New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
proteus F7 hard fault on "enable self stimulation" #5354
Comments
Same using "self stimulate" button in TS
|
i feel lucky so I make random changes
master build https://github.com/rusefi/rusefi/actions/runs/5370417084 https://github.com/rusefi/rusefi/suites/13847069639/artifacts/769103925 AFFECTED - fatal with self-stim
GCC12 NOT AFFECTED WOW @mck1117
GCC11 no optimization AFFECTED - fatal with self-stim
|
@dron0gus says |
addr2line 0x20029F00 -e build/rusefi.elf |
ctx 5 = lr |
Still an issue even with gcc 12 :(
|
why don't you attach a debugger instead of shooting in the dark |
that's exactly what I am currently failing to do! I am failing to compile a debug firmware see #5354 (comment) #5432 At the moment I desperately need a ELI5 on how to figure out what features consume most of flash so that I can reduce binary size at least in a special build. Another option is a true 2MB chip. |
Lua is huge. Output/config value lookup are both huge. |
I was suspecting that Output/config value lookup is part of the deal - let me try that. Still an ELI5 in general would be great :) |
#5434 created to make Output/config conditional compilation |
making configurations visible for HW tests
See #5462 |
|
Reproduced with gcc 12.2.rel1. Not reproduced with 11.3.rel1
|
Does not look like stack overflow...
|
what's the exact pc where it crashed? |
optimized out
|
@dron0gus you build locally right not master? and that's you EG33 F7 board not proteus right? I know 4chan4F was affected as long as it's -0s |
Master. Build localy. Only optimization level is changed. |
Not much details even locals are preserved from optimization
|
are you building with |
Yes. To reproduce: start TS, pres "Enable internal trigger simulation", close TS, open Console. Done, you are in HardFault. |
Minor related issue #5467 |
so what was the bug? |
hard fault due to misaligned big buffer on F7 firmware #5354
OK, I went and debugged it myself. Here's what was happening: The tooth logger uses an instance of If you force misalignment of that buffer, you can force this to happen on all ARM processors at any optimization level. |
on next boot
The text was updated successfully, but these errors were encountered: