aboutsummaryrefslogtreecommitdiffstats
path: root/backends
Commit message (Collapse)AuthorAgeFilesLines
* smtbmc: Do not assume skipped assertions when loading a witness traceJannis Harder2022-10-201-3/+0
| | | | | | This is not valid when the prefix of a trace already violates assertions. This can happen when the trace generating solver doesn't look for a minimal length counterexample.
* Temporal induction counterexample loop detection (#3504)Emil J2022-10-191-1/+36
| | | I have added an optional flag to smtbmc that causes failed temporal induction counterexample traces to be checked for duplicate states and reported to the user, since loops in the counterexample mean that increasing the induction depth won't help prove a design's safety properties.
* smtbmc: Fix witness handling for k-induction failuresJannis Harder2022-10-181-1/+1
| | | | | The "uninitialized" value is a _list_ of chunks that are part of the initial state for the witness trace.
* smt2/smtbmc: Fix FF witness data for fine grained or multi chunk FFsJannis Harder2022-10-123-5/+22
| | | | | | | The witness metadata was missing fine grained FFs completely and for coarse grained FFs where the output connection has multiple chunks it lacked the offset of the chunk within the SMT expression. This fixes both, the later by adding an "smtoffset" field to the metadata.
* backends: protobuf: removed protobuf backendAki Van Ness2022-10-103-383/+0
|
* smtbmc: Avoid unnecessary string copies when parsing solver outputJannis Harder2022-09-021-12/+9
|
* Fitting help messages to 80 character widthKrystalDelusion2022-08-244-9/+10
| | | | | | | | | Uses the regex below to search (using vscode): ^\t\tlog\("(.{10,}(?<!\\n)|.{81,}\\n)"\); Finds any log messages double indented (which help messages are) and checks if *either* there are is no newline character at the end, *or* the number of characters before the newline is more than 80.
* write_aiger: Fix non-$_FF_ FFsJannis Harder2022-08-181-1/+1
| | | | This broke while switching sby's formal flows to always use $_FF_'s.
* smtbmc: Set step range for --yw and dont skip steps for --check-witnessJannis Harder2022-08-161-2/+14
|
* yosys-witness: Add stats commandJannis Harder2022-08-161-0/+18
|
* smtbmc: Add --check-witness modeJannis Harder2022-08-161-1/+22
| | | | | This verifies that the given constraints force an assertion failure. This is useful to debug witness trace conversion (and minimization).
* aiger: Add yosys-witness supportJannis Harder2022-08-162-2/+317
| | | | | Adds a new json based aiger map file and yosys-witness converters to us this to convert between native and AIGER witness files.
* smtbmc: Add native json based witness format + smt2 backend supportJannis Harder2022-08-166-113/+968
| | | | | | | | | | | | | | | | | | | | This adds a native json based witness trace format. By having a common format that includes everything we support, and providing a conversion utility (yosys-witness) we no longer need to implement every format for every tool that deals with witness traces, avoiding a quadratic opportunity to introduce subtle bugs. Included: * smt2: New yosys-smt2-witness info lines containing full hierarchical paths without lossy escaping. * yosys-smtbmc --dump-yw trace.yw: Dump results in the new format. * yosys-smtbmc --yw trace.yw: Read new format as constraints. * yosys-witness: New tool to convert witness formats. Currently this can only display traces in a human-readable-only format and do a passthrough read/write of the new format. * ywio.py: Small python lib for reading and writing the new format. Used by yosys-smtbmc and yosys-witness to avoid duplication.
* btor: Support $anyinit cellsJannis Harder2022-08-161-1/+1
|
* aiger: Support $anyinit cellsJannis Harder2022-08-161-0/+11
|
* smt2: Support $anyinit cellsJannis Harder2022-08-161-10/+11
|
* formalff: Set new replaced_by_gclk attribute on removed dff's clksJannis Harder2022-08-162-0/+21
| | | | | | This attribute can be used by formal backends to indicate which clocks were mapped to the global clock. Update the btor and smt2 backend which already handle clock inputs to understand this attribute.
* Switched to utf-8 in smtio.pyMiodrag Milanovic2022-08-091-2/+2
|
* properly encode string in rtlilMiodrag Milanovic2022-08-091-1/+1
|
* Merge pull request #3432 from YosysHQ/aki/jny_updatesMiodrag Milanović2022-08-031-10/+33
|\ | | | | jny: Added JNY schema and additional information to JNY output file
| * backend: jny: updated the `JnyWriter` to emite a new "invocation" entry as ↵Aki Van Ness2022-08-021-10/+33
| | | | | | | | well as a "$schema" entry to point to the location the schema will be at
* | smt2: Fix $shift/$shiftx with negative shift ammountsJannis Harder2022-08-021-4/+4
|/ | | | Fixes #3431, fixes #3344
* Add support for GHDL modfloor operatorMichael Nolan2022-07-052-1/+22
|
* smt2, btor: Revert calling memory_map -rom-onlyJannis Harder2022-06-292-2/+0
| | | | | | | | This approach had major issues with ROMs whose initialization was not fully defined. If required, memory_map -rom-only -keepdc should be called early in a formal flow instead. (This does require a careful choice of optimization passes though. Sby's scripts will be updated accordingly.)
* memory_map: -keepdc option for formalJannis Harder2022-06-272-2/+2
| | | | Use it when invoking memory_map -rom-only from write_{smt2,btor}.
* btor: add support for $pos cellKevin Läufer2022-06-201-8/+11
|
* smt2, btor: Use memory_map -rom-only to make ROMs usable for k-inductionJannis Harder2022-06-172-0/+2
| | | | | | This avoids provability regressions now that we infer more ROMs. This fixes #3378
* smtbmc: noincr: keep solver running for post check-sat unrollingJannis Harder2022-06-081-2/+7
|
* Merge pull request #3357 from jix/smtbmc-cvc5Jannis Harder2022-06-081-3/+13
|\ | | | | smtbmc: recognize cvc5 and fix unrolling for cvc4/cvc5
| * smtbmc: recognize cvc5 and fix unrolling for cvc4/cvc5Jannis Harder2022-06-031-3/+13
| |
* | smt2: emit smtlib2_comb_expr outputs after all inputsJannis Harder2022-06-071-5/+9
| |
* | Merge pull request #3319 from programmerjake/smtlib2-expr-supportJannis Harder2022-06-071-6/+57
|\ \ | | | | | | add smtlib2_comb_expr
| * | smt2: Add smtlib2_comb_expr attribute to allow user-selected smtlib2 expressionsJacob Lifshay2022-06-021-6/+57
| |/
* / smtbmc: Force nonincremental mode when yices is used with forallJannis Harder2022-06-031-1/+4
|/
* Use proper operatorMiodrag Milanovic2022-05-271-4/+4
|
* add $divfloor support to write_smt2Jacob Lifshay2022-05-241-0/+21
| | | | Fixes: #3330
* Add propagated clock signals into btor info fileClaire Xenia Wolf2022-05-041-0/+2
|
* smt2: Make write port array stores conditional on nonzero write maskJannis Harder2022-04-201-2/+4
|
* pass jny: flipped the defaults for the inclusion of various bits of metadataAki Van Ness2022-04-081-30/+30
|
* pass jny: ensured the cell collection is cleared between modulesAki Van Ness2022-04-081-0/+1
|
* pass jny: fixed missing quotes around the type value for the cell sortAki Van Ness2022-04-081-1/+1
|
* pass jny: fixed the backslash escape for stringsAki Van Ness2022-04-081-2/+1
|
* pass jny: removed the invalid json escapesAki Van Ness2022-04-081-12/+0
|
* pass jny: added some todo comments about things that need to be done before ↵Aki Van Ness2022-04-081-0/+5
| | | | a proper merge, but it should be enough for the PoC at the moment
* pass jny: changed the constructor initializers to use parens rather than ↵Aki Van Ness2022-04-081-2/+2
| | | | curly-braces to hopefully make GCC 4.8 happy
* pass jny: fixed the string escape method to be less jank and more properAki Van Ness2022-04-081-21/+58
|
* pass jny: fixed the signed output for param value outputAki Van Ness2022-04-081-1/+1
|
* pass jny: added connection outputAki Van Ness2022-04-081-4/+88
|
* pass jny: added filter options for including connections, attributes, and ↵Aki Van Ness2022-04-081-25/+125
| | | | properties
* pass jny: large chunk of refactoring to make the JSON output more pretty and ↵Aki Van Ness2022-04-081-75/+89
| | | | the internals less of a spaghetti nightmare
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-01 08:09:46 +0000