Skip to content

[STM32H7] Provisioning fs reformat#957

Closed
andreagilardoni wants to merge 5 commits into
arduino:mainfrom
andreagilardoni:provisioning-fs-reformat
Closed

[STM32H7] Provisioning fs reformat#957
andreagilardoni wants to merge 5 commits into
arduino:mainfrom
andreagilardoni:provisioning-fs-reformat

Conversation

@andreagilardoni

Copy link
Copy Markdown
Contributor

Changing flash partitioning scheme in order to make room for a KV store

@andreagilardoni andreagilardoni changed the title Provisioning fs reformat [STM32H7] Provisioning fs reformat Sep 23, 2024

@pennam pennam left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There is no need to erase the whole flash before repartitioning; you only need to erase the MBR.

root.init();
root.erase(0x0, root.get_erase_size());

Furthermore if you keep the KVstorage in partition 3 for both schemes you should be able switch between them without erase the MBR at all

mbed::MBRBlockDevice::partition(&root, 1, 0x0B, 0, 1024 * 1024);
if(default_scheme) {
  mbed::MBRBlockDevice::partition(&root, 4, 0x0B, 14 * 1024 * 1024, 14 * 1024 * 1024);
  mbed::MBRBlockDevice::partition(&root, 3, 0x0B, 13 * 1024 * 1024, 14 * 1024 * 1024);
  mbed::MBRBlockDevice::partition(&root, 2, 0x0B, 1024 * 1024, 13 * 1024 * 1024);
  // use space from 15.5MB to 16 MB for another fw, memory mapped
} else {
  mbed::MBRBlockDevice::partition(&root, 2, 0x0B, 1024 * 1024, 6 * 1024 * 1024);
  mbed::MBRBlockDevice::partition(&root, 3, 0x0B, 6 * 1024 * 1024, 7 * 1024 * 1024);
  mbed::MBRBlockDevice::partition(&root, 4, 0x0B, 7 * 1024 * 1024, 14 * 1024 * 1024);
  // use space from 15.5MB to 16 MB for another fw, memory mapped
}

@andreagilardoni

Copy link
Copy Markdown
Contributor Author

Furthermore if you keep the KVstorage in partition 3 for both schemes you should be able switch between them without erase the MBR at all

I tried switching from this scheme to the older one and something bad happened and the only way to make it work properly was to erase the partitions. I am double checking this.

@andreagilardoni

Copy link
Copy Markdown
Contributor Author

I confirm that without the MBR erase I cannot use the old partitioning scheme.
@pennam do you think it is better to split this PR into 2? the first one with just the MBR erase command and the other with the new partitioning scheme, so that one can rollback with this fix present.

@andreagilardoni andreagilardoni force-pushed the provisioning-fs-reformat branch 2 times, most recently from 1d22ce2 to 6a309b7 Compare March 11, 2025 16:46
@andreagilardoni

andreagilardoni commented Mar 11, 2025

Copy link
Copy Markdown
Contributor Author

@pennam

There is no need to erase the whole flash before repartitioning; you only need to erase the MBR.

What do you think about the way I wrote the MBR table clean?

Furthermore if you keep the KVstorage in partition 3 for both schemes you should be able switch between them without erase the MBR at all

Now partition 3 is being used for user data, this would mean that we need either to switch user data partition to 4, or distinguish between the two cases when using kvstore. I prefer to keep it to partition number 4.

@andreagilardoni andreagilardoni force-pushed the provisioning-fs-reformat branch from 6a309b7 to 392cc64 Compare March 11, 2025 16:52
@megacct

megacct commented Mar 11, 2025

Copy link
Copy Markdown
Contributor

@andreagilardoni @pennam Hi, a couple of us have tried, unsuccessfully, to scrub the MBR in order to change the partition layout. Detail here https://forum.arduino.cc/t/how-to-partition-giga-flash-similar-to-original-factory-defaults/1349759. I'm sure we were doing something wrong but it may be worth double-double-checking that the erase in this commit is doing what's expected.

@andreagilardoni

Copy link
Copy Markdown
Contributor Author

I would like to point out that when repartitioning, for the user or ota block devices some files can be corrupted if they lie in the area that will be used for partition 4.

@andreagilardoni andreagilardoni force-pushed the provisioning-fs-reformat branch from 41d9a90 to 9e05f2b Compare March 12, 2025 12:00
@pennam

pennam commented May 6, 2025

Copy link
Copy Markdown
Contributor

superseded by #1058

@pennam pennam closed this May 6, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants