Two issues found with the locking system used to prevent access to
stale connectivity data during the zone fill process:
1) a std::mutex has undefined behavior if you try to use it to guard
against access from the same thread. Because of the use of wx event
loops (and coroutines) it is entirely possible, and in some situations
inevitable, that the same thread will try to redraw the ratsnest in the
middle of zone refilling.
2) The mutex was only guarding the ZONE_FILLER::Fill method, but the callers
of that method also do connectivity updates as part of the COMMIT::Push.
Redrawing the ratsnest after the Fill but before the Push will result in
stale connectivity pointers to zone filled areas.
Fixed (1) by switching to a trivial spinlock implementation. Spinlocks would
generally not be desirable if the contention for the connectivity data crossed
thread boundaries, but at the moment I believe it's guaranteed that the reads
and writes to connectivity that are guarded by this lock happen from the main
UI thread. The writes are also quite rare compared to reads, and reads are
generally fast, so I'm not really worried about the UI thread spinning for any
real amount of time.
Fixed (2) by moving the locking location up to the call sites of
ZONE_FILLER::Fill.
This issue was quite difficult to reproduce, but I found a fairly reliable way:
It only happens (for me) on Windows, MSYS2 build, with wxWidgets 3.0
It also only happens if I restrict PcbNew to use 2 CPU cores.
With those conditions, I can reproduce the issue described in #6471 by
repeatedly editing a zone properties and changing its net. The crash is
especially easy to trigger if you press some keys (such as 'e' for edit)
while the progress dialog is displayed. It's easiest to do this in a debug
build as the slower KiCad is running, the bigger the window is to trigger this
bug.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/6471
Fixes https://gitlab.com/kicad/code/kicad/-/issues/7048
Prevents line splitting behavior needed in v5 when drawing polygons on
courtyards. Also provide property dialog with some smarts to handle
special layers
Fixes https://gitlab.com/kicad/code/kicad/issues/6900
These objects can now be used in advanced DRC rules and
not just for keeping things out. Also remove the restriction
that at least one of the "basic" keepout rules must be set,
so that these areas can be used for more advanced rules.
1) better load-balancing for deferred zones
2) sort zones by priority before filling
3) retire BOARD::GetZoneList() which had a horrible performance profile
4) implement a zone bounding box cache
5) better checks for IsCancelled() so long fills can be exited
Fixes https://gitlab.com/kicad/code/kicad/issues/5738
Note: this is a basic implementation but it could be
improved once we include bus information in the netlist
and pcbnew can natively keep track of buses and nets
instead of just nets.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/4158
The constrained draw was constraining the polygon size to fit the
existing points rather than extending the existing points to fit the
user's intent.
This adds a 90° double-joint for the completion polygon that allows the
user to complete common constrained polygons visually based on the
preview with a double-click to finish.
1) Intermediate states might be self-intersecting, and we shouldn't
be policing our users on what order to do things in
2) The polygon might already be self-intersecting, at which point we're
preventing the user from fixing it.
Also includes better const management for SHAPE_POLY_SET API.
Fixes: lp:1833831
* https://bugs.launchpad.net/kicad/+bug/1833831
We were running into various corner conditions where a tool's event
loop would exit while the tool was still active, or the tool would
get popped while we were still in the event loop. (A lot of these
had to do with the POINT_EDITOR's, but not all of them.)
The new architecture:
1) tools always do a Push()/Pop()
2) everyone is responsible for their own pops; no more stack-clearing
on a cancel
3) CancelInteractive events go to all tools to facilitate (2)
When creating the zone cutout, we delete the old zone and add a new one
with each cutout. This requires resetting the source zone and clearing
our previous selection if we would like to continue using the cutout
tool on the new zone.
Fixes: lp:1812339
* https://bugs.launchpad.net/kicad/+bug/1812339
This ensures that when starting to draw a zone or graphic element, the
preview is drawn in the same color/on the same layer as the element that
will be created. Previously, beginning a graphic polygon while a copper
layer was active would result in the preview being drawn in the color of
the active copper layer.
Also includes a bunch of changes to simplifiy the terminology
and unify the copper, non-copper and keepout versions.
Also removes some legacy features:
Removes contol for fill method. If the fill method is currently
segments the dialog will ask if you want to convert to polygons
on OK.
Removes control of boundary resolution. We've done this with trig
since 5.0.
(cherry picked from commit 487aaeb)
Jeff Young
Dominik Wernberger
Jon Evans
Seth Hillbrand
Michael Kavanagh
jean-pierre charras
Ross Schlaikjer
Seth Hillbrand
Tomasz Wlostowski
Maciej Suminski
Wayne Stambaugh
Marvin Schmidt