* [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused
@ 2020-01-07 22:14 Arnd Bergmann
2020-01-07 22:15 ` [PATCH 2/2] ata: brcm: fix reset controller API usage Arnd Bergmann
2020-01-07 22:51 ` [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Florian Fainelli
0 siblings, 2 replies; 4+ messages in thread
From: Arnd Bergmann @ 2020-01-07 22:14 UTC (permalink / raw)
To: Jens Axboe, Florian Fainelli, Hans de Goede
Cc: Arnd Bergmann, Greg Kroah-Hartman, Thomas Gleixner, linux-ide,
linux-kernel
The new shutdown callback causes a link failure:
drivers/ata/ahci_brcm.c: In function 'brcm_ahci_shutdown':
drivers/ata/ahci_brcm.c:552:8: error: implicit declaration of function 'brcm_ahci_suspend'; did you mean 'brcm_ahci_shutdown'? [-Werror=implicit-function-declaration]
ret = brcm_ahci_suspend(&pdev->dev);
^~~~~~~~~~~~~~~~~
Remove the incorrect #ifdef and use __maybe_unused annotations
instead to make this more robust.
Fixes: 7de9b1688c1d ("ata: ahci_brcm: Add a shutdown callback")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
drivers/ata/ahci_brcm.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/ata/ahci_brcm.c b/drivers/ata/ahci_brcm.c
index 13ceca687104..239333d11b88 100644
--- a/drivers/ata/ahci_brcm.c
+++ b/drivers/ata/ahci_brcm.c
@@ -338,7 +338,6 @@ static const struct ata_port_info ahci_brcm_port_info = {
.port_ops = &ahci_brcm_platform_ops,
};
-#ifdef CONFIG_PM_SLEEP
static int brcm_ahci_suspend(struct device *dev)
{
struct ata_host *host = dev_get_drvdata(dev);
@@ -348,7 +347,10 @@ static int brcm_ahci_suspend(struct device *dev)
brcm_sata_phys_disable(priv);
- ret = ahci_platform_suspend(dev);
+ if (IS_ENABLED(CONFIG_PM_SLEEP))
+ ret = ahci_platform_suspend(dev);
+ else
+ ret = 0;
if (!IS_ERR_OR_NULL(priv->rcdev))
reset_control_assert(priv->rcdev);
@@ -356,7 +358,7 @@ static int brcm_ahci_suspend(struct device *dev)
return ret;
}
-static int brcm_ahci_resume(struct device *dev)
+static int __maybe_unused brcm_ahci_resume(struct device *dev)
{
struct ata_host *host = dev_get_drvdata(dev);
struct ahci_host_priv *hpriv = host->private_data;
@@ -405,7 +407,6 @@ static int brcm_ahci_resume(struct device *dev)
ahci_platform_disable_clks(hpriv);
return ret;
}
-#endif
static struct scsi_host_template ahci_platform_sht = {
AHCI_SHT(DRV_NAME),
--
2.20.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH 2/2] ata: brcm: fix reset controller API usage
2020-01-07 22:14 [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Arnd Bergmann
@ 2020-01-07 22:15 ` Arnd Bergmann
2020-01-07 22:52 ` Florian Fainelli
2020-01-07 22:51 ` [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Florian Fainelli
1 sibling, 1 reply; 4+ messages in thread
From: Arnd Bergmann @ 2020-01-07 22:15 UTC (permalink / raw)
To: Jens Axboe, Philipp Zabel, Florian Fainelli
Cc: Arnd Bergmann, Hans de Goede, Allison Randal, Thomas Gleixner,
linux-ide, linux-kernel
While fixing another issue in this driver I noticed it uses
IS_ERR_OR_NULL(), which is almost always a mistake.
Change the driver to use the proper devm_reset_control_get_optional()
interface instead and remove the checks except for the one that
checks for a failure in that function.
Fixes: 2b2c47d9e1fe ("ata: ahci_brcm: Allow optional reset controller to be used")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
drivers/ata/ahci_brcm.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/drivers/ata/ahci_brcm.c b/drivers/ata/ahci_brcm.c
index 239333d11b88..7ac1141c6ad0 100644
--- a/drivers/ata/ahci_brcm.c
+++ b/drivers/ata/ahci_brcm.c
@@ -352,8 +352,7 @@ static int brcm_ahci_suspend(struct device *dev)
else
ret = 0;
- if (!IS_ERR_OR_NULL(priv->rcdev))
- reset_control_assert(priv->rcdev);
+ reset_control_assert(priv->rcdev);
return ret;
}
@@ -365,8 +364,7 @@ static int __maybe_unused brcm_ahci_resume(struct device *dev)
struct brcm_ahci_priv *priv = hpriv->plat_data;
int ret = 0;
- if (!IS_ERR_OR_NULL(priv->rcdev))
- ret = reset_control_deassert(priv->rcdev);
+ ret = reset_control_deassert(priv->rcdev);
if (ret)
return ret;
@@ -454,9 +452,11 @@ static int brcm_ahci_probe(struct platform_device *pdev)
else
reset_name = "ahci";
- priv->rcdev = devm_reset_control_get(&pdev->dev, reset_name);
- if (!IS_ERR_OR_NULL(priv->rcdev))
- reset_control_deassert(priv->rcdev);
+ priv->rcdev = devm_reset_control_get_optional(&pdev->dev, reset_name);
+ if (IS_ERR(priv->rcdev))
+ return PTR_ERR(priv->rcdev);
+
+ reset_control_deassert(priv->rcdev);
hpriv = ahci_platform_get_resources(pdev, 0);
if (IS_ERR(hpriv)) {
@@ -520,8 +520,7 @@ static int brcm_ahci_probe(struct platform_device *pdev)
out_disable_clks:
ahci_platform_disable_clks(hpriv);
out_reset:
- if (!IS_ERR_OR_NULL(priv->rcdev))
- reset_control_assert(priv->rcdev);
+ reset_control_assert(priv->rcdev);
return ret;
}
--
2.20.0
^ permalink raw reply related [flat|nested] 4+ messages in thread* Re: [PATCH 2/2] ata: brcm: fix reset controller API usage
2020-01-07 22:15 ` [PATCH 2/2] ata: brcm: fix reset controller API usage Arnd Bergmann
@ 2020-01-07 22:52 ` Florian Fainelli
0 siblings, 0 replies; 4+ messages in thread
From: Florian Fainelli @ 2020-01-07 22:52 UTC (permalink / raw)
To: Arnd Bergmann, Jens Axboe, Philipp Zabel
Cc: Hans de Goede, Allison Randal, Thomas Gleixner, linux-ide,
linux-kernel
On 1/7/20 2:15 PM, Arnd Bergmann wrote:
> While fixing another issue in this driver I noticed it uses
> IS_ERR_OR_NULL(), which is almost always a mistake.
>
> Change the driver to use the proper devm_reset_control_get_optional()
> interface instead and remove the checks except for the one that
> checks for a failure in that function.
>
> Fixes: 2b2c47d9e1fe ("ata: ahci_brcm: Allow optional reset controller to be used")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
This is nearly equivalent to:
https://lore.kernel.org/lkml/20200107183022.26224-2-f.fainelli@gmail.com/
which was just submitted this morning. Thanks!
> ---
> drivers/ata/ahci_brcm.c | 17 ++++++++---------
> 1 file changed, 8 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/ata/ahci_brcm.c b/drivers/ata/ahci_brcm.c
> index 239333d11b88..7ac1141c6ad0 100644
> --- a/drivers/ata/ahci_brcm.c
> +++ b/drivers/ata/ahci_brcm.c
> @@ -352,8 +352,7 @@ static int brcm_ahci_suspend(struct device *dev)
> else
> ret = 0;
>
> - if (!IS_ERR_OR_NULL(priv->rcdev))
> - reset_control_assert(priv->rcdev);
> + reset_control_assert(priv->rcdev);
>
> return ret;
> }
> @@ -365,8 +364,7 @@ static int __maybe_unused brcm_ahci_resume(struct device *dev)
> struct brcm_ahci_priv *priv = hpriv->plat_data;
> int ret = 0;
>
> - if (!IS_ERR_OR_NULL(priv->rcdev))
> - ret = reset_control_deassert(priv->rcdev);
> + ret = reset_control_deassert(priv->rcdev);
> if (ret)
> return ret;
>
> @@ -454,9 +452,11 @@ static int brcm_ahci_probe(struct platform_device *pdev)
> else
> reset_name = "ahci";
>
> - priv->rcdev = devm_reset_control_get(&pdev->dev, reset_name);
> - if (!IS_ERR_OR_NULL(priv->rcdev))
> - reset_control_deassert(priv->rcdev);
> + priv->rcdev = devm_reset_control_get_optional(&pdev->dev, reset_name);
> + if (IS_ERR(priv->rcdev))
> + return PTR_ERR(priv->rcdev);
> +
> + reset_control_deassert(priv->rcdev);
>
> hpriv = ahci_platform_get_resources(pdev, 0);
> if (IS_ERR(hpriv)) {
> @@ -520,8 +520,7 @@ static int brcm_ahci_probe(struct platform_device *pdev)
> out_disable_clks:
> ahci_platform_disable_clks(hpriv);
> out_reset:
> - if (!IS_ERR_OR_NULL(priv->rcdev))
> - reset_control_assert(priv->rcdev);
> + reset_control_assert(priv->rcdev);
> return ret;
> }
>
>
--
Florian
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused
2020-01-07 22:14 [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Arnd Bergmann
2020-01-07 22:15 ` [PATCH 2/2] ata: brcm: fix reset controller API usage Arnd Bergmann
@ 2020-01-07 22:51 ` Florian Fainelli
1 sibling, 0 replies; 4+ messages in thread
From: Florian Fainelli @ 2020-01-07 22:51 UTC (permalink / raw)
To: Arnd Bergmann, Jens Axboe, Hans de Goede
Cc: Greg Kroah-Hartman, Thomas Gleixner, linux-ide, linux-kernel
On 1/7/20 2:14 PM, Arnd Bergmann wrote:
> The new shutdown callback causes a link failure:
>
> drivers/ata/ahci_brcm.c: In function 'brcm_ahci_shutdown':
> drivers/ata/ahci_brcm.c:552:8: error: implicit declaration of function 'brcm_ahci_suspend'; did you mean 'brcm_ahci_shutdown'? [-Werror=implicit-function-declaration]
> ret = brcm_ahci_suspend(&pdev->dev);
> ^~~~~~~~~~~~~~~~~
>
> Remove the incorrect #ifdef and use __maybe_unused annotations
> instead to make this more robust.
>
> Fixes: 7de9b1688c1d ("ata: ahci_brcm: Add a shutdown callback")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
good catch, thanks!
--
Florian
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2020-01-07 22:53 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-01-07 22:14 [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Arnd Bergmann
2020-01-07 22:15 ` [PATCH 2/2] ata: brcm: fix reset controller API usage Arnd Bergmann
2020-01-07 22:52 ` Florian Fainelli
2020-01-07 22:51 ` [PATCH 1/2] ata: brcm: mark PM functions as __maybe_unused Florian Fainelli
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox