Details:
- Added sgemm and dgemm micro-kernels, which employ 256-bit AVX vectors
and FMA instructions. (Complex support is currently provided by default
induced method, 4m1a.)
- Added a 'haswell' configuration, which uses the aforementioned kernels.
- Inserted auto-detection support for haswell configuration in
build/auto-detect/cpuid_x86.c.
- Modified configure script to explicitly echo when automatic or manual
configuration is in progress.
- Changed beta scalar in test_gemm.c module of test suite to -1.0 to 0.9.
Details:
- Added a new configuration for AMD Excavator-based hardware also known
as Carrizo when referring to the entire APU. This configuration uses
the same micro-kernels as the piledriver, but with different
cache blocksizes.
Details:
- Replaced the old memory allocator, which was based on statically-
allocated arrays, with one based on a new internal pool_t type, which,
combined with a new bli_pool_*() API, provides a new abstract data
type that implements the same memory pool functionality but with blocks
from the heap (ie: malloc() or equivalent). Hiding the details of the
pool in a separate API also allows for a much simpler bli_mem.c family
of functions.
- Added a new internal header, bli_config_macro_defs.h, which enables
sane defaults for the values previously found in bli_config. Those
values can be overridden by #defining them in bli_config.h the same
way kernel defaults can be overridden in bli_kernel.h. This file most
resembles what was previously a typical configuration's bli_config.h.
- Added a new configuration macro, BLIS_POOL_ADDR_ALIGN_SIZE, which
defaults to BLIS_PAGE_SIZE, to specify the alignment of individual
blocks in the memory pool. Also added a corresponding query routine to
the bli_info API.
- Deprecated (once again) the micro-panel alignment feature. Upon further
reflection, it seems that the goal of more predictable L1 cache
replacement behavior is outweighed by the harm caused by non-contiguous
micro-panels when k % kc != 0. I honestly don't think anyone will even
miss this feature.
- Changed bli_ukr_get_funcs() and bli_ukr_get_ref_funcs() to call
bli_cntl_init() instead of bli_init().
- Removed query functions from bli_info.c that are no longer applicable
given the dynamic memory allocator.
- Removed unnecessary definitions from configurations' bli_config.h files,
which are now pleasantly sparse.
- Fixed incorrect flop counts in addv, subv, scal2v, scal2m testsuite
modules. Thanks to Devangi Parikh for pointing out these
miscalculations.
- Comment, whitespace changes.
Details:
- Added conditional code that returns early from the API-level _init()
routines if the API is already initialized. Actually meant for this to
be included in 5f93cbe8.
Details:
- Added API-level initialization state to _const, _error, _mem, _thread,
_ind, and _cntl APIs. While this functionality will mostly go unused,
adding miniscule overhead at init-time, there will be at least once
instance in the near future where, in order to avoid an infinite loop,
a certain portion of the initialization will call a query function that
itself attempts to call bli_init(). API-level initialization will allow
this later stage to verify that an earlier stage of initialization has
completed, even if the overall call to bli_init() has not yet returned.
- Added _is_initialized() functions for each API, setting the underlying
bool_t during _init() and unsetting it during _finalize().
- Comment, whitespace changes.
Details:
- Fixed some bugs that only manifested in multithreaded instances of
some (non-gemm) level-3 operations. The bugs were related to invalid
allocation of "edge" cases to thread subpartitions. (Here, we define
an "edge" case to be one where the dimension being partitioned for
parallelism is not a whole multiple of whatever register blocksize
is needed in that dimension.) In BLIS, we always require edge cases
to be part of the bottom, right, or bottom-right subpartitions.
(This is so that zero-padding only has to happen at the bottom, right,
or bottom-right edges of micro-panels.) The previous implementations
of bli_get_range() and _get_range_weighted() did not adhere to this
implicit policy and thus produced bad ranges for some combinations of
operation, parameter cases, problem sizes, and n-way parallelism.
- As part of the above fix, the functions bli_get_range() and
_get_range_weighted() have been renamed to use _l2r, _r2l, _t2b,
and _b2t suffixes, similar to the partitioning functions. This is
an easy way to make sure that the variants are calling the right
version of each function. The function signatures have also been
changed slightly.
- Comment/whitespace updates.
- Removed unnecessary '/' from macros in bli_obj_macro_defs.h.
Details:
- Spun-off initialization of global scalar constants to bli_const_init()
and of threading stuff to bli_thread_init().
- Added some missing _finalize() functions, even when there is nothing
to do.
Details:
- Fixed a sort-of bug in bli_init.c whereby the wrong pthread mutex
was used to lock access to initialization/finalization actions.
But everything worked out okay as long as bli_init() was called by
single-threaded code.
- Changed to static initialization for memory allocator mutex in
bli_mem.c, and moved mutex to that file (from bli_init.c).
- Fixed some type mismatches in bli_threading_pthreads.c that resulted
in compiler warnings.
- Fixed a small memory leak with allocated-but-never-freed (and unused)
pthread_attr_t objects.
- Whitespace changes to bli_init.c and bli_mem.c.
Details:
- Updated test/3m4m driver files to build test drivers that allow
comparision of real "asm_blis" results to complex "asm_blis" results,
except with the latter's problem sizes adjusted so that problems are
generated with equal flop counts.
Details:
- Fixed a bug that caused a memory leak in the contiguous memory
allocator. Because packm_init() was using simple aliasing when
a subpartition object was marked as zeros by bli_acquire_mpart_*(),
the "destination" pack object's mem_t entry was being overwritten
by the corresponding field of the "source" object (which was likely
NULL). This prevented the block from being released back to the
memory allocator. But this bug only manifested when changing the
location of packing B from outside the var1 loop to inside the
var3 loop, and only for trsm with triangular B (side = right). The
bug was fixed by changing the type of alias used in packm_init()
when handling zero partition cases. Specifically, we now use
bli_obj_alias_for_packing(), which does not clobber the destination
(pack) object's mem_t field. Thanks to Devangi Parikh for this bug
report.
Details:
- Added a line of output that forces matlab to allocate the entire array
up-front.
- Re-enabled real domain benchmarks in runme.sh, which were temporarily
disabled.
Details:
- Fixed incorrect number of mc_x_kc memory blocks in
sandybridge/bli_config.h.
- Enabled OpenMP multithreding in piledriver/bli_config.h.
- More updates to test/3m4m driver files.
Details:
- Enable OpenMP support by default in sandybridge's bli_config.h.
- Reorganized sandybridge's bli_kernel.h.
- Updated 3m4m Makefile, runme.sh to also test MKL implementation.
Details:
- Defined a new "3ms" (separated 3m) pack schema and added appropriate
support in packm_init(), packm_blk_var2().
- Generalized packm_struc_cxk_3mi to take the imaginary stride (is_p)
as an argument instead of computing it locally. Exception: for trmm,
is_p must be computed locally, since it changes for triangular
packed matrices. Also exposed is_p in interface to dt-specific
packm_blk_var2 (and _var1, even though it does not use imaginary
stride).
- Renamed many functions/variables from _3mi to _3mis to indicate that
they work for either interleaved or separated 3m pack schemas.
- Generalized gemm and herk macro-kernels to pass in imaginary stride
rather than compute them locally.
- Added support for 3m2 and 3m3 algorithms to frame/ind, including 3m2-
and 3m3-specific virtual micro-kernels.
- Added special gemm macro-kernels to support 3m2 and 3m3.
- Added support for 3m2 and 3m3 to testsuite.
- Corrected the type of the panel dimension (pd_) in various macro-
kernels from inc_t to dim_t.
- Renamed many functions defined in bli_blocksize.c.
- Moved most induced-related macro defs from frame/include to
frame/ind/include.
- Updated the _ukernel.c files so that the micro-kernel function pointers
are obtained from the func_t objects rather than the cpp macros that
define the function names.
- Updated test/3m4m driver, Makefile, and run script.
Details:
- Consolidated most of the code relating to induced complex methods
(e.g. 4mh, 4m1, 3mh, 3m1, etc.) into frame/ind. Induced methods
are now enabled on a per-operation basis. The current "available"
(enabled and implemented) implementation can then be queried on
an operation basis. Micro-kernel func_t objects as well as blksz_t
objects can also be queried in a similar maner.
- Redefined several micro-kernel and operation-related functions in
bli_info_*() API, in accordance with above changes.
- Added mr and nr fields to blksz_t object, which point to the mr
and nr blksz_t objects for each cache blocksize (and are NULL for
register blocksizes). Renamed the sub-blocksize field "sub" to
"mult" since it is really expressing a blocksize multiple.
- Updated bli_*_determine_kc_[fb]() for gemm/hemm/symm, trmm, and
trsm to correctly query mr and nr (for purposes of nudging kc).
- Introduced an enumerated opid_t in bli_type_defs.h that uniquely
identifies an operation. For now, only level-3 id values are defined,
along with a generic, catch-all BLIS_NOID value.
- Reworked testsuite so that all induced methods that are enabled
are tested (one at a time) rather than only testing the first
available method.
- Reformated summary at the beginning of testsuite output so that
blocksize and micro-kernel info is shown for each induced method
that was requested (as well as native execution).
- Reduced the number of columns needed to display non-matlab
testsuite output (from approx. 90 to 80).
Details:
- Fixed a bug that affects all level-2 and level-3 blocked variants. The
bug only manifested, however, if the packing of operands (A and B in
gemm, for example) spanned multiple nodes in the control tree. Until
recently, the main consumers of packm were level-3 operations, all of
which packed both input operands from blocked variant 1 (B outside of
the loop, and A within the loop). This particular usage masked a flaw
in the code whereby bli_obj_release_pack() would always release the
underlying mem_t buffer (provided it was allocated), even if the buffer
was not allocated in the current variant. This has been fixed by
replacing all calls to bli_obj_release_pack() with calls to a new
function, bli_packm_release(), which takes the same control tree node
argument passed into the object's corresponding call to packm_init()
or packv_init(). bli_packm_release() then proceeds to invoke
bli_obj_release_pack() only if the control tree node indicates that
packing was requested. Thanks to Devangi Parikh for identifying this
bug.
Details:
- Renamed all remaining 3m/4m packing files and symbols to 3mi/4mi
('i' for "interleaved"). Similar changes to 3M/4M macros.
- Renamed all 3m/4m files and functions to 3m1/4m1.
- Whitespace changes.
Details:
- Relocated bli_4mh.c, bli_4mb.c, bli_4m.c, bli_3mh.c, bli_3m.c (and
associated headers) from frame/base to frame/base/induced.
- Added bli_xm.? to frame/base/induced, which implements
bli_xm_is_enabled(), which detects whether ANY induced complex method
is currently enabled.
- The new function bli_xm_is_enabled() is now used in bli_info.c to
detect when an induced complex method is used, so we know when to
return blocksizes from one of the induced methods' blocksize objects.
Details:
- Changed the static memory definitions to compute the maximum register
blocksize for each datatype and add it to kc when computing the size
of blocks of A and B. This formally accounts for the nudging of kc
up to a multiple of mr or nr at runtime for triangular operations
(e.g. trmm).
Details:
- Renamed several variables and macros from 3m/4m to 3mi/4mi. This is
because those packing schemas were always implicitly "interleaved".
This new naming scheme will make way for new schemas that separate
instead of interleve the real and imaginary (and summed) parts.
- Expanded the pack format sub-field of the pack schema field of the
info_t to 4 bits (from 3). This will allow for more schema types
going forward.
- Removed old _cntl.c files for herk3m, herk4m, trmm3m, trmm4m.
Details:
- Fixed a bug that only affected trmm3 when performed via 3mh or 4mh,
whereby micro-panels of the triangular matrix were packed with "dead
space" between them due to failing to adjust for the fact that pointer
arithmetic was occurring in units of complex elements while the data
being packed consisted of real elements. It turns out that the macro-
kernel suffered from the same bug, meaning the panels were actually
being packed and read consistently. The only way I was able to
discover the bug in the first place was because the packed block of A
was overflowing into the beginning of the packed row panel of B using
the sandybridge configuration.
Details:
- Modified bli_packm_blk_var1.c and _var2.c to increase the triangular
case's panel increment by 1 if it would otherwise be odd. This is
particularly necessary in _var2.c when handling the interleaved 3m
or ro/io/rpi pack schemas, since division of an odd number by 2 can
happen if both the panel length and the panel packing dimension
(register packing blocksize) are odd, thus making their product odd.
- Modified bli_packm_init.c so that panel strides are increased by 1
if they would otherwise be odd, even for non-3m related packing.
- Modified the trmm and trsm macro-kernels so that triangular packed
micro-panels are traversed with this new "increment by 1 if odd"
policy.
- Added sanity checks in trmm and trsm macro-kernels that would result
in an abort() if the conditions that would lead to a "divide odd
integer by 2" scenario ever manifest.
- Defined bli_is_odd(), _is_even() macros in bli_scalar_macro_defs.h.
Details:
- Added an imaginary stride field ("is") to obj_t.
- Renamed bli_obj_set_incs() macro to bli_obj_set_strides().
- Defined bli_obj_imag_stride() and bli_obj_set_imag_stride() and
added invocations in key locations.
- Added some basic error-checking related to imaginary stride.
- For now, imaginary stride will not be exposed into the most-used
BLIS APIs such as bli_obj_create(), and certainly not the
computational APIs such as bli_dgemm().
Details:
- Added a new function, bli_gemm_cntl_ukrs(), that returns the func_t*
for the gemm micro-kernels from the leaf node of the control tree.
This allows all the func_t* fields from higher-level nodes in the tree
to be NULL, which makes the function that builds the control trees
slightly easier to read.
- Call bli_gemm_cntl_ukrs() instead of the cntl_gemm_ukrs() macro in
all bli_*_front() functions (which is needed to apply the row/column
preference optimization).
- In all level-3 bli_*_cntl_init() functions, changed the _obj_create()
function arguments corresponding to the gemm_ukrs fields in higher-
level cntl tree nodes to NULL.
- Removed some old her2k macro-kernels.
