[PATCH 0/2] scsi: ufs: convert probe to use dev_err_probe()

[PATCH 0/2] scsi: ufs: convert probe to use dev_err_probe()

[Brian Masney]

The following two log messages are shown on bootup due to an
-EPROBE_DEFER when booting on a Qualcomm sa8775p development board:

    ufshcd-qcom 1d84000.ufs: ufshcd_variant_hba_init: variant qcom init
        failed err -517
    ufshcd-qcom 1d84000.ufs: Initialization failed

This patch series converts the relevant two probe functions over to use
dev_err_probe() so that these messages are not shown on bootup.

Brian Masney (2):
  scsi: ufs: core: convert to dev_err_probe() in hba_init
  scsi: ufs: host: convert to dev_err_probe() in pltfrm_init

 drivers/ufs/core/ufshcd.c        | 17 +++++++++--------
 drivers/ufs/host/ufshcd-pltfrm.c |  2 +-
 2 files changed, 10 insertions(+), 9 deletions(-)

-- 
2.41.0

[PATCH 1/2] scsi: ufs: core: convert to dev_err_probe() in hba_init

[Brian Masney]

Convert ufshcd_variant_hba_init() over to use dev_err_probe() to avoid
log messages like the following on bootup:

    ufshcd-qcom 1d84000.ufs: ufshcd_variant_hba_init: variant qcom init
        failed err -517

While changes are being made here, let's go ahead and clean up the rest
of that function.

Signed-off-by: Brian Masney <bmasney@redhat.com>
---
 drivers/ufs/core/ufshcd.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index 129446775796..90d87cf5e25e 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -9228,17 +9228,18 @@ static int ufshcd_init_clocks(struct ufs_hba *hba)
 
 static int ufshcd_variant_hba_init(struct ufs_hba *hba)
 {
-	int err = 0;
+	int ret;
 
 	if (!hba->vops)
-		goto out;
+		return 0;
 
-	err = ufshcd_vops_init(hba);
-	if (err)
-		dev_err(hba->dev, "%s: variant %s init failed err %d\n",
-			__func__, ufshcd_get_var_name(hba), err);
-out:
-	return err;
+	ret = ufshcd_vops_init(hba);
+	if (ret)
+		dev_err_probe(hba->dev, ret,
+			      "%s: variant %s init failed with error %d\n",
+			      __func__, ufshcd_get_var_name(hba), ret);
+
+	return ret;
 }
 
 static void ufshcd_variant_hba_exit(struct ufs_hba *hba)
-- 
2.41.0

[PATCH 2/2] scsi: ufs: host: convert to dev_err_probe() in pltfrm_init

[Brian Masney]

Convert ufshcd_pltfrm_init() over to use dev_err_probe() to avoid
the following log message on bootup due to an -EPROBE_DEFER return
code:

    ufshcd-qcom 1d84000.ufs: Initialization failed

Signed-off-by: Brian Masney <bmasney@redhat.com>
---
 drivers/ufs/host/ufshcd-pltfrm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ufs/host/ufshcd-pltfrm.c b/drivers/ufs/host/ufshcd-pltfrm.c
index 0b7430033047..f2c50b78efbf 100644
--- a/drivers/ufs/host/ufshcd-pltfrm.c
+++ b/drivers/ufs/host/ufshcd-pltfrm.c
@@ -373,7 +373,7 @@ int ufshcd_pltfrm_init(struct platform_device *pdev,
 
 	err = ufshcd_init(hba, mmio_base, irq);
 	if (err) {
-		dev_err(dev, "Initialization failed\n");
+		dev_err_probe(dev, err, "Initialization failed\n");
 		goto dealloc_host;
 	}
 
-- 
2.41.0

Re: [PATCH 1/2] scsi: ufs: core: convert to dev_err_probe() in hba_init

[Hugo Villeneuve]

On Tue,  8 Aug 2023 10:26:49 -0400
Brian Masney <bmasney@redhat.com> wrote:

> Convert ufshcd_variant_hba_init() over to use dev_err_probe() to avoid
> log messages like the following on bootup:
> 
>     ufshcd-qcom 1d84000.ufs: ufshcd_variant_hba_init: variant qcom init
>         failed err -517
> 
> While changes are being made here, let's go ahead and clean up the rest
> of that function.

Hi,
you should not combine code cleanup and fixes/improvements in the same
patch, split them.

Hugo Villeneuve


> Signed-off-by: Brian Masney <bmasney@redhat.com>
> ---
>  drivers/ufs/core/ufshcd.c | 17 +++++++++--------
>  1 file changed, 9 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> index 129446775796..90d87cf5e25e 100644
> --- a/drivers/ufs/core/ufshcd.c
> +++ b/drivers/ufs/core/ufshcd.c
> @@ -9228,17 +9228,18 @@ static int ufshcd_init_clocks(struct ufs_hba *hba)
>  
>  static int ufshcd_variant_hba_init(struct ufs_hba *hba)
>  {
> -	int err = 0;
> +	int ret;
>  
>  	if (!hba->vops)
> -		goto out;
> +		return 0;
>  
> -	err = ufshcd_vops_init(hba);
> -	if (err)
> -		dev_err(hba->dev, "%s: variant %s init failed err %d\n",
> -			__func__, ufshcd_get_var_name(hba), err);
> -out:
> -	return err;
> +	ret = ufshcd_vops_init(hba);
> +	if (ret)
> +		dev_err_probe(hba->dev, ret,
> +			      "%s: variant %s init failed with error %d\n",
> +			      __func__, ufshcd_get_var_name(hba), ret);
> +
> +	return ret;
>  }
>  
>  static void ufshcd_variant_hba_exit(struct ufs_hba *hba)
> -- 
> 2.41.0
> 

Re: [PATCH 1/2] scsi: ufs: core: convert to dev_err_probe() in hba_init

[Brian Masney]

On Tue, Aug 08, 2023 at 04:29:29PM -0400, Hugo Villeneuve wrote:
> On Tue,  8 Aug 2023 10:26:49 -0400
> Brian Masney <bmasney@redhat.com> wrote:
> 
> > Convert ufshcd_variant_hba_init() over to use dev_err_probe() to avoid
> > log messages like the following on bootup:
> > 
> >     ufshcd-qcom 1d84000.ufs: ufshcd_variant_hba_init: variant qcom init
> >         failed err -517
> > 
> > While changes are being made here, let's go ahead and clean up the rest
> > of that function.
> 
> Hi,
> you should not combine code cleanup and fixes/improvements in the same
> patch, split them.

This is a pretty simple patch as is, and split up the code clean up is
not very useful on its own. I'll just skip doing the code cleanup and
only post the dev_err_probe() change in v2.

Brian