Skip to contentSkip to conversation

drivers/soc/litex: drop obsolete dependency on COMPILE_TEST

PATCH In review Jean Delvare 2 participants 5 messages Nov 24, 2022 – Dec 20, 2023 Litex platform

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drivers/soc/litex: drop obsolete dependency on COMPILE_TEST

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed.

It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues.

As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work.

Cc Karol Gugala <[email protected]>
Cc Mateusz Holenko <[email protected]>
Cc Gabriel Somlo <[email protected]>
Cc Joel Stanley <[email protected]>
---
 drivers/soc/litex/Kconfig          |    2 +-
 drivers/soc/litex/litex_soc_ctrl.c |    4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)
drivers/soc/litex/Kconfig +1 -1
@@ -7,7 +7,7 @@ config LITEX
 
 config LITEX_SOC_CONTROLLER
 	tristate "Enable LiteX SoC Controller driver"
-	depends on OF || COMPILE_TEST
+	depends on OF
 	depends on HAS_IOMEM
 	select LITEX
 	help
drivers/soc/litex/litex_soc_ctrl.c +1 -3
@@ -82,13 +82,11 @@ static int litex_reset_handler(struct no
 	return NOTIFY_DONE;
 }
 
-#ifdef CONFIG_OF
 static const struct of_device_id litex_soc_ctrl_of_match[] = {
 	{.compatible = "litex,soc-controller"},
 	{},
 };
 MODULE_DEVICE_TABLE(of, litex_soc_ctrl_of_match);
-#endif /* CONFIG_OF */
 
 static int litex_soc_ctrl_probe(struct platform_device *pdev)
 {
@@ -131,7 +129,7 @@ static int litex_soc_ctrl_remove(struct
 static struct platform_driver litex_soc_ctrl_driver = {
 	.driver = {
 		.name = "litex-soc-controller",
-		.of_match_table = of_match_ptr(litex_soc_ctrl_of_match)
+		.of_match_table = litex_soc_ctrl_of_match,
 	},
 	.probe = litex_soc_ctrl_probe,
 	.remove = litex_soc_ctrl_remove,
Signature · 2 lines
Jean Delvare
SUSE L3 Support
Acked-by on v1 1/1
#1 lore raw
Quoted 13 lines · Jean Delvare

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed.

It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues.

Quoted 3 lines

As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work.

Thanks,
--G

Quoted 51 lines
Cc Karol Gugala <[email protected]>
Cc Mateusz Holenko <[email protected]>
Cc Gabriel Somlo <[email protected]>
Cc Joel Stanley <[email protected]>
---
 drivers/soc/litex/Kconfig          |    2 +-
 drivers/soc/litex/litex_soc_ctrl.c |    4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)
drivers/soc/litex/Kconfig +1 -1
@@ -7,7 +7,7 @@ config LITEX
 
 config LITEX_SOC_CONTROLLER
 	tristate "Enable LiteX SoC Controller driver"
-	depends on OF || COMPILE_TEST
+	depends on OF
 	depends on HAS_IOMEM
 	select LITEX
 	help
drivers/soc/litex/litex_soc_ctrl.c +1 -3
@@ -82,13 +82,11 @@ static int litex_reset_handler(struct no
 	return NOTIFY_DONE;
 }
 
-#ifdef CONFIG_OF
 static const struct of_device_id litex_soc_ctrl_of_match[] = {
 	{.compatible = "litex,soc-controller"},
 	{},
 };
 MODULE_DEVICE_TABLE(of, litex_soc_ctrl_of_match);
-#endif /* CONFIG_OF */
 
 static int litex_soc_ctrl_probe(struct platform_device *pdev)
 {
@@ -131,7 +129,7 @@ static int litex_soc_ctrl_remove(struct
 static struct platform_driver litex_soc_ctrl_driver = {
 	.driver = {
 		.name = "litex-soc-controller",
-		.of_match_table = of_match_ptr(litex_soc_ctrl_of_match)
+		.of_match_table = litex_soc_ctrl_of_match,
 	},
 	.probe = litex_soc_ctrl_probe,
 	.remove = litex_soc_ctrl_remove,
Signature · 3 lines
-- 
Jean Delvare
SUSE L3 Support
#2 lore raw

Hi Gabriel,

Quoted 14 lines · Gabriel L. Somlo
Quoted 13 lines · Jean Delvare

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed.

It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues.

Quoted 5 lines
Quoted 3 lines

As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work.

Despite your ack, this patch was never committed. Was it forgotten somehow? Should I resubmit?

Thanks,

Signature · 2 lines
Jean Delvare
SUSE L3 Support
#3 lore raw
Quoted 17 lines · Jean Delvare

Hi Gabriel,

Quoted 14 lines · Gabriel L. Somlo
Quoted 13 lines · Jean Delvare

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed.

It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues.

Quoted 8 lines
Quoted 5 lines
Quoted 3 lines

As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work.

Despite your ack, this patch was never committed. Was it forgotten somehow? Should I resubmit?

AFAIK, LiteX is too small to have its own direct path into Linus's upstream tree, and so far any changes to LiteX specific kernel code were filtered upstream through the respective dedicated subsystems affected (e.g., mmc, networking, block, etc.).

IIRC Joel (cc-ed) might have been involved in the upstreaming of the original LiteX soc driver -- is that correct? If so, which way did it end up going upstream, and can we replicate that for Jean's patch?

Thanks much,
--Gabriel

Quoted 4 lines

Thanks,

Signature · 2 lines
Jean Delvare
SUSE L3 Support
Reviewed-by on v1 1/1
#4 lore raw
In reply to Jean Delvare
Quoted 17 lines · Jean Delvare

Since commit 0166dc11be91 ("of: make CONFIG_OF user selectable"), it is possible to test-build any driver which depends on OF on any architecture by explicitly selecting OF. Therefore depending on COMPILE_TEST as an alternative is no longer needed.

It is actually better to always build such drivers with OF enabled, so that the test builds are closer to how each driver will actually be built on its intended target. Building them without OF may not test much as the compiler will optimize out potentially large parts of the code. In the worst case, this could even pop false positive warnings. Dropping COMPILE_TEST here improves the quality of our testing and avoids wasting time on non-existent issues.

As a minor optimization, this also lets us drop of_match_ptr() and ifdef-guarding, as we now know what they will resolve to, we might as well save cpp some work.

Quoted 5 lines
Cc Karol Gugala <[email protected]>
Cc Mateusz Holenko <[email protected]>
Cc Gabriel Somlo <[email protected]>
Cc Joel Stanley <[email protected]>

Arnd: since it's been a while (sorry about that), please advise on whether Jean should re-send the original patch, unless you can grab it directly from here and pull it through the soc tree:
https://lore.kernel.org/lkml/[email protected]/

Thanks much,
--Gabriel

Quoted 47 lines
---
 drivers/soc/litex/Kconfig          |    2 +-
 drivers/soc/litex/litex_soc_ctrl.c |    4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)
drivers/soc/litex/Kconfig +1 -1
@@ -7,7 +7,7 @@ config LITEX
 
 config LITEX_SOC_CONTROLLER
 	tristate "Enable LiteX SoC Controller driver"
-	depends on OF || COMPILE_TEST
+	depends on OF
 	depends on HAS_IOMEM
 	select LITEX
 	help
drivers/soc/litex/litex_soc_ctrl.c +1 -3
@@ -82,13 +82,11 @@ static int litex_reset_handler(struct no
 	return NOTIFY_DONE;
 }
 
-#ifdef CONFIG_OF
 static const struct of_device_id litex_soc_ctrl_of_match[] = {
 	{.compatible = "litex,soc-controller"},
 	{},
 };
 MODULE_DEVICE_TABLE(of, litex_soc_ctrl_of_match);
-#endif /* CONFIG_OF */
 
 static int litex_soc_ctrl_probe(struct platform_device *pdev)
 {
@@ -131,7 +129,7 @@ static int litex_soc_ctrl_remove(struct
 static struct platform_driver litex_soc_ctrl_driver = {
 	.driver = {
 		.name = "litex-soc-controller",
-		.of_match_table = of_match_ptr(litex_soc_ctrl_of_match)
+		.of_match_table = litex_soc_ctrl_of_match,
 	},
 	.probe = litex_soc_ctrl_probe,
 	.remove = litex_soc_ctrl_remove,
Signature · 3 lines
-- 
Jean Delvare
SUSE L3 Support