2020-11-02 08:59:23

by Yejune Deng

Subject: [PATCH] drm/panfrost: Add support for non-existent reset node

Some dts hasn't no reset node, is should ok.

Signed-off-by: Yejune Deng <[email protected]>
---
drivers/gpu/drm/panfrost/panfrost_device.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c
index ea8d318..a982878 100644
--- a/drivers/gpu/drm/panfrost/panfrost_device.c
+++ b/drivers/gpu/drm/panfrost/panfrost_device.c
@@ -18,7 +18,12 @@

static int panfrost_reset_init(struct panfrost_device *pfdev)
{
- pfdev->rstc = devm_reset_control_array_get(pfdev->dev, false, true);
+ pfdev->rstc = devm_reset_control_array_get_optional_exclusive(pfdev->dev);
+ if (!pfdev->rstc) {
+ dev_info(pfdev->dev, "reset does not exist\n");
+ return 0;
+ }
+
if (IS_ERR(pfdev->rstc)) {
dev_err(pfdev->dev, "get reset failed %ld\n", PTR_ERR(pfdev->rstc));
return PTR_ERR(pfdev->rstc);
@@ -29,7 +34,8 @@ static int panfrost_reset_init(struct panfrost_device *pfdev)

static void panfrost_reset_fini(struct panfrost_device *pfdev)
{
- reset_control_assert(pfdev->rstc);
+ if (pfdev->rstc)
+ reset_control_assert(pfdev->rstc);
}

static int panfrost_clk_init(struct panfrost_device *pfdev)
--
1.9.1

2020-11-02 09:15:47

by Steven Price

Subject: Re: [PATCH] drm/panfrost: Add support for non-existent reset node

On 02/11/2020 08:54, Yejune Deng wrote:
> Some dts hasn't no reset node, is should ok.
>
> Signed-off-by: Yejune Deng <[email protected]>
> ---
> drivers/gpu/drm/panfrost/panfrost_device.c | 10 ++++++++--
> 1 file changed, 8 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/panfrost/panfrost_device.c b/drivers/gpu/drm/panfrost/panfrost_device.c
> index ea8d318..a982878 100644
> --- a/drivers/gpu/drm/panfrost/panfrost_device.c
> +++ b/drivers/gpu/drm/panfrost/panfrost_device.c
> @@ -18,7 +18,12 @@
>
> static int panfrost_reset_init(struct panfrost_device *pfdev)
> {
> - pfdev->rstc = devm_reset_control_array_get(pfdev->dev, false, true);
> + pfdev->rstc = devm_reset_control_array_get_optional_exclusive(pfdev->dev);

This change is fine - it should have no functional change and makes the
code more readable.

> + if (!pfdev->rstc) {
> + dev_info(pfdev->dev, "reset does not exist\n");

I don't think this output is particularly useful.

> + return 0;

The early return isn't necessary. IS_ERR(NULL) is false, and
reset_control_deassert() deals fine with NULL, so the path below would
be fine.

> + }
> +
> if (IS_ERR(pfdev->rstc)) {
> dev_err(pfdev->dev, "get reset failed %ld\n", PTR_ERR(pfdev->rstc));
> return PTR_ERR(pfdev->rstc);
> @@ -29,7 +34,8 @@ static int panfrost_reset_init(struct panfrost_device *pfdev)
>
> static void panfrost_reset_fini(struct panfrost_device *pfdev)
> {
> - reset_control_assert(pfdev->rstc);
> + if (pfdev->rstc)
> + reset_control_assert(pfdev->rstc);

reset_control_assert() deals fine with a NULL input, so this is unneeded.

> }
>
> static int panfrost_clk_init(struct panfrost_device *pfdev)
>

Have you actually seen a problem on a platform without a reset node - it
works fine for me and I can't see any functional change with this patch.

Steve