llext: export DT devices to extensions #78508
Merged
Conversation
This file contains 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
The error message for Z_DEVICE_NAME_CHECK was needlessly calling DEVICE_NAME_GET() on the "name" string, resulting in a confusing message being displayed. Remove the DEVICE_NAME_GET call so that the error message shown by the compiler contains the actual name. Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
pillo79
force-pushed
the
pr-llext-export-dev
branch
2 times, most recently
from
031a747
to
0d1c9a1
Compare
zephyrbot
requested review from
dcpleung,
gmarull,
lyakh,
nashif,
tbursztyka and
teburd
lyakh
reviewed
Sep 18, 2024
The comment for Z_DEVICE_DT_DEV_ID() is misleading: device_get_binding() compares its inputs with values generated by the DEVICE_DT_NAME() macro, which expands to the node's DT label or to the node's full name. This patch updates the comment to reflect the actual behavior of Z_DEVICE_DT_DEV_ID(). Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
pillo79
force-pushed
the
pr-llext-export-dev
branch
from
0d1c9a1
to
14fb69f
Compare
|
@gmarull can you take a look? |
gmarull
reviewed
Sep 20, 2024
include/zephyr/device.h
Outdated
| @@ -197,7 +207,8 @@ typedef int16_t device_handle_t; | |||
| DEVICE_DT_NAME(node_id), init_fn, pm, data, config, \ | |||
| level, prio, api, \ | |||
| &Z_DEVICE_STATE_NAME(Z_DEVICE_DT_DEV_ID(node_id)), \ | |||
| __VA_ARGS__) | |||
| __VA_ARGS__); \ | |||
| Z_DEVICE_EXPORT(node_id) | |||
There was a problem hiding this comment.
you can use IF_ENABLED(CONFIG_LLEXT_EXPORT_DEVICES, (EXPORT_SYMBOL(DEVICE_DT_NAME_GET(node_id))))
There was a problem hiding this comment.
I could, but the separate macro is easier to extend in the (short) future; I have a draft of a more complex proposal (#77799) where there are two different implementations for Z_DEVICE_EXPORT.
For this reason I'm proposing to keep the separate macro, but included within an IF_ENABLED in the parent.
include/zephyr/device.h
Outdated
Comment on lines
103
to
110
| #if defined(CONFIG_LLEXT_EXPORT_DEVICES) | ||
| /* Export device identifiers using the builtin name */ | ||
| #define Z_DEVICE_EXPORT(node_id) EXPORT_SYMBOL(DEVICE_DT_NAME_GET(node_id)) | ||
| #else | ||
| /* Do not export device identifiers */ | ||
| #define Z_DEVICE_EXPORT(node_id) | ||
| #endif | ||
|
|
include/zephyr/device.h
Outdated
| @@ -197,7 +207,8 @@ typedef int16_t device_handle_t; | |||
| DEVICE_DT_NAME(node_id), init_fn, pm, data, config, \ | |||
| level, prio, api, \ | |||
| &Z_DEVICE_STATE_NAME(Z_DEVICE_DT_DEV_ID(node_id)), \ | |||
| __VA_ARGS__) | |||
| __VA_ARGS__); \ | |||
There was a problem hiding this comment.
shouldn't trailing ; be conditioned to the next statement? if not, move export above.
This patch adds an indirection level moving the current code from inside the EXPORT_SYMBOL and LL_EXTENSION_SYMBOL macros to private Z_* macros. This is done to allow for the official APIs to properly expand more complex arguments, such as the DT function-like macros. Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
This change adds a new configuration option, LLEXT_EXPORT_DEVICES, which enables exporting all devices defined in the device tree to llext extensions. When enabled, all devices are made available to extensions via the standard DT_ / DEVICE_* macros. Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
Add a new test to the simple llext test suite that verifies that DT devices are properly exported to extensions. This is done by obtaining a device handle from the DT and then trying to get the same handle at runtime using the device_get_binding() API. This test is optional because the current qemu_xtensa platform does not have any usable DT devices to test with. The LLEXT heap size is also increased to 64 kbytes to avoid heap exhaustion when running the new test on mps2_an385. Signed-off-by: Luca Burelli <l.burelli@arduino.cc>
pillo79
force-pushed
the
pr-llext-export-dev
branch
from
14fb69f
to
689da27
Compare
|
v2:
|
teburd
approved these changes
Sep 24, 2024
This was referenced Sep 25, 2024
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.
This PR adds support for automatic export of all Zephyr devices to extensions via a new Kconfig option,
CONFIG_LLEXT_EXPORT_DEVICES. When this option is enabled, all Devicetree macros inzephyr/device.hcan be used in extensions as well to inspect and interact with the devices configured in the current build.Calling
EXPORT_SYMBOLwith anything more complicated than a simple identifier resulted in macro expansion issues. This has been fixed by adding a level of indirection, moving the actual symbol definition code to privateZ_macros. A number of related comments have been updated.New test code has been added to properly verify the new support. The test retrieves the device object associated with the Zephyr console via
DT_CHOSENmacros and verifies it matches with the one obtained at runtime by callingdevice_get_binding(). This check is currently unsupported on Xtensa since theqemu_xtensaplatform defines no usable DT devices.Also fixes 2 minor issues in
device.h:Z_DEVICE_NAME_CHECKincluded a call toDEVICE_NAME_GET, but the argument was already a string. This resulted in a confusing error message being displayed when the check failed.Z_DEVICE_DT_DEV_IDwas used indevice_get_binding()instead of the DT path, and must thus be limited toZ_DEVICE_MAX_NAME_LENchars.This is not the case (anymore?):
device_get_binding()matches neither paths nor identifiers, but either DT labels or DT full names as returned by theDEVICE_DT_NAMEmacro here.