Details:
- Added logic to configure that checks the version of the compiler
against known version ranges that could cause problems later in the
build process. For example, versions of gcc older than 4.9.0 use
different -march labels than version 4.9.0 or later
('-march=corei7-avx' vs '-march=sandybridge', respectively).
Similarly, before 6.1, compilation on Zen was possible, but you
need to start with -march=bdver4 and then disable instruction sets
that were discarded during the transition from Excavator to Zen. So
now, configure substitutes 'yes'/'no' values into anchors in
config.mk.in, which sets various make variables (e.g. GCC_OT_4_9_0),
which can be accessed and branched upon by the various
configurations' make_defs.mk files when setting their compiler flags.
- Updated config/haswell/make_defs.mk to branch on GCC_OT_4_9_0.
- Updated config/sandybridge/make_defs.mk to branch on GCC_OT_4_9_0.
- Updated config/zen/make_defs.mk to branch on GCC_OT_6_1_0.
Details:
- Added a new standalone test driver directory named '1m4m' that can
build and run performance experiments for BLIS 1m, 4m1a, assembly,
OpenBLAS, and the vendor library (MKL). This new driver directory
was used to regenerate performance results for the 1m paper.
- Added alternate (commented-out) cache blocksizes to
config/haswell/bli_cntx_init_haswell.c. These blocksizes tend to
work well on an a 12-core Intel Xeon E5-2650 v3.
Details:
- Updated the default MC cache blocksizes used by the haswell subconfig
for both row-preferential (the default) and column-preferential
microkernels.
Details:
- Updated the '-march=corei7-avx' flag in the sandybridge subconfig
to '-march=sandybridge' and the '-march=core-avx2' flag in the
haswell subconfig to '-march=haswell'. The older flags were used
by older versions of gcc and should have been updated to the newer
forms a long time ago. (The older flags were clearly working, even
though they are no longer documented in the gcc man page.)
Details:
- Fine-tuned the double-precision real MT threshold (which controls
whether the sup implementation kicks for smaller m dimension values)
from 180 to 201 for haswell and 180 to 256 for zen.
- Updated octave scripts in test/sup/octave to include a seventh column
to display performance for m = n = k.
Details:
- Increased the double-precision real MT threshold (which controls
whether the sup implementation kicks for smaller m dimension values)
from 80 to 180, and this change was made for both haswell and zen
subconfigurations. This is less about the m dimension in particular
and more about facilitating a smoother performance transition when
m = n = k.
Details:
- Removed already limited use of the BLIS_ENABLE_SUP_MR_EXT and
BLIS_ENABLE_SUP_NR_EXT macros in bli_gemmsup_ref_var1n() and
bli_gemmsup_ref_var2m(). Their purpose was merely to avoid a long
conditional that would determine whether to allow the last iteration
to be merged with the second-to-last iteration. Functionally, the
macros were not needed, and they ended up causing problems when
building configuration families such as intel64 and x86_64.
Details:
- Implemented a new sub-framework within BLIS to support the management
of code and kernels that specifically target matrix problems for which
at least one dimension is deemed to be small, which can result in long
and skinny matrix operands that are ill-suited for the conventional
level-3 implementations in BLIS. The new framework tackles the problem
in two ways. First the stripped-down algorithmic loops forgo the
packing that is famously performed in the classic code path. That is,
the computation is performed by a new family of kernels tailored
specifically for operating on the source matrices as-is (unpacked).
Second, these new kernels will typically (and in the case of haswell
and zen, do in fact) include separate assembly sub-kernels for
handling of edge cases, which helps smooth performance when performing
problems whose m and n dimension are not naturally multiples of the
register blocksizes. In a reference to the sub-framework's purpose of
supporting skinny/unpacked level-3 operations, the "sup" operation
suffix (e.g. gemmsup) is typically used to denote a separate namespace
for related code and kernels. NOTE: Since the sup framework does not
perform any packing, it targets row- and column-stored matrices A, B,
and C. For now, if any matrix has non-unit strides in both dimensions,
the problem is computed by the conventional implementation.
- Implemented the default sup handler as a front-end to two variants.
bli_gemmsup_ref_var2() provides a block-panel variant (in which the
2nd loop around the microkernel iterates over n and the 1st loop
iterates over m), while bli_gemmsup_ref_var1() provides a panel-block
variant (2nd loop over m and 1st loop over n). However, these variants
are not used by default and provided for reference only. Instead, the
default sup handler calls _var2m() and _var1n(), which are similar
to _var2() and _var1(), respectively, except that they defer to the
sup kernel itself to iterate over the m and n dimension, respectively.
In other words, these variants rely not on microkernels, but on
so-called "millikernels" that iterate along m and k, or n and k.
The benefit of using millikernels is a reduction of function call
and related (local integer typecast) overhead as well as the ability
for the kernel to know which micropanel (A or B) will change during
the next iteration of the 1st loop, which allows it to focus its
prefetching on that micropanel. (In _var2m()'s millikernel, the upanel
of A changes while the same upanel of B is reused. In _var1n()'s, the
upanel of B changes while the upanel of A is reused.)
- Added a new configure option, --[en|dis]able-sup-handling, which is
enabled by default. However, the default thresholds at which the
default sup handler is activated are set to zero for each of the m, n,
and k dimensions, which effectively disables the implementation. (The
default sup handler only accepts the problem if at least one dimension
is smaller than or equal to its corresponding threshold. If all
dimensions are larger than their thresholds, the problem is rejected
by the sup front-end and control is passed back to the conventional
implementation, which proceeds normally.)
- Added support to the cntx_t structure to track new fields related to
the sup framework, most notably:
- sup thresholds: the thresholds at which the sup handler is called.
- sup handlers: the address of the function to call to implement
the level-3 skinny/unpacked matrix implementation.
- sup blocksizes: the register and cache blocksizes used by the sup
implementation (which may be the same or different from those used
by the conventional packm-based approach).
- sup kernels: the kernels that the handler will use in implementing
the sup functionality.
- sup kernel prefs: the IO preference of the sup kernels, which may
differ from the preferences of the conventional gemm microkernels'
IO preferences.
- Added a bool_t to the rntm_t structure that indicates whether sup
handling should be enabled/disabled. This allows per-call control
of whether the sup implementation is used, which is useful for test
drivers that wish to switch between the conventional and sup codes
without having to link to different copies of BLIS. The corresponding
accessor functions for this new bool_t are defined in bli_rntm.h.
- Implemented several row-preferential gemmsup kernels in a new
directory, kernels/haswell/3/sup. These kernels include two general
implementation types--'rd' and 'rv'--for the 6x8 base shape, with
two specialized millikernels that embed the 1st loop within the kernel
itself.
- Added ref_kernels/3/bli_gemmsup_ref.c, which provides reference
gemmsup microkernels. NOTE: These microkernels, unlike the current
crop of conventional (pack-based) microkernels, do not use constant
loop bounds. Additionally, their inner loop iterates over the k
dimension.
- Defined new typedef enums:
- stor3_t: captures the effective storage combination of the level-3
problem. Valid values are BLIS_RRR, BLIS_RRC, BLIS_RCR, etc. A
special value of BLIS_XXX is used to denote an arbitrary combination
which, in practice, means that at least one of the operands is
stored according to general stride.
- threshid_t: captures each of the three dimension thresholds.
- Changed bli_adjust_strides() in bli_obj.c so that bli_obj_create()
can be passed "-1, -1" as a lazy request for row storage. (Note that
"0, 0" is still accepted as a lazy request for column storage.)
- Added support for various instructions to bli_x86_asm_macros.h,
including imul, vhaddps/pd, and other instructions related to integer
vectors.
- Disabled the older small matrix handling code inserted by AMD in
bli_gemm_front.c, since the sup framework introduced in this commit
is intended to provide a more generalized solution.
- Added test/sup directory, which contains standalone performance test
drivers, a Makefile, a runme.sh script, and an 'octave' directory
containing scripts compatible with GNU Octave. (They also may work
with matlab, but if not, they are probably close to working.)
- Reinterpret the storage combination string (sc_str) in the various
level-3 testsuite modules (e.g. src/test_gemm.c) so that the order
of each matrix storage char is "cab" rather than "abc".
- Comment updates in level-3 BLAS API wrappers in frame/compat.
Details:
- Changed -funsafe-loop-optimizations (re-)introduced in 7690855 for
make_defs.mk files' CRVECFLAGS to -funsafe-math-optimizations (to
account for a miscommunication in issue #300). Thanks to Dave Love
for this suggestion and Jeff Hammond for his feedback on the topic.
Details:
- Restored use of -funsafe-loop-optimizations in the definitions of
CRVECFLAGS (when using gcc), but only for sub-configurations (and
not configuration families such as amd64, intel64, and x86_64).
This more or less reverts 5190d05 and 6cf1550.
Details:
- Per Dave Love's recommendation in issue #300, this commit defines
COPTFLAGS := -03
and
CRVECFLAGS := $(CKVECFLAGS) -funsafe-loop-optimizations
in the make_defs.mk for all Intel- and AMD-based configurations.
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:
- Renamed the microkernels in kernels/zen/3 to kernels/haswell/3 and
then updated the file contents to use the 'haswell' infix.
- Updated bli_cntx_init_zen.c and bli_cntx_init_haswell.c according to
above function renames.
- Moved/updated the corresponding prototypes in bli_kernels_zen.h to
bli_kernels_haswell.h.
- Updated config_registry according to above changes.
- NOTE: This rename reflects the fact that haswell microkernels are
specifically written to overcome the floating-point latency for FMA
instructions on Intel Haswell-like architectures, which can issue two
FMA instructions per cycle. These ukernels happen to work fine on AMD
Zen-based architectures. However, Zen only issues one FMA per cycle,
which, while halving its floating-point throughput, gives it extra
flexibility in the design of its microkernels--namely, mr and nr can
be smaller and still overcome the floating-point latency for those
single-issue cores. A smaller value of mr and nr allows for a larger
value of kc, which may be useful in some situations. In the future,
we may write such Zen-specific microkernels to take advantage of this
additional flexibility.
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:
- Added two presently-disabled cpp blocks in bli_cntx_init_haswell.c to
easily allow one to switch to a set of column-preferential gemm
microkernels (in the haswell subconfiguration). The second column-
preferring block sets the the register blocksizes to their appropriate
values. However, cache blocksizes are left unchanged, and therefore are
likely suboptimal. This should be addressed later.
Details:
- Historically, the compiler selection has happened statically in the
various make_defs.mk and would only be overriden by setting CC (either
prior to running configure or as a configure argument). However, in
the last couple months, configure has evolved to contain rather
sophisticated compiler detection logic for the purposes of blacklisting
sub-configurations. It only makes sense that configure now fully take
over the responsibility of selecting a compiler from the GNU make side
of the build system. Thanks to Alex Arslan for his help exposing this
issue.
- Substitute found_cc into CC in config.mk via configure.
- Set a new variable, CC_VENDOR, in config.mk via substitution from
configure, and disable the corresponding CC_VENDOR code in common.mk.
- Disabled default compiler selection (usually gcc) in the sub-configs'
various make_def.mk files.
Details:
- Removed some old conditional code in config/knl/make_defs.mk that
added -lmemkind to LDFLAGS if DEBUG_TYPE was not 'sde' and inserted
code into common.mk that affirmatively filters out -lmemkind from
LDFLAGS if DEBUG_TYPE is 'sde'. (Thanks to Dave Love for reporting
this issue.) Other minor cleanups to neighboring code in common.mk.
- Updated CRVECFLAGS in knl/make_defs.mk to be based on -march=knl,
and then AVX-512 functionality is manually removed via various
-mno-avx512* flags. Also, make the setting of CRVECFLAGS conditional
on CC_VENDOR. Similar change to skx/make_defs.mk.
- Comment/whitespace updates.
Details:
- Renamed CVECFLAGS variables in sub-configurations' make_defs.mk files
to CKVECFLAGS.
- Added default defintions of two new make variables to most sub-
configurations' make_defs.mk files--CROPTFLAGS and CRVECFLAGS--
which correspond to reference kernel analogues of the CKOPTFLAGS
and CKVECFLAGS, which track optimization and vectorization flags for
optimized kernels. Currently, two sub-configurations (knl and skx)
explicitly set CRVECFLAGS to non-default values (using AVX2 instead of
AVX-512 for reference kernels. Thanks to Jeff Hammond, whose feedback
prompted me to make this change (issue #187).
- Changed common.mk so that the get-refkern-cflags-for function returns
the flags associated with the given sub-configuration's CROPTFLAGS
and CRVECFLAGS (instead of CKOPTFLAGS and CKVECFLAGS).
Details:
- Register use of level-1v zen intrinsic kernels for amaxv, axpyv, dotv,
dotxv, and scalv, as well asl level-1f zen intrinsic kernels for axpyf
and dotxf. This works because these kernels simply target AVX/AVX2,
and therefore work without modification on haswell hardware.
- Switch to use of zen microkernels in bli_cntx_init_haswell.c. The zen
kernels are essentially identical to those used by haswell, except that
now zen kernels are a bit more up-to-date. In the future, I may
continue to maintain duplicates, or I may keep the kernels named after
one architecture (zen or haswell) but used by both sub-configurations.
- In config_registry, enable use of both haswell and zen kernels for the
haswell sub-configuration. This is necessary in order to make zen
kernels visible when registering kernels in bli_cntx_init_haswell.c.
- Enable use of assembly-based complex gemm microkernels for zen,
bli_cgemm_zen_asm_3x8() and bli_zgemm_zen_asm_3x4(), in
bli_cntx_init_zen.c. This was actually intended for 1681333.
Details:
- Rewrote code that selects the compiler for the purposes of compiling
the auto-detection executable. CC (if specified) is tried first. Then
gcc. Then clang. The absolute fallback is cc. The previous code was
sort of broken, and seemed to unintentionally always use gcc.
- Moved various configuration-agnostic flags from config/*/make_defs.mk
files to common.mk. The new mechanism appends the configuration-
agnostic flags to the various compiler flag variables initialized in
make_defs.mk. Flags specific to the sub-configuration are still set
in make_defs.mk.
- Added -Wno-tautological-compare to CMISCFLAGS when clang is in use.
Also added the flag to the compiler instantiation during configure-
time hardware detection (when clang is selected).
- Added some missing (but mostly-optional) quotes to configure script.
Details:
- Added a "generic" configuration that leaves the default blocksizes and
kernels unchanged. This replaces the older "reference" configuration.
Updated auto-detect script and code accordingly.
- Added support for generic configuration to arch_t (bli_type_defs.h),
bli_gks_init() (bli_gks.c), and bli_arch_config.h
- Moved bli_arch_query_id() to bli_arch.c (and prototype to bli_arch.h).
- Whitespace changes to configurations' make_defs.mk files.
Details:
- Renamed the various configurations' "bli_arch_<configname>.h" header files
(replacing "arch" with "family") to free up the 'bli_arch' namespace for a
different purpose (hardware detection).
- Renamed "bli_arch.h" and "bli_arch_pre_macro_defs.h" in frame/include to
"bli_arch_config.h" and "bli_arch_config_pre.h", respectively.
Details:
- Reworked the build system around a configuration registry file, named
config_registry', that identifies valid configuration targets, their
constituent sub-configurations, and the kernel sets that are needed by
those sub-configurations. The build system now facilitates the building
of a single library that can contains kernels and cache/register
blocksizes for multiple configurations (microarchitectures). Reference
kernels are also built on a per-configuration basis.
- Updated the Makefile to use new variables set by configure via the
config.mk.in template, such as CONFIG_LIST, KERNEL_LIST, and KCONFIG_MAP,
in determining which sub-configurations (CONFIG_LIST) and kernel sets
(KERNEL_LIST) are included in the library, and which make_defs.mk files'
CFLAGS (KCONFIG_MAP) are used when compiling kernels.
- Reorganized 'kernels' directory into a "flat" structure. Renamed kernel
functions into a standard format that includes the kernel set name
(e.g. 'haswell'). Created a "bli_kernels_<kernelset>.h" file in each
kernels sub-directory. These files exist to provide prototypes for the
kernels present in those directories.
- Reorganized reference kernels into a top-level 'ref_kernels' directory.
This directory includes a new source file, bli_cntx_ref.c (compiled on
a per-configuration basis), that defines the code needed to initialize
a reference context and a context for induced methods for the
microarchitecture in question.
- Rewrote make_defs.mk files in each configuration so that the compiler
variables (e.g. CFLAGS) are "stored" (renamed) on a per-configuration
basis.
- Modified bli_config.h.in template so that bli_config.h is generated with
#defines for the config (family) name, the sub-configurations that are
associated with the family, and the kernel sets needed by those
sub-configurations.
- Deprecated all kernel-related information in bli_kernel.h and transferred
what remains to new header files named "bli_arch_<configname>.h", which
are conditionally #included from a new header bli_arch.h. These files
are still needed to set library-wide parameters such as custom
malloc()/free() functions or SIMD alignment values.
- Added bli_cntx_init_<configname>.c files to each configuration directory.
The files contain a function, named the same as the file, that initializes
a "native" context for a particular configuration (microarchitecture). The
idea is that optimized kernels, if available, will be initialized into
these contexts. Other fields will retain pointers to reference functions,
which will be compiled on a per-configuration basis. These bli_cntx_init_*()
functions will be called during the initialization of the global kernel
structure. They are thought of as initializing for "native" execution, but
they also form the basis for contexts that use induced methods. These
functions are prototyped, along with their _ref() and _ind() brethren, by
prototype-generating macros in bli_arch.h.
- Added a new typedef enum in bli_type_defs.h to define an arch_t, which
identifies the various sub-configurations.
- Redesigned the global kernel structure (gks) around a 2D array of cntx_t
structures (pointers to cntx_t, actually). The first dimension is indexed
over arch_t and the inner dimension is the ind_t (induced method) for
each microarchitecture. When a microarchitecture (configuration) is
"registered" at init-time, the inner array for that configuration in the
2D array is initialized (and allocated, if it hasn't been already). The
cntx_t slot for BLIS_NAT is initialized immediately and those for other
induced method types are initialized and cached on-demand, as needed. At
cntx_t registration, we also store function pointers to cntx_init functions
that will initialize (a) "reference" contexts and (b) contexts for use with
induced methods. We don't cache the full contexts for reference contexts
since they are rarely needed. The functions that initialize these two kinds
of contexts are generated automatically for each targeted sub-configuration
from cpp-templatized code at compile-time. Induced method contexts that
need "stage" adjustments can still obtain them via functions in
bli_cntx_ind_stage.c.
- Added new functions and functionality to bli_cntx.c, such as for setting
the level-1f, level-1v, and packm kernels, and for converting a native
context into one for executing an induced method.
- Moved the checking of register/cache blocksize consistency from being cpp
macros in bli_kernel_macro_defs.h to being runtime checks defined in
bli_check.c and called from bli_gks_register_cntx() at the time that the
global kernel structure's internal context is initialized for a given
microarchitecture/configuration.
- Deprecated all of the old per-operation bli_*_cntx.c files and removed
the previous operation-level cntx_t_init()/_finalize() invocations.
Instead, we now query the gks for a suitable context, usually via
bli_gks_query_cntx().
- Deprecated support for the 3m2 and 3m3 induced methods. (They required
hackery that I was no longer willing to support.)
- Consolidated the 1e and 1r packm kernels for any given register blocksize
into a single kernel that will branch on the schema and support packing
to both formats.
- Added the cntx_t* argument to all packm kernel signatures.
- Deprecated the local function pointer array in all bli_packm_cxk*.c files
and instead obtain the packm kernel from the cntx_t.
- Added bli_calloc_intl(), which serves as the calloc-equivalent to to
bli_malloc_intl(). Useful when we wish to allocate and initialize to
zero/NULL.
- Converted existing cpp macro functions defined in bli_blksz.h, bli_func.h,
bli_cntx.h into static functions.
Details:
- Dropped 'u' from the list of modifiers passed into the library archiver
ar. Previously, "cru" was used, while now we employ only "cr". This
change was prompted by a warning observed on Ubuntu 16.04:
ar: `u' modifier ignored since `D' is the default (see `U')
This caused me to realize that the default mode causes timestamps to be
zero, and thus the 'u' option, which causes only changed object files to
be inserted, is not applicable.
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:
- Replaced permutation-based implementations in bli_gemm_asm_d4x12.c, which
defines 4x24 single real and 4x12 double real gemm microkernels, with
broadcast-based implementations. (The previous microkernel file has been
moved to an 'old' subdirectory.)
Details:
- Updated s and d microkernels in bli_gemm_asm_d8x6.c to relax alignment
constraints.
- Added missing c and z microkernels, which are based on the corresponding
kernels in the d6x8 set.
- This completes the d8x6 set (which may be used for situations when it
is desirable to have a microkernel with a column preference).
Intel compilers include a highly optimized math library (libimf) that
should be used instead of GNU libm.
yes, this change is for ALL targets, including those that are not
supported by the Intel compiler. there is no harm in doing this, and it
is future-proof in the event that the Intel compilers support other
architectures.
Details:
- Defined cgemm (3x8) and zgemm (3x4) micro-kernels for haswell-based
architectures. As with their real domain brethren, these kernels perfer
row storage, (though this doesn't affect most users due to high-level
optimizations in most level-3 operations that induce a transpose to
whatever storage preference the kernel may have).
Details:
- Added two new sets of [sd]gemm micro-kernels for haswell architectures,
one that is 4x24/4x12 (s and d) and one that is 6x16/6x8.
- Changed the haswell configuration to use the 6x16/6x8 micro-kernels
by default.
- Updated various Makefiles, in test, test/3m4m, and testsuite.
Options to configure have been added for:
- Setting the internal BLIS and BLAS/CBLAS integer sizes.
- Enabling and disabling the BLAS and CBLAS layers.
Additionally, configure options which require defining macros (the above plus the threading model), write their macros to the automatically-generated bli_config.h file in the top-level build directory. The old bli_config.h files in the config dirs were removed, and any kernel-related macros (SIMD size and alignment etc.) were moved to bli_kernel.h. The Makefiles were also modified to find the new bli_config.h file.
Lastly, support for OMP in clang has been added (closes#56).
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.
1) Add -- options.
2) Add -d/--enable-debug option to enable debugging symbols with and without optimization.
3) Allow user to specify CC at configure time, and determine vendor (gcc/icc/etc.). For now configurations enforce a particular vendor.
4) Add make V=[0,1] option to control build verbosity.
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.
