[PATCH resend 3] staging: rtl8188eu: Add rtw_led_enable module parameter

Greg Kroah-Hartman gregkh at linuxfoundation.org
Tue Mar 24 12:48:45 UTC 2020


On Tue, Mar 24, 2020 at 01:38:53PM +0100, Bastien Nocera wrote:
> On Tue, 2020-03-24 at 13:32 +0100, Greg Kroah-Hartman wrote:
> > On Tue, Mar 24, 2020 at 12:47:01PM +0100, Bastien Nocera wrote:
> > > On Tue, 2020-03-24 at 12:38 +0100, Greg Kroah-Hartman wrote:
> > > > On Tue, Mar 24, 2020 at 11:36:00AM +0100, Bastien Nocera wrote:
> > > > > Make it possible to disable the LED, as it can be pretty
> > > > > annoying
> > > > > depending on where it's located.
> > > > > 
> > > > > See also https://github.com/lwfinger/rtl8188eu/pull/304 for the
> > > > > out-of-tree version.
> > > > > 
> > > > > Signed-off-by: Bastien Nocera <hadess at hadess.net>
> > > > > ---
> > > > >  drivers/staging/rtl8188eu/core/rtw_led.c      | 6 ++++++
> > > > >  drivers/staging/rtl8188eu/include/drv_types.h | 2 ++
> > > > >  drivers/staging/rtl8188eu/os_dep/os_intfs.c   | 5 +++++
> > > > >  3 files changed, 13 insertions(+)
> > > > > 
> > > > > diff --git a/drivers/staging/rtl8188eu/core/rtw_led.c
> > > > > b/drivers/staging/rtl8188eu/core/rtw_led.c
> > > > > index d1406cc99768..75a859accb7e 100644
> > > > > --- a/drivers/staging/rtl8188eu/core/rtw_led.c
> > > > > +++ b/drivers/staging/rtl8188eu/core/rtw_led.c
> > > > > @@ -467,10 +467,16 @@ void blink_handler(struct LED_871x *pLed)
> > > > >  
> > > > >  void led_control_8188eu(struct adapter *padapter, enum
> > > > > LED_CTL_MODE LedAction)
> > > > >  {
> > > > > +	struct registry_priv *registry_par;
> > > > > +
> > > > >  	if (padapter->bSurpriseRemoved || padapter-
> > > > > >bDriverStopped ||
> > > > >  	    !padapter->hw_init_completed)
> > > > >  		return;
> > > > >  
> > > > > +	registry_par = &padapter->registrypriv;
> > > > > +	if (!registry_par->led_enable)
> > > > > +		return;
> > > > > +
> > > > >  	if ((padapter->pwrctrlpriv.rf_pwrstate != rf_on &&
> > > > >  	     padapter->pwrctrlpriv.rfoff_reason >
> > > > > RF_CHANGE_BY_PS) &&
> > > > >  	    (LedAction == LED_CTL_TX || LedAction == LED_CTL_RX
> > > > > ||
> > > > > diff --git a/drivers/staging/rtl8188eu/include/drv_types.h
> > > > > b/drivers/staging/rtl8188eu/include/drv_types.h
> > > > > index 35c0946bc65d..4ca828141d3f 100644
> > > > > --- a/drivers/staging/rtl8188eu/include/drv_types.h
> > > > > +++ b/drivers/staging/rtl8188eu/include/drv_types.h
> > > > > @@ -67,6 +67,8 @@ struct registry_priv {
> > > > >  	u8	wmm_enable;
> > > > >  	u8	uapsd_enable;
> > > > >  
> > > > > +	u8	led_enable;
> > > > > +
> > > > >  	struct wlan_bssid_ex    dev_network;
> > > > >  
> > > > >  	u8	ht_enable;
> > > > > diff --git a/drivers/staging/rtl8188eu/os_dep/os_intfs.c
> > > > > b/drivers/staging/rtl8188eu/os_dep/os_intfs.c
> > > > > index 8907bf6bb7ff..ba55ae741215 100644
> > > > > --- a/drivers/staging/rtl8188eu/os_dep/os_intfs.c
> > > > > +++ b/drivers/staging/rtl8188eu/os_dep/os_intfs.c
> > > > > @@ -47,6 +47,8 @@ static int rtw_acm_method;/*  0:By SW 1:By
> > > > > HW. */
> > > > >  static int rtw_wmm_enable = 1;/*  default is set to enable the
> > > > > wmm. */
> > > > >  static int rtw_uapsd_enable;
> > > > >  
> > > > > +static int rtw_led_enable = 1;
> > > > > +
> > > > >  static int rtw_ht_enable = 1;
> > > > >  /* 0 :disable, bit(0): enable 2.4g, bit(1): enable 5g */
> > > > >  static int rtw_cbw40_enable = 3;
> > > > > @@ -98,6 +100,7 @@ module_param(rtw_channel, int, 0644);
> > > > >  module_param(rtw_wmm_enable, int, 0644);
> > > > >  module_param(rtw_vrtl_carrier_sense, int, 0644);
> > > > >  module_param(rtw_vcs_type, int, 0644);
> > > > > +module_param(rtw_led_enable, int, 0644);
> > > > 
> > > > Ick, really?  No, no nee module parameters, this is not the
> > > > 1990's.
> > > > 
> > > > This should be done on a per-device basis, using the correct
> > > > apis.
> > > 
> > > What API?
> > 
> > Documentation/leds/index.rst should give you a good start :)
> 
> Given how much work it'd be, I'll give it a miss and carry on using the
> out-of-tree driver.

Huh?  Why not fix this properly, as that's the only way this driver is
ever going to be fixed up correctly and get out of staging at all.

thanks,

greg k-h


More information about the devel mailing list