Skip to content
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

mtd/*: drop .write() if .write_page() is implemented #15380

Merged
merged 11 commits into from
Dec 13, 2023

Conversation

benpicco
Copy link
Contributor

@benpicco benpicco commented Nov 4, 2020

Contribution description

The old .write() function is only used as a fall-back if .write_page() is not implemented.
We can drop it.

Testing procedure

Everything will fall back to .write_page now, so no user visible change.
On the driver level, .write_page can be much simpler than .write as all the common logic is handled by MTD and we can do partial writes (only write to page boundary, MTD will split up the transaction).

  • native (tested by CI)
  • ESP8266/ESP32
  • AT25xxx
  • MTD Mapper (tested by CI)
  • MTD SDcard
AT25xxx

tests/mtd_raw

2023-03-09 23:26:41,366 # init MTD_0… OK (128 kiB)
2023-03-09 23:26:43,208 # > test 0
2023-03-09 23:26:43,208 # [START]
2023-03-09 23:26:43,286 # [SUCCESS]

2023-03-09 23:28:20,306 # > read 0 0 32
2023-03-09 23:28:20,312 # 00000000  31  32  33  34  35  36  FF  F7  6C  69  74  74  6C  65  66  73
2023-03-09 23:28:20,318 # 00000010  2F  E0  00  10  00  00  02  00  00  01  00  00  00  02  00  00

2023-03-09 23:28:52,800 # > read 0 0 32
2023-03-09 23:28:52,806 # 00000000  31  61  62  63  64  65  66  67  68  69  74  74  6C  65  66  73
2023-03-09 23:28:52,812 # 00000010  2F  E0  00  10  00  00  02  00  00  01  00  00  00  02  00  00
ESP32

tests/mtd_raw

2023-03-09 23:31:24,456 # > test 0
2023-03-09 23:31:24,457 # [START]
2023-03-09 23:31:24,603 # [SUCCESS]

2023-03-09 23:32:43,227 # > read 0 0 32
2023-03-09 23:32:43,234 # 00000000  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF
2023-03-09 23:32:43,240 # 00000010  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF

2023-03-09 23:32:48,918 # > write 0 1 abcdefgh
2023-03-09 23:32:50,963 # 00000000  FF  61  62  63  64  65  66  67  68  FF  FF  FF  FF  FF  FF  FF
2023-03-09 23:32:50,969 # 00000010  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF

2023-03-09 23:32:50,957 # > read 0 0 32
2023-03-09 23:32:50,963 # 00000000  FF  61  62  63  64  65  66  67  68  FF  FF  FF  FF  FF  FF  FF
2023-03-09 23:32:50,969 # 00000010  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF  FF
MTD SDcard

tests/mtd_raw

2023-12-13 16:55:18,923 - INFO # main(): This is RIOT! (Version: 2024.01-devel-369-ga45d2-mtd_drop_write)
2023-12-13 16:55:18,923 - INFO # Manual MTD test
2023-12-13 16:55:19,034 - INFO # init MTD_0… OK (15223 MiB)
test 0
2023-12-13 16:55:26,746 - INFO # > test 0
2023-12-13 16:55:26,747 - INFO # 
2023-12-13 16:55:26,747 - INFO # [START]
2023-12-13 16:55:27,706 - INFO # [SUCCESS]
MTD spi_nor, at24mac, MTD SDMMC (with #20180)

tests/mtd_raw

2023-12-13 17:07:07,125 - INFO # Manual MTD test
2023-12-13 17:07:07,128 - INFO # init MTD_0… OK (8 MiB)
2023-12-13 17:07:07,131 - INFO # init MTD_1… OK (256 byte)
2023-12-13 17:07:07,342 - INFO # init MTD_2… OK (15223 MiB)
test 0
2023-12-13 17:07:10,339 - INFO # > test 0
2023-12-13 17:07:10,339 - INFO # 
2023-12-13 17:07:10,339 - INFO # [START]
2023-12-13 17:07:11,608 - INFO # [SUCCESS]
test 1
2023-12-13 17:07:14,593 - INFO # > test 1
2023-12-13 17:07:14,593 - INFO # 
2023-12-13 17:07:14,594 - INFO # [START]
2023-12-13 17:07:14,648 - INFO # [SUCCESS]
test 2
2023-12-13 17:07:16,275 - INFO # > test 2
2023-12-13 17:07:16,275 - INFO # 
2023-12-13 17:07:16,275 - INFO # [START]
2023-12-13 17:07:16,963 - INFO # [SUCCESS]

Issues/PRs references

includes #15378
probably also needs #14953 and #19258

@benpicco benpicco added Area: fs Area: File systems Type: cleanup The issue proposes a clean-up / The PR cleans-up parts of the codebase / documentation labels Nov 4, 2020
@benpicco benpicco requested a review from jue89 November 4, 2020 11:05
@benpicco benpicco changed the title mtd/*: drop .write() if .write_page() is implemnted mtd/*: drop .write() if .write_page() is implemented Nov 4, 2020
@jue89
Copy link
Contributor

jue89 commented Nov 4, 2020

Do you an idea how test this without owning the respective hardware?

Or can we prove that mtd's translation works in 100% of all cases?

@benpicco benpicco requested a review from miri64 as a code owner November 4, 2020 15:17
@benpicco benpicco added the State: waiting for other PR State: The PR requires another PR to be merged first label Nov 4, 2020
@benpicco
Copy link
Contributor Author

benpicco commented Nov 4, 2020

I don't think we have to test this on every implementation, the functionality is the same everywhere.
However, we should make sure nothing breaks when layering mtd implementations, e.g. with mtd_mapper.

I didn't do any tests yet, this is more a PoC, but the idea is to only have one write implementation per MTD backend.

@bergzand
Copy link
Member

Please rebase!

@benpicco benpicco added CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR and removed State: waiting for other PR State: The PR requires another PR to be merged first labels Nov 17, 2020
@benpicco benpicco requested a review from dylad January 29, 2021 20:50
@benpicco
Copy link
Contributor Author

rebased

@MrKevinWeiss MrKevinWeiss added this to the Release 2021.07 milestone Jun 22, 2021
@MrKevinWeiss
Copy link
Contributor

Any blockers for this, does it just need some testing?

@fjmolinas
Copy link
Contributor

I don't think we have to test this on every implementation, the functionality is the same everywhere.
However, we should make sure nothing breaks when layering mtd implementations, e.g. with mtd_mapper.

I didn't do any tests yet, this is more a PoC, but the idea is to only have one write implementation per MTD backend.

Las tcomment sated this was a PoC, is this still the case?

@benpicco
Copy link
Contributor Author

If .write_page() is implemented .write will not be used, so we can drop it and save some ROM.

@MrKevinWeiss MrKevinWeiss removed this from the Release 2021.07 milestone Jul 15, 2021
@fjmolinas
Copy link
Contributor

Do you have a test procedure for this?

@fjmolinas fjmolinas added the Process: removal Integration Process: The PR is removing a deprecated feature or API label Aug 24, 2021
@fjmolinas
Copy link
Contributor

If .write_page() is implemented .write will not be used, so we can drop it and save some ROM.

I only worry that eventually, some external code could be relying on this function, so at least it needs to be tagged as removal. We could also conditionally include (with a pseudo module) and then remove it in a release or two, but maybe it's not worth the effort. Let's just make sure it will be tagged in release notes.

bors bot added a commit that referenced this pull request Mar 7, 2023
@bors
Copy link
Contributor

bors bot commented Mar 7, 2023

try

Build succeeded:

The old .write() function is only used as a fall-back if .write_page()
is not implemented.
We can drop it.
The old .write() function is only used as a fall-back if .write_page()
is not implemented.
We can drop it.
The old .write() function is only used as a fall-back if .write_page()
is not implemented.
We can drop it.
The old .write() function is only used as a fall-back if .write_page()
is not implemented.
We can drop it.
The old .write() function is only used as a fall-back if .write_page()
is not implemented.
We can drop it.
The MTD layer now takes care of splitting up writes, so it
should always work (as long as the destination address is
within the memory region)
@benpicco benpicco requested review from maribu and removed request for gschorcht December 13, 2023 16:39
All MTD drivers should now implement the .write_page() function instead.
@benpicco
Copy link
Contributor Author

All backends are tested now, the virtual backends are tested by CI.

@benpicco benpicco added this pull request to the merge queue Dec 13, 2023
@benpicco
Copy link
Contributor Author

Thank you! 😃

@benpicco benpicco added the Process: API change Integration Process: PR contains or issue proposes an API change. Should be handled with care. label Dec 13, 2023
Merged via the queue into RIOT-OS:master with commit 6eac1e1 Dec 13, 2023
27 checks passed
@benpicco benpicco deleted the mtd_drop_write branch December 13, 2023 23:09
@MrKevinWeiss MrKevinWeiss added this to the Release 2024.01 milestone Feb 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Area: cpu Area: CPU/MCU ports Area: drivers Area: Device drivers Area: fs Area: File systems Area: tests Area: tests and testing framework CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR Platform: ESP Platform: This PR/issue effects ESP-based platforms Platform: native Platform: This PR/issue effects the native platform Process: API change Integration Process: PR contains or issue proposes an API change. Should be handled with care. Process: removal Integration Process: The PR is removing a deprecated feature or API Type: cleanup The issue proposes a clean-up / The PR cleans-up parts of the codebase / documentation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

10 participants