aboutsummaryrefslogtreecommitdiffstats
path: root/passes
Commit message (Collapse)AuthorAgeFilesLines
...
* | memory_unpack: Use Mem helpers.Marcelina Kościelnicka2020-10-211-106/+10
| |
* | memory_collect: Use Mem helpers.Marcelina Kościelnicka2020-10-211-223/+9
| |
* | memory_nordff: Use Mem helpers.Marcelina Kościelnicka2020-10-211-63/+9
|/
* opt_clean: Better memory handling.Marcelina Kościelnicka2020-10-081-8/+45
| | | | | | | | Previously, `$memwr` and `$meminit` cells were always preserved (along with the memory itself). With this change, they are instead part of the main cell mark-and-sweep pass: a memory (and its `$meminit` and `$memwr` cells) is only preserved iff any associated `$memrd` cell needs to be preserved.
* Validate parameters only when they are usedMiodrag Milanovic2020-09-251-5/+7
|
* use the new isPublic() in a few placesN. Engelhardt2020-09-149-17/+17
|
* Merge pull request #2356 from whitequark/flatten-techmap-no-tpl_driven-sigmapwhitequark2020-08-272-9/+6
|\ | | | | flatten, techmap: don't canonicalize tpl driven bits via sigmap
| * flatten, techmap: don't canonicalize tpl driven bits via sigmap.whitequark2020-08-262-9/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For connection `assign a = b;`, `sigmap(a)` returns `b`. This is exactly the opposite of the desired canonicalization for driven bits. Consider the following code: module foo(inout a, b); assign a = b; endmodule module bar(output c); foo f(c, 1'b0); endmodule Before this commit, the inout ports would be swapped after flattening (and cause a crash while attempting to drive a constant value). This issue was introduced in 9f772eb9. Fixes #2183.
* | Merge pull request #2358 from whitequark/rename-ilang-to-rtlilwhitequark2020-08-277-24/+24
|\ \ | | | | | | Replace "ILANG" with "RTLIL" everywhere
| * | Replace "ILANG" with "RTLIL" everywhere.whitequark2020-08-267-24/+24
| |/ | | | | | | | | | | | | | | | | | | The only difference between "RTLIL" and "ILANG" is that the latter is the text representation of the former, as opposed to the in-memory graph representation. This distinction serves no purpose but confuses people: it is not obvious that the ILANG backend writes RTLIL graphs. Passes `write_ilang` and `read_ilang` are provided as aliases to `write_rtlil` and `read_rtlil` for compatibility.
* / dfflegalize: Fix decision tree for adffe.Marcelina Kościelnicka2020-08-271-1/+5
|/ | | | | | | | When an adffe is being legalized, and is not natively supported, prioritize unmapping to adff over converting to dffsre if dffsre is not natively supported itself. Fixes #2361.
* proc: Add -nomux switchPeder Bergebakken Sundt2020-08-201-1/+10
| | | | running proc -nomux will ommit the proc_mux pass
* Merge pull request #2344 from YosysHQ/mwk/opt_share-fixesclairexen2020-08-201-223/+138
|\ | | | | opt_share: Refactor, fix some bugs.
| * opt_share: Refactor, fix some bugs.Marcelina Kościelnicka2020-08-171-223/+138
| | | | | | | | | | | | Fixes #2334. Fixes #2335. Fixes #2336.
* | Merge pull request #2337 from YosysHQ/mwk/clean-keep-wireclairexen2020-08-201-4/+9
|\ \ | | | | | | opt_clean: Fix module keep rules.
| * | opt_clean: Fix module keep rules.Marcelina Kościelnicka2020-08-091-4/+9
| |/ | | | | | | | | | | - wires with keep attribute now force a module to be kept - presence of $memwr and $meminit cells no longer forces a module to be kept
* | Merge pull request #2333 from YosysHQ/mwk/peepopt-shiftmul-signedclairexen2020-08-201-5/+1
|\ \ | | | | | | peeopt.shiftmul: Add a signedness check.
| * | peeopt.shiftmul: Add a signedness check.Marcelina Kościelnicka2020-08-051-5/+1
| | | | | | | | | | | | Fixes #2332.
* | | Merge pull request #2328 from YosysHQ/mwk/opt_dff-cleanupclairexen2020-08-208-1468/+0
|\ \ \ | | | | | | | | Remove passes redundant with opt_dff
| * | | Remove now-redundant dff2dffe pass.Marcelina Kościelnicka2020-08-072-415/+0
| | | |
| * | | Remove now-redundant dff2dffs pass.Marcelina Kościelnicka2020-08-072-166/+0
| | | |
| * | | peepopt: Remove now-redundant dffmux pattern.Marcelina Kościelnicka2020-08-073-175/+0
| | | |
| * | | Remove now-redundant opt_rmdff pass.Marcelina Kościelnicka2020-08-072-712/+0
| | |/ | |/|
* | | Merge pull request #2327 from YosysHQ/mwk/techmap-constmap-fixclairexen2020-08-201-1/+22
|\ \ \ | | | | | | | | techmap.CONSTMAP: Handle outputs before inputs.
| * | | techmap.CONSTMAP: Handle outputs before inputs.Marcelina Kościelnicka2020-08-051-1/+22
| | |/ | |/| | | | | | | Fixes #2321.
* | | Merge pull request #2326 from YosysHQ/mwk/peeopt-muldiv-signclairexen2020-08-201-1/+4
|\ \ \ | | | | | | | | peepopt.muldiv: Add a signedness check.
| * | | peepopt.muldiv: Add a signedness check.Marcelina Kościelnicka2020-08-041-1/+4
| |/ / | | | | | | | | | Fixes #2318.
* | | Merge pull request #2319 from YosysHQ/mwk/techmap-celltype-patternclairexen2020-08-201-4/+32
|\ \ \ | | | | | | | | techmap: Add support for [] wildcards in techmap_celltype.
| * | | techmap: Add support for [] wildcards in techmap_celltype.Marcelina Kościelnicka2020-08-021-4/+32
| |/ / | | | | | | | | | Fixes #1826.
* | / Ensure \A_SIGNED is never used with $shiftxXiretza2020-08-181-0/+4
| |/ |/| | | | | | | It has no effect on the output ($shiftx doesn't perform any sign extension whatsoever), so an attempt to use it should be caught early.
* | Replace opt_rmdff with opt_dff.Marcelina Kościelnicka2020-08-072-9/+17
|/
* Add dffunmap pass.Marcelina Kościelnicka2020-07-312-0/+108
| | | | | To be used with backends that cannot deal with fancy FF types (like blif or smt).
* opt_expr: Remove -clkinv option, make it the default.Marcelina Kościelnicka2020-07-312-14/+14
| | | | | Adds -noclkinv option just in case the old behavior was actually useful to someone.
* synth_ice40: Use opt_dff.Marcelina Kościelnicka2020-07-302-247/+86
| | | | | | | | | The main part is converting ice40_dsp to recognize the new FF types created in opt_dff instead of trying to recognize the mux patterns on its own. The fsm call has been moved upwards because the passes cannot deal with $dffe/$sdff*, and other optimizations don't help it much anyway.
* synth_xilinx: Use opt_dff.Marcelina Kościelnicka2020-07-305-861/+200
| | | | | | | | | The main part is converting xilinx_dsp to recognize the new FF types created in opt_dff instead of trying to recognize the patterns on its own. The fsm call has been moved upwards because the passes cannot deal with $dffe/$sdff*, and other optimizations don't help it much anyway.
* async2sync: Support all FF types.Marcelina Kościelnicka2020-07-302-145/+162
|
* Add opt_dff pass.Marcelina Kościelnicka2020-07-302-0/+876
|
* opt_expr: Fix handling of $_XNOR_ cells with A = B.Marcelina Kościelnicka2020-07-291-1/+1
| | | | Fixes #2311.
* equiv_induct: Fix up assumption for $equiv cells in -undef mode.Marcelina Kościelnicka2020-07-271-1/+3
| | | | | | | | | | | | | | | | Before this fix, equiv_induct only assumed that one of the following is true: - defined value of A is equal to defined value of B - A is undefined This lets through valuations where A is defined, B is undefined, and the defined (meaningless) value of B happens to match the defined value of A. Instead, tighten this up to OR of the following: - defined value of A is equal to defined value of B, and B is not undefined - A is undefined
* async2sync: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-53/+11
|
* memory_dff: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-12/+5
|
* proc_dlatch: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-28/+8
|
* pmux2shift: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-15/+4
|
* wreduce: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-37/+7
|
* techmap: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-41/+4
|
* shregmap: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-39/+10
|
* abc: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-25/+6
|
* dffinit: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-41/+7
|
* zinit: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-43/+11
|
* dfflegalize: Refactor to use FfInitVals.Marcelina Kościelnicka2020-07-241-80/+25
|
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-25 11:51:22 +0000