aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Merge pull request #1984 from YosysHQ/eddie/getParam_exceptionEddie Hung2020-04-227-21/+34
|\ | | | | kernel: Cell::getParam() to throw exception again if not found
| * ecp5: ecp5_gsr to skip cells that don't have GSR parameter againEddie Hung2020-04-221-1/+1
| |
| * tests: read +/xilinx/cell_sim.v before xilinx_dsp testEddie Hung2020-04-221-0/+1
| |
| * xilinx: xilinx_dsp_cascade to check CREG for DSP48E1 onlyEddie Hung2020-04-221-1/+1
| |
| * test: ice40_dsp test to read +/ice40/cells_sim.v for default paramsEddie Hung2020-04-221-0/+1
| |
| * xilinx: improve xilinx_dffopt messageEddie Hung2020-04-221-3/+6
| |
| * xilinx: xilinx_dffopt to read cells_sim.v; fix testEddie Hung2020-04-221-13/+22
| |
| * kernel: Cell::getParam() to throw exception again if not foundEddie Hung2020-04-221-3/+2
|/ | | | As it did before #1945
* Merge pull request #1949 from YosysHQ/eddie/select_blackboxEddie Hung2020-04-222-9/+54
|\ | | | | select: do not select inside black-/white- boxes unless '=' prefix used
| * Update passes/cmds/select.ccClaire Wolf2020-04-221-2/+2
| | | | | | Co-Authored-By: Eddie Hung <eddie@fpgeh.com>
| * tests: update select black/white-box testsEddie Hung2020-04-221-0/+7
| |
| * select: do not select black/white boxes by default, '=' prefix to do soEddie Hung2020-04-221-5/+5
| |
| * Add '=' selection pattern prefix for non-blackbox only patternsClaire Wolf2020-04-211-12/+26
| | | | | | | | Signed-off-by: Claire Wolf <claire@symbioticeda.com>
| * select: add test for not selecting inside black/white boxesEddie Hung2020-04-161-0/+21
| |
| * select: do not select inside blackboxesEddie Hung2020-04-161-0/+3
| |
* | Merge pull request #1983 from YosysHQ/eddie/use_default_paramEddie Hung2020-04-229-60/+54
|\ \ | | | | | | Cleanup use of hard-coded default parameters in light of #1945
| * | Cleanup use of hard-coded default parameters in light of #1945Eddie Hung2020-04-229-60/+54
| | |
* | | Merge pull request #1982 from AsuMagic/asu/cxxrtl-memory-queue-optwhitequark2020-04-221-3/+5
|\ \ \ | | | | | | | | cxxrtl: keep the memory write queue sorted on insertion.
| * | | cxxrtl: keep the memory write queue sorted on insertion.Asu2020-04-221-3/+5
| |/ / | | | | | | | | | | | | | | | | | | Strategically inserting the pending memory write in memory::update to keep the queue sorted allows us to skip the queue sort in memory::commit. The Minerva SRAM SoC runs ~7% faster as a result.
* | | Merge pull request #1969 from boqwxp/pool_emplaceEddie Hung2020-04-221-2/+32
|\ \ \ | |/ / |/| | kernel: Add `pool` support for rvalue references and C++11 move semantics.
| * | pool: add emplace() functionEddie Hung2020-04-221-0/+6
| | |
| * | kernel: Rename arguments to rvalue-reference-accepting functions.Alberto Gonzalez2020-04-211-8/+8
| | |
| * | Add rvalue-reference-accepting `entry_t` constructor for `pool`.Alberto Gonzalez2020-04-201-0/+1
| | |
| * | In `pool`, construct `entry_t`s in-place and add an ↵Alberto Gonzalez2020-04-201-2/+25
| | | | | | | | | | | | rvalue-accepting-and-forwarding `insert()` method.
* | | Merge pull request #1973 from YosysHQ/eddie/fix1966Eddie Hung2020-04-222-2/+4
|\ \ \ | | | | | | | | tests: fix various/plugin.sh when PREFIX != /usr/local/share
| * | | yosys-config: spellingEddie Hung2020-04-221-1/+1
| | | |
| * | | tests: use `yosys-config --datdir` instead of hard-codedEddie Hung2020-04-221-1/+3
| | | |
* | | | Merge pull request #1950 from YosysHQ/eddie/design_importEddie Hung2020-04-223-7/+30
|\ \ \ \ | | | | | | | | | | design: -import to not count black/white-boxes as candidates for top
| * | | | design: add testEddie Hung2020-04-162-5/+22
| | | | |
| * | | | design: -import to not count black/white-boxes as candidates for topEddie Hung2020-04-161-2/+8
| | |_|/ | |/| |
* | | | Merge pull request #1976 from YosysHQ/dave/fix-sim-constClaire Wolf2020-04-222-1/+18
|\ \ \ \ | | | | | | | | | | sim: Fix handling of constant-connected cell inputs at startup
| * | | | sim: Fix handling of constant-connected cell inputs at startupDavid Shah2020-04-212-1/+18
| | | | | | | | | | | | | | | | | | | | Signed-off-by: David Shah <dave@ds0.me>
* | | | | Merge pull request #1979 from whitequark/cxxrtl-go-fasterClaire Wolf2020-04-222-184/+396
|\ \ \ \ \ | | | | | | | | | | | | cxxrtl: Gas gas gas! I'm gonna step on the gas! Tonight I'll fly!
| * | | | | cxxrtl: run edge detectors only once in eval().whitequark2020-04-221-6/+22
| | | | | | | | | | | | | | | | | | | | | | | | As a result, Minerva SRAM SoC runs ~15% faster.
| * | | | | cxxrtl: add an unsupported knob for manipulating clock trees.whitequark2020-04-221-0/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is quite possibly the worst way to implement this, but it does work for a subset of well-behaved designs, and can be used to measure how much performance is lost simulating the inactive edge of a clock. It should be replaced with a clock tree analyzer generating safe code once it is clear how should such a thing look like.
| * | | | | cxxrtl: use log_id() where appropriate. NFC.whitequark2020-04-211-4/+4
| | | | | |
| * | | | | cxxrtl: add (*cxxrtl.{comb,sync}*) annotations on black box outputs.whitequark2020-04-211-65/+186
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the annotations are not used, this commit does not alter semantics at all, other than removing elision of outputs of black box cells. (Elision of such outputs is expected to be too rare to have any noticeable benefit, and the implementation was somewhat of a hack.) The (* cxxrtl.comb *) annotation alters the semantics of the output of the black box it is applied to such that, if the black box converges immediately, no additional delta cycle is necessary to propagate the computed combinatorial value upwards in hierarchy. The (* cxxrtl.sync *) annotation alters the semantics of the output of the black box it is applied to such as to remove any uses of the black box by the wires connected to this output, and break false feedback arcs arising from conservative modeling of dependencies of the black box. Although currently these attributes are only recognized on black boxes, if separate compilation is added in the future, it could also emit and consume them.
| * | | | | cxxrtl: s/sync_{wire,type}/edge_{wire,type}/. NFC.whitequark2020-04-211-23/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The attribute for this is called (* cxxrtl.edge *), and there is a planned attribute (* cxxrtl.sync *) that would cause blackbox cell outputs to be added to sync defs rather than comb defs. Rename the edge detector related stuff to avoid confusion.
| * | | | | cxxrtl: use one delta cycle for immediately converging netlists.whitequark2020-04-212-11/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If it is statically known that eval() will converge in one delta cycle (that is, the second commit() will always return `false`) because the design contains no feedback or buffered wires, then there is no need to run the second delta cycle at all. After this commit, the case where eval() always converges immediately is detected and the second delta cycle is omitted. As a result, Minerva SRAM SoC runs ~25% faster.
| * | | | | cxxrtl: add -O6, a shortcut for running `proc; flatten`.whitequark2020-04-211-4/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | People judge a compiler backend by the first impression, and the metric they judge it for is speed. -O6 does severely impact debuggability, but it provides equally massive gains in performance, so use it by default.
| * | | | | cxxrtl: unbuffer module input wires.whitequark2020-04-211-31/+61
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Module input wires are never set by the module, so it is unnecessary to buffer them. Although important for all inputs, this is especially critical for clocks, since after this commit, hierarchy levels no longer add delta cycles. As a result, Minerva SRAM SoC runs ~73% faster when flattened, and ~264% (!!) faster when hierarchical.
| * | | | | cxxrtl: simplify generated edge detection logic.whitequark2020-04-211-56/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit changes the way edge detectors are represented in generated code from a variable that is set in commit() and reset in eval() to a function that considers .curr and .next of the clock wire. Behavior remains the same. Besides being simpler to generate and providing more opportunities for optimization, this commit paves way for unbuffering module inputs.
| * | | | | cxxrtl: localize wires with multiple comb drivers, too.whitequark2020-04-211-32/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Before this commit, any wire that was not driven by an output port of exactly one comb cell would not be localized, even if there were no feedback arcs through that wire. This would cause the wire to become buffered and require (often quite a few) extraneous delta cycles during evaluation. To alleviate this problem, -O5 was running `splitnets -driver`. However, this solution was mistaken. Because `splitnets -driver` followed by `opt_clean -purge` would produce more nets with multiple drivers, it would have to be iterated to fixpoint. Moreover, even if this was done, it would not be sufficient because `opt_clean -purge` does not currently remove wires with the `\init` attribute (and it is not desirable to remove such wires, since they correspond to registers and may be useful for debugging). The proper solution is to consider the condition in which a wire may be localized. Specifically, if there are no feedback arcs through this wire, and no part of the wire is driven by an output of a sync cell, then the wire holds no state and is localizable. After this commit, the original condition for not localizing a wire is replaced by a check for any sync cell driving it. This makes it unnecessary to run `splitnets -driver` in the majority of cases to get a design with no buffered wires, and -O5 no longer includes that pass. As a result, Minerva SRAM SoC no longer has any buffered wires, and runs ~27% faster. In addition, this commit prepares the flow graph for introduction of sync outputs of black boxes. Co-authored-by: Jean-François Nguyen <jf@lambdaconcept.com>
| * | | | | cxxrtl: detect buffered comb wires, not just feedback wires.whitequark2020-04-211-5/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Any buffered combinatorial wires (including, as a subset, feedback wires) will prevent the design from always converging in one delta cycle. Before this commit, only feedback wires were detected. After this commit, any buffered combinatorial wires, including feedback wires, are detected. Co-authored-by: Jean-François Nguyen <jf@lambdaconcept.com>
* | | | | | bugpoint: Don't remove modules or cells while iterating over them.Marcelina Kościelnicka2020-04-221-4/+14
| | | | | | | | | | | | | | | | | | | | | | | | Reported by @ZirconiumX.
* | | | | | intel_alm: Documentation improvementsDan Ravensloft2020-04-213-14/+127
| | | | | |
* | | | | | write_json: dump default parameter valuesMarcelina Kościelnicka2020-04-211-0/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes #1823. This will allow nextpnr to reuse the default value information already present in yosys cells_sim.v and avoid duplicating (and probably desyncing) this information.
* | | | | | Use default parameter value in getParamMarcelina Kościelnicka2020-04-212-4/+13
| | | | | | | | | | | | | | | | | | | | | | | | Fixes #1822.
* | | | | | hierarchy: Convert positional parameters to named.Marcelina Kościelnicka2020-04-212-3/+50
| | | | | | | | | | | | | | | | | | | | | | | | Fixes #1821.
* | | | | | ilang, ast: Store parameter order and default value information.Marcelina Kościelnicka2020-04-216-9/+27
| | | | | | | | | | | | | | | | | | | | | | | | Fixes #1819, #1820.
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-09 20:24:55 +0000