Large footprint arrays cause unhadled exception error

Bug #1711697 reported by Art
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
KiCad
Fix Released
Undecided
Unassigned

Bug Description

Trying to create a simple array. I manually place a single footprint on a brand new board and try to create an array 20x20 it starts executing and after awhile NewPcb crashes with an an handled exception error message.

Application: kicad
Version: (2017-07-18 revision f509ccc7c)-makepkg, release build
Libraries: wxWidgets 3.0.2
           libcurl/7.54.0 OpenSSL/1.0.2k zlib/1.2.11 libssh2/1.8.0 nghttp2/1.19.0 librtmp/2.3
Platform: Windows 8 (build 9200), 64-bit edition, 64 bit, Little endian, wxMSW
- Build Info -
wxWidgets: 3.0.2 (wchar_t,wx containers,compatible with 2.8)
Boost: 1.63.0
Curl: 7.54.0
KiCad - Compiler: GCC 6.3.0 with C++ ABI 1010
        Settings: USE_WX_GRAPHICS_CONTEXT=OFF
                  USE_WX_OVERLAY=OFF
                  KICAD_SCRIPTING=ON
                  KICAD_SCRIPTING_MODULES=ON
                  KICAD_SCRIPTING_WXPYTHON=ON
                  KICAD_SCRIPTING_ACTION_MENU=ON
                  BUILD_GITHUB_PLUGIN=ON
                  KICAD_USE_OCE=ON

Revision history for this message
Art (diametrix) wrote :

Interestingly enough doing the same by hand makes PcbNew hang up or crash with the same error message. If I place the same footprint on board manually and try to create the same array of 20x20 by duplicating the footprint on the grid it causes NewPCB to crash. Interestingly enough if you hit ignore on the error message and then save the board, it will never open again. It will cause the same error right off the bat when you try to open the board.

Revision history for this message
jean-pierre charras (jp-charras) wrote :

I do not have this issue.

Although creating the array is really very badly optimized (too long calculation time for a basic operation) it works.

What is the error message (a error message is not necessary a crash)?

Revision history for this message
Tomasz Wlostowski (twlostow) wrote :

Can you provide the footprint which causes the crash?

Tom

Changed in kicad:
status: New → Incomplete
Revision history for this message
Art (diametrix) wrote :

>>>What is the error message (a error message is not necessary a crash)?

See picture attached

Revision history for this message
Art (diametrix) wrote :

>>>Can you provide the footprint which causes the crash?

See footprint file attached. If I attempt to create a 20x20 array with this footprint, KiCad will crash.

Art (diametrix)
Changed in kicad:
status: Incomplete → New
Revision history for this message
jean-pierre charras (jp-charras) wrote :

Just for my info, how the footprint was designed?
It is really strange for me, and looks like very badly designed.

A 20x20 array creates (with this footprint) more than 45000 pads with 400 footprints!

Changed in kicad:
status: New → Incomplete
Revision history for this message
Art (diametrix) wrote :

It was designed per manufacturers specs around KiCad limitations. So if you are going to blame it to bad design, blame it to the design of the footprint editor. Be it as it may, regardless of the number of pads in a footprint, creating an array should not lead to the program crashing. Besides, what if you create an array of an actual board? A regular board would most likely have more than 112 pads per board. So I don't thing an array with 400 single footprints really deserves and exclamation point.

Here is another example - BGA-1295-1mm from KiCad library SMD_Packages causes PCBnew to exhibit similar behavior. So you can't really chuck it up to bad footprint design.

And do me a favor, can you please stop messing with the bug report status. I believe the bug description is complete and reproducible. If something in the description is not clear, feel free to ask for additional information, I will provide it promptly. If you are not feeling like fixing it just say so, may be somebody else will be up to the task?

Changed in kicad:
status: Incomplete → New
Revision history for this message
Art (diametrix) wrote :

I just tried to create and 5x5 array with the BGA-1295-1mm footprint which caused the program to crash with the same error message.

Revision history for this message
Art (diametrix) wrote :

One more example: QFN-64-1EP_9x9mm_Pitch0.5mm from Housings_QFN_DFN library. An array 18x18 causes NewPCB to crash

Revision history for this message
Seth Hillbrand (sethh) wrote :

High pad count is currently addressed in recent masters. My guess would be 63a952d23

Making large arrays of this footprint functions correctly at the moment. Art, if you still experience this issue, please re-open.

Changed in kicad:
status: New → Fix Committed
Changed in kicad:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.