Hello Tero,
a few comments.
On Tue, 16 Sep 2008, Tero Kristo wrote:
> Hardware supervised control for clockdomain power state transitions now
> enabled in omap3_pm_init(). Also fixed set_pwrdm_state() to allow state
> changes between sleep states (i.e. RET<->OFF.)
>
> Signed-off-by: Tero Kristo <tero.kristo@xxxxxxxxx>
> ---
> arch/arm/mach-omap2/pm34xx.c | 38 +++++++++++++++++++-------------------
> 1 files changed, 19 insertions(+), 19 deletions(-)
>
> diff --git a/arch/arm/mach-omap2/pm34xx.c b/arch/arm/mach-omap2/pm34xx.c
> index 01b637a..2bccdab 100644
> --- a/arch/arm/mach-omap2/pm34xx.c
> +++ b/arch/arm/mach-omap2/pm34xx.c
> @@ -227,28 +227,13 @@ static int omap3_can_sleep(void)
> return 1;
> }
>
> -/* _clkdm_deny_idle - private callback function used by set_pwrdm_state() */
> -static int _clkdm_deny_idle(struct powerdomain *pwrdm,
> - struct clockdomain *clkdm)
> -{
> - omap2_clkdm_deny_idle(clkdm);
> - return 0;
> -}
> -
> -/* _clkdm_allow_idle - private callback function used by set_pwrdm_state() */
> -static int _clkdm_allow_idle(struct powerdomain *pwrdm,
> - struct clockdomain *clkdm)
> -{
> - omap2_clkdm_allow_idle(clkdm);
> - return 0;
> -}
> -
> /* This sets pwrdm state (other than mpu & core. Currently only ON &
> * RET are supported. Function is assuming that clkdm doesn't have
> * hw_sup mode enabled. */
> static int set_pwrdm_state(struct powerdomain *pwrdm, u32 state)
> {
> u32 cur_state;
> + int sleep_switch = 0;
> int ret = 0;
>
> if (pwrdm == NULL || IS_ERR(pwrdm))
> @@ -259,8 +244,12 @@ static int set_pwrdm_state(struct powerdomain *pwrdm, u32 state)
> if (cur_state == state)
> return ret;
>
> - pwrdm_for_each_clkdm(pwrdm, _clkdm_deny_idle);
> -
> + if (pwrdm_read_pwrst(pwrdm) < PWRDM_POWER_ON) {
> + omap2_clkdm_wakeup(pwrdm->pwrdm_clkdms[0]);
In general, shouldn't this happen for each clkdm in the pwrdm? i.e.,
you should probably create some _clkdm_wakeup() (similar to
_clkdm_allow_idle() ) and call that via pwrdm_for_each_clkdm().
This may not have any functional effect at present with OMAP3, but
avoids the assumption that each powerdomain will have only one
software-controllable clockdomain (and that it will always appear at the 0
element in the array).
> + sleep_switch = 1;
> + pwrdm_wait_transition(pwrdm);
> + }
> +
> ret = pwrdm_set_next_pwrst(pwrdm, state);
> if (ret) {
> printk(KERN_ERR "Unable to set state of powerdomain: %s\n",
> @@ -268,7 +257,10 @@ static int set_pwrdm_state(struct powerdomain *pwrdm, u32 state)
> goto err;
> }
>
> - pwrdm_for_each_clkdm(pwrdm, _clkdm_allow_idle);
> + if (sleep_switch) {
> + omap2_clkdm_allow_idle(pwrdm->pwrdm_clkdms[0]);
Same comment as previous - shouldn't _clkdm_allow_idle() be kept, and
called for each clkdm in the pwrdm, instead?
> + pwrdm_wait_transition(pwrdm);
> + }
>
> err:
> return ret;
> @@ -540,6 +532,12 @@ static int __init pwrdms_setup(struct powerdomain *pwrdm)
> return set_pwrdm_state(pwrst->pwrdm, pwrst->next_state);
> }
>
> +static int __init clkdms_setup(struct clockdomain *clkdm)
> +{
> + omap2_clkdm_allow_idle(clkdm);
> + return 0;
> +}
> +
> int __init omap3_pm_init(void)
> {
> struct power_state *pwrst;
> @@ -566,6 +564,8 @@ int __init omap3_pm_init(void)
> goto err2;
> }
>
> + (void) clkdm_for_each(clkdms_setup);
> +
> mpu_pwrdm = pwrdm_lookup("mpu_pwrdm");
> if (mpu_pwrdm == NULL) {
> printk(KERN_ERR "Failed to get mpu_pwrdm\n");
> --
> 1.5.4.3
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
- Paul
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
