[BUG] EEPROM error with M502 then M500

[ellensp]

Did you test the latest bugfix-2.1.x code?

Yes, and the problem still exists.

Bug Description

With 2 extruders

With debugging added to see what is going on

Send M502

echo:Hardcoded Default Settings Loaded
ok

No problems so far

Send: M500

echo:eeprom_error: 0 (my added debugging)
echo:eeprom_index: 607 (my added debugging
echo:offset: 603 (my added debugging
Error:Field motor_current_setting mismatch.
ok

So position in eeprom is out by 4

Bug Timeline

4 weeks

Expected behavior

This should not error

Actual behavior

It errors!

Steps to Reproduce

1. Compile firmware with provided Configs and upload
2. execute M502 them M500
3. pull your hair out

Version of Marlin Firmware

Bugfix-2.1.x

Electronics

mks tinybee

Don't forget to include

• A ZIP file containing your Configuration.h and Configuration_adv.h.

Additional information & file uploads

Configuration.zip

Debugging I added

diff --git a/Marlin/src/module/settings.cpp b/Marlin/src/module/settings.cpp
index f76b5afe74..c20761ac1a 100644
--- a/Marlin/src/module/settings.cpp
+++ b/Marlin/src/module/settings.cpp
@@ -1429,6 +1429,12 @@ void MarlinSettings::postprocess() {
     // Motor Current PWM
     //
     {
+      //EEPROM_ASSERT( eeprom_error || eeprom_index == offsetof(SettingsData, motor_current_setting) + EEPROM_OFFSET, "Field " STRINGIFY(motor_current_setting) " 
+
+      SERIAL_ECHO_MSG("eeprom_error: ",eeprom_error);
+      SERIAL_ECHO_MSG("eeprom_index: ",eeprom_index);
+      SERIAL_ECHO_MSG("offset: ",      offsetof(SettingsData, motor_current_setting) + EEPROM_OFFSET);
+
       _FIELD_TEST(motor_current_setting);
 
       #if HAS_MOTOR_CURRENT_SPI || HAS_MOTOR_CURRENT_PWM

Note this issue is not with motor_current_setting but with planner_extruder_advance_K just before motor_current_setting

we have

  //
  // LIN_ADVANCE
  //
  float planner_extruder_advance_K[DISTINCT_E]; // M900 K  planner.extruder_advance_K

where DISTINCT_E = 1

later we have

    //
    // Linear Advance
    //
      _FIELD_TEST(planner_extruder_advance_K);

      #if ENABLED(LIN_ADVANCE)
        EEPROM_WRITE(planner.extruder_advance_K);
      #else
        dummyf = 0;
        for (uint8_t q = _MAX(EXTRUDERS, 1); q--;) EEPROM_WRITE(dummyf);
      #endif
    }

where the else is run but _MAX(EXTRUDERS, 1) returns 2
These are in conflict
The structure allocates 1 float, but 2 floats are written to eeprom.

The first block of code was changed 4 weeks ago in https://github.com/MarlinFirmware/Marlin/pull/24821/files

where
float planner_extruder_advance_K[_MAX(EXTRUDERS, 1)]; // M900 K planner.extruder_advance_K
became
float planner_extruder_advance_K[DISTINCT_E]; // M900 K planner.extruder_advance_K

I suspect the fix is as simple as changing
for (uint8_t q = _MAX(EXTRUDERS, 1); q--;) EEPROM_WRITE(dummyf);
to
for (uint8_t q = DISTINCT_E; q--;) EEPROM_WRITE(dummyf);

This compiles and seems to work.. But I don't have time to do exhaustive testing right now.

[GMagician]

why not use COUNT(planner_extruder_advance_K) (sizeof(planner_extruder_advance_K)/sizeof(planner_extruder_advance_K[0])?

[valterfc]

I have a similar bug.

When active "EXTRUDERS=2" and "SINGLENOZZLE", the error "EEPROM DATASIZE ERROR" occurs, testing with M502, after M500.

If "EXTRUDERS=1" and "SINGLENOZZLE" off, the error does not occur.

Board: BIGTREE_OCTOPUS_PRO_V1_F429_USB
Marlin Version: Bugfix-2.1.x.

[thinkyhead]

I have a similar bug.

If the committed patch doesn't help, you can narrow down the cause by doing a build with #define DEBUG_EEPROM_READWRITE added to your config. This will add extra debugging to emit the name of the EEPROM field following the one that is causing the size problem.

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.