[PATCH] staging: rtl8188eu: replace two ODM_RT_TRACE calls in hal/phy.c

[PATCH] staging: rtl8188eu: replace two ODM_RT_TRACE calls in hal/phy.c

[Phillip Potter]

Within rtl88eu_dm_txpower_track_adjust function, retrieve the struct
net_device pointer, and replace both calls to the ODM_RT_TRACE macro
with equivalent netdev_dbg calls, as well as modifying layout, wording
and spacing slightly. The purpose of this, and patches like it, is to
eventually remove the need for include/odm_debug.h, which is an overly
complex way of printing debug/tracing information about the driver.

Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
---
 drivers/staging/rtl8188eu/hal/phy.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/staging/rtl8188eu/hal/phy.c b/drivers/staging/rtl8188eu/hal/phy.c
index a664bff370bb..a8caf618f588 100644
--- a/drivers/staging/rtl8188eu/hal/phy.c
+++ b/drivers/staging/rtl8188eu/hal/phy.c
@@ -301,11 +301,12 @@ void rtl88eu_dm_txpower_track_adjust(struct odm_dm_struct *dm_odm, u8 type,
 				     u8 *direction, u32 *out_write_val)
 {
 	u8 pwr_value = 0;
+	struct net_device *pnetdev = dm_odm->Adapter->pnetdev;
+
 	/*  Tx power tracking BB swing table. */
 	if (type == 0) { /* For OFDM adjust */
-		ODM_RT_TRACE(dm_odm, ODM_COMP_TX_PWR_TRACK, ODM_DBG_LOUD,
-			     ("BbSwingIdxOfdm = %d BbSwingFlagOfdm=%d\n",
-			     dm_odm->BbSwingIdxOfdm, dm_odm->BbSwingFlagOfdm));
+		netdev_dbg(pnetdev, "%s(): BbSwingIdxOfdm = %d BbSwingFlagOfdm = %d\n",
+			   __func__, dm_odm->BbSwingIdxOfdm, dm_odm->BbSwingFlagOfdm);
 
 		if (dm_odm->BbSwingIdxOfdm <= dm_odm->BbSwingIdxOfdmBase) {
 			*direction = 1;
@@ -318,9 +319,8 @@ void rtl88eu_dm_txpower_track_adjust(struct odm_dm_struct *dm_odm, u8 type,
 		}
 
 	} else if (type == 1) { /* For CCK adjust. */
-		ODM_RT_TRACE(dm_odm, ODM_COMP_TX_PWR_TRACK, ODM_DBG_LOUD,
-			     ("dm_odm->BbSwingIdxCck = %d dm_odm->BbSwingIdxCckBase = %d\n",
-			     dm_odm->BbSwingIdxCck, dm_odm->BbSwingIdxCckBase));
+		netdev_dbg(pnetdev, "%s(): BbSwingIdxCck = %d BbSwingIdxCckBase = %d\n",
+			   __func__, dm_odm->BbSwingIdxCck, dm_odm->BbSwingIdxCckBase);
 
 		if (dm_odm->BbSwingIdxCck <= dm_odm->BbSwingIdxCckBase) {
 			*direction = 1;
-- 
2.30.2

Re: [PATCH] staging: rtl8188eu: replace two ODM_RT_TRACE calls in hal/phy.c

[Dan Carpenter]

On Wed, Jun 09, 2021 at 12:56:09AM +0100, Phillip Potter wrote:
> Within rtl88eu_dm_txpower_track_adjust function, retrieve the struct
> net_device pointer, and replace both calls to the ODM_RT_TRACE macro
> with equivalent netdev_dbg calls, as well as modifying layout, wording
> and spacing slightly. The purpose of this, and patches like it, is to
> eventually remove the need for include/odm_debug.h, which is an overly
> complex way of printing debug/tracing information about the driver.
> 

In the original code DebugComponents is always zero so the ODM_RT_TRACE()
stuff was dead code and could never be printed.  I would prefer we just
delete it all instead of fixing it.

regards,
dan carpenter

Re: [PATCH] staging: rtl8188eu: replace two ODM_RT_TRACE calls in hal/phy.c

[Greg KH]

On Wed, Jun 09, 2021 at 12:56:09AM +0100, Phillip Potter wrote:
> Within rtl88eu_dm_txpower_track_adjust function, retrieve the struct
> net_device pointer, and replace both calls to the ODM_RT_TRACE macro
> with equivalent netdev_dbg calls, as well as modifying layout, wording
> and spacing slightly. The purpose of this, and patches like it, is to
> eventually remove the need for include/odm_debug.h, which is an overly
> complex way of printing debug/tracing information about the driver.
> 
> Signed-off-by: Phillip Potter <phil@philpotter.co.uk>
> ---
>  drivers/staging/rtl8188eu/hal/phy.c | 12 ++++++------
>  1 file changed, 6 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/staging/rtl8188eu/hal/phy.c b/drivers/staging/rtl8188eu/hal/phy.c
> index a664bff370bb..a8caf618f588 100644
> --- a/drivers/staging/rtl8188eu/hal/phy.c
> +++ b/drivers/staging/rtl8188eu/hal/phy.c
> @@ -301,11 +301,12 @@ void rtl88eu_dm_txpower_track_adjust(struct odm_dm_struct *dm_odm, u8 type,
>  				     u8 *direction, u32 *out_write_val)
>  {
>  	u8 pwr_value = 0;
> +	struct net_device *pnetdev = dm_odm->Adapter->pnetdev;
> +
>  	/*  Tx power tracking BB swing table. */
>  	if (type == 0) { /* For OFDM adjust */
> -		ODM_RT_TRACE(dm_odm, ODM_COMP_TX_PWR_TRACK, ODM_DBG_LOUD,
> -			     ("BbSwingIdxOfdm = %d BbSwingFlagOfdm=%d\n",
> -			     dm_odm->BbSwingIdxOfdm, dm_odm->BbSwingFlagOfdm));
> +		netdev_dbg(pnetdev, "%s(): BbSwingIdxOfdm = %d BbSwingFlagOfdm = %d\n",
> +			   __func__, dm_odm->BbSwingIdxOfdm, dm_odm->BbSwingFlagOfdm);

Along with what Dan said, please note that the dev_dbg() and
netdev_dbg() calls allow the user to see the function where the message
came from automatically, so there's never a need to add a __func__
string in the output.

thanks,

greg k-h

Re: [PATCH] staging: rtl8188eu: replace two ODM_RT_TRACE calls in hal/phy.c

[Phillip Potter]

On Wed, Jun 09, 2021 at 12:21:06PM +0300, Dan Carpenter wrote:
> On Wed, Jun 09, 2021 at 12:56:09AM +0100, Phillip Potter wrote:
> > Within rtl88eu_dm_txpower_track_adjust function, retrieve the struct
> > net_device pointer, and replace both calls to the ODM_RT_TRACE macro
> > with equivalent netdev_dbg calls, as well as modifying layout, wording
> > and spacing slightly. The purpose of this, and patches like it, is to
> > eventually remove the need for include/odm_debug.h, which is an overly
> > complex way of printing debug/tracing information about the driver.
> > 
> 
> In the original code DebugComponents is always zero so the ODM_RT_TRACE()
> stuff was dead code and could never be printed.  I would prefer we just
> delete it all instead of fixing it.
> 
> regards,
> dan carpenter
> 

Dear Dan (and Greg),

Thank you for your feedback. I will just remove all the ODM_RT_TRACE
calls in a follow up patch set in that case.

Regards,
Phil