dts: drop HAS_DTS #63696
dts: drop HAS_DTS #63696
Conversation
gmarull
requested review from
carlocaione,
fabiobaltieri,
henrikbrixandersen and
carlescufi
gmarull
requested review from
jakub-uC,
nordic-krch,
sjg20,
tbursztyka,
nashif,
MaureenHelm and
tejlmand
as code owners
zephyrbot
added
area: Shell
There was a problem hiding this comment.
afaik there are still users not having DTS downstream.
Would like review from @mped-oticon / @daor-oti first.
Considering that all architectures require DTS, and that we do not support oot archs, I suspect this must be a heavily patched Zephyr. What this PR shows is that Zephyr as is now, can live without this option because it's doing nothing in practice. |
We support out of tree architectures :) |
|
@tejlmand thanks for the heads-up and inclusion! @gmarull Yes, there are are indeed platforms without (That said, I think DTS is great and we internally should use DTS -- but that would require some non-trivial work from our side. And that would require some more official longer-runway announcement; perhaps brought up in TSC) |
There was a problem hiding this comment.
As @aescolar writes, out of tree architectures are supported and used. We would not be able to use Zephyr without this.
We also cannot support DTS and therefor the HAS_DTS configure option is important to us, so please do not remove that.
Well but that's already the case since all platforms enables the option, if there's a regression related to If the option is supported there should be some test cases to ensure a minimum level or stability, but if all platforms have been moved it may make sense to make up a deadline to mark it as required and move on. |
|
@fabiobaltieri That would be an ideal solution, unfortunately we have a custom toolchain consisting if in-house developed tools, which we cannot share. Unless I am misunderstanding what you are suggesting? We are though looking into the issues and trying to see if we can come up with a solution that will allow us to not block this PR for too long, or possibly suggest a workaround. |
Not really, I understand that some tooling is custom but stuff like the
Sure, it's just that if we can have an upstream test that reproduces the specific condition in isolation we could have some confidence that it won't break again. Also, is there a linker error you could paste here? |
|
There are no linker errors and the target seems to be running, but is apparently missing the device drivers. We are investigating and will update here as soon as we have anything useful. Sorry for the holdup and the delays. |
When we have an empty Devicetree, ie,
```
/dts-v1/;
/ {
};
```
The node's dep_ordinal is never initialized because the node graph is
empty. This ends up with invalid ordinal tokens (-1) in
devicetree_generated.h which in turn produce some cryptic compiler
errors, see e.g.
```
token
95 | #define Z_DEVICE_DT_DEV_ID(node_id) _CONCAT(dts_ord_,
DT_DEP_ORD(node_id))
...
include/zephyr/devicetree.h:2498:41:
2498 | #define DT_FOREACH_STATUS_OKAY_NODE(fn)
DT_FOREACH_OKAY_HELPER(fn)
|
^~~~~~~~~~~~~~~~~~~~~~
include/zephyr/device.h:1022:1:
1022 |
DT_FOREACH_STATUS_OKAY_NODE(Z_MAYBE_DEVICE_DECLARE_INTERNAL)
```
(devicetree_generated.h)
```
...
#define DT_N_ORD -1
#define DT_N_ORD_STR_SORTABLE 000-1
...
```
This patch makes sure root node is always inserted (without any target)
so that it gets initialized later.
Discovered as part of
zephyrproject-rtos/zephyr#63696
(cherry picked from commit 7772cec)
Original-error: pasting "dts_ord_" and "-" does not give a valid preprocessing
Original-note: in expansion of macro 'DT_FOREACH_OKAY_HELPER'
Original-note: in expansion of macro 'DT_FOREACH_STATUS_OKAY_NODE'
Original-Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
GitOrigin-RevId: 7772cec
Change-Id: I8a3717a58e889c0c87734fc73e27a40eaf7c6f4c
Reviewed-on: https://chromium-review.googlesource.com/c/chromiumos/third_party/zephyr/+/4933942
Commit-Queue: Keith Short <keithshort@chromium.org>
Tested-by: ChromeOS Prod (Robot) <chromeos-ci-prod@chromeos-bot.iam.gserviceaccount.com>
Tested-by: Keith Short <keithshort@chromium.org>
Reviewed-by: Keith Short <keithshort@chromium.org>
By providing a devicetree stub file, we make sure some internal macros required by devicetree.h are generated in devicetree_generated.h. This makes sure that systems without devicetree can continue working without extra ifdeffery. Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
HAS_DTS has become a redundant option. All Zephyr architectures now select this option, meaning devicetree has become a de-facto requirement. In fact, if any board does not provide a devicetree source, the build system uses an empty stub, meaning the devicetree machinery always runs. Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
gmarull
dismissed stale reviews from fabiobaltieri, carlescufi, henrikbrixandersen, aescolar, and tejlmand
via
74f33ae
gmarull
requested review from
mbolivar-ampere,
daor-oti,
henrikbrixandersen,
tejlmand,
fabiobaltieri,
carlescufi and
aescolar
Please provide any logs or information, so it can be investigated. It looks like a bug that was being papered by HAS_DT, same as the one that this PR fixes. |
When we have an empty Devicetree, ie,
```
/dts-v1/;
/ {
};
```
The node's dep_ordinal is never initialized because the node graph is
empty. This ends up with invalid ordinal tokens (-1) in
devicetree_generated.h which in turn produce some cryptic compiler
errors, see e.g.
```
error: pasting "dts_ord_" and "-" does not give a valid preprocessing
token
95 | #define Z_DEVICE_DT_DEV_ID(node_id) _CONCAT(dts_ord_,
DT_DEP_ORD(node_id))
...
include/zephyr/devicetree.h:2498:41:
note: in expansion of macro 'DT_FOREACH_OKAY_HELPER'
2498 | #define DT_FOREACH_STATUS_OKAY_NODE(fn)
DT_FOREACH_OKAY_HELPER(fn)
|
^~~~~~~~~~~~~~~~~~~~~~
include/zephyr/device.h:1022:1:
note: in expansion of macro 'DT_FOREACH_STATUS_OKAY_NODE'
1022 |
DT_FOREACH_STATUS_OKAY_NODE(Z_MAYBE_DEVICE_DECLARE_INTERNAL)
```
(devicetree_generated.h)
```
...
#define DT_N_ORD -1
#define DT_N_ORD_STR_SORTABLE 000-1
...
```
This patch makes sure root node is always inserted (without any target)
so that it gets initialized later.
Discovered as part of
zephyrproject-rtos#63696
Signed-off-by: Gerard Marull-Paretas <gerard@teslabs.com>
HAS_DTS has become a redundant option. All Zephyr architectures now select this option, and no out-of-tree architectures are supported, meaning devicetree has become a de-facto requirement. There are actually many Zephyr areas that are not guarded, and expect devicetree to always be available regardless of HAS_DTS.
Depends on #63799