Details:
- Updated the typed and object APIs to include language on the rntm_t
parameters in the expert interfaces.
- Updated README to include link to object API.
Details:
- Modified a few sections to take advantage of a feature of markdown
that allows a bullet or enumeration to have multiple paragraphs. This
is a trial run to make sure the indentation looks good when rendered
in a web browser.
Details:
- Filled in remaining section on object creation function reference
of BLISObjectAPI.md. All object management functions demonstrated as
part of the example code in examples/oapi are now documented, as well
as some other functions that are not shown in the example code.
- Updated variuos links (mostly in function index) to correctly point to
the object API reference instead of the typed API reference.
- Added documentation to getijm, setijm.
Details:
- Defined a new struct datatype, rntm_t (runtime), to house the thrloop
field of the cntx_t (context). The thrloop array holds the number of
ways of parallelism (thread "splits") to extract per level-3
algorithmic loop until those values can be used to create a
corresponding node in the thread control tree (thrinfo_t structure),
which (for any given level-3 invocation) usually happens by the time
the macrokernel is called for the first time.
- Relocating the thrloop from the cntx_t remedies a thread-safety issue
when invoking level-3 operations from two or more application threads.
The race condition existed because the cntx_t, a pointer to which is
usually queried from the global kernel structure (gks), is supposed to
be a read-only. However, the previous code would write to the cntx_t's
thrloop field *after* it had been queried, thus violating its read-only
status. In practice, this would not cause a problem when a sequential
application made a multithreaded call to BLIS, nor when two or more
application threads used the same parallelization scheme when calling
BLIS, because in either case all application theads would be using
the same ways of parallelism for each loop. The true effects of the
race condition were limited to situations where two or more application
theads used *different* parallelization schemes for any given level-3
call.
- In remedying the above race condition, the application or calling
library can now specify the parallelization scheme on a per-call basis.
All that is required is that the thread encode its request for
parallelism into the rntm_t struct prior to passing the address of the
rntm_t to one of the expert interfaces of either the typed or object
APIs. This allows, for example, one application thread to extract 4-way
parallelism from a call to gemm while another application thread
requests 2-way parallelism. Or, two threads could each request 4-way
parallelism, but from different loops.
- A rntm_t* parameter has been added to the function signatures of most
of the level-3 implementation stack (with the most notable exception
being packm) as well as all level-1v, -1d, -1f, -1m, and -2 expert
APIs. (A few internal functions gained the rntm_t* parameter even
though they currently have no use for it, such as bli_l3_packm().)
This required some internal calls to some of those functions to
be updated since BLIS was already using those operations internally
via the expert interfaces. For situations where a rntm_t object is
not available, such as within packm/unpackm implementations, NULL is
passed in to the relevant expert interfaces. This is acceptable for
now since parallelism is not obtained for non-level-3 operations.
- Revamped how global parallelism is encoded. First, the conventional
environment variables such as BLIS_NUM_THREADS and BLIS_*_NT are only
read once, at library initialization. (Thanks to Nathaniel Smith for
suggesting this to avoid repeated calls getenv(), which can be slow.)
Those values are recorded to a global rntm_t object. Public APIs, in
bli_thread.c, are still available to get/set these values from the
global rntm_t, though now the "set" functions have additional logic
to ensure that the values are set in a synchronous manner via a mutex.
If/when NULL is passed into an expert API (meaning the user opted to
not provide a custom rntm_t), the values from the global rntm_t are
copied to a local rntm_t, which is then passed down the function stack.
Calling a basic API is equivalent to calling the expert APIs with NULL
for the cntx and rntm parameters, which means the semantic behavior of
these basic APIs (vis-a-vis multithreading) is unchanged from before.
- Renamed bli_cntx_set_thrloop_from_env() to bli_rntm_set_ways_for_op()
and reimplemented, with the function now being able to treat the
incoming rntm_t in a manner agnostic to its origin--whether it came
from the application or is an internal copy of the global rntm_t.
- Removed various global runtime APIs for setting the number of ways of
parallelism for individual loops (e.g. bli_thread_set_*_nt()) as well
as the corresponding "get" functions. The new model simplifies these
interfaces so that one must either set the total number of threads, OR
set all of the ways of parallelism for each loop simultaneously (in a
single function call).
- Updated sandbox/ref99 according to above changes.
- Rewrote/augmented docs/Multithreading.md to document the three methods
(and two specific ways within each method) of requesting parallelism
in BLIS.
- Removed old, disabled code from bli_l3_thrinfo.c.
- Whitespace changes to code (e.g. bli_obj.c) and docs/BuildSystem.md.
Details:
- Added documentation to docs/BLISObjectAPI.md for a handful of
commonly-used obj_t accessor functions.
- Minor updates to docs/BLISTypedAPI.md.
Details:
- Deleted some lines from the API function signatures that did not
belong (and were only left over from the copy-paste of the typed API).
- Fixed some paragraph-in-bullet indentation.
Details:
- Added first draft of BLISObjectAPI.md. (Object management section is
still missing.)
- Small fixes to BLISTypedAPI.md found while writing BLISObjectAPI.md.
- In various .md files, changed ``` verbatim blocks to language
attributes (e.g. ```c for C code).
Details:
- Added missing pthreads library linking to example makefile in
docs/BuildSystem.md, as well as similar language to build requirements
at the beginning of the document. Thanks to Stefanos Mavros for
bringing this to our attention.
- Updated CREDITS file.
Details:
- Minor updates to BLISTypedAPI.md, mostly to bring terminology
up-to-date with the new "typed API" classification.
- Added contents section to KernelsHowTo.md.
Details:
- Added missing level-1v operations to BLISTypedAPI (e.g. axpbyv,
xpbyv).
- Updated broken linkes in KernelsHowTo.md based on misnamed anchors.
- Other minor changes.
Details:
- Within the documents in docs/*.md, reverted links to other local
documents to relative paths.
- Fixed some links/documents that did not yet have the '.md' suffix.
- Testing whether we can use relative links ('docs/BLISTypedAPI.md')
from within README.md.
Details:
- Updated most markdown links in the documents/wikis to use absolute
paths instead of the relative paths that were in use previously.
A few links were not updated, except for adding a ".md" to reflect
the documents' new names, in order to test whether relative
linking still works.
Details:
- Exported all github wikis to a new 'docs' directory.
- Renamed 'BLISAPIQuickReference' wiki to 'BLISTypedAPI' and removed
all cntx_t* arguments from the (now non-expert) APIs (with the
exception of the kernel APIs).
- Added section to BuildSystem documenting new ARG_MAX hack.
