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
Before processing the select tool actions, we check if the router is
currently active and if it is, pass the event for processing in other
tools
Fixes https://gitlab.com/kicad/code/kicad/issues/7021
Check for footprint editor
Promote pad edits to footprint edits iff
- Pad is locked
- Footprint is not
Otherwise, handle the pad normally and prompt for locked items if found
Many, many KIDIALOGs use OK/Cancel and then rename both buttons to
confirm or deny some action. In those cases we *do* want to store
the deny actions if they check "Do Not Show Again".
Fixes https://gitlab.com/kicad/code/kicad/issues/6979
Prompting to unlock is obtrusive and blocks the common action of
dragging from a pad to align footprints. The less common action of
editing pads in layout can be accomplished after unlocking the pads
Fixes https://gitlab.com/kicad/code/kicad/issues/6997
Our rendering pipeline is not set up for this kind of speed,
and this is a blocking operation at the moment.
Note that we will get an update at the end of the move
because OnModify does it.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/6819
During the first expansion, we are supposed to stop at vias and branches
(where three track ends connect). This needs to account for the number
of vias and tracks not just tracks
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
CHANGED: If you flip or rotate an item or selection of items, move
the mouse, press 'M' and start moving the mouse, the selection jumps
to where the mouse is. This MR fixes that so it doesn't happen.
This bug was reported against the 5.1.x branch but also affects the
5.99 branch. This MR is for the 5.99 branch.
Fixes https://gitlab.com/kicad/code/kicad/issues/4069
It's much more useful to use the locked items as targets.
Also fixes some issues with commits getting the wrong items when
pads are promoted to footprints.
Fixes https://gitlab.com/kicad/code/kicad/issues/6859
Locked pads are a bit different from other locked items
and there are sufficient reasons to need to select them
that it makes sense to introduce this inconsistency.
Fixes https://gitlab.com/kicad/code/kicad/-/issues/6837
CHANGED: Ctrl-middle mouse button double click should perform
'Zoom to Objects' in Pcbnew, but a change in modifier key handling
broke it. This MR restores the functionality.
Fixes https://gitlab.com/kicad/code/kicad/issues/6740
CHANGED: manually-placed (stitching) vias won't have their nets automatically updated
(unless the via is placed directly on a track segment)
CHANGED: stitching vias can be placed on footprint pads and pick up their nets
Fixes https://gitlab.com/kicad/code/kicad/-/issues/5484
Don't use the same "post_*" icon in the treeview. The badge gives an
action that doesn't apply in the treeview.
Also renames different file blobs to the "file_*" prefix.
Also removes EditToolSelectionFilter which was misused in more places
than it was used correctly. The original point of the client filter
was to move the logic to the point of use, which the
EditToolSelectionFilter sort of obviated anyway.
Fixes https://gitlab.com/kicad/code/kicad/issues/6751
In this case it's reactivated when the move command ends; only the
move command was ended because of activating the measure tool so it
gets activated twice.
Fixes https://gitlab.com/kicad/code/kicad/issues/6321
This unifies everything under a single architecture with a "don't
show again" dialog. Since everything now goes through the same
path it should be reasonably easy to make it do whatever we want
in the future.
Right now it presents 3 options: modify only unlocked items, override
locks and modify all items, or cancel command.
It doesn't pay to try and be too smart as it just confuses people.
When the bugs come in that it doesn't dim I'll just be marking them
as "as designed". :)
Fixes https://gitlab.com/kicad/code/kicad/issues/6367
For some reason, a footprint loaded from the fp viewer did not have its
view parameters fully updated (view bounding box not moved to the right place).
The fix ensure these parameters are updated.
If all the uncovered space is at the very edge it doesn't feel like
there's anywhere to click but the disambiguation menu doesn't come
up.
Fixes https://gitlab.com/kicad/code/kicad/issues/6584
1) make sure we get the default netclass when we want it
2) escape for HTML (particularly important for "<invalid>", but also
for reporting user rule names, netclass names, etc.)
Do not use the alt modifier key to force the display of the heuristic
context menu. This causes a conflict with menu selection on windows
and immediately dismisses the disambiguation context menu.
Fixes: https://gitlab.com/kicad/code/kicad/-/issues/1873
Also a rationalization of text polygon generators, with the "standard"
version inherited from BOARD_ITEM now giving the bounding box. This
requires callers who want the (much) more expensive stroke-based one
to call it explicitly (and brings PCB_TEXT in line with the was FP_TEXT
already was.
Fixes https://gitlab.com/kicad/code/kicad/issues/6525
The new shape architecture gives us the opportunity to make text
selection much more intuitive by actually looking at the glyph
shapes. Before text would be selected when you clicked in the
descenders area (which was usually blank given uppercase letters
and digits).
Fixes https://gitlab.com/kicad/code/kicad/issues/6525
Groups do not own their children; they only reference them. That
being said, they do own *selection* and *moving* of their children.
This might address some of 6493, but in any case will be a
pre-requisite for fixing it.
Fixes https://gitlab.com/kicad/code/kicad/issues/6493
- ref and value texts were converted to user texts.
- fpid was left empty, thus creating a potential issue if pasted in the board editor.
Now:
- ref and value texts are no changed, and not copied is pasted in the footprint editor
(these texts already exist)
- a dummy fpid (clipboard:<md5 string>) is added if pasted in board editor
(Pasting a (partial) footprint from fp editor to a board editor is certainly not
a good idea but is possible)
It shows much more detail. Removes some nag dialogs and places
hypertext links in others.
Also fixes the auto-layer-showing to correctly show Edge.Cuts or
F.CrtYd or B.CrtYd for errors relating to them.
Fixes https://gitlab.com/kicad/code/kicad/issues/6446
Jeff Young
Michael Kavanagh
Seth Hillbrand
Jon Evans
Roberto Fernandez Bautista
Mikolaj Wielgus
hoijui
PJM
Wayne Stambaugh
Marek Roszko
Ian McInerney
jean-pierre charras
Dominik Wernberger