DTS API close out for 2.3

[mbolivar-nordic]

DTS changes for 2.3.

This PR is attempting to close out the remaining issues discussed in the last dev-review, namely being able to access data for all nodes regardless of their status.

This has knock-on effects in the API which cause various renames.

We're going to defer a few more of them until after code freeze to avoid breaking CI any more.

[zephyrbot]

All checks are passing now.

checkpatch (informational only, not a failure)

-:898: WARNING:LEADING_SPACE: please, no spaces at the start of a line
#898: FILE: drivers/can/can_stm32.c:24:
+    DT_NODE_HAS_COMPAT_STATUS(DT_NODELABEL(can2), st_stm32_can, okay)$

-:2550: WARNING:TRAILING_SEMICOLON: macros should not use a trailing semicolon
#2550: FILE: drivers/sensor/nxp_kinetis_temp/temp_kinetis.c:186:
+#define TEMP_KINETIS_INIT(inst)						\
+	BUILD_ASSERT(DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, sensor) <	\
+		     DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, bandgap),	\
+		     "This driver assumes sensor ADC channel to come before "\
+		     "bandgap ADC channel");				\
+									\
+	static struct temp_kinetis_data temp_kinetis_data_0;		\
+									\
+	static const struct temp_kinetis_config temp_kinetis_config_0 = {\
+		.adc_dev_name =						\
+			DT_INST_IO_CHANNELS_LABEL_BY_IDX(inst, 0),	\
+		.sensor_adc_ch =					\
+			DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, sensor),\
+		.bandgap_adc_ch =					\
+			DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, bandgap),\
+		.bandgap_mv = DT_INST_PROP(0, bandgap_voltage) / 1000,	\
+		.vtemp25_mv = DT_INST_PROP(0, vtemp25) / 1000,		\
+		.slope_cold_uv = DT_INST_PROP(0, sensor_slope_cold),	\
+		.slope_hot_uv = DT_INST_PROP(0, sensor_slope_hot),	\
+		.adc_seq = {						\
+			.options = NULL,				\
+			.channels =					\
+		BIT(DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, sensor)) |	\
+		BIT(DT_INST_IO_CHANNELS_INPUT_BY_NAME(inst, bandgap)),	\
+			.buffer = &temp_kinetis_data_0.buffer,		\
+			.buffer_size = sizeof(temp_kinetis_data_0.buffer),\
+			.resolution = CONFIG_TEMP_KINETIS_RESOLUTION,	\
+			.oversampling = CONFIG_TEMP_KINETIS_OVERSAMPLING,\
+			.calibrate = false,				\
+		},							\
+	};								\
+									\
+	DEVICE_AND_API_INIT(temp_kinetis, DT_INST_LABEL(inst),		\
+			    temp_kinetis_init, &temp_kinetis_data_0,	\
+			    &temp_kinetis_config_0, POST_KERNEL,	\
+			    CONFIG_SENSOR_INIT_PRIORITY,		\
+			    &temp_kinetis_driver_api);

-:2957: WARNING:LONG_LINE: line over 80 characters
#2957: FILE: drivers/usb/device/usb_dc_stm32.c:60:
+#if DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otgfs) && DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otghs)

-:2992: WARNING:LONG_LINE: line over 80 characters
#2992: FILE: drivers/usb/device/usb_dc_stm32.c:138:
+#elif DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otgfs) || DT_HAS_COMPAT_STATUS_OKAY(st_stm32_usb)

-:3012: WARNING:LONG_LINE: line over 80 characters
#3012: FILE: drivers/usb/device/usb_dc_stm32.c:313:
+#if DT_HAS_COMPAT_STATUS_OKAY(st_stm32_usbphyc) && DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otghs)

-:3021: WARNING:LONG_LINE: line over 80 characters
#3021: FILE: drivers/usb/device/usb_dc_stm32.c:324:
+#if DT_HAS_COMPAT_STATUS_OKAY(st_stm32_usbphyc) && DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otghs)

-:3038: WARNING:LONG_LINE: line over 80 characters
#3038: FILE: drivers/usb/device/usb_dc_stm32.c:360:
+#if DT_HAS_COMPAT_STATUS_OKAY(st_stm32_usbphyc) && DT_HAS_COMPAT_STATUS_OKAY(st_stm32_otghs)

-:3187: WARNING:COMPLEX_MACRO: Macros with complex values should be enclosed in parentheses
#3187: FILE: include/devicetree.h:1013:
+#define DT_NODE_HAS_COMPAT_STATUS(node_id, compat, status) \
+	DT_NODE_HAS_COMPAT(node_id, compat) && DT_NODE_HAS_STATUS(node_id, status)

-:3188: WARNING:LONG_LINE: line over 80 characters
#3188: FILE: include/devicetree.h:1014:
+	DT_NODE_HAS_COMPAT(node_id, compat) && DT_NODE_HAS_STATUS(node_id, status)

-:3680: WARNING:LONG_LINE: line over 80 characters
#3680: FILE: tests/lib/devicetree/src/main.c:215:
+	zassert_equal(DT_NODE_HAS_STATUS(DT_NODELABEL(test_no_status), disabled),

-:3705: WARNING:LONG_LINE_STRING: line over 80 characters
#3705: FILE: tests/lib/devicetree/src/main.c:299:
+	zassert_equal(DT_ANY_INST_ON_BUS_STATUS_OKAY(spi), 1, "no spi is on spi");

-:3706: WARNING:LONG_LINE_STRING: line over 80 characters
#3706: FILE: tests/lib/devicetree/src/main.c:300:
+	zassert_equal(DT_ANY_INST_ON_BUS_STATUS_OKAY(i2c), 0, "a spi is on i2c");

-:3723: WARNING:LONG_LINE_STRING: line over 80 characters
#3723: FILE: tests/lib/devicetree/src/main.c:315:
+	zassert_equal(DT_ANY_INST_ON_BUS_STATUS_OKAY(i2c), 1, "no i2c is on i2c");

-:3724: WARNING:LONG_LINE_STRING: line over 80 characters
#3724: FILE: tests/lib/devicetree/src/main.c:316:
+	zassert_equal(DT_ANY_INST_ON_BUS_STATUS_OKAY(spi), 0, "an i2c is on spi");

-:3878: WARNING:SINGLE_STATEMENT_DO_WHILE_MACRO: Single statement macros should not use a do {} while (0) loop
#3878: FILE: tests/lib/devicetree/src/main.c:1268:
+#define INC(inst_ignored) do { val++; } while (0);

-:3878: WARNING:DO_WHILE_MACRO_WITH_TRAILING_SEMICOLON: do {} while (0) macros should not be semicolon terminated
#3878: FILE: tests/lib/devicetree/src/main.c:1268:
+#define INC(inst_ignored) do { val++; } while (0);

-:3884: WARNING:LONG_LINE_COMMENT: line over 80 characters
#3884: FILE: tests/lib/devicetree/src/main.c:1273:
+	 * Make sure DT_INST_FOREACH_STATUS_OKAY works with 0 instances, and does

- total: 0 errors, 17 warnings, 2947 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.

Your patch has style problems, please review.

NOTE: Ignored message types: AVOID_EXTERNS BRACES CONFIG_EXPERIMENTAL CONST_STRUCT DATE_TIME FILE_PATH_CHANGES MINMAX NETWORKING_BLOCK_COMMENT_STYLE PRINTK_WITHOUT_KERN_LEVEL SPDX_LICENSE_TAG SPLIT_STRING VOLATILE

NOTE: If any of the errors are false positives, please report
      them to the maintainers.

Tip: The bot edits this comment instead of posting a new one, so you can check the comment's history to see earlier messages.

[galak]

Need decide if we want to phase out compat2enabled or tweak its semantics. I think changing the semantics is ok since the only status values we've really dealt with are (no status, status = "okay", status = "ok", and status = "disabled". I'm also good if we want to deprecate this and introduce something new.

[mbolivar-nordic]

Need decide if we want to phase out compat2enabled or tweak its semantics. I think changing the semantics is ok since the only status values we've really dealt with are (no status, status = "okay", status = "ok", and status = "disabled". I'm also good if we want to deprecate this and introduce something new.

We agreed in dev-review that whatever we do, we can't break gen_legacy_defines.py. It uses compat2enabled, so we can't change its semantics without also making changes to gen_legacy_defines.py, which seems a bit reckless this late in the release cycle.

I'm adding new functionality specifically to be able to leave the legacy scripts alone.

[galak]

Need decide if we want to phase out compat2enabled or tweak its semantics. I think changing the semantics is ok since the only status values we've really dealt with are (no status, status = "okay", status = "ok", and status = "disabled". I'm also good if we want to deprecate this and introduce something new.

We agreed in dev-review that whatever we do, we can't break gen_legacy_defines.py. It uses compat2enabled, so we can't change its semantics without also making changes to gen_legacy_defines.py, which seems a bit reckless this late in the release cycle.

I'm adding new functionality specifically to be able to leave the legacy scripts alone.

We should probably add a comment in compat2enabled and node.enabled about these only existing for gen_legacy_defines.py. Do we want to call this compat2avail to match the linux API?

Reported CAN issue solved

[pabigot]

Haven't looked closely at the documentation yet but figure that can be tuned if necessary.

[mbolivar-nordic]

Haven't looked closely at the documentation yet but figure that can be tuned if necessary.

Yep, I promise to go over the documentation with a fine tooth comb before the release and correct any errors I find during the stabilization/documentation period. Thanks.

[ioannisg]

remove "is"

[ioannisg]

remove second either in the line

[pabigot]

No, this is correct. There are two optional subordinate clauses, one of which requires its own either.

[pabigot]

Or rather, it's correct with two eithers, just the first, just the second, or neither. IMO. English is pretty flexible.

[ioannisg]

Looks ok to me
Pls, address some minor doc-related comments.