aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Fix typoEddie Hung2019-09-301-1/+1
|
* Update doc for equiv_optEddie Hung2019-09-301-2/+3
|
* Merge pull request #1406 from whitequark/connect_rpcwhitequark2019-09-3011-0/+1767
|\ | | | | rpc: new frontend
| * rpc: new frontend.whitequark2019-09-309-0/+744
| | | | | | | | | | | | | | | | | | | | | | | | | | A new pass, connect_rpc, allows any HDL frontend that can read/write JSON from/to stdin/stdout or an unix socket or a named pipe to participate in elaboration as a first class citizen, such that any other HDL supported by Yosys directly or indirectly can transparently instantiate modules handled by this frontend. Recognizing that many HDL frontends emit Verilog, it allows the RPC frontend to direct Yosys to process the result of instantiation via any built-in Yosys frontend. The resulting RTLIL is then hygienically integrated into the overall design.
| * libs: import json11.whitequark2019-09-303-0/+1023
| | | | | | | | | | This commit imports the code from upstream commit dropbox/json11@8ccf1f0c5ecab6151a65f216e7eeccd8588e5457.
* | Merge pull request #1397 from btut/fix/python_wrappers_inline_constructorsEddie Hung2019-09-301-0/+2
|\ \ | | | | | | Generate Python wrappers for inline constructors
| * | Generate Python wrappers for inline constructorsBenedikt Tutzer2019-09-231-0/+2
| | | | | | | | | | | | Fixes: #1353
* | | Merge pull request #1416 from YosysHQ/mmicko/frontend_binary_inMiodrag Milanović2019-09-304-6/+10
|\ \ \ | | | | | | | | Open aig frontend as binary file
| * | | Fix reading aig files on windowsMiodrag Milanovic2019-09-291-1/+5
| | | |
| * | | Open aig frontend as binary fileMiodrag Milanovic2019-09-294-5/+5
| | | |
* | | | Bump versionClifford Wolf2019-09-301-1/+1
| | | | | | | | | | | | | | | | Signed-off-by: Clifford Wolf <clifford@clifford.at>
* | | | Merge pull request #1412 from YosysHQ/eddie/equiv_opt_async2syncClifford Wolf2019-09-301-0/+2
|\ \ \ \ | | | | | | | | | | equiv_opt to call async2sync when not -multiclock like SymbiYosys
| * | | | equiv_opt to call async2sync when not -multiclock like SymbiYosysEddie Hung2019-09-271-0/+2
| | | | |
* | | | | Merge pull request #1417 from YosysHQ/clifford/fixasync2syncClifford Wolf2019-09-301-0/+1
|\ \ \ \ \ | | | | | | | | | | | | Fix $dlatch handling in async2sync
| * | | | | Fix $dlatch handling in async2syncClifford Wolf2019-09-301-0/+1
|/ / / / / | | | | | | | | | | | | | | | Signed-off-by: Clifford Wolf <clifford@clifford.at>
* | | | | Add latch test modified from #1363Eddie Hung2019-09-302-0/+73
| | | | |
* | | | | Add LDCE/LDPE sim library, remove from *cells_xtra.{v,py}Eddie Hung2019-09-306-122/+46
| | | | |
* | | | | synth_xilinx: Support latches, remove used-up FF init values.Marcin Kościelnicki2019-09-303-2/+77
| | | | | | | | | | | | | | | | | | | | Fixes #1387.
* | | | | Merge pull request #1414 from hzeller/improve-replace-with-empty-mapEddie Hung2019-09-291-0/+2
|\ \ \ \ \ | | | | | | | | | | | | Avoid work in replace() if rules empty.
| * | | | | Avoid work in replace() if rules empty.Henner Zeller2019-09-291-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This speeds up processing when number of bits are large but there is actually nothing to replace. Adresses part of #1382. Signed-off-by: Henner Zeller <h.zeller@acm.org>
* | | | | | Merge pull request #1359 from YosysHQ/xc7dspEddie Hung2019-09-2944-281/+6234
|\ \ \ \ \ \ | |_|_|/ / / |/| | | | | DSP inference for Xilinx (improved for ice40, initial support for ecp5)
| * | | | | Re-orderEddie Hung2019-09-272-2/+2
| | | | | |
| * | | | | Missing (* mul2dsp *) for sliceBEddie Hung2019-09-271-2/+2
| | | | | |
| * | | | | Ooops AREG and BREG to default to -1Eddie Hung2019-09-271-2/+2
| | | | | |
| * | | | | Update doc with max cascade chain of 20Eddie Hung2019-09-261-2/+4
| | | | | |
| * | | | | Do not always zero out C (e.g. during cascade breaks)Eddie Hung2019-09-262-7/+3
| | | | | |
| * | | | | Update docEddie Hung2019-09-261-1/+2
| | | | | |
| * | | | | Zero out portsEddie Hung2019-09-261-2/+2
| | | | | |
| * | | | | xilinx_dsp_cascade to also cascade AREG and BREGEddie Hung2019-09-262-454/+172
| | | | | |
| * | | | | Try recursive pmgen for P cascadeEddie Hung2019-09-261-88/+118
| | | | | |
| * | | | | Combine 'flatten' & 'coarse' labels in synth_ecp5 so proc run onceEddie Hung2019-09-261-9/+4
| | | | | |
| * | | | | TypoEddie Hung2019-09-261-1/+1
| | | | | |
| * | | | | CREG to check for \keepEddie Hung2019-09-261-0/+3
| | | | | |
| * | | | | Remove newlineEddie Hung2019-09-261-1/+0
| | | | | |
| * | | | | select onceEddie Hung2019-09-262-8/+12
| | | | | |
| * | | | | Stop trying to be too smart by prematurely optimisingEddie Hung2019-09-263-38/+14
| | | | | |
| * | | | | mul2dsp.v slice namesEddie Hung2019-09-251-5/+5
| | | | | |
| * | | | | Do not die if DSP48E1.P has no users (would otherwise get 'clean'-ed)Eddie Hung2019-09-251-1/+5
| | | | | |
| * | | | | Reject if (* init *) presentEddie Hung2019-09-252-0/+6
| | | | | |
| * | | | | Remove unnecessary check for A_SIGNED != B_SIGNED; be more explicitEddie Hung2019-09-251-3/+1
| | | | | |
| * | | | | Revert "Remove _TECHMAP_CELLTYPE_ check since all $mul"Eddie Hung2019-09-251-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 234738b103d4f2b3d937ed928fd89bc4e31627f1.
| * | | | | Revert "No need for $__mul anymore?"Eddie Hung2019-09-251-8/+8
| | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 1d875ac76a354f654f28b9632d83f6b43542e827.
| * | | | | Rework xilinx_dsp postAdd for new wreduce callEddie Hung2019-09-251-3/+3
| | | | | |
| * | | | | Only wreduce on t:$addEddie Hung2019-09-251-1/+1
| | | | | |
| * | | | | Remove _TECHMAP_CELLTYPE_ check since all $mulEddie Hung2019-09-251-6/+2
| | | | | |
| * | | | | Fix memory issue since SigSpec& could be invalidatedEddie Hung2019-09-251-6/+10
| | | | | |
| * | | | | No need for $__mul anymore?Eddie Hung2019-09-251-8/+8
| | | | | |
| * | | | | unextend only used in initEddie Hung2019-09-251-2/+1
| | | | | |
| * | | | | Call 'wreduce' after mul2dsp to avoid unextend()Eddie Hung2019-09-252-5/+5
| | | | | |
| * | | | | Oops. Actually use __NAME__ in ABC_DSP48E1 macroEddie Hung2019-09-251-1/+1
| | | | | |
generated by cgit v1.2.3 (git 2.25.1) at 2024-09-18 06:57:51 +0000