* fix remaining flake8 errors
* imports: sort imports everywhere in Spack
We enabled import order checking in #23947, but fixing things manually drives
people crazy. This used `spack style --fix --all` from #24071 to automatically
sort everything in Spack so PR submitters won't have to deal with it.
This should go in after #24071, as it assumes we're using `isort`, not
`flake8-import-order` to order things. `isort` seems to be more flexible and
allows `llnl` mports to be in their own group before `spack` ones, so this
seems like a good switch.
* util.tty.log: read up to 100 lines if ready
Rework to read up to 100 lines from the captured stdin as long as data
is ready to be read immediately. Adds a helper function to poll with
`select` for ready data. This showed a roughly 5-10x perf improvement
for high-rate writes through the logger with relatively short lines.
* util.tty.log: Defer flushes to end of ready reads
Rather than flush per line, flush per set of reads. Since this is a
non-blocking loop, the total perceived wait is short.
* util.tty.log: only scan each line once, usually
Rather than always find all control characters then substitute them all,
use `subn` to count the number of control characters replaced. Only if
control characters exist find out what they are. This could be made
truly single pass with sub with a function, but it's a more intrusive
change and this got 99%ish of the performance improvement (roughly
another 2x in some cases).
* util.tty.log: remove check for `readable`
Python < 3 does not support a readable check on streams, should not be
necessary here since we control the only use and it's explicitly a
stream to be read.
The loading protocol mandates that the the module we are going
to import needs to be already in sys.modules before its code is
executed, so to prevent unbounded recursions and multiple loading.
Loading a module from file exits early if the module is already
in sys.modules
In debug mode, processes taking an exclusive lock write out their node name to
the lock file. We were using `getfqdn()` for this, but it seems to produce
inconsistent results when used from within some github actions containers.
We get this error because getfqdn() seems to return a short name in one place
and a fully qualified name in another:
```
File "/home/runner/work/spack/spack/lib/spack/spack/test/llnl/util/lock.py", line 1211, in p1
assert lock.host == self.host
AssertionError: assert 'fv-az290-764....cloudapp.net' == 'fv-az290-764'
- fv-az290-764.internal.cloudapp.net
+ fv-az290-764
!!!!!!!!!!!!!!!!!!!! Interrupted: stopping after 1 failures !!!!!!!!!!!!!!!!!!!!
== 1 failed, 2547 passed, 7 skipped, 22 xfailed, 2 xpassed in 1238.67 seconds ==
```
This seems to stem from https://bugs.python.org/issue5004.
We don't really need to get a fully qualified hostname for debugging, so use
`gethostname()` because its results are more consistent. This seems to fix the
issue.
Signed-off-by: vsoch <vsoch@users.noreply.github.com>
We set LC_ALL=C to encourage a build process to generate ASCII
output (so our logger daemon can decode it). Most packages
respect this but it appears that intel-oneapi-compilers does
not in some cases (see #22813). This reads the output of the build
process as UTF-8, which still works if the build process respects
LC_ALL=C but also works if the process generates UTF-8 output.
For Python >= 3.7 all files are opened with UTF-8 encoding by
default. Python 2 does not support the encoding argument on
'open', so to support Python 2 the files would have to be
opened in byte mode and explicitly decoded (as a side note,
this would be the only way to handle other encodings without
being informed of them in advance).
We have been using the `@llnl.util.lang.key_ordering` decorator for specs
and most of their components. This leverages the fact that in Python,
tuple comparison is lexicographic. It allows you to implement a
`_cmp_key` method on your class, and have `__eq__`, `__lt__`, etc.
implemented automatically using that key. For example, you might use
tuple keys to implement comparison, e.g.:
```python
class Widget:
# author implements this
def _cmp_key(self):
return (
self.a,
self.b,
(self.c, self.d),
self.e
)
# operators are generated by @key_ordering
def __eq__(self, other):
return self._cmp_key() == other._cmp_key()
def __lt__(self):
return self._cmp_key() < other._cmp_key()
# etc.
```
The issue there for simple comparators is that we have to bulid the
tuples *and* we have to generate all the values in them up front. When
implementing comparisons for large data structures, this can be costly.
This PR replaces `@key_ordering` with a new decorator,
`@lazy_lexicographic_ordering`. Lazy lexicographic comparison maps the
tuple comparison shown above to generator functions. Instead of comparing
based on pre-constructed tuple keys, users of this decorator can compare
using elements from a generator. So, you'd write:
```python
@lazy_lexicographic_ordering
class Widget:
def _cmp_iter(self):
yield a
yield b
def cd_fun():
yield c
yield d
yield cd_fun
yield e
# operators are added by decorator (but are a bit more complex)
There are no tuples that have to be pre-constructed, and the generator
does not have to complete. Instead of tuples, we simply make functions
that lazily yield what would've been in the tuple. If a yielded value is
a `callable`, the comparison functions will call it and recursively
compar it. The comparator just walks the data structure like you'd expect
it to.
The ``@lazy_lexicographic_ordering`` decorator handles the details of
implementing comparison operators, and the ``Widget`` implementor only
has to worry about writing ``_cmp_iter``, and making sure the elements in
it are also comparable.
Using this PR shaves another 1.5 sec off the runtime of `spack buildcache
list`, and it also speeds up Spec comparison by about 30%. The runtime
improvement comes mostly from *not* calling `hash()` `_cmp_iter()`.
Sometimes we need to patch a file that is a dependency for some other
automatically generated file that comes in a release tarball. As a
result, make tries to regenerate the dependent file using additional
tools (e.g. help2man), which would not be needed otherwise.
In some cases, it's preferable to avoid that (e.g. see #21255). A way
to do that is to save the modification timestamps before patching and
restoring them afterwards. This PR introduces a context wrapper that
does that.
* sbang pushed back to callers;
star moved to util.lang
* updated unit test
* sbang test moved; local tests pass
Co-authored-by: Nathan Hanford <hanford1@llnl.gov>
- [x] add `concretize.lp`, `spack.yaml`, etc. to licensed files
- [x] update all licensed files to say 2013-2021 using
`spack license update-copyright-year`
- [x] appease mypy with some additions to package.py that needed
for oneapi.py
I lost my mind a bit after getting the completion stuff working and
decided to get Mypy working for spack as well. This adds a
`.mypy.ini` that checks all of the spack and llnl modules, though
not yet packages, and fixes all of the identified missing types and
type issues for the spack library.
In addition to these changes, this includes:
* rename `spack flake8` to `spack style`
Aliases flake8 to style, and just runs flake8 as before, but with
a warning. The style command runs both `flake8` and `mypy`,
in sequence. Added --no-<tool> options to turn off one or the
other, they are on by default. Fixed two issues caught by the tools.
* stub typing module for python2.x
We don't support typing in Spack for python 2.x. To allow 2.x to
support `import typing` and `from typing import ...` without a
try/except dance to support old versions, this adds a stub module
*just* for python 2.x. Doing it this way means we can only reliably
use all type hints in python3.7+, and mypi.ini has been updated to
reflect that.
* add non-default black check to spack style
This is a first step to requiring black. It doesn't enforce it by
default, but it will check it if requested. Currently enforcing the
line length of 79 since that's what flake8 requires, but it's a bit odd
for a black formatted project to be quite that narrow. All settings are
in the style command since spack has no pyproject.toml and I don't
want to add one until more discussion happens. Also re-format
`style.py` since it no longer passed the black style check
with the new length.
* use style check in github action
Update the style and docs action to use `spack style`, adding in mypy
and black to the action even if it isn't running black right now.
Users can add test() methods to their packages to run smoke tests on
installations with the new `spack test` command (the old `spack test` is
now `spack unit-test`). spack test is environment-aware, so you can
`spack install` an environment and then run `spack test run` to run smoke
tests on all of its packages. Historical test logs can be perused with
`spack test results`. Generic smoke tests for MPI implementations, C,
C++, and Fortran compilers as well as specific smoke tests for 18
packages.
Inside the test method, individual tests can be run separately (and
continue to run best-effort after a test failure) using the `run_test`
method. The `run_test` method encapsulates finding test executables,
running and checking return codes, checking output, and error handling.
This handles the following trickier aspects of testing with direct
support in Spack's package API:
- [x] Caching source or intermediate build files at build time for
use at test time.
- [x] Test dependencies,
- [x] packages that require a compiler for testing (such as library only
packages).
See the packaging guide for more details on using Spack testing support.
Included is support for package.py files for virtual packages. This does
not change the Spack interface, but is a major change in internals.
Co-authored-by: Tamara Dahlgren <dahlgren1@llnl.gov>
Co-authored-by: wspear <wjspear@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
- [x] Solver now uses the Python interface to clingo
- [x] can extract unsatisfiable cores from problems when things go wrong
- [x] use Python callbacks for versions instead of choice rules (this may
ultimately hurt performance)
Spack creates a separate process to do package installation. Different
operating systems and Python versions use different methods to create
it but up until Python 3.8 both Linux and Mac OS used "fork" (which
duplicates process memory, file descriptor table, etc.).
Python >= 3.8 on Mac OS prefers creating an entirely new process
(referred to as the "spawn" start method) because "fork" was found to
cause issues (in other words "spawn" is the default start method used
by multiprocessing.Process). Spack was dependent on the particular
behavior of fork to replicate process memory and transmit file
descriptors.
This PR refactors the Spack internals to support starting a child
process with the "spawn" method. To achieve this, it makes the
following changes:
- ensure that the package repository and other global state are
transmitted to the child process
- ensure that file descriptors are transmitted to the child process in
a way that works with multiprocessing and spawn
- make all the state needed for the build process and tests picklable
(package, stage, etc.)
- move a number of locally-defined functions into global scope so that
they can be pickled
- rework tests where needed to avoid using local functions
This PR also reworks sbang tests to work on macOS, where temporary
directories are deeper than the Linux sbang limit. We make the limit
platform-dependent (macOS supports 512-character shebangs)
See: #14102
* allow environments to specify dev-build packages
* spack develop and spack undevelop commands
* never pull dev-build packges from bincache
* reinstall dev_specs when code has changed; reinstall dependents too
* preserve dev info paths and versions in concretization as special variant
* move install overwrite transaction into installer
* move dev-build argument handling to package.do_install
now that specs are dev-aware, package.do_install can add
necessary args (keep_stage=True, use_cache=False) to dev
builds. This simplifies driving logic in cmd and env._install
* allow 'any' as wildcard for variants
* spec: allow anonymous dependencies
raise an error when constraining by or normalizing an anonymous dep
refactor concretize_develop to remove dev_build variant
refactor tests to check for ^dev_path=any instead of +dev_build
* fix variant class hierarchy
Fixes#18441
When writing an environment, there are cases where the lock file for
the environment may be removed. In this case there was a period
between removing the lock file and writing the new manifest file
where an exception could leave the manifest in its old state (in
which case the lock and manifest would be out of sync).
This adds a context manager which is used to restore the prior lock
file state in cases where the manifest file cannot be written.
As detailed in https://bugs.python.org/issue33725, starting new
processes with 'fork' on Mac OS is not guaranteed to work in general.
As of Python 3.8 the default process spawning mechanism was changed
to avoid this issue.
Spack depends on the fork-based method to preserve file descriptors
transparently, to preserve global state, and to avoid pickling some
objects. An effort is underway to remove dependence on fork-based
process spawning (see #18205). In the meantime, this allows Spack to
run with Python 3.8 on Mac OS by explicitly choosing to use 'fork'.
Co-authored-by: Peter Josef Scheibel <scheibel1@llnl.gov>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Todd Gamblin <tgamblin@llnl.gov>
Some of the feature flags are named differently and clwb is missing on
my i7-1065G7. cascadelake and cannonlake might have similar problems but
I do not have access to those architectures to test.
The YAML config for paths and modules of external packages has
changed: the new format allows a single spec to load multiple
modules. Spack will automatically convert from the old format
when reading the configs (the updates do not add new essential
properties, so this change in Spack is backwards-compatible).
With this update, Spack cannot modify existing configs/environments
without updating them (e.g. “spack config add” will fail if the
configuration is in a format that predates this PR). The user is
prompted to do this explicitly and commands are provided. All
config scopes can be updated at once. Each environment must be
updated one at a time.
* switch from bool to int debug levels
* Added debug options and changed lock logging to use more detailed values
* Limit installer and timestamp PIDs to standard debug output
* Reduced verbosity of fetch/stage/install output, changing most to debug level 1
* Combine lock log methods; change build process install to debug
* Changed binary cache install messages to extraction messages
* Separate Apple Clang from LLVM Clang
Apple Clang is a compiler of its own. All places
referring to "-apple" suffix have been updated.
* Hack to use a dash in 'apple-clang'
To be able to use autodoc from Sphinx we need
a valid Python name for the module that contains
Apple's Clang code.
* Updated packages to account for the existence of apple-clang
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Added unit test for XCode related functions
Co-authored-by: Gregory Becker <becker33@llnl.gov>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
After migrating to `travis-ci.com`, we saw I/O issues in our tests --
tests that relied on `capfd` and `capsys` were failing. We've also seen
this in GitHub actions, and it's kept us from switching to them so far.
Turns out that the issue is that using streams like `sys.stdout` as
default arguments doesn't play well with `pytest` and output redirection,
as `pytest` changes the values of `sys.stdout` and `sys.stderr`. if these
values are evaluated before output redirection (as they are when used as
default arg values), output won't be captured properly later.
- [x] replace all stream default arg values with `None`, and only assign stream
values inside functions.
- [x] fix tests we didn't notice were relying on this erroneous behavior
Makes the following changes:
* (Fixes#15620) tty configuration was failing when stdout was
redirected. The implementation now creates a pseudo terminal for
stdin and checks stdout properly, so redirections of stdin/out/err
should be handled now.
* Handles terminal configuration when the Spack process moves between
the foreground and background (possibly multiple times) during a
build.
* Spack adjusts terminal settings to allow users to to enable/disable
build process output to the terminal using a "v" toggle, abnormal
exit cases (like CTRL-C) could leave the terminal in an unusable
state. This is addressed here with a special-case handler which
restores terminal settings.
Significantly extend testing of process output logger:
* New PseudoShell object for setting up a master and child process
and configuring file descriptor inheritance between the two
* Tests for "v" verbosity toggle making use of the added PseudoShell
object
* Added `uniq` function which takes a list of elements and replaces
any consecutive sequence of duplicate elements with a single
instance (e.g. "112211" -> "121")
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Spack currently cannot run as a background process uninterrupted because some of the logging functions used in the install method (especially to create the dynamic verbosity toggle with the v key) cause the OS to issue a SIGTTOU to Spack when it's backgrounded.
This PR puts the necessary gatekeeping in place so that Spack doesn't do anything that will cause a signal to stop the process when operating as a background process.
Sometimes one needs to preserve the (relative order) of
mtimes on installed files. So it's better to just copy
over all the metadata from the source tree to the install
tree. If permissions need fixing, that will be done anyway
afterwards.
One major use case are resource()s:
They're unpacked in one place and then copied to their
final place using install_tree(). If the resource is a
source tree using autoconf/automake, resetting mtimes
uncorrectly might force unwanted autoconf/etc calls.
Fixes#9394Closes#13217.
## Background
Spack provides the ability to enable/disable parallel builds through two options: package `parallel` and configuration `build_jobs`. This PR changes the algorithm to allow multiple, simultaneous processes to coordinate the installation of the same spec (and specs with overlapping dependencies.).
The `parallel` (boolean) property sets the default for its package though the value can be overridden in the `install` method.
Spack's current parallel builds are limited to build tools supporting `jobs` arguments (e.g., `Makefiles`). The number of jobs actually used is calculated as`min(config:build_jobs, # cores, 16)`, which can be overridden in the package or on the command line (i.e., `spack install -j <# jobs>`).
This PR adds support for distributed (single- and multi-node) parallel builds. The goals of this work include improving the efficiency of installing packages with many dependencies and reducing the repetition associated with concurrent installations of (dependency) packages.
## Approach
### File System Locks
Coordination between concurrent installs of overlapping packages to a Spack instance is accomplished through bottom-up dependency DAG processing and file system locks. The runs can be a combination of interactive and batch processes affecting the same file system. Exclusive prefix locks are required to install a package while shared prefix locks are required to check if the package is installed.
Failures are communicated through a separate exclusive prefix failure lock, for concurrent processes, combined with a persistent store, for separate, related build processes. The resulting file contains the failing spec to facilitate manual debugging.
### Priority Queue
Management of dependency builds changed from reliance on recursion to use of a priority queue where the priority of a spec is based on the number of its remaining uninstalled dependencies.
Using a queue required a change to dependency build exception handling with the most visible issue being that the `install` method *must* install something in the prefix. Consequently, packages can no longer get away with an install method consisting of `pass`, for example.
## Caveats
- This still only parallelizes a single-rooted build. Multi-rooted installs (e.g., for environments) are TBD in a future PR.
Tasks:
- [x] Adjust package lock timeout to correspond to value used in the demo
- [x] Adjust database lock timeout to reduce contention on startup of concurrent
`spack install <spec>` calls
- [x] Replace (test) package's `install: pass` methods with file creation since post-install
`sanity_check_prefix` will otherwise error out with `Install failed .. Nothing was installed!`
- [x] Resolve remaining existing test failures
- [x] Respond to alalazo's initial feedback
- [x] Remove `bin/demo-locks.py`
- [x] Add new tests to address new coverage issues
- [x] Replace built-in package's `def install(..): pass` to "install" something
(i.e., only `apple-libunwind`)
- [x] Increase code coverage
VSX alitvec extensions are supported by PowerISA from v2.06 (Power7+), but might
not be listed in features.
FMA has been supported by PowerISA since Power1, but might not be listed in
features.
This commit adds these features to all the power ISA family sets.
This PR adds a `--format=bash` option to `spack commands` to
auto-generate the Bash programmable tab completion script. It can be
extended to work for other shells.
Progress:
- [x] Fix bug in superclass initialization in `ArgparseWriter`
- [x] Refactor `ArgparseWriter` (see below)
- [x] Ensure that output of old `--format` options remains the same
- [x] Add `ArgparseCompletionWriter` and `BashCompletionWriter`
- [x] Add `--aliases` option to add command aliases
- [x] Standardize positional argument names
- [x] Tests for `spack commands --format=bash` coverage
- [x] Tests to make sure `spack-completion.bash` stays up-to-date
- [x] Tests for `spack-completion.bash` coverage
- [x] Speed up `spack-completion.bash` by caching subroutine calls
This PR also necessitates a significant refactoring of
`ArgparseWriter`. Previously, `ArgparseWriter` was mostly a single
`_write` method which handled everything from extracting the information
we care about from the parser to formatting the output. Now, `_write`
only handles recursion, while the information extraction is split into a
separate `parse` method, and the formatting is handled by `format`. This
allows subclasses to completely redefine how the format will appear
without overriding all of `_write`.
Co-Authored-by: Todd Gamblin <tgamblin@llnl.gov>
`PackagePrefs` has had a class-level cache of data from `packages.yaml` for
a long time, but it complicates testing and leads to subtle errors,
especially now that we frequently manipulate custom config scopes and
environments.
Moving the cache to instance-level doesn't slow down concretization or
the test suite, and it just caches for the life of a `PackagePrefs`
instance (i.e., for a single cocncretization) so we don't need to worry
about global state anymore.
- [x] Remove class-level caches from `PackagePrefs`
- [x] Add a cached _spec_order object on each `PackagePrefs` instance
- [x] Remove all calls to `PackagePrefs.clear_caches()`
Commands like `spack blame` were printig poorly when redirected to files,
as colify reverts to a single column when redirected. This works for
list data but not tables.
- [x] Force a table by always passing `tty=True` from `colify_table()`
This commit removes the `python_version.py` unit test module
and the vendored dependencies `pyqver2.py` and `pyqver3.py`.
It substitutes them with an equivalent check done using
`vermin` that is run as a separate workflow via Github Actions.
This allows us to delete 2 vendored dependencies that are unmaintained
and substitutes them with a maintained tool.
Also, updates the list of vendored dependencies.
Our `LockTransaction` class was reading overly aggressively. In cases
like this:
```
1 with spack.store.db.read_transaction():
2 with spack.store.db.write_transaction():
3 ...
```
The `ReadTransaction` on line 1 would read in the DB, but the
WriteTransaction on line 2 would read in the DB *again*, even though we
had a read lock the whole time. `WriteTransaction`s were only
considering nested writes to decide when to read, but they didn't know
when we already had a read lock.
- [x] `Lock.acquire_write()` return `False` in cases where we already had
a read lock.
If a write transaction was nested inside a read transaction, it would not
write properly on release, e.g., in a sequence like this, inside our
`LockTransaction` class:
```
1 with spack.store.db.read_transaction():
2 with spack.store.db.write_transaction():
3 ...
4 with spack.store.db.read_transaction():
...
```
The WriteTransaction on line 2 had no way of knowing that its
`__exit__()` call was the last *write* in the nesting, and it would skip
calling its write function.
The `__exit__()` call of the `ReadTransaction` on line 1 wouldn't know
how to write, and the file would never be written.
The DB would be correct in memory, but the `ReadTransaction` on line 4
would re-read the whole DB assuming that other processes may have
modified it. Since the DB was never written, we got stale data.
- [x] Make `Lock.release_write()` return `True` whenever we release the
*last write* in a nest.
Lock transactions were actually writing *after* the lock was
released. The code was looking at the result of `release_write()` before
writing, then writing based on whether the lock was released. This is
pretty obviously wrong.
- [x] Refactor `Lock` so that a release function can be passed to the
`Lock` and called *only* when a lock is really released.
- [x] Refactor `LockTransaction` classes to use the release function
instead of checking the return value of `release_read()` / `release_write()`
Vendors for ARM come out of `/proc/cpuinfo` as hex numbers instead of readable strings.
- Add support for associating vendor names with the hex numbers.
- Also move these mappings from Python code to `microarchitectures.json`
- Move darwin feature name mappings to `microarchitectures.json` as well
* CUDA HeaderList: Unit Test
* Spec Header Dirs: Only first include/
Avoid matching recurringly nested include paths that usually
refer to internally shipped libraries in packages.
Example in CUDA Toolkit, shipping a libc++ fork internally
with libcu++ since 10.2.89:
`<prefix>/include/cuda/some/more/details/include/` or
`<prefix>/include/cuda/std/detail/libcxx/include`
regex: non-greedy first match of include
Co-Authored-By: Massimiliano Culpo <massimiliano.culpo@gmail.com>
* CUDA: Re-Enable 10.2.89 as Default
* Fixed x86-64 optimization flags for clang
* Fixed expected results in unit tests
Before the flags used where the one for llc, the underlying compiler from LLVM IR to machine assembly. It turns out that the semantic of `-march`, `-mtune` and `-mcpu` changes from clang front-end to llc.
I found no definitive reference for the flags submitted in this PR, but I checked the assembly on a vectorizable function using Godbolt's web-site.
* Add process to determine aarch64 microarchitecture
* add microarchitectures for thunderx2 and a64fx
* Add optimize flags for gcc on aarch64 family processors thunderx2 and a64fx.
* Add optimize flags for clang on aarch64 family processors thunderx2 and a64fx
* Add testing for thunderx2 and a64fx microarchitectures
