To ensure that unit tests include the right version of the code, aml.h
should include local header. This also works for installed programs.

[refactor] use autoconf + m4 for version mngmt

See merge request !36

Use m4 to define autoconf-level version variables, following the naming
scheme of semver.org

To make use of these variables in the headers and sources, a
generated-header is added in aml/utils/version.h

Also add a simple test for that part of the lib.

Aml version

See merge request !35

Continue splitting headers.

See merge request !34

The overall strategy for now is to split the implementations into
distinct headers, but keep the generic APIs inside the main aml.h

Related to #27

The overall strategy for now is to split the implementations into
distinct headers, but keep the generic APIs inside the main aml.h

Related to #27

[doc] add documentation on design/workflow

Closes #29

See merge request !33

Document design principles, coding guidelines, contribution workflow.

[refactor] make bitmasks closer to a real type

See merge request !32

Nicolas Denoyelle authored Mar 14, 2019 and Swann Perarnau committed Mar 14, 2019

While the current bitmask management is heavily inspired by libnuma, it is not
as easy to use has the libnuma API. This patch is an attempt to refactor
to code towards something cleaner.

Add No-op request capability to Scratch

Closes #18

See merge request !31

If an operation should not trigger any actual work, create a no-op
request to deal with it properly.

Related to #18.

Reorganisation of aml files

See merge request !30

Nicolas Denoyelle authored Mar 12, 2019 and Swann Perarnau committed Mar 13, 2019

- create one directory per building block in src and include
- keep one directory for tests, \
  otherwise automake make them "test suites"
- move to AC_OPENMP, which is from autoconf 2.62 (2008)

[ci] URL fix for gitlab CI.

See merge request !29

[fix] Embed custom jemalloc into libaml

Closes #26

See merge request !28

Force libtool to static link the PIC version of our jemalloc import into
libaml, making libaml standalone. This requires us to test some
additional libraries in our own configure (pthread, and dlopen).

This also solves the long-standing issue of `make check` only working after
`make install`, while removing our custom jemalloc from the installed
libraries.

Fixes #26.

Mbind is giving us trouble again, will need to spend time looking at it
carefully.

Modify the benchmarks to use the new functions.

Instead of asking the user to provide the offsets into a tiling, add a
function providing a tileid. This tileid corresponds to the in-memory
order of tiles, making the tilestart functions a lot simpler.

We still need to split the tileid for tilestart because scratchpads
create requests based on tileids.

Also add a unit test for tiling_2d, to make sure we're not doing
anything crazy.

Tiling 2d and its interfaces wasn't the right way of looking at 2d grids
of tiles. Rename the contig ones to provide the required features.

This patch provides aligned allocations for all areas. Simple tests
included.

Note that I haven't tested if it conflicts with arena-wide alignements.

We are going to need more of those flags, and keeping track of the
conversions is tricky. So let's use a copy of the macros.

mbind requires that the input ptr be aligned on a page.

NOTE: we could also figure out a way to ask jemalloc
for page-aligned allocations, but that would probably
be too much for each alloc.

Prevents issue with the new arena/alloc logic. This way, no allocation
can end up overlapping the pages of another one on a different binding.

The way jemalloc handles big allocations can often result in surprising
calls to mmap/mbind (splitting allocations, rounded up sizes). It also
makes the path between an aml_alloc and mbind quite difficult to see.
More worrying, if jemalloc reuses a previous allocation, the mbind will
not be called again, which might result in the wrong binding happening.

To fix those issues, we move the mbind logic to be around the
allocations returned from jemalloc. This will ensure that we always bind
properly. The only issue is that it might slow down allocations.

It can also cause issues if the same arena is used by multiple areas, as
allocations might be overlapping a page. We will move away from sharing
arenas for benchmarks from now on.

Duplicate changes to dgemm_nofetch.c from commits
6a0d1cbd and
9326c388.

We had a race condition where OpenMP threads were accidentally reusing
the same variables on stack, resulting in races and incorrect results.
The number of FP operations was probably correct, although the memory
accesses to the arrays may have been wrong.

The A, B, and C matrices are tiled (tiles in A are also transposed).
Add initialization code for A and B and conversion code for C that
respects the tiling, thus enabling a direct comparison of results with
mkl and vanilla.

Benchmarks were crashing due to bad parameters.

Use the new 2D tilings for dgemm_prefetch, also refactor the code to
match the rest of the benchmarks. The code should be a lot more cleaner
now.

Fix dgemm_noprefetch to match pattern from @suchyb in #19.
In order to do so we split our 2d tiling into column-major and
row-major ones. Note that those are refering to the order of the tiles,
not the internal data of a tile, as a tiling should be agnostic to it.

1. refactor overall main function to match intended benchmark interface.
2. Use the new tiling type to cleanup the noprefetch version. Careful
inspection unearthed some bad offset computations, which are fixed here.
3. double checked the way we were spawning threads, new code should be
straightforward.

I believe that code should be easier to read and to play with.
Converting the prefetch versions might not be as easy.