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
- remove the old LGPL license headers from all files in Spack
- add SPDX headers to all files
- core and most packages are (Apache-2.0 OR MIT)
- a very small number of remaining packages are LGPL-2.1-only
- A package can depend on a special patched version of its dependencies.
- The `Spec` YAML (and therefore the hash) now includes the sha256 of
the patch in the `Spec` YAML, which changes its hash.
- The special patched version will be built separately from a "vanilla"
version of the same package.
- This allows packages to maintain patches on their dependencies
without affecting either the dependency package or its dependents.
This could previously be accomplished with special variants, but
having to add variants means the hash of the dependency changes
frequently when it really doesn't need to. This commit allows the
hash to change *just* for dependencies that need patches.
- Patching dependencies shouldn't be the common case, but some packages
(qmcpack, hpctoolkit, openspeedshop) do this kind of thing and it
makes the code structure mirror maintenance responsibilities.
- Note that this commit means that adding or changing a patch on a
package will change its hash. This is probably what *should* happen,
but we haven't done it so far.
- Only applies to `patch()` directives; `package.py` files (and their
`patch()` functions) are not hashed, but we'd like to do that in the
future.
- The interface looks like this: `depends_on()` can optionally take a
patch directive or a list of them:
depends_on(<spec>,
patches=patch(..., when=<cond>),
when=<cond>)
# or
depends_on(<spec>,
patches=[patch(..., when=<cond>),
patch(..., when=<cond>)],
when=<cond>)
- Previously, the `patch()` directive only took an `md5` parameter. Now
it only takes a `sha256` parameter. We restrict this because we want
to be consistent about which hash is used in the `Spec`.
- A side effect of hashing patches is that *compressed* patches fetched
from URLs now need *two* checksums: one for the downloaded archive and
one for the content of the patch itself. Patches fetched uncompressed
only need a checksum for the patch. Rationale:
- we include the content of the *patch* in the spec hash, as that is
the checksum we can do consistently for patches included in Spack's
source and patches fetched remotely, both compressed and
uncompressed.
- we *still* need the patch of the downloaded archive, because we want
to verify the download *before* handing it off to tar, unzip, or
another decompressor. Not doing so is a security risk and leaves
users exposed to any arbitrary code execution vulnerabilities in
compression tools.
