update with last modifications - #168
Open
rchotin wants to merge 278 commits into
Open
Conversation
It seems that hTracksReservedMin is actually more important than hTracksReservedLocal, which is just a threshold value.
…ity is better handled
* Fixed Verilog netlist export with presence of unconnected plugs * Made export port be output by default when direction undefined. --------- Co-authored-by: Serge Rabyking <serge.rabyking@chipflow.io>
with square brackets Co-authored-by: Serge Rabyking <serge.rabyking@chipflow.io>
Signed-off-by: gatecat <gatecat@ds0.me>
The gold rule is that a fixed segment is inserted in it's track at the first session revalidation and *must* stay inserted. Allowing to remove it if they are of null length create cascading problems. * Change: In AutoSegment::canReduce(), never reduce allow to reduce fixed segments, except in the final driving state. * Change: In Session::revalidate(), no longer skip track insertion of fixed segments when they are reduceds. * Change: In Katana::printCompletion(), no longer skip fixed segments while checking for unrouteds. In case of bugs in the algorithm they may end up non routed.
* The "selection" button appears in multiple widgets and needs to be kept synchronous when one is changed. We use a "star" model where one widget changes it, then send a signal to all widgets that the setting has changed. We will loop over the one that has made the modification (so the setting is made twice) *but* we temporarily block him from emitting signals, hence we break the loop. It was too complex and contained bugs, now fixed for the integration of the short-circuit widget. * Add support to draw DRCError (used as well for short circuit, may have to change the name...). Still the problem of "text.reference" beigng activated as a sub-setting of "marker". * Change the way the full redraw is called. No longer check for a change in selection state but use a dedicated flag. * In select(Occurrence), when it's a Net, select all net's components, but not the Net object itself (for stability of selection). * Bug in, SelectionModel::getSelector(), check for *negative* row indexes... * Make a incremental clear in SelectionModel, to manage the activation/deactivation of the selection. * New Netset widget to manage a defined set<Net*> instead of all the nets od a Cell.
Note: this does not solve the channel routing issues, but allows to
debug it using the standard procedure.
* Enhanced and more readable log/error messages in the recursives calls of Tramontana. * Add a new NoExtractConsistent flag on the Cell to signal a Cell which is not in a coherent state to be extracted. Typically a sub-model which is placed (uniquified) but contains no routing. Same goes for core cell inside a corona (only the corona makes sense). * In KatanaEngine::_postCreate(), flag cells that will not be consistent for the extractor.
New setting to activate this feature (deactivated by default):
conf.doLvx = True
…nerateds). * Move extractable layer list and cut/contact connexions info from SweepLine to Tramontana, as we need them both in SweepLine *and* Tile. * In ShortCiruit, add a failsafe if the two occurrences bounding boxes do not overlap. This must not occurs, but just in case, issues a warning and uses the union instead of the intersection, so at least we could see both faulty components. * In Tile::create(), check that we do not create tiles with empty bounding box. If so, throw an exception. When forcing the layer, like when creating metal parts from a lone cut, use the overall bounding box, not the one in the metal (wich will be *empty* !). * In Equipotential::add(), when counting the number of components for a net to be fully included in the Equi, discard those in a layer which is *not* extractable. Was generating false positive on open nets.
Update the documentation in rst mode
…tion on library mangement
|
A preview of is uploaded and can be seen here: ✨ https://lip6.github.io/coriolis/pull/168/ ✨ Changes may take a few minutes to propagate. Since this is a preview of production, content with |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.