Linker ignoring partitions.csv (worked 1 day ago)

[Serverguy]

I have been loading up code and a huge LittleFS partition for a couple of weeks. I recently upgraded to 3.2.0 board definition and, now, the partitions.csv is being ignored. I have tried two formats, both of which worked yesterday, now both are ignored in v1 and v2.

Sketch uses 1317296 bytes (100%) of program storage space. Maximum is 1310720 bytes.
Global variables use 70612 bytes (21%) of dynamic memory, leaving 257068 bytes for local variables. Maximum is 327680 bytes.
Sketch too big; see https://support.arduino.cc/hc/en-us/art ... 0013825179 for tips on reducing it.
text section exceeds available space in board
Error compiling for board ESP32S3 Dev Module.

Yes, the partitions.csv is in the sketch folder like it always has been but, now, it is ignored. It is being copied into the temp area where the actual compile is done in Appdata (Windows) so it is being used so it is "there" but it is not effective any longer. This is broken on V1 and V2 of the IDE.

I uninstalled both then reinstalled. Rebooted. Shutdown. Nothing has changed the failure. This is set for ESP32-S3. I can't load to any board since it is breaking the partition size. It was going to a T-Display S3 but no longer.

Here are the two versions of the csv file that have worked, previously.

Code: Select all

# Name,   Type, SubType, Offset,  Size, Flags
nvs,      data, nvs,      0x9000,  0x5000,
otadata,  data, ota,      0xe000,  0x2000,
app0,     app,  ota_0,    0x10000,  3M,
spiffs,   data, spiffs,          , 10M,
coredump, data, coredump,        , 65K,

...and...

Code: Select all

# Name,   Type, SubType,  Offset,  Size,    Flags
nvs,      data, nvs,      0x9000,  0x5000,
otadata,  data, ota,      0xe000,  0x2000,
app0,     app,  ota_0,    0x10000, 0x300000,
spiffs,   data, spiffs,          , 0x9E0000,
coredump, data, coredump,        , 0x10000,

[lbernstone]

Did you not like this answer or did you just not read it?

[Serverguy]

Although this file is shown as a Comma Separated Values file, it seems to be actually strictly setup depending on columns. I find no documentation on the column requirements but when I moved one column of numbers one row to the right, it totally confused the linker. So you have to be very careful editing this file, it seems. After putting the numbers back into their apparently required columns, I have my large app partition back. In process of testing if the larger data partition is going to be right. The large data partition came back, too. I don't know where the flags would go, or what they are, since the header for it is shorter than some of the numbers written in hex. Someone needs to document the columns that are required so that this will work.

For now, it seems like 1, 11, 17, 26, 35, ?? (Flags). You can see the minor change I made in the columns and it crashed the setup.

Close this as done but needs additional documentation as to the actual required columns since this is not a csv file in actuality. Only down for 24 sleepless hours. It bothered me all night!

[lbernstone]

https://docs.espressif.com/projects/ard ... table.html
https://docs.espressif.com/projects/esp ... ables.html
https://github.com/serifpersia/esp32partitiontool