On Thu, Mar 14, 2013 at 12:54:08PM +0000, Philip, Avinash wrote: > On Thu, Mar 14, 2013 at 17:09:04, Nori, Sekhar wrote: > > On 3/14/2013 4:02 PM, Philip Avinash wrote: > > > Add EHRPWM and ECAP support build support for DAVINCI_DA850 platforms. > > > > > > Also, since DAVINCI platforms doesn't support TI-PWM-Subsystem module, > > > remove the select option for CONFIG_PWM_TIPWMSS. > > > > > > Also, update CONFIG_PWM_TIPWMSS compiler directive appropriately in > > > pwm-tipwmss.h to fix the below compiler error upon removal of > > > CONFIG_PWM_TIPWMSS for Davinci platforms. > > > > > > drivers/pwm/pwm-tiecap.c: In function 'ecap_pwm_probe': > > > drivers/pwm/pwm-tiecap.c:263:4: error: 'PWMSS_ECAPCLK_EN' undeclared > > > (first use in this function) > > > drivers/pwm/pwm-tiecap.c:263:4: note: each undeclared identifier > > > is reported only once for each function it appears in > > > drivers/pwm/pwm-tiecap.c:264:17: error: 'PWMSS_ECAPCLK_EN_ACK' > > > undeclared (first use in this function) > > > drivers/pwm/pwm-tiecap.c: In function 'ecap_pwm_remove': > > > drivers/pwm/pwm-tiecap.c:291:49: error: 'PWMSS_ECAPCLK_STOP_REQ' > > > undeclared (first use in this function) > > > make[2]: *** [drivers/pwm/pwm-tiecap.o] Error 1 > > > make[1]: *** [drivers/pwm] Error 2 > > > make: *** [drivers] Error 2 > > > > > > Signed-off-by: Philip Avinash <avinashphilip@xxxxxx> > > > > > config PWM_TIECAP > > > tristate "ECAP PWM support" > > > - depends on SOC_AM33XX > > > - select PWM_TIPWMSS > > > + depends on SOC_AM33XX || ARCH_DAVINCI_DA850 > > > > Having such narrow dependencies is wrong. The same device is present on > > DaVinci DA830 too. A depends on should not be required at all since the > > driver should build on all architectures. But I have seen resistance to > > doing that since users don't like to see configuration options totally > > irrelevant for the architecture they are building for. So may be take a > > middle path and do 'depends on ARCH_ARM'? > > I will add dependency only on ARCH_ARM and submit another version. Perhaps SOC_AM33XX || ARCH_DAVINCI_DA8XX would be an alternative? Thierry
Attachment:
pgpBB3ruooEHA.pgp
Description: PGP signature
