SPI: allow forcing CPU-driven blocking transfers#5350
Open
bugadani wants to merge 1 commit intoesp-rs:mainfrom
Open
SPI: allow forcing CPU-driven blocking transfers#5350bugadani wants to merge 1 commit intoesp-rs:mainfrom
bugadani wants to merge 1 commit intoesp-rs:mainfrom
Conversation
bugadani
force-pushed
the
spi-force-blocking-config
branch
3 times, most recently
from
70f6776 to
9f22983
Compare
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
9f22983 to
7454c8e
Compare
Contributor
Author
|
/hil full |
|
Triggered full HIL run for #5350. Run: https://github.com/esp-rs/esp-hal/actions/runs/24341117212 Status update: ❌ HIL (full) run failed (conclusion: failure). |
Contributor
Author
|
/test-size embassy_spi esp32s3 |
Binary Size Report
|
github-actions
Bot
added
the
merge-conflict
|
New commits in main has made this PR unmergable. Please resolve the conflicts. |
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
7454c8e to
a10fea5
Compare
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
a10fea5 to
57d0584
Compare
github-actions
Bot
removed
the
merge-conflict
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
57d0584 to
036e67e
Compare
Contributor
There was a problem hiding this comment.
Pull request overview
Adds a new SPI master configuration knob to optionally bypass async/DMA overhead for small transfers by forcing CPU-driven (blocking) transfers under a configurable length threshold.
Changes:
- Introduces
Config::min_async_transfer_size(unstable setter) and stores it in per-peripheralState. - Uses the threshold in
Spi<Async>andSpiBusAsyncimplementations to fall back to blocking transfers for small buffers. - Extends the same threshold behavior to
SpiDma(blocking + async), disabling DMA and executing CPU-driven transfers when below the threshold.
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| esp-hal/src/spi/master/mod.rs | Adds config/state for the threshold and applies it in async SPI paths. |
| esp-hal/src/spi/master/dma.rs | Applies the same threshold to DMA driver paths, with a CPU fallback. |
| esp-hal/CHANGELOG.md | Documents the new SPI master config option. |
Comment on lines
+475
to
+487
| /// Minimum transfer size in bytes below which CPU-driven (blocking) I/O | ||
| /// is used instead of async or DMA transfers. | ||
| /// | ||
| /// This can reduce overhead for small transfers where DMA setup or | ||
| /// async context-switch cost exceeds the benefit. For | ||
| /// [`SpiDma`][crate::spi::master::dma::SpiDma], the threshold applies in | ||
| /// both blocking and async DMA modes: when met, DMA is disabled and the | ||
| /// transfer is performed by the CPU. | ||
| /// | ||
| /// A value of `0` (the default) disables the threshold — all transfers use | ||
| /// the driver's default method. | ||
| #[builder_lite(unstable)] | ||
| min_async_transfer_size: usize, |
Contributor
Author
There was a problem hiding this comment.
We can have a mile-long field name, or just accept this...
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
036e67e to
943d031
Compare
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
943d031 to
f3b2d37
Compare
bugadani
force-pushed
the
spi-force-blocking-config
branch
from
f3b2d37 to
e833770
Compare
Contributor
Author
|
/hil full --test spi_full_duplex |
|
Triggered full HIL run for #5350. Run: https://github.com/esp-rs/esp-hal/actions/runs/25435607324 Status update: ❌ HIL (full) run failed (conclusion: failure). |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Transferring only a few times may be faster than setting up the DMA or rescheduling an async task. This PR adds a configuration that lets users force CPU-driven blocking mode below a certain transfer length, to avoid these setup/reschedule times.