* Begin work on new OpenQASM 3 importer
* Tidy up public interface construction
* Add load-from-file entry point
* Create public experimental interface
* Credit John
While I've been the author of the Qiskit side of this, John wrote the
separate Rust crates that this depends on, so in principle this
contribution to Qiskit is from both of us.
Co-authored-by: John Lapeyre <jlapeyre@users.noreply.github.com>
* Fix lint
* Bump version (git hash) of dependency on openqasm3_parser
The qasm3 crate fails to build on Windows. This commit should fix this.
* Add support for barrier
* Handle measurement with new parser version
Measures, including broadcasted measures, are now supported following
increased support on the parser side. The updated version also includes
changes to a few APIs, and the include-path settings are now
configurable at the entry point, so we switch to those.
* Fix Rust 1.75 code
* Depend on released version of oq3_seamntics
Change crates/qasm3/Cargo.toml to depend on a released version of the parser.
that is the crate oq3_semantics, rather than a commit of the github repo.
All of the commits added since the commit previously specified in the
dependency are either updating the README or tweaking the github actions.
* Add tests of interface successes
* Add dummy library to manifest
* Depend on lastest version, 0.0.7, of openqasm3_parser crates
This commit does two things
* Change the dependency on oq3_semantcis from 0.0.5 to 0.0.7
* Change in build.rs to reflect chang in API for gate modifiers from
`gatecall.modifier()` returning `Option<GateModifier>` to
`gatecall.modifiers()` returning `&[GateModifier]`
* Expand on readme
* Centralise dependencies
* Fix typing and error typing
* Credit Matt
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Fix clippy
* Clone using GIL
* Format
* Improve error for unexpected angle
* Add test of gate broadcasting
* Add additional warnings about experimental interface
* Add release note
* Add comment about diagnostics in docs
* Correct release series comment
---------
Co-authored-by: John Lapeyre <jlapeyre@users.noreply.github.com>
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Move build configuration to static configuration files
This removes the vast majority of the dynamic code from our `setup.py`
files, and instead uses static configuration to specify them. As part
of this, we also use a single `VERSION.txt` as a single source of truth
for both the `qiskit-terra` main package and the `qiskit` "metapackage".
This should avoid a situation where we let the two get out of sync
during a release, which can cause us severe problems for fixing the
problem with respect to git tags.
* Update components changed in rebase
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Update contact email
* Remove reference to 'qiskit_pkg'
---------
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Remove legacy qasm2 parser
This commit removes the legacy qasm2 parser. This was the original
qasm parser in qiskit dating back to the very beginning of the project,
but it has been superseded in recent releases by the faster more
correct/strict rust parser. We no longer need to keep around two parsers
for qasm2. This commit removes the legacy one and it's associated
functions (like the ast converter).
* Remove api docs too
* Remove ast converter benchmark
* Remove QuantumCircuit.qasm() exporter method
This commit removes the QuantumCircuit.qasm() method as part of the
legacy parser removal. The exporter method was tied to some of the
internals in the now removed `qiskit.qasm` module. The qasm() method was
already queued up for eventual removal, this commit just pushes that
forward so we don't have to add more comptaibility code into it to
keep it working as is without the qiskit.qasm module existing anymore.
* Remove jupyter qasm widget
* Update tests
* Remove BlueprintCircuit .qasm() override
* Fix rst syntax in release note
* Apply suggestions from code review
Co-authored-by: Jake Lishman <jake@binhbar.com>
---------
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Replace `qiskit` metapackage with `qiskit-terra`
This commit completely removes the concept of the `qiskit-terra` package
from Qiskit main. The hitherto metapackage `qiskit` is promoted to be
the concrete code package.
This is a completely breaking change for packaging purposes for users,
as there is no clean upgrade path from `qiskit-terra` to `qiskit`; PyPI
and pip do not give us the tools to mark that the former obsoletes and
supersedes the latter. We intend to follow this PR with a technical
blog post somewhere explaining the changes, how users should adapt
("to install Qiskit 1.0, you must start a new venv"), and why we needed
to do this.
The "why", in part, is:
- the metapackage legacy causes awkward upgrade paths on every release;
some packages still depend on `qiskit-terra`, some on `qiskit`, and
pip will happily make those two get out-of-sync when upgrading a
transitive dependency with only a warning that users are used to
ignoring.
- with the 1.0 release, we (believe we) will have some leeway from users
to make such a breaking change.
- having the metapackage split makes it difficult for downstream
projects and developers to test against `main` - they always must
install both `qiskit-terra` and `qiskit`, and the latter has no
meaning in editable installs, so needs re-installing after each
version bump. Problems surrounding this have already broken CI for
Qiskit, for at least one internal IBM repo, and at least my dev
install of Qiskit. This could be improved a bit with more education,
but it's still always going to increase the barrier to entry and make
it much harder to do the right thing.
We will not publish any 1.0 or above version of `qiskit-terra`. All
dependents on Qiskit should switch their requirements to `qiskit`.
* Add manual uninstall for Neko
* Fix Windows paths
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Refer to stdlib documentation for odd shells
---------
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Add support for Python 3.12
Python 3.12.0 was released on 10-02-2023, this commit marks the start of
support for Python 3.12 in Qiskit. It adds the supported Pythonv ersion
in the package metadata and updates the CI configuration to run test
jobs on Python 3.12 and build Python 3.12 wheels on release.
Fixes: #10887
* Add release note
* Avoid deprecated `datetime.datetime.utcnow()` usage
In Python 3.12 `datetime.datetime.utcnow()` has been deprecated, being
replaced by: `datetime.datetime.now(datetime.UTC)`. This commit updates
the usage of `utcnow()` to follow the new convention.
* Adjust UTC usage to support Python 3.8
The recommended alternative for using utcnow() in the deprecation
warnings emitted by Python 3.12 are not compatible with Python 3.8. The
datetime.UTC alias was not added to Python until Python 3.11. To ensure
that the code is compatible with Python < 3.11 this commit updates all
the usage of datetime.UTC to use datetime.timezone.utc instead, which is
what datetime.UTC aliases to in Python >=3.11.
Pull request #11059 bumped the version on `main` to `1.0.0`, but
it missed the version in the toplevel `setup.py`. This PR
fixes that oversight.
In the current state, the `qiskit` and `qiskit-terra` python packages
cannot be simultaneously installed from `main`. This led to a CI
failure for us in the circuit knitting toolbox, which can be seen here:
https://github.com/Qiskit-Extensions/circuit-knitting-toolbox/actions/runs/6582190865/job/17883338486
For the 0.45.0 release we're going to push release candidates prior to
the release to enable testing before we cut the final release. In
preparation for tagging the first release candidate this commit updates
the version string to indicate it's a release candidate. This commit
should be the final commit on main for 0.45.0rc1 and what gets tagged as
0.45.0rc1 post-merge.
This commit updates the preset pass manager construction to only use
plugins for the init stage. To accomplish this the previously hard
coded built-in pass manager used for each optimization level are
refactored to be in a plugin named "default". One thing that is changed
in this PR is that the use of `generate_control_flow_options_check()`
is moved to the `pre_init` stage. The reason for this is because the way
the init stage was being constructed in the preset pass managers was to
do initial checking of any methods, and this was unconditionally being
run. This is a more logical fit for pre_init stage because it should run
before any specified plugin.
Fixes: #10687Fixes: #8661
* Initial: Optimization stage as a plugin.
* CI: Added fix to preset_passmanager test
* Simplify inner translation stage creation logic
This commit changes the plugin construction logic slightly to adjust
how the embedded translation stage is created. After #10621 merged we
only need to use the plugin interface to get the translation stage pass
manager.
* Update documentation
---------
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Migrate layout stage to plugins
This commit updates the preset pass manager construction to only use
plugins for the layout stage. To accomplish this the previously hard
coded built-in layout methods, trivial, dense, noise adpative, and
sabre are migrated to be exposed as built-in plugins. Additionally, the
special case of layout_method=None has been centralized into a standard
default method plugin, as the pass construction in this case involved
extra steps for each optimization level. This simplifies the preset pass
manager construction as now the layout stage is solely built via
plugins.
Fixes#9455
* Remove unnecessary if statement
* Handle invalid optimization levels in plugins
* Remove unused variable post-rebase
This commit updates the preset pass manager construction to only use
plugins for the translation stage. To accomplish this the previously
hard coded built-in translation methods, unroller, translator, and
synthesis are migrated to be exposed as built-in plugins. This
simplifies the preset pass manager construction as now the translation
stage is solely built via plugins.
Fixes#9456
* Move metapackage shim for combined releases
Now that Qiskit 0.44.0 has been released, the Qiskit project is now what
was formerly qiskit-terra. However, because Python packaging lacks a
clean mechanism to enable one package superseding another we're not able
to stop shipping a qiskit-terra package that owns the qiskit python
namespace without introducing a lot of potential user friction. So
moving forward the qiskit project will release 2 packages an inner
qiskit-terra which still contains all the library code and public facing
qiskit package which installs that inner package only. To enable this
workflow this commit migrates the metapackage setup.py into the terra
repository and setups build automation to publish a qiskit package in
addition to the inner terra package on each release tag.
* some follow up on https://github.com/Qiskit/qiskit-terra/pull/10530 (#19)
* some follow up on https://github.com/Qiskit/qiskit-terra/pull/10530
* extend some badges
* This Qiskit contains the building blocks for creating and working with quantum circuits, programs, and algorithms. -> This framework allows for building, transforming, and visualizing quantum circuits.
* The explanation of the Bell state is moving to IBM Quantun learning platform
* I think examples/ should eventually be replaced by proper tutorials
* Add StackOverflow as a forum
* Remove link to https://github.com/Qiskit/qiskit-tutorials
* Update README.md
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Update README.md
* Update README.md
* broken lines in badges
* doi
---------
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Expand lint checks to entire qiskit_pkg dir
* Unify extras in terra setup.py
* Update CI lint job
* Remove unused json imports
* Cleanup manifest file
* Finish comment
---------
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
* Add no-optionals and full-optionals test runs
This restructures the CI slightly to perform a complete "no-optionals"
run, and a complete "all optionals" run (for the optionals that are
readily accessible as Python packages, without complex additional
setup). Previously, we only did a partial test with some of the oldest
optional components, which could have allowed for behaviour to
accidentally require optional components without us noticing.
This splits the `requirements-dev.txt` file into two; lines that remain
in the file are what is actually _required_ to run the test suite,
run the style checks, and do the documentation build. The rest (and
everything that was missing) is added to a new
`requirements-optional.txt` file, which can be used to pull in (almost)
all of the packages that Terra can use to provide additional /
accelerated functionality.
Several tests needed to gain additional skips to account for this
change. There is a good chance that some tests are missing skips for
some libraries that are not the first point of failure, but it's hard to
test explicitly for these in one go.
* Fix typo in coverage workflow
* Try relaxing ipython constraints
* Squash newly exposed lint failures
* Fix typo in tutorials pipeline
* Update the 'also update' comments
* Remove unneeded qiskit-toqm dependency
* Section requirements-optional.txt
* Test all optionals on min not max Python version
Optionals are generally more likely to have been made available on the
older Pythons, and some may take excessively long to provide wheels for
the latest versions of Python.
* Add missing test skip
* Fix optional call
* Use correct boolean syntax
* Fix tests relying on Jupyter
* Install ipykernel in tutorials
* Remove HAS_PDFLATEX skip from quantum_info tests
For simple LaTeX tests, IPython/Jupyter can handle the compilation
internally using MathJax, and doesn't actually need a `pdflatex`
installation. We only need that when we're depending on LaTeX libraries
that are beyond what MathJax can handle natively.
* Include additional tutorials dependencies
* Install all of Terra's optionals in tutorial run
* Do not install all optionals in docs build
* Use class-level skips where appropriate
* Do not install ibmq-provider in tutorials run
* Include matplotlib in docs requirements
* Remove unnecessary whitespace
* Split long pip line
* Only install graphviz when installOptionals
Co-authored-by: Eric Arellano <14852634+Eric-Arellano@users.noreply.github.com>
* Install visualization extras in docs build
* Don't `--upgrade` when installing optionals
This is to prevent any optionals that have a dependency on Terra from
potentially upgrading it to some PyPI version during their installation.
This shouldn't happen with the current development model of having only
one supported stable branch and the main branch, but the `-U` is
unnecessary, and not having it is safer for the future.
* Update secondary installation of Terra in docs build
* Install all optionals during the docs build
I yoyo-ed on this, not wanting it to be too onerous to build the
documentation, but in the end, we need to have the optional features
installed in order to build most of the documentation of those features,
and some unrelated parts of the documentation may use these features to
enhance their own output.
* Fix test setup job
* Remove duplication of matplotlib in requirements files
* Update image test installation command
* Restore editable build
* Move `pip check` to `pip` section
* Remove redundant "post-install" description
* Expand comment on first-stage choices
* pytohn lol
---------
Co-authored-by: Eric Arellano <14852634+Eric-Arellano@users.noreply.github.com>
This opens development on `main` for the next minor version of *Qiskit*
(no longer to be called Terra). The version number is jumping to 0.45
to bring the version of `qiskit-terra` in line with `qiskit` (currently
the metapackge); starting from "Qiskit 0.45", this repository will be
called `qiskit` only, and the packages `qiskit` and `qiskit-terra` will
be unified, at least as far as Python packaging allows us to do.
For the 0.25.0 release we're going to push release candidates prior to
the release to enable testing before we cut the final release. In
preparation for tagging the first release candidate this commit updates
the version string to indicate it's a release candidate. This commit
should be what gets tagged as 0.25.0rc1.
* ruff executable check
* Ruff add rule name to pyproject.toml
* Unset the executable bit on setup.py
There is no shebang line anyway. If you try to run it, sh will try
to understand it.
---------
Co-authored-by: John Lapeyre <jlapeyre@users.noreply.github.com>
* Use stable Python C API for building Rust extension
This commit tweaks the rust extension code to start using the PyO3 abi3
flag to build binaries that are compatible with all python versions, not
just a single release. Previously, we were building against the version
specific C API and that resulted in needing abinary file for each
supported python version on each supported platform/architecture. By
using the abi3 feature flag and marking the wheels as being built with
the limited api we can reduce our packaging overhead to just having one
wheel file per supported platform/architecture.
The only real code change needed here was to update the memory
marginalization function. PyO3's abi3 feature is incompatible with
returning a big int object from rust (the C API they use for that
conversion isn't part of the stable C API). So this commit updates the
function to convert to create a python int manually using the PyO3 api
where that was being done before.
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
* Set minimum version on abi3 flag to Python 3.8
* Fix lint
* Use py_limited_api="auto" on RustExtension
According to the docs for the setuptools-rust RustExtension class:
https://setuptools-rust.readthedocs.io/en/latest/reference.html#setuptools_rust.RustExtension
The best setting to use for the py_limited_api argument is `"auto"` as
this will use the setting in the PyO3 module to determine the correct
value to set. This commit updates the setup.py to follow the
recommendation in the docs.
* Update handling of phase input to expval rust calls
The pauli_expval module in Rust that Statevector and DensityMatrix
leverage when computing defines the input type of the phase argument as
Complex64. Previously, the quantum info code in the Statevector and
DensityMatrix classes were passing in a 1 element ndarray for this
parameter. When using the the version specific Python C API in pyo3 it
would convert the single element array to a scalar value. However when
using abi3 this handling was not done (or was not done correctly) and
this caused the tests to fail. This commit updates the quantum info
module to pass the phase as a complex value instead of a 1 element numpy
array to bypass this behavior change in PyO3 when using abi3.
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
* Set py_limited_api explicitly to True
* DNM: Test cibuildwheel works with abi3
* Add abi3audit to cibuildwheel repair step
* Force setuptools to use abi3 tag
* Add wheel to sdist build
* Workaround abiaudit3 not moving wheels and windows not having a default repair command
* Add source of setup.py hack
* Add comment about pending pyo3 abi3 bigint support
* Revert "DNM: Test cibuildwheel works with abi3"
This reverts commit 8ca24cf1e4.
* Add release note
* Simplify setting abi3 tag in built wheels
* Update releasenotes/notes/use-abi3-4a935e0557d3833b.yaml
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Update release note
* Update releasenotes/notes/use-abi3-4a935e0557d3833b.yaml
---------
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Drop support for Python 3.7
This updates our documentation and build processes to set Python 3.8 as
the oldest support version of Python, consistent with our policy on
Python support.
This commit also removes remaining Python-version gating, since we were
principally using this to account for differences between 3.7 and 3.8.
Several backport packages are no longer required for any suppported
Python version, because of this change.
* Remove test for multiprocessing on Mac
With Python 3.7 support now dropped, there are no versions of Python on
macOS that we expect to support with multiprocessing again. There's no
guarantee that the hypothetical Python 10.11 (???) that this test was
previously checking would be any more valid than existing versions.
---------
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
Now that the first release candidate for the 0.24.0 release has been
tagged, the stable branch for the 0.24 series has been created and we
can start developing the 0.25.0 release on main. This commit bumps all
the version strings from 0.24.0 to 0.25.0 (and the backport branch for
mergify to 0.24) accordingly to differentiate the main branch from
0.24.*.
For the 0.24.0 release we're going to push release candidates
prior to the release to enable testing before we cut the final release.
In preparation for tagging the first release candidate this commit
updates the version string to indicate it's a release candidate. This
commit should be what gets tagged as 0.24.0rc1.
* Add Rust-based OpenQASM 2 converter
This is a vendored version of qiskit-qasm2
(https://pypi.org/project/qiskit-qasm2), with this initial commit being
equivalent (barring some naming / documentation / testing conversions to
match Qiskit's style) to version 0.5.3 of that package.
This adds a new translation layer from OpenQASM 2 to Qiskit, which is
around an order of magnitude faster than the existing version in Python,
while being more type safe (in terms of disallowing invalid OpenQASM 2
programs rather than attempting to construction `QuantumCircuit`s that
are not correct) and more extensible.
The core logic is a hand-written lexer and parser combination written in
Rust, which emits a bytecode stream across the PyO3 boundary to a small
Python interpreter loop. The main bulk of the parsing logic is a simple
LL(1) recursive-descent algorithm, which delegates to more specific
recursive Pratt-based algorithm for handling classical expressions.
Many of the design decisions made (including why the lexer is written by
hand) are because the project originally started life as a way for me to
learn about implementations of the different parts of a parser stack;
this is the principal reason there are very few external crates used.
There are a few inefficiencies in this implementation, for example:
- the string interner in the lexer allocates twice for each stored
string (but zero times for a lookup). It may be possible to
completely eliminate allocations when parsing a string (or a file if
it's read into memory as a whole), but realistically there's only a
fairly small number of different tokens seen in most OpenQASM 2
programs, so it shouldn't be too big a deal.
- the hand-off from Rust to Python transfers small objects frequently.
It might be more efficient to have a secondary buffered iterator in
Python space, transferring more bytecode instructions at a time and
letting Python resolve them. This form could also be made
asynchronous, since for the most part, the Rust components only need
to acquire the CPython GIL at the API boundary.
- there are too many points within the lexer that can return a failure
result that needs unwrapping at every site. Since there are no tokens
that can span multiple lines, it should be possible to refactor so
that almost all of the byte-getter and -peeker routines cannot return
error statuses, at the cost of the main lexer loop becoming
responsible for advancing the line buffer, and moving the non-ASCII
error handling into each token constructor.
I'll probably keep playing with some of those in the `qiskit-qasm2`
package itself when I have free time, but at some point I needed to draw
the line and vendor the package. It's still ~10x faster than the
existing one:
In [1]: import qiskit.qasm2
...: prog = """
...: OPENQASM 2.0;
...: include "qelib1.inc";
...: qreg q[2];
...: """
...: prog += "rz(pi * 2) q[0];\ncx q[0], q[1];\n"*100_000
...: %timeit qiskit.qasm2.loads(prog)
...: %timeit qiskit.QuantumCircuit.from_qasm_str(prog)
2.26 s ± 39.5 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
22.5 s ± 106 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)
`cx`-heavy programs like this one are actually the ones that the new
parser is (comparatively) slowest on, because the construction time of
`CXGate` is higher than most gates, and this dominates the execution
time for the Rust-based parser.
* Work around docs failure on Sphinx 5.3, Python 3.9
The version of Sphinx that we're constrained to use in the docs build
can't handle the `Unpack` operator, so as a temporary measure we can
just relax the type hint a little.
* Remove unused import
* Tweak documentation
* More specific PyO3 usage
* Use PathBuf directly for paths
* Format
* Freeze dataclass
* Use type-safe id types
This should have no impact on runtime or on memory usage, since each of
the new types has the same bit width and alignment as the `usize` values
they replace.
* Documentation tweaks
* Fix comments in lexer
* Fix lexing version number with separating comments
* Add test of pathological formatting
* Fixup release note
* Fix handling of u0 gate
* Credit reviewers
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
Co-authored-by: Kevin Hartman <kevin@hart.mn>
Co-authored-by: Eric Arellano <14852634+Eric-Arellano@users.noreply.github.com>
* Add test of invalid gate-body statements
* Refactor custom built-in gate definitions
The previous system was quite confusing, and required all accesses to
the global symbol table to know that the `Gate` symbol could be present
but overridable. This led to confusing logic, various bugs and
unnecessary constraints, such as it previously being (erroneously)
possible to provide re-definitions for any "built-in" gate.
Instead, we keep a separate store of instructions that may be redefined.
This allows the logic to be centralised to only to the place responsible
for performing those overrides, and remains accessible for error-message
builders to query in order to provide better diagnostics.
* Credit Sasha
Co-authored-by: Alexander Ivrii <alexi@il.ibm.com>
* Credit Matthew
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Remove dependency on `lazy_static`
For a hashset of only 6 elements that is only checked once, there's not
really any point to pull in an extra dependency or use a hash set at
all.
* Update PyO3 version
---------
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
Co-authored-by: Kevin Hartman <kevin@hart.mn>
Co-authored-by: Eric Arellano <14852634+Eric-Arellano@users.noreply.github.com>
Co-authored-by: Alexander Ivrii <alexi@il.ibm.com>
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Adding barebone HLS plugins for linear functions and cliffods
* adding clifford.lnn method and improving docstrings
* improving release notes with table and example
* docstring improvements following review
* Fix `tox -e docs` for macOS
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Fix two issues. Duh, thanks Jake
* Update Azure job to set RUST_DEBUG
Not strictly necessary since Tox already sets it. But makes things consistent and avoids accidentally
removing this in the future
---------
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Add structure for multiplie Rust extension crates
This is a precursor to adding an entirely separate and self-contained
crate to handle parsing of OpenQASM 2 (not 3 - this is kind of like a
trial run for that). It could conceivably still live within
`qiskit._accelerate`, but having separate crates helps us enforce more
sane API barriers, and has improvements in the incremental build time
when working on only one of the Rust extensions.
The intent after this commit is still to have `qiskit._accelerate` as an
easy catch-all for accelerators for Python. Developers should not need
to be concerned with defining a new crate for every new feature, and for
the most part `_accelerate` is still logically interoperative within
itself (for example, `NLayout` is used all over). This is just laying
out the groundwork so more complex crate additions _can_ also be made.
Some of the niceties to do with Cargo workspaces only became stabilised
in Rust 1.64. In particular, inheritance from the workspace root for
things like the package version, Rust version, and dependencies only
came in then. For now, the `workspace.packages` key in the root
`Cargo.toml` is ignored with a small warning during the build, but I've
put it in place mostly to keep track of what we can change once the MSRV
becomes 1.64 or greater (likely not til at least Terra 0.26).
* Add README.md to crates/accelerate
* Correct licence metadata
* Bump main branch version post 0.23.0rc1 tag
Now that the first release candidate for the 0.23.0 release has been
tagged, the stable branch for the 0.23 series has been created and we
can start developing the 0.24.0 release on main. This commit bumps all
the version strings from 0.23.0 to 0.24.0 (and the backport branch for
mergify to 0.23) accordingly to differentiate the main branch from
0.23.*.
* Bump version number in cargo lock file too
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
For the 0.23.0 release we're going to push release candidates
prior to the release to enable testing before we cut the final release.
In preparation for tagging the first release candidate this commit
updates the version string to indicate it's a release candidate. This
commit should be what gets tagged as 0.23.0rc1.
* Add importer for OpenQASM 3
This adds an interface for importing OpenQASM 3 into Qiskit Terra.
Right now, this is provided only as an optional dependency (the
`qasm3-import` optional), using the package `qiskit_qasm3_import`, which
in turn depends on the ANTLR runtime `antlr4-python3-runtime`.
The importer code may in the future be vendored into Terra, and will
likely be replaced completely by a more efficient importer in the
future, once a more concrete strategy has been decided for handling of
classical components in quantum programs.
* Expand capabilities section of release note
* Expand OQ3 import documentation
* Relax exact pin in optional requirement
* Remove superfluous word
* Update link to Qiskit org
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Adding permutation synthesis algorithm for LNN
* release notes
* Checking that the synthesized permutation adheres to the LNN connectivity and has depth guaranteed by the algorithm
* Adding tests for 15 qubits
Co-authored-by: Nir Gavrielov <nirgavrielov@gmail.com>
* Changing Permutation to be a Gate rather than QuantumCircuit
* Adding the property pattern to Permutation class
* fixing assert
* improving description message for _get_ordered_swap
* applying suggestions from code review
* minor
* attempt to fix docstring
* Another attempt to fix docsting
* another attempt to fix docstring
* temporarily simplifying docstring to see if this passes docs build
* adding blank line
* another attempt
* Restoring docstring
* removing extra line
* adding __array__ method for permutation + tests
* HLS permutation plugin based on the original synthesis algorithm for permutations
* speeding up _get_ordered_swap based on review comments
* Adding depth-2 synthesis algorithm for permutations for all-to-all connectivity; including tests and plugin
* release notes
* Adding example to release notes
* Update documentation of Permutation
* add missing import
* drawing decomposed circuit with permutations
* forgot parenthesis
* restoring qasm for circuits containing Permutations
* Adding permutation method to QuantumCircuit
* Adding test for quantum circuit with permutations
* pylint
* adding inverse() method to permutations
* qpy support for permutations
* tests for quantum circuits with permutations
* checking depth bound on the ACG method
* Adding tests for new Permutation functionality
* black
* Following review, keeping the old Permutation quantum circuit for backward compatibility, and naming the new permutation gate class as PermutationGate
* additional fixes
* updating release notes
* docs fix
* Removing permutation method from QuantumCircuit
* Adding QPY test for circuits with permutation gates
* Update qiskit/circuit/quantumcircuit.py
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Set default qasm name override to None
Co-authored-by: Nir Gavrielov <nirgavrielov@gmail.com>
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* add sk pass
* add utils
* add sk skeleton
* add example test file
* Add implementations initial version
* make the test run
* fix lint
* fix phase, add test
* fix the phase calculation
* add accuracy test
* cleanup of unused methods, append more efficient
* Add unittests for Solovay Kitaev Utils
* Add unitttests for commutator_decompose
* Remove unittests for non-public functions
* Fix pylint issues
* refactor simplify
* fix the test imports
* improve and add docstrings
* fix lint in testfile
* fix lint in utils
* fix lint in sk
* Add unittests
* fix phase, lint and allow basis gates as strings
* rm print
* fix adjoint and dot
* Fix unittests and add docstring to them
* move to circuit to GateSequence
* Fix assertion
* allow caching the basic approxs
* Fix bug in append from GateSequence
* Remove unnecesssary code
* Fix bug in test setup
* Add unittests for multiqubit QFT-circuit
* Add unittests for QFT with more qubits
* make compatible with current tests
* Remove unnecessary testcases
* Remove unnecessary unittests
* Fix bug in unittests
* Add release notes
* import scipy; scipy.optimize.fsolve() does not work
scipy.optimize has to be imported, since it is not imported when scipy is imported
* numpy
* Update qiskit/transpiler/passes/synthesis/solovay_kitaev_utils.py
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
* document ValueError
* rm unused methods, add GateSequence tests
* fix lint
* try fixing lint and docs
* cleanup
* improve readability of math
* move commutator_decompose into utils
* rm tests that are never run
* rm trailing todos and commented code
* add default dataset, cleanup tests
* rm unused imports
* replace diamond norm by trace distance
* rm unused imports
* fix deprecated use of DAGNode.type
* black
* really black!
* fail nicely
* test
* attempt to fix default approx path
* Move decompositions into py file dict
* speed up the basic approx generation
* add candidate checking by string
further reduces the runtime by a factor of about 1.8!
* use a tree for basic approx generation
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
* fix tests
* more speedups!
* use kdtree!
* refactor: generation as sep. function
* remove this masssssive file!
* start plugin work
* first attempt at synthesis plugin
* plugin itself works
-- but not when called via transpile or the UnitarySynthesis
* unitary synth works!
* use class-level method for SolovayKitaev
* docstrings
* add tests which cover still missing feature
* rename to SKSynth and better reno
* re-organize files
- algorithm to qiskit.synthesis
- plugin to qiskit.transpiler.passes.synthesis
* cover sklearn-free case
* missing future import
* move sk pass to transpiler subdirectory
* reno & try fixing slow optional module-level imports
* attempt 2 to fix sklearn import
* comments from Matthew's review
* directly construct dag
* (previous commit incomplete: missed this here)
* add SK to plugin toctree
* try fixing sphinx
* Apply Jake's comments
Removed the SK plugin from the plugin.py docs for now
* Fix doc example
* Update docs
* Fix tests
* Fix docs error
Co-authored-by: Julien Gacon <gaconju@gmail.com>
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
* Add support for Python 3.11
Python 3.11.0 was released on 10-24-2022, this commit marks the start of
support for Python 3.11 in qiskit. It adds the supported Python version in
the package metadata and updates the CI configuration to run test jobs
on Python 3.11 and build Python 3.11 wheels on release.
* Fix inspect.Parameter usage for API change in 3.11
Per the Python 3.11.0 release notes inspect.Parameter now raises a
ValueError if the name argument is a Python identifier. This was causing a
test failure in one case where a parameter named `lambda` was used.
This commit adjusts the parameter name in the tests to be lam to avoid
this issue.
* Set a version cap on the jax dev requirement
Currently jax doesn't publish Python 3.11 wheels which is blocking test
runs with python 3.11. Since jax is an optional package only used for
the gradient package we can just skip it as isn't a full blocker for
using python 3.11. This commit sets an environment marker on the jax
dev requirements to only try to install it on Python < 3.11.
* Set python version cap on cplex in CI
* DNM: Test wheel builds work
* Skip tests on i686/win32 wheel buids with python 3.11
* Revert "DNM: Test wheel builds work"
This reverts commit 725c21b465.
* Run QPY backwards compat tests on trailing edge Python version
This commit moves the qpy backwards compatibility testing from the
leading edge python version, which in this PR branch is Python 3.11, to
the trailing edge Python version which is currently 3.7. Trying to add
support for a new Python version has demonstrated that we can't use the
leading edge version as historical versions of Qiskit used to generate
old QPY payloads are not going to be generally installable with newer
Python versions. So by using the trailing edge version instead we can
install all the older versions of Qiskit as there is Python
compatibility for those Qiskit versions. Eventually we will need to
raise the minimum Qiskit version we use in the QPY tests, when Python
3.9 goes EoL in October 2025 and Qiskit Terra 0.18.0 no longer has any
supported versions of Python it was released for. We probably could
get by another year until Python 3.10 goes EoL in 2026 it just means
we're building 0.18.x and 0.19.x from source for the testing, but when
Python 3.11 becomes our oldest supported version we'll likely have to
bump the minimum version.
This does go a bit counter to the intent of the test matrix to make the
first stage return fast and do a more through check in the second stage.
But, in this case the extra runtime is worth the longer term stability
in the tests.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Bump main branch version post rc tag
Now that the first release candidate for the 0.22.0 release has been
tagged, the stable branch for the 0.22 series has been created and we
can start developing the 0.23.0 release on main. This commit bumps all
the version strings from 0.22.0 to 0.23.0 (and the backport branch for
mergify to 0.22) accordingly to differentiate the main branch from
0.22.*.
* Move release notes to dedicated directory too
* Update Cargo too
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
For the 0.22.0 release we're going to push release candidates
prior to the release to enable testing before we cut the final release.
In preparation for tagging the first release candidate this commit
updates the version string to indicate it's a release candidate. This
commit should be what gets tagged as 0.22.0rc1.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Initial implementation of pluggable HighLevelSynthesis
* typo
* black
* fix merge error
* Reorganizing entry points for high-level-synthesis plugins based on the comments
* fixing some typos
* HLSConfig now accepts a list of methods for each operation; also allowing to specify these lists in config's __init__
* running black
* removing hls_config from __str__ in PassManagerConfig
* Adding abc interface for high-level-synthesis plugins
* improving docs and comments
* fix
* adding tests; removing print statements
* removing external test file
* remove redundant print
* Passing HLSConfig in all transpiler optimization levels
* Fix for is_parameterized as Operations don't have to support this
* Replacing node.op.copy by copy.deepcopy(node.op) to avoid Operations implementing copy method
* release notes
* adding a link to HLSConfig class
* Temporarily removing high-level-synthesis from optimization loop
* Adding example from release notes to HighLevelSynthesis documentation
* fixing random typo
* adding references to documentation:
* Further trying to improve documentation following the review comments
* removing usused import
* minor docs changes
* Docs consolidation following review comments
* added test
* fixing title underline in docs
* minor output tweaks
* Update setup.py
Co-authored-by: Kevin Krsulich <kevin@krsulich.net>
Co-authored-by: Kevin Krsulich <kevin@krsulich.net>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Add stage plugin interface for transpile
This commit adds a new plugin interface to qiskit for enabling external
packages to write plugins that will replace a stage in the transpilation
pipeline. For example, if an external package had a custom layout pass
that they wanted to integrate into transpile() they could export a
plugin using this new interface and then users would just need to run
transpile(.., layout_method=foo)
This adds long asked for extensibility to the transpiler so that to
cleanly integrate new transpiler passes we're no longer required to
merge the features into terra. This should hopefully make it easier for
downstream pass authors to integrate their passes into terra and make it
easier for the terra maintainers to evaluate new transpiler passes.
* Fix docs builds
* Fix doc warning
* Make routing methods all plugins
This commit converts all the built-in routing method options into
separate plugins and also adds a default plugin for the default behavior
at each optimization level. To support using plugins for routing method
adding the optimization_level to the passmanager config was necessary so
that the plugin has sufficient context on how to construct the routing
pass used for the routing stage. As depending on the optimization level
the settings for each pass differs. For example on stochastic swap the
number of stochastic trials increases for level 3 to try and find a
better solution at the cost of more runtime.
* Add plugin usage to level 3
* Add plugin support to level 0 preset pass manager
* Add default plugin to reserved routing methods list
* Add release notes
* Add tests
* Apply suggestions from code review
Co-authored-by: Alexander Ivrii <alexi@il.ibm.com>
* Apply suggestions from code review
Co-authored-by: Alexander Ivrii <alexi@il.ibm.com>
* Remove raise on non-builtin layout method argument
* Fix typo
* Deduplicate code in built-in routing plugins
This commit deduplicates the code in the built-in routing stage plugins.
First it removes the default plugin which was duplicated with the
stochastic and sabre plugins. There was no functional difference between
just setting the implicit default method name and using a per method
plugin and having a standalone default plugin. Secondly all the vf2 call
limit code is abstracted into a helper function which reduces code
duplication.
* Make vf2 call limit function private
* Deduplicate code in stochastic swap plugin
* Expand example plugin documentation to show more complete use case
* Update qiskit/transpiler/preset_passmanagers/level1.py
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
* Add missing TODO comment
* Unify vf2 call limit handling
* Remove default plugin from entry points
* Simplify level 3 optimization stage logic
* Update qiskit/transpiler/preset_passmanagers/plugin.py
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
* Prefer toqm plugin if one is available
The qiskit-toqm project will be one of the first users of this plugin
interface. Once this is released in qiskit, qiskit-toqm will likely
publish their own plugin soon after and if they do we want that plugin
to be used instead of the hardcoded stage in terra. This commit updates
the logic for toqm handling to only use the built-in toqm if a version
of qiskit-toqm is installed without a plugin present.
* Apply suggestions from code review
Co-authored-by: Kevin Hartman <kevin@hart.mn>
Co-authored-by: Toshinari Itoko <15028342+itoko@users.noreply.github.com>
* Fix lint
* Remove unnecessary elses in builtin plugins
* Apply suggestions from code review
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
* Make optimization level a plugin argument
* Add test coverage for all built-in routing plugins
* Reorder stage variables to execution order
Co-authored-by: Alexander Ivrii <alexi@il.ibm.com>
Co-authored-by: Luciano Bello <bel@zurich.ibm.com>
Co-authored-by: Kevin Hartman <kevin@hart.mn>
Co-authored-by: Toshinari Itoko <15028342+itoko@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Bump version strings post pre-release
Now that qiskit-terra 0.21.0rc0 is out the door we should bump the version
string on main to show the source version we're installing is newer
than what's been pre-released.
* Move 0.21 release notes into a separate directory
* Update cargo.toml too
* Set version number to 0.21.0rc1 for first release candidate
For the 0.21.0 release we're going to start pushing release candidates
prior to the release to enable testing before we cut the final release.
In preparation for tagging the first release candidate this commit
updates the version string to indicate it's a release candidate.
* Update cargo version to reflect 0.21.0
* Fix qpy handling of prerelease tags
* Fix pre-release handling in qpy compat tests
* Inline version parsing in qpy compat test
* Fix qpy version handling
* Integrate qiskit-toqm as an optional layout and routing pass.
* Run formatting.
* Add release note.
* S
* Exclude Linux aarch64 for qiskit-toqm in requirements-dev.txt.
This is currently necessary since Qiskit TOQM isn't published to PyPI
for Linux aarch64 due to a bug.
* qiskit-toqm performs layout changes always.
...when configured using one of its optimization strategies. This is because
non-optimal configurations (i.e. anything using GreedyMapper under the
hood) aren't compatible with the layout search parameter, and in fact
break when it is provided. These configurations happen to always perform
layout (without the explicit search) through a different means. To make
everything consistent, it should be assumed that no matter which optimization
strategy is used, layout changes will be made via routing.
qiskit-toqm users can still create a custom strategy if they wish to use
an optimal configuration that does not make layout changes, and then
use their own pass manager.
* Add basic testing for TOQM integration.
* Update to use qiskit-toqm 0.0.3 API.
* Add testing of 5 qubit device.
This exercises the optimal mapper code paths of qiskit-toqm
opt levels.
* Update for target awareness.
* Add extra to setup.py for toqm.
* Bump requirement-dev.txt TOQM version.
* Make python-constraint optional
Since #7213 we no longer have been using the CSPLayout pass by default
in the preset passmanagers or transpile(). This is because it has been
superseded by the VF2Layout pass which is now used everywhere. While we
will keep the CSPLayout pass around for the forseeable future there is
no need to install python-constraint by default anymore since it's only
user is the CSPLayout pass, which isn't going to be commonly used
anymore now that it's not used in the default compilation path anymore.
This commit removes the python-constraint library from the requirements
list and makes it an optional dependency.
Fixes#7726
* Add docstring for HAS_CONSTRAINT
* Add private module to avoid module level optional import
This commit splits the custom solver class definition out into a
separate private module that is not imported until runtime. This enables
us to avoid a module level import for python-constraint meaning we only
try to import if something is actually using CSPLayout.
* Fix rebase issue
* Fix lint
* Fix typo in release note
Co-authored-by: Steve Wood <40241007+woodsp-ibm@users.noreply.github.com>
* Alphabetize optionals list
Co-authored-by: Steve Wood <40241007+woodsp-ibm@users.noreply.github.com>
Now that qiskit-terra 0.20.0 is out the door we should bump the version
string on main to show the source version we're installing is newer
than what's been released.
* Add dummy extra for `CSPLayout`
This currently has no effect, but since we no longer use `CSPLayout` in
the preset pass managers, we are moving to demote its dependency
`python-constraint` from a requirement to an optional. This is the
preliminary step of the process, which allows packages to depend on
`qiskit-terra[csp-layout-pass]` before we actually make the switch.
* Add redundant requirement rather than TODO
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Replace pauli expectation value cython with rust implementation
This commit replaces the cython implementation of the pauli expectation
value functions with a multithreaded rust implementation. This was done
primarily for two reasons, the first and primary reason for this change
is because after #7658 this module was the only cython code left in the
qiskit-terra repository so unifying on a single compiled language will
reduce the maintanence burden in qiskit-terra. The second reason is
similar to the rationale in #7658 around why using rust over cython for
multi-threaded hybrid python module. The difference here though is
unlike in stochastic swap this module isn't as performance critical as
it's not nearly as widely used.
* Tune single threaded performance for rust sum
This commit tunes the sum for the single threaded path. Using the
iterator sum() method is very convienent but for the single threaded
path it doesn't create the most efficient output. This was causing a
regression in performance over the previous cython version. To address
that issue, this commit adds a new tuned function which does a chunked
sum which the compiler can handle better. It more closely models how
we'd do this with vectorized SIMD instructions. As a future step we can
look at using simdeez https://github.com/jackmott/simdeez
to further optimize this by doing runtime CPU feature detection and
leveraging SIMD instrinsics (we might want to look at using `fast_sum()`
in the multithreaded path if we do that too).
* Add release notes
* Fix lint
* Add docstring and signature to rust functions
* Define parallel threshold as a constant
* Add attribution comment to fast_sum()
* Rename eval_parallel_env -> getenv_use_multiple_threads
* Use inline literal type for size
Co-authored-by: Kevin Hartman <kevin@hart.mn>
* Add overflow check on num_qubits
The functions only work for at most for number of qubits < usize bits
anything larger would cause an overflow. While rust provides overflow
checking in debug mode it disables this for performance in release mode.
Sice we ship binaries in release mode this commit adds an overflow check
for the num_qubits argument to ensure that we don't overflow and produce
incorrect results.
* Remove unecessary setup_requires field from setup.py
The setup_requires field in the setup.py is deprecated and
has been superseded by the pyproject.toml to define build
system dependencies. Since we're already relying on the
pyproject.toml to install setuptools-rust for us having the
setup_requires line will do nothing but potentially cause
issues as it will use an older install mechanism that will potentially conflict with people's environments.
* Drop `.iter().take(LANES)`.
* Fix typo.
Co-authored-by: Kevin Hartman <kevin@hart.mn>
* Implement multithreaded stochastic swap in rust
This commit is a rewrite of the core swap trials functionality in the
StochasticSwap transpiler pass. Previously this core routine was written
using Cython (see #1789) which had great performance, but that
implementation was single threaded. The core of the stochastic swap
algorithm by it's nature is well suited to be executed in parallel, it
attempts a number of random trials and then picks the best result
from all the trials and uses that for that layer. These trials can
easily be run in parallel as there is no data dependency between the
trials (there are shared inputs but read-only). As the algorithm
generally scales exponentially the speed up from running the trials in
parallel can offset this and improve the scaling of the pass. Running
the pass in parallel was previously tried in #4781 using Python
multiprocessing but the overhead of launching an additional process and
serializing the input arrays for each trial was significantly larger
than the speed gains. To run the algorithm efficiently in parallel
multithreading is needed to leverage shared memory on shared inputs.
This commit rewrites the cython routine using rust. This was done for
two reasons. The first is that rust's safety guarantees make dealing
with and writing parallel code much easier and safer. It's also
multiplatform because the rust language supports native threading
primatives in language. The second is while writing parallel cython
code using open-mp there are limitations with it, mainly on windows. In
practice it was also difficult to write and maintain parallel cython
code as it has very strict requirements on python and c code
interactions. It was much faster and easier to port it to rust and the
performance for each iteration (outside of parallelism) is the same (in
some cases marginally faster) in rust. The implementation here reuses
the data structures that the previous cython implementation introduced
(mainly flattening all the terra objects into 1d or 2d numpy arrays for
efficient access from C).
The speedups from this PR can be significant, calling transpile() on a
400 qubit (with a depth of 10) QV model circuit targetting a 409 heavy
hex coupling map goes from ~200 seconds with the single threaded cython
to ~60 seconds with this PR locally on a 32 core system, When transpiling
a 1000 qubit (also with a depth of 10) QV model circuit targetting a 1081
qubit heavy hex coupling map goes from taking ~6500 seconds to ~720
seconds.
The tradeoff with this PR is for local qiskit-terra development a rust
compiler needs to be installed. This is made trivial using rustup
(https://rustup.rs/), but it is an additional burden and one that we
might not want to make. If so we can look at turning this PR into a
separate repository/package that qiskit-terra can depend on. The
tradeoff here is that we'll be adding friction to the api boundary
between the pass and the core swap trials interface. But, it does ease
the dependency on development for qiskit-terra.
* Sanitize packaging to support future modules
This commit fixes how we package the compiled rust module in
qiskit-terra. As a single rust project only gives us a single compiled
binary output we can't use the same scheme we did previously with cython
with a separate dynamic lib file for each module. This shifts us to
making the rust code build a `qiskit._accelerate` module and in that we
have submodules for everything we need from compiled code. For this PR
there is only one submodule, `stochastic_swap`, so for example the
parallel swap_trials routine can be imported from
`qiskit._accelerate.stochastic_swap.swap_trials`. In the future we can
have additional submodules for other pieces of compiled code in qiskit.
For example, the likely next candidate is the pauli expectation value
cython module, which we'll likely port to rust and also make parallel
(for sufficiently large number of qubits). In that case we'd add a new
submodule for that functionality.
* Adjust random normal distribution to use correct mean
This commit corrects the use of the normal distribution to have the mean
set to 1.0. Previously we were doing this out of band for each value by
adding 1 to the random value which wasn't necessary because we could
just generate it with a mean of 1.0.
* Remove unecessary extra scope from locked read
This commit removes an unecessary extra scope around the locked read for
where we store the best solution. The scope was previously there to
release the lock after we check if there is a solution or not. However
this wasn't actually needed as we can just do the check inline and the
lock will release after the condition block.
* Remove unecessary explicit type from opt_edges variable
* Fix indices typo in NLayout constructor
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Remove explicit lifetime annotation from swap_trials
Previously the swap_trials() function had an explicit lifetime
annotation `'p` which wasn't necessary because the compiler can
determine this on it's own. Normally when dealing with numpy views and a
Python object (i.e. a GIL handle) we need a lifetime annotation to tell
the rust compiler the numpy view and the python gil handle will have the
same lifetime. But since swap_trials doesn't take a gil handle and
operates purely in rust we don't need this lifetime and the rust
compiler can deal with the lifetime of the numpy views on their own.
* Use sum() instead of fold()
* Fix lint and add rust style and lint checks to CI
This commit fixes the python lint failures and also updates the ci
configuration for the lint job to also run rust's style and lint
enforcement.
* Fix returned layout mapping from NLayout
This commit fixes the output list from the `layout_mapping()`
method of `NLayout`. Previously, it incorrectly would return the
wrong indices it should be a list of virtual -> physical to
qubit pairs. This commit corrects this error
Co-authored-by: georgios-ts <45130028+georgios-ts@users.noreply.github.com>
* Tweak tox configuration to try and reliably build rust extension
* Make swap_trials parallelization configurable
This commit makes the parallelization of the swap_trials() configurable.
This is dones in two ways, first a new argument parallel_threshold is
added which takes an optional int which is the number of qubits to
switch between a parallel and serial version. The second is that it
takes into account the the state of the QISKIT_IN_PARALLEL environment
variable. This variable is set to TRUE by parallel_map() when we're
running in a multiprocessing context. In those cases also running
stochastic swap in parallel will likely just cause too much load as
we're potentially oversubscribing work to the number of available CPUs.
So, if QISKIT_IN_PARALLEL is set to True we run swap_trials serially.
* Revert "Make swap_trials parallelization configurable"
This reverts commit 57790c84b0. That
commit attempted to sovle some issues in test running, mainly around
multiple parallel dispatch causing exceess load. But in practice it was
broken and caused more issues than it fixed. We'll investigate and add
control for the parallelization in a future commit separately after all
the tests are passing so we have a good baseline.
* Add docs to swap_trials() and remove unecessary num_gates arg
* Fix race condition leading to non-deterministic behavior
Previously, in the case of circuits that had multiple best possible
depth == 1 solutions for a layer, there was a race condition in the fast
exit path between the threads which could lead to a non-deterministic
result even with a fixed seed. The output was always valid, but which
result was dependent on which parallel thread with an ideal solution
finished last and wrote to the locked best result last. This was causing
weird non-deterministic test failures for some tests because of #1794 as
the exact match result would change between runs. This could be a bigger
issue because user expectations are that with a fixed seed set on the
transpiler that the output circuit will be deterministically
reproducible.
To address this is issue this commit trades off some performance to
ensure we're always returning a deterministic result in this case. This
is accomplished by updating/checking if a depth==1 solution has been
found in another trial thread we only act (so either exit early or
update the already found depth == 1 solution) if that solution already
found has a trial number that is less than this thread's trial number.
This does limit the effectiveness of the fast exit, but in practice it
should hopefully not effect the speed too much.
As part of this commit some tests are updated because the new
deterministic behavior is slightly different from the previous results
from the cython serial implementation. I manually verified that the
new output circuits are still valid (it also looks like the quality
of the results in some of those cases improved, but this is strictly
anecdotal and shouldn't be taken as a general trend with this PR).
* Apply suggestions from code review
Co-authored-by: georgios-ts <45130028+georgios-ts@users.noreply.github.com>
* Fix compiler errors in previous commit
* Revert accidental commit of parallel reduction in compute_cost
This was only a for local testing to prove it was a bad idea and was
accidently included in the branch. We should not nest the parallel
execution like this.
* Eliminate short circuit for depth == 1 swap_trial() result
This commit eliminates the short circuit fast return in swap_trial()
when another trial thread has found an ideal solution. Trying to do this
in a parallel context is tricky to make deterministic because in cases
of >1 depth == 1 solutions there is an inherent race condition between
the threads for writing out their depth == 1 result to the shared
location. Different strategies were tried to make this reliably
deterministic but there wa still a race condition. Since this was just a
performance optimization to avoid doing unnecessary work this commit
removes this step. Weighing improved performance against repeatability
in the output of the compiler, the reproducible results are more
important. After we've adopted a multithreaded stochastic swap we can
investigate adding this back as a potential future optimization.
* Add missing docstrings
* Add section to contributing on installing form source
* Make rust python classes pickleable
* Add rust compiler install to linux wheel jobs
* Try more tox changes to fix docs builds
* Revert "Eliminate short circuit for depth == 1 swap_trial() result"
This reverts commit c510764a77. The
removal there was premature and we had a fix for the non-determinism in
place, ignoring a typo which was preventing it from working.
Co-Authored-By: Georgios Tsilimigkounakis <45130028+georgios-ts@users.noreply.github.com>
* Fix submodule declaration and module attribute on rust classes
* Fix rust lint
* Fix docs job definition
* Disable multiprocessing parallelism in unit tests
This commit disables the multiprocessing based parallelism when running
unittest jobs in CI. We historically have defaulted the use of
multiprocessing in environments only where the "fork" start method is
available because this has the best performance and has no caveats
around how it is used by users (you don't need an
`if __name__ == "__main__"` guard). However, the use of the "fork"
method isn't always 100% reliable (see
https://bugs.python.org/issue40379), which we saw on Python 3.9 #6188.
In unittest CI (and tox) by default we use stestr which spawns (not using
fork) parallel workers to run tests in parallel. With this PR this means
in unittest we're now running multiple test runner subprocesses, which
are executing parallel dispatched code using multiprocessing's fork
start method, which is executing multithreaded rust code. This three layers
of nesting is fairly reliably hanging as Python's fork doesn't seem to
be able to handle this many layers of nested parallelism. There are 2
ways I've been able to fix this, the first is to change the start method
used by `parallel_map()` to either "spawn" or "forkserver" either of
these does not suffer from random hanging. However, doing this in the
unittest context causes significant overhead and slows down test
executing significantly. The other is to just disable the
multiprocessing which fixes the hanging and doesn't impact runtime
performance signifcantly (and might actually help in CI so we're not
oversubscribing the limited resources.
As I have not been able to reproduce `parallel_map()` hanging in
a standalone context with multithreaded stochastic swap this commit opts
for just disabling multiprocessing in CI and documenting the known issue
in the release notes as this is the simpler solution. It's unlikely that
users will nest parallel processes as it typically hurts performance
(and parallel_map() actively guards against it), we only did it in
testing previously because the tests which relied on it were a small
portion of the test suite (roughly 65 tests) and typically did not have
a significant impact on the total throughput of the test suite.
* Fix typo in azure pipelines config
* Remove unecessary extension compilation for image tests
* Add test script to explicitly verify parallel dispatch
In an earlier commit we disabled the use of parallel dispatch in
parallel_map() to avoid a bug in cpython associated with their fork()
based subprocess launch. Doing this works around the bug which was
reliably triggered by running multiprocessing in parallel subprocesses.
It also has the side benefit of providing a ~2x speed up for test suite
execution in CI. However, this meant we lost our test coverage in CI for
running parallel_map() with actual multiprocessing based parallel
dispatch. To ensure we don't inadvertandtly regress this code path
moving forward this commit adds a dedicated test script which runs a
simple transpilation in parallel and verifies that everything works as
expected with the default parallelism settings.
* Avoid multi-threading when run in a multiprocessing context
This commit adds a switch on running between a single threaded and a
multithreaded variant of the swap_trials loop based on whether the
QISKIT_IN_PARALLEL flag is set. If QISKIT_IN_PARALLEL is set to TRUE
this means the `parallel_map()` function is running in the outer python
context and we're running in multiprocessing already. This means we do
not want to be running in multiple threads generally as that will lead
to potential resource exhaustion by spawn n processes each potentially
running with m threads where `n` is `min(num_phys_cpus, num_tasks)` and
`m` is num_logical_cpus (although only
`min(num_logical_cpus, num_trials)` will be active) which on the typical
system there aren't enough cores to leverage both multiprocessing and
multithreading. However, in case a user does have such an environment
they can set the `QISKIT_FORCE_THREADS` env variable to `TRUE` which
will use threading regardless of the status of `QISKIT_IN_PARALLEL`.
* Apply suggestions from code review
Co-authored-by: Jake Lishman <jake@binhbar.com>
* Minor fixes from review comments
This commits fixes some minor details found during code review. It
expands the section on building from source to explain how to build a
release optimized binary with editable mode, makes the QISKIT_PARALLEL
env variable usage consistent across all jobs, and adds a missing
shebang to the `install_rush.sh` script which is used to install rust in
the manylinux container environment.
* Simplify tox configuration
In earlier commits the tox configuration was changed to try and fix the
docs CI job by going to great effort to try and enforce that
setuptools-rust was installed in all situations, even before it was
actually needed. However, the problem with the docs ci job was unrelated
to the tox configuration and this reverts the configuration to something
that works with more versions of tox and setuptools-rust.
* Add missing pieces of cargo configuration
Co-authored-by: Jake Lishman <jake@binhbar.com>
Co-authored-by: georgios-ts <45130028+georgios-ts@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Add support for Python 3.10
Python 3.10.0 was released on 10-04-2021, this commit marks the support
of Python 3.10 in qiskit-terra. It adds the supported python version in
the package metadata and updates the CI configuration to run test jobs
on Python 3.10 and build Python 3.10 wheels.
* Fix typo
* Update default envlist in tox.ini to include 3.10
* Bump cibuildwheel to the latest version
This also takes the opportunity to deduplicate the cibuildwheel
configuration using the pyproject.toml support in newer versions of
cibuildwheel. The common options for all builds are put there and per
build overrides (which are only for cross compiling arm wheels) are left
as environment variables in the CI configuration.
* Add missing cibuildwheel config to pyproject.toml
* Ignore internal deprecation warning emitted by jupyter in ci
* Fix black
Co-authored-by: Jake Lishman <jake.lishman@ibm.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
* Drop support for Python 3.6
As has been advertised for some time the Qiskit 0.19.0 release is the
last release which supports Python 3.6 (which goes End of Life in
Decemeber). This commit drops support accordingly so that the minimum
supported Python version is 3.7.
* Add release note
Now that qiskit-terra 0.19.0 is out the door we should bump the version
string on master to show the source version we're installing is newer
than what's been released.
* Fix compatibility with Matplotlib 3.5
The problem was `qiskit.visualization.bloch.Arrow3D`, which subclassed
`matplotlib.patches.FancyArrowPatch` (a 2D patch), but not the 3D
machinery. It is now made an unholy multiple-inherited abomination of
both the 2D patch and `mpl_toolkits.mplot3d.art3d.Patch3D`; the latter
is a relatively thin wrapper (in terms of attributes) around the 2D
patch, so this is not too terrible.
Matplotlib 3.5 calls the `Patch3D.do_3d_projection` method using a
deprecated parameter, triggering two warnings, unless the artist's
module appears to have come from `mpl_toolkits.mplot3d.art3d`, even if
the new calling convention is respected. The warning is only triggered
when the figure is drawn, which may well be outside of our control, so
we cannot suppress the warnings. Instead, we just lie about the module
the arrow patch was defined in, to trick it into not warning, because we
use the new calling convention. This is supported at least as far back
as Matplotlib 3.3, which is the current minimum supported version. The
nasty hack should be removable once Matplotlib 3.6 is the minimum
version because the deprecation period will expire.
* Remove useless word
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
Co-authored-by: Matthew Treinish <mtreinish@kortar.org>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Jim Garrison