luke
/
yosys
mirror of https://github.com/YosysHQ/yosys.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
16,619 Commits 128 Branches 64 Tags 68 MiB
Commit Graph

2004 Commits

Author SHA1 Message Date
Robert O'Callahan f34c6fec19 Add unit-tests for `ParallelDispatchThread` and friends 2026-02-08 22:52:17 +00:00
Robert O'Callahan c8298aae02 Parallelize `Design::check()` 2026-02-05 21:22:29 +00:00
Robert O'Callahan 63aa31172c Parallelize `collect_garbage()` 2026-02-05 21:22:29 +00:00
Robert O'Callahan e551e30fa3 Add `FfInitVals::set_parallel()` method 
We'll use this later in the PR.
 2026-02-05 21:22:29 +00:00
Robert O'Callahan d5950a6c03 Add `MonotonicFlag` 
We'll use this later in the PR.
 2026-02-05 21:22:29 +00:00
Robert O'Callahan b18ca8710e Add `ConcurrentWorkQueue` 
We'll use this later in the PR.
 2026-02-05 21:22:29 +00:00
Robert O'Callahan d0276169d3 Add `ShardedHashSet` 
We'll use this later in the PR.
 2026-02-05 21:22:29 +00:00
Robert O'Callahan b06e903906 Add `ShardedVector` 
We'll use this later in the PR.
 2026-02-05 18:58:17 +00:00
Robert O'Callahan 0004708177 Add `ParallelDispatchThreadPool` 
We'll use this later in the PR.
 2026-02-05 18:58:17 +00:00
Robert O'Callahan fb24763a15 Add `work_pool_size`, `IntRange`, `item_range_for_worker`, and `ThreadIndex` 
We'll use these later in this PR.
 2026-02-05 18:58:17 +00:00
Robert O'Callahan 3b563b877e Make `log_error()` work in a `Multithreaded` context. 
`log_error()` causes an exit so we don't have to try too hard here. The main
thing is to ensure that we normally are able to exit without causing a stack
overflow due to recursive asserts about not being in a `Multithreaded` context.
 2026-02-05 18:58:17 +00:00
Robert O'Callahan 81fd03c6e4 Add `IdString::unescape()` method 
We've already talked about adding this as an alternative to `log_id()`, and we'll
need it later in this PR.
 2026-02-05 18:58:17 +00:00
Emil J 2aa0e1d009
Merge pull request #5629 from rocallahan/remove-zero-wires 
Avoid scanning entire module in `Module::remove()` if there are no wires to remove
 2026-02-04 17:44:24 +01:00
Emil J 59653da599
Merge pull request #5609 from nataliakokoromyti/upstream-design-run-pass 
Add Design::run_pass()
 2026-02-02 19:30:18 +01:00
nella 8f6c4d40e4
Merge pull request #5623 from YosysHQ/nella/opt-dff-rewrite 
opt_dff restructure.
 2026-01-28 14:41:40 +01:00
Krystine Sherwin aaebce7adc
log_help: Don't reformat codeblocks 2026-01-28 08:07:44 +13:00
nella 9367090763 OptDff more accurate ctrl/pattern desc. 2026-01-26 22:19:36 +01:00
nella 5803461c24 opt_dff pattern extraction. 2026-01-26 22:10:10 +01:00
Robert O'Callahan dcd7742d52 Avoid scanning entire module if there are no wires to remove 
It's pretty common for `opt_clean` to find no wires to remove. In that case,
there is no point scanning the entire design, which can be significantly
expensive for huge designs.
 2026-01-23 01:38:20 +00:00
Robert O'Callahan 2c0448a81b Avoid spurious copy in `IdStringCollector::trace_named()` 2026-01-21 03:31:56 +00:00
Emil J. Tywoniak befadf6d4d consteval: describe 2026-01-19 12:00:18 +01:00
Natalia cf511628b0 modify generator for pyosys/wrappers.cc instead of headers 2026-01-18 02:11:09 -08:00
Natalia fb864e91ee Add Design::run_pass() API for programmatic pass execution 
This commit adds a new run_pass() method to the RTLIL::Design class,
providing a convenient API for executing Yosys passes programmatically.

This is particularly useful for PyYosys users who want to run passes
on a design object without needing to manually construct Pass::call()
invocations. The method wraps Pass::call() with appropriate logging
to maintain consistency with command-line pass execution.

Example usage (from Python):
    design = ys.Design()
    # ... build or load design ...
    design.run_pass("hierarchy")
    design.run_pass("proc")
    design.run_pass("opt")

Changes:
- kernel/rtlil.h: Add run_pass() method declaration
- kernel/rtlil.cc: Implement run_pass() method
- tests/unit/kernel/test_design_run_pass.cc: Add unit tests
 2026-01-14 17:35:45 -08:00
nella 763001885f
Merge pull request #5608 from YosysHQ/nella/rtlil-to-string 
Add rtlil string getters
 2026-01-14 19:00:47 +01:00
nella 210b733555 Add rtlil string getters 2026-01-14 15:37:18 +01:00
Emil J. Tywoniak 8e2038c419 Use digit separators for large decimal integers 2026-01-13 16:38:12 +01:00
Miodrag Milanovic 0e6973037d Update year in banner and license 2026-01-13 14:23:51 +01:00
Robert O'Callahan 8da919587d Parallelize `opt_merge`. 
I'm not sure why but this is actually faster than existing `opt_merge` even with
YOSYS_MAX_THREADS=1, for the jpeg synthesis test. 16.0s before, 15.5s after for
end-to-end synthesis.
 2026-01-08 04:21:39 +00:00
Emil J 0ab967b036
Merge pull request #5564 from rocallahan/pass-fuzz 
Add support for fuzz-test comparison of two passes intended to give identical RTLIL results
 2026-01-06 20:07:31 +01:00
Robert O'Callahan a6d696ba2b Give `IdString` a default move constructor and make it a POD type. 
Now that we're not refcounting `IdString`, it can use the default move constructor.
This lets us make `IdString` a POD type so it can be passed in registers
in the standard C++ ABI.
 2025-12-30 22:35:14 +00:00
Robert O'Callahan 48cdb499f2 Remove `IdString::id_string()`. 
This was needed for the short time when `ID()` could return a value of `StaticIdString`.
That is no longer a problem.
 2025-12-22 01:57:30 +00:00
Robert O'Callahan 46cb05c471 Pass IdString by value instead of by const reference. 
When IdString refcounting was expensive, it made sense to pass it by const reference
instead of by value, to avoid refcount churn. Now that IdString is not refcounted,
it's slightly more efficient to pass it by value.
 2025-12-22 01:52:59 +00:00
Robert O'Callahan ddd6a16ee0 Add -legalize option to read_rtlil 2025-12-21 21:47:48 +00:00
Emil J f003eca615
Merge pull request #5526 from YosysHQ/emil/fix-cellaigs-function-arg-eval-order 
cellaigs: fix function argument evaluation order
 2025-12-12 10:00:09 +01:00
Emil J 5594b817cd
Merge pull request #5524 from rocallahan/single-thread-log 
Check that we don't use logging during multithreading
 2025-12-10 13:54:09 +01:00
Emil J e08e9119ee
Merge pull request #5516 from rocallahan/limit-threads 
Limit thread usage in tests
 2025-12-10 13:45:07 +01:00
Emil J. Tywoniak 99e873efc9 cellaigs: fix AOI and OAI ordering 2025-12-10 12:41:13 +01:00
Emil J. Tywoniak d932ce7f47 cellaigs: formatting 2025-12-10 12:33:17 +01:00
Emil J. Tywoniak 882001cb01 cellaigs: fix adder function argument evaluation order 2025-12-10 12:33:17 +01:00
Emil J. Tywoniak 7f3ea41103 cellaigs: fix function argument evaluation order 2025-12-10 11:36:37 +01:00
Emil J 1cceaa2a80
Merge pull request #5538 from YosysHQ/emil/driver-git-hash 
driver: add --git-hash
 2025-12-09 15:00:36 +01:00
Emil J. Tywoniak 6acb79afa2 driver: add --git-hash 2025-12-09 11:58:57 +01:00
Emil J. Tywoniak 23e1b0656c version: add git hash string 2025-12-09 11:58:43 +01:00
Robert O'Callahan d274ff8627 Delete prefix strings on shutdown to avoid triggering leak warnings. 
Fixes #5532
 2025-12-05 09:45:47 +00:00
Robert O'Callahan 638e904f91 Remove cover() coverage tracking 2025-12-04 16:27:13 +01:00
Robert O'Callahan 7219ac94b3 Add YOSYS_MAX_THREADS 2025-12-04 12:09:04 +01:00
Robert O'Callahan 7e75200b2a Check that we don't use logging during multithreading 2025-11-29 22:47:23 +00:00
Robert O'Callahan b23dc345ae Make it safe to access .c_str() for autoidx IDs in a multithreaded context 2025-11-25 22:49:41 +00:00
Robert O'Callahan 948001f39f Merge the two autoidx hashtables into one 
When something calls `IdString::c_str()` on an autoidx ID, we need to cache the
full string in a thread-safe way. If we need to allocate an entry in some data
structure to do that, it's difficult to do in a thread-safe no-performance-hazard way.
So instead, store the cached string pointer in the same hashtable as the prefix
pointer.
 2025-11-25 21:57:46 +00:00
Robert O'Callahan 8f0ecce53f Forbid creating IdStrings and incrementing autoidx during multithreaded phases, and add dynamic checks for that 
We could make it safe to increment autoidx during multithreaded passes, but that's
actually undesirable because it would lead to nondeterminism. If/when we need new
IDs during parallel passes, we'll have to figure out how to allocate them in a
deterministic way, and that will depend on the details of what the pass does.
So don't try to tackle that now.
 2025-11-25 21:57:46 +00:00