[PATCH v2 6/8] Add debug log info to msm_csid_subdev_init function

George Chan via B4 Relay posted 8 patches 1 year, 5 months ago
There is a newer version of this series
[PATCH v2 6/8] Add debug log info to msm_csid_subdev_init function
Posted by George Chan via B4 Relay 1 year, 5 months ago
From: George Chan <gchan9527@gmail.com>

Print out missing clock's name when doing msm_csid_subdev_init().

Signed-off-by: George Chan <gchan9527@gmail.com>
---
 drivers/media/platform/qcom/camss/camss-csid.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/media/platform/qcom/camss/camss-csid.c b/drivers/media/platform/qcom/camss/camss-csid.c
index eb27d69e89a1..02162bf9dbc0 100644
--- a/drivers/media/platform/qcom/camss/camss-csid.c
+++ b/drivers/media/platform/qcom/camss/camss-csid.c
@@ -628,8 +628,10 @@ int msm_csid_subdev_init(struct camss *camss, struct csid_device *csid,
 		struct camss_clock *clock = &csid->clock[i];
 
 		clock->clk = devm_clk_get(dev, res->clock[i]);
-		if (IS_ERR(clock->clk))
+		if (IS_ERR(clock->clk)) {
+			dev_err(dev, "missing clk %s", res->clock[i]);
 			return PTR_ERR(clock->clk);
+		}
 
 		clock->name = res->clock[i];
 

-- 
2.34.1
Re: [PATCH v2 6/8] Add debug log info to msm_csid_subdev_init function
Posted by Bryan O'Donoghue 1 year, 5 months ago
On 24/06/2024 00:22, George Chan via B4 Relay wrote:
> +			dev_err(dev, "missing clk %s", res->clock[i]);

Please don't abbreviate - the other error message says "clock" here we 
say "clk"

Once fixed to clock please add

Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Re: [PATCH v2 6/8] Add debug log info to msm_csid_subdev_init function
Posted by Krzysztof Kozlowski 1 year, 5 months ago
On 24/06/2024 12:08, Bryan O'Donoghue wrote:
> On 24/06/2024 00:22, George Chan via B4 Relay wrote:
>> +			dev_err(dev, "missing clk %s", res->clock[i]);
> 
> Please don't abbreviate - the other error message says "clock" here we 
> say "clk"
> 

Is it a probe path? If yes, then this should be dev_err_probe. If not,
then are you sure that non-probe paths should acquire resources? How
does it handle deferred probe in such case?

Best regards,
Krzysztof