- Notifications
You must be signed in to change notification settings - Fork13.3k
Postmortem - allow more control over crash dump#9271
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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.
Already on GitHub?Sign in to your account
base:master
Are you sure you want to change the base?
Uh oh!
There was an error while loading.Please reload this page.
Conversation
ref.esp8266#9193,esp8266#9194,esp8266#9196- persistent & public REASON_USER_... for user callback rst_info reason- public postmortem callback functions, don't overly hide the implementation- separate reasons for panic, assert, abort and c++ exceptions- put postmortem func one level down in the function chain, allow to override both the function itself or individual internals- stack offset for -fexceptions, skip unwind data to the actual point of origin
| auto *rtc_sys =reinterpret_cast<uint32_t *>(0x60001100ul); | ||
| memset(rst_info,0,sizeof(*rst_info)); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
slightly cheaper thansystem_rtc_... and no need to remember its logic
| #ifdef__cplusplus | ||
| #defineESP8266_POSTMORTEM_NOEXCEPT noexcept | ||
| #else | ||
| #defineESP8266_POSTMORTEM_NOEXCEPT | ||
| #endif |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
General idea here is to restrict crash opportunities in the functions. Even when exceptions are on.
| USF(1) = c; | ||
| } | ||
| voidcrash_system_prepare_rst_info(rst_info *rst_info,uint32_t user_reset_reason)noexcept { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others.Learn more.
Output is optional, so rst_info has to be modified earlier
ref.#9193
resolves#9194
resolves#9196
@DRSDavidSoft
this got somewhat forgotten... my draft patch was just replacing the restart, but I figured it would be better to also replace/refactor internals to make things more consistent
possibly a minor breakage for anyone tracking reason==254 specifically, but very easily noticed
override both the function itself or individual internals
also separates rst_info modifications and output, since the idea is to strip .o from the needless PSTR(...) strings when output is disabled (~1..2KiB)
as top comment states, tested by throwing a dummy exception with just what(). this would stop at the
~Foo()(hm) and the line throwing it