the NDUP was still non-unit.
By encoding both pieces of information into one constant in _kernel.h,
it seems somewhat less likely others will encounter this bug in the
future.
- Added level-2 cache blocksizes to _kernel.h for reference configuration,
and defined blocksizes in _cntl.c files to these default values.
- Changed semantics of her2k and syr2k such that these operations no longer
expect the B matrix to already be conjugate-transposed (or just transposed
for syr2k). However, these semantics are preserved for the internal
mechanics of the implementations, including the internal back-end and all
blocked variants.
- Inserted checks for real-valued alpha and beta for herk/her2k and herk,
respectively.
- Relaxed general object structure constraints in _basic_check() for gemv, ger.
- Changed her front-end to NOT copy-cast to real projection; instead, this is
replaced by selecting either the real part or both parts within the unblocked
algorithm implementation, depending on the value of conjh.
- Added conjh to all _check routines for her so that the code knows when to
verify that alpha has an imaginary component equal to zero (for her, but
not syr).
- Changed control tree for her to forgo packing.
- Added unit diagonal support to fnormm.
- Redefined real versions of abval2s macros in terms of fabs(), fabsf().
- Redefined complex versions of sqrt2s macros using the actual "complex square
root" formula.
- Created new level-0 object-based routines, suffixed with "sc" (for "scalar").
- Defined new level-1v, -1d, and -1m versions of add and sub operations
(two-operand add and subtract).
- Added new scalar macros:
- getris: acquire real and imaginary components.
- setris: set real and imaginary components.
- addjs: addition with conjugated x.
- subjs: subtraction with conjugated x.
- Defined new utility operations:
- absumv: element-wise sum of absolute values for vector elements.
- absumm: element-wise sum of absolute values for matrix elements.
- mkherm: convert existing matrix to Hermitian.
- mksymm: convert existing matrix to symmetric.
- mktrim: convert existing matrix to triangular.
- Added various error checking routines.
- Added bl2_clock_min_diff(), which is used to more cleanly measure the
wall clock time of a code block.
- Added general stride support to bl2_obj_alloc_buffer().
- Added bl2_obj_init_scalar().
- Updated parameter mapping in bl2_param_map.c.
- Added support for queriable version string.
- Fixed a bug in the her2k macro-kernels (which currently are simply
implemented in terms of two invocations of herk) whereby beta was being
applied to both the first and second rank-k updates, rather than only
the first.
- Fixed a bug in trmm/trsm whereby transpose and right side cases were not
properly implemented due to erroneous assumptions regarding aliasing and
root objects.
- Fixed a bug in the upper triangular trsm macro-kernel in which the wrong
MR x NR block of B was being updated.
- Fixed a bug in the inverts macro in the double real case whereby the
value was typecast to float before inversion. This affected non-unit cases
of dtrsm.
- Fixed a bug in the reference kernels for gemmtrsm whereby the minus one
constant was being applied incorrectly.
- Fixed a bug in the overall treatment of non-unit alpha for trsm. The code
now mimics the rank-k strategy of gemm, whereby alpah is applied during
the first iteration of variant 3, with BLIS_ONE passed in instead for
subsequent iterations. This also required passing alpha into the macro-
kernels as well as the fused gemmtrsm micro-kernels.
- Fixed a bug in trsm_u_blk_var1 whereby the gemm macro-kernel was being
called for blocks strictly above the diagonal. While this sounds good in
theory, this cannot be done because gemm_ker_var2 expects row panels of
A to be packed from top to bottom, while for trsm_u, A is actually packed
from bottom to top due to the reverse (BR->TL) nature of the algorithm.
- Fixed a bug in packm_cxk() whereby panel packings with unit panel
dimensions were mishandled due to incorrect arguments to the copyv kernel.
Also changed the copyv kernel invocation to scal2v so that these edge
cases are properly handled when scaling is requested.
- Fixed a bug in packv_int() whereby an uninitialized object is passed in
instead of the source object.
- Fixed a bug whereby level-2 code could allocate memory dynamically via
bl2_malloc() and then attempt to free it via bl2_mm_release(). Also fixed
a potential future bug whereby a mem_t object that is actually no longer
"allocated" from the static pool is mistaken for being allocated due to
failure to NULLify the buffer when the block was most recently released.
- Fixed a bug in bl2_acquire_mpart_*() whreby the uplo field was mistakenly
toggled when the requested subpartition needed to be "reflected" due to it
residing in an unstored region.
Details:
- Added level-0 grabis macro operation to grab imaginary component of one
variable and copy it to the real component of another variable.
- Defined sumsqv operation, which computes the sum of the absolute squares
of the elements of a vector. This implementation is modeled after ?lassq
in netlib LAPACK.
- Defined fnormv and fnormm operations, which compute the Frobenius norm on
vectors and matrices, respectively. These operations are treated as one-
operand operations where the output norm value is the real projection of
the datatype of the input operand. Both operations are implemented in terms
of sumsqv.
