riscv
/
yosys
mirror of https://github.com/YosysHQ/yosys.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
12,769 Commits 92 Branches 49 Tags 39 MiB
Commit Graph

1686 Commits

Author SHA1 Message Date
Martin Povišer 554d8caef7 quicklogic: Add basic k6n10f tests 2023-12-04 15:52:03 +01:00
Martin Povišer 6672b6c1b3 quicklogic: Move pp3 tests one level down 2023-12-04 15:52:02 +01:00
N. Engelhardt 98769010af synth_quicklogic: rearrange files to prepare for adding more architectures 2023-12-04 15:52:02 +01:00
Catherine 62bbd086b1 cxxrtl: reorganize runtime component files. 
In preparation for substantial expansion of CXXRTL's runtime, this commit
reorganizes the files used by the implementation. Only minimal changes are
required in a consumer.

First, change:
  -I$(yosys-config --datdir)/include
to:
  -I$(yosys-config --datdir)/include/backends/cxxrtl/runtime

Second, change:
  #include <backends/cxxrtl/cxxrtl.h>
to:
  #include <cxxrtl/cxxrtl.h>
(and do the same for cxxrtl_vcd.h, etc.)
 2023-11-28 15:32:36 +00:00
Lofty 7ae4041e20 ice40, ecp5, gowin: enable ABC9 by default 2023-11-13 15:28:13 +00:00
N. Engelhardt 63cec22a0c
Merge pull request #3883 from phsauter/peepopt-shiftadd 
peepopt: Add `shiftadd` pattern
 2023-11-07 10:42:15 +01:00
phsauter c3b8de54da test: add tests for `shiftadd` and `shiftmul` 
This expands the part-select tests with one additional module.
It specifically tests the different variants of the `peepopt`
optimizations `shiftadd` and `shiftmul`.
Not all these cases are actually transformed using `shiftadd`,
including them also checks if the correct variants are rejected.
 2023-11-06 14:01:37 +01:00
Lofty b8b47f7c6c
Revert "ice40, ecp5: enable ABC9 by default" 2023-11-03 14:52:52 +00:00
Lofty 32082477b5 ice40, ecp5: enable ABC9 by default 2023-11-03 08:52:54 +00:00
N. Engelhardt 833b67af80 verific: import attributes on ports 
Co-authored-by: Miodrag Milanović <mmicko@gmail.com>
 2023-10-20 18:31:41 +02:00
N. Engelhardt 1b6d1e9419 memory_libmap: look for ram_style attributes on surrounding signals 2023-10-19 19:23:35 +02:00
Martin Povišer 62d6338688 quicklogic: Fix pp3 `dffs` test 
Fix name confusion which was making the test look into the vendor's cell
blackbox rather than into the synthesis results.
 2023-10-12 12:45:40 +02:00
Martin Povišer 4506e11d0f booth: Extend test to catch bug from previous commit 2023-10-04 23:30:29 +02:00
Jannis Harder c174597014 Fix sva_value_change_changed test for updated verific 2023-10-03 11:46:43 +02:00
Martin Povišer b0045300fd booth: Cut down the test 
Cut the test down from taking ~25 s to ~3 s.
 2023-09-28 11:55:51 +02:00
Martin Povišer c4762d930e
Merge pull request #3930 from povik/verific-test-memsemantics 
verific: Add test of accurate semantics in memory inference
 2023-09-20 11:46:42 +02:00
Martin Povišer 99a5773911
Merge pull request #3920 from zachjs/asgn-expr 
sv: support assignments within expressions
 2023-09-20 11:30:14 +02:00
Zachary Snow 28e99f2b8c fix width of post-increment/decrement expressions 2023-09-18 23:46:06 -04:00
Zachary Snow 7d07615dee allow attributes in front of ++/-- statements 2023-09-18 23:46:02 -04:00
Martin Povišer 8222121164 verific: Add test of accurate semantics in memory inference 2023-09-18 16:37:15 +02:00
andyfox-rushc 6d29dc659b renamed passname to booth, replaced connect_sigSpecToWire with connect, updated test script 2023-09-08 15:34:56 -07:00
Martin Povišer 25a33d4082 techmap: Make the Booth test deterministic 2023-09-07 14:56:56 +02:00
Martin Povišer 0c2a99ca47 techmap: Test the Booth multiplier 2023-09-07 14:46:59 +02:00
Zachary Snow 4edb1a1921 sv: support assignments within expressions 
- Add support for assignments within expressions, e.g., `x[y++] = z;` or
  `x = (y *= 2) - 1;`. The logic is handled entirely within the parser
  by injecting statements into the current procedural block.
- Add support for pre-increment/decrement statements, which are
  behaviorally equivalent to post-increment/decrement statements.
- Fix non-standard attribute position used for post-increment/decrement
  statements.
 2023-09-05 22:27:55 -04:00
Miodrag Milanovic a42c630264 put back previous test state, due to default change 2023-08-29 10:21:58 +02:00
Miodrag Milanovic 3b9ebfa672 Addressed code review comments 2023-08-25 11:10:20 +02:00
Miodrag Milanovic ea50d96135 fixed tests 2023-08-23 10:54:29 +02:00
Asherah Connor 4a475fa7a2 cxxrtl: include iostream when prints are used 2023-08-17 07:08:22 +02:00
Charlotte 860e3e4056 proc_clean: only consider fully-defined switch operands too. 2023-08-12 02:46:31 +02:00
Charlotte bf84861fc2 proc_clean: only consider fully-defined case operands. 2023-08-12 02:46:31 +02:00
Charlotte ce245b5105 cxxrtl_backend: respect sync `$print` priority 
We add a new flow graph node type, PRINT_SYNC, as they don't get handled
with regular CELL_EVALs.  We could probably move this grouping out of
the dump method.
 2023-08-11 04:46:52 +02:00
Charlotte 04582f2fb7 verilog_backend: emit sync `$print` cells with same triggers together 
Sort by PRIORITY, ensuring output order.
 2023-08-11 04:46:52 +02:00
Charlotte 4ffdee65e0 cxxrtl: store comb $print cell last EN/ARGS in module 
statics were obviously wrong -- may be multiple instantiations of any
given module.  Extend test to cover this.
 2023-08-11 04:46:52 +02:00
Charlotte 843ad9331b cxxrtl: WIP: adjust comb display cells to only fire on change 
Naming and use of statics to be possibly revised.
 2023-08-11 04:46:52 +02:00
Charlotte eb0fb4d662 tests: -std=c++11 not optional 2023-08-11 04:46:52 +02:00
Charlotte 992a728ec7 tests: CXX may be e.g. gcc, so use CC and link stdc++ explicitly 2023-08-11 04:46:52 +02:00
Charlotte f9b149fa7b cxxrtl: add "-print-output" option, test in fmt 2023-08-11 04:46:52 +02:00
Charlotte a1de898fcc fmt: merge fuzzers since we don't rely on BigInteger logic 
This is per fmt's (effective) use, as it turns out, so we're not losing
any fidelity in the comparison.
 2023-08-11 04:46:52 +02:00
Charlotte 3571bf2c2d fmt: fuzz, remove some unnecessary busywork 
Removing some signed checks and logic where we've already guaranteed the
values to be positive.  Indeed, in these cases, if a negative value got
through (per my realisation in the signed fuzz harness), it would cause
an infinite loop due to flooring division.
 2023-08-11 04:46:52 +02:00
Charlotte 2ae551c0af fmt: fuzz, fix (remove extraneous + incorrect fill) 
"blk + chunks" is often an overrun, plus the fill is unnecessary; we
throw blk away immediately.
 2023-08-11 04:46:52 +02:00
Charlotte 9f9561379b fmt: format %t consistently at initial 2023-08-11 04:46:52 +02:00
Charlotte 75b44f21d1 fmt: rudimentary %m support (= %l) 2023-08-11 04:46:52 +02:00
Charlotte c382d7d3ac fmt: %t/$time support 2023-08-11 04:46:52 +02:00
Charlotte b0f69f2cd5 tests: test cxxrtl against iverilog (and uncover bug!) 2023-08-11 04:46:52 +02:00
Charlotte 51d9b73107 fmt: tests completing again 
We need to invoke "read_verilog" manually, since the default action on
input files is to defer processing.  Under such conditions, we never
simplify the AST, and initial $prints never execute.
 2023-08-11 04:46:52 +02:00
Charlotte 1eff84cb92 fmt: ensure test exits on fail 
shebang not honoured when directly called with "bash run-test.sh".
 2023-08-11 04:46:52 +02:00
whitequark c285880684 fmt: add tests for Verilog round trip of format expressions. 2023-08-11 04:46:52 +02:00
whitequark 67052f62ec fmt: add tests for Yosys evaluation of format expressions. 2023-08-11 04:46:52 +02:00
whitequark 9f8e039a4b ast: use new format string helpers. 2023-08-11 04:46:52 +02:00
Martin Povišer f8325f66b7 opt_expr: Fix 'signed X>=0' replacement for wide output ports 
If the `$ge` cell we are replacing has wide output port, the upper bits
on the port should be driven to zero. That's not what a `$not` cell with
a single-bit input does. Instead opt for a `$logic_not` cell, which does
zero-pad its output.

Fixes #3867.
 2023-08-01 13:50:12 +01:00