.gitlab-ci.yml: allow "test-wsl" to fail until runner problems are sorted out

See merge request !1145

Increasing buffer size from 4 kB to 64 kB speeds up read() bandwidth by
factor 4, according to a very simple benchmark.

Increasing buffer size from 4 kB to 64 kB speeds up read() bandwidth by
factor 4, according to a very simple benchmark.

Increase read buffer size to improve performance

See merge request !1142

At present, there isn't an easy way to print anything from `bst show`
that would give the users an idea of what the dependency graph looks
like. One could use things like `--deps build`, but that will just print
a list, without any information about the dependency edges.

Add `%{deps}`, `%{build-deps}` and `%{runtime-deps}` format strings to
`bst show` that would simply print the list of all dependencies, build
dependencies and runtime dependencies respectively.

Summary of changes:

  * buildstream/_frontend/cli.py: Add help for new format symbols.
  * buildstream/_frontend/widget.py: Add support for new format symbols
                                     for dependencies.
  * tests/frontend/show.py: Add tests for new format symbols.

_frontend: Allow printing dependencies using `bst show`

Closes #890

See merge request !1121

Fedora 29 includes python 3.7 so this is not needed anymore

.gitlab-ci.yml: Test with current fedora release: 29

See merge request !1137

We were indiscriminately retrieving the node's provenance data in the
`node_get()` function which was accounting for approximately a third of
the total runtime of `node_get()` which dominates pre-scheduler time in
`bst build`.  This change ameliorates that situation by only retrieving
the provenance data when it's actually needed.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

We know that nodes are typically one of:
str, list, dict, bool, tuple, NoneType or our ChainMap
Of these, dict and ChainMap are Mapping, only list is list
and the rest are returned unchanged.  We can reduce/defer our use
of isinstance here, dramatically, improving performance.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

A non-trivial proportion of time pre-scheduler in `bst build` is spent
copying (or chain-copying) nodes.  Approximately a quarter of the time
spent in that effort is in `isinstance()`.  This removes that CPU load.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

By re-using the isinstance replacements from earlier commits and
using a tuple of the string constants for checking for composition
markers, we reduce the cost of node_final_assertions by two thirds
in basic testing.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

While get_strict() doesn't look expensive per-se, it is called so many
times that it is valuable to cache the result once computed.  Since I
don't think it can change once it is computable, cache it immediately
that becomes possible and we save 20s in my test case.

Signed-off-by: Daniel Silverstone <daniel.silverstone@codethink.co.uk>

Further optimisations

See merge request !1131

Valentin David authored Feb 11, 2019 and Valentin David committed Feb 12, 2019

This was broken by 5ef19a0b.