| Message ID | 1450770677-13557-1-git-send-email-henryc.chen@mediatek.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 22/12/15 08:51, Henry Chen wrote: > Sometimes PMIC is too busy to send data in time to cause pmic wrap timeout, > because pmic wrap is waiting for FSM_VLDCLR after finishing WACS2_CMD. It > just return error when issue happened, so the state machine will stay on > FSM_VLDCLR state when data send back later by PMIC and timeout again in next > time because pmic wrap waiting for FSM_IDLE state at the begining of the > read/write function. > > Clear the vldclr when timeout if state machine is stay on FSM_VLDCLR. > > Signed-off-by: Henry Chen <henryc.chen@mediatek.com> > --- > drivers/soc/mediatek/mtk-pmic-wrap.c | 16 ++++++++++++++-- > 1 file changed, 14 insertions(+), 2 deletions(-) > > diff --git a/drivers/soc/mediatek/mtk-pmic-wrap.c b/drivers/soc/mediatek/mtk-pmic-wrap.c > index 105597a..ccd5337 100644 > --- a/drivers/soc/mediatek/mtk-pmic-wrap.c > +++ b/drivers/soc/mediatek/mtk-pmic-wrap.c > @@ -443,10 +443,16 @@ static int pwrap_wait_for_state(struct pmic_wrapper *wrp, > static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) > { > int ret; > + u32 val; > > ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); > - if (ret) > + if (ret) { > + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ > + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); > + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) > + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); > return ret; > + } I would prefer to have this encapsulated in a (inline) function. Maybe with better description then just the one line comment. Thanks, Matthias > > pwrap_writel(wrp, (1 << 31) | ((adr >> 1) << 16) | wdata, > PWRAP_WACS2_CMD); > @@ -457,10 +463,16 @@ static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) > static int pwrap_read(struct pmic_wrapper *wrp, u32 adr, u32 *rdata) > { > int ret; > + u32 val; > > ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); > - if (ret) > + if (ret) { > + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ > + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); > + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) > + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); > return ret; > + } > > pwrap_writel(wrp, (adr >> 1) << 16, PWRAP_WACS2_CMD); > >
Hi Matthias, On Wed, 2015-12-30 at 19:02 +0100, Matthias Brugger wrote: > > On 22/12/15 08:51, Henry Chen wrote: > > Sometimes PMIC is too busy to send data in time to cause pmic wrap timeout, > > because pmic wrap is waiting for FSM_VLDCLR after finishing WACS2_CMD. It > > just return error when issue happened, so the state machine will stay on > > FSM_VLDCLR state when data send back later by PMIC and timeout again in next > > time because pmic wrap waiting for FSM_IDLE state at the begining of the > > read/write function. > > > > Clear the vldclr when timeout if state machine is stay on FSM_VLDCLR. > > > > Signed-off-by: Henry Chen <henryc.chen@mediatek.com> > > --- > > drivers/soc/mediatek/mtk-pmic-wrap.c | 16 ++++++++++++++-- > > 1 file changed, 14 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/soc/mediatek/mtk-pmic-wrap.c b/drivers/soc/mediatek/mtk-pmic-wrap.c > > index 105597a..ccd5337 100644 > > --- a/drivers/soc/mediatek/mtk-pmic-wrap.c > > +++ b/drivers/soc/mediatek/mtk-pmic-wrap.c > > @@ -443,10 +443,16 @@ static int pwrap_wait_for_state(struct pmic_wrapper *wrp, > > static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) > > { > > int ret; > > + u32 val; > > > > ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); > > - if (ret) > > + if (ret) { > > + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ > > + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); > > + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) > > + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); > > return ret; > > + } > > I would prefer to have this encapsulated in a (inline) function. Maybe > with better description then just the one line comment. > > Thanks, > Matthias Ok, I will make the description more clear, do you means write the function like below and used it on pwrap_write/pwrap_read. static inline void pwrap_leave_fsm_vldclr(struct pmic_wrapper *wrp) { if (pwrap_is_fsm_vldclr(wrp)) pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); } Thanks, Henry > > > > > pwrap_writel(wrp, (1 << 31) | ((adr >> 1) << 16) | wdata, > > PWRAP_WACS2_CMD); > > @@ -457,10 +463,16 @@ static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) > > static int pwrap_read(struct pmic_wrapper *wrp, u32 adr, u32 *rdata) > > { > > int ret; > > + u32 val; > > > > ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); > > - if (ret) > > + if (ret) { > > + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ > > + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); > > + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) > > + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); > > return ret; > > + } > > > > pwrap_writel(wrp, (adr >> 1) << 16, PWRAP_WACS2_CMD); > > > >
On 31/12/15 03:09, Henry Chen wrote: > Hi Matthias, > > On Wed, 2015-12-30 at 19:02 +0100, Matthias Brugger wrote: >> >> On 22/12/15 08:51, Henry Chen wrote: >>> Sometimes PMIC is too busy to send data in time to cause pmic wrap timeout, >>> because pmic wrap is waiting for FSM_VLDCLR after finishing WACS2_CMD. It >>> just return error when issue happened, so the state machine will stay on >>> FSM_VLDCLR state when data send back later by PMIC and timeout again in next >>> time because pmic wrap waiting for FSM_IDLE state at the begining of the >>> read/write function. >>> >>> Clear the vldclr when timeout if state machine is stay on FSM_VLDCLR. >>> >>> Signed-off-by: Henry Chen <henryc.chen@mediatek.com> >>> --- >>> drivers/soc/mediatek/mtk-pmic-wrap.c | 16 ++++++++++++++-- >>> 1 file changed, 14 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/soc/mediatek/mtk-pmic-wrap.c b/drivers/soc/mediatek/mtk-pmic-wrap.c >>> index 105597a..ccd5337 100644 >>> --- a/drivers/soc/mediatek/mtk-pmic-wrap.c >>> +++ b/drivers/soc/mediatek/mtk-pmic-wrap.c >>> @@ -443,10 +443,16 @@ static int pwrap_wait_for_state(struct pmic_wrapper *wrp, >>> static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) >>> { >>> int ret; >>> + u32 val; >>> >>> ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); >>> - if (ret) >>> + if (ret) { >>> + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ >>> + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); >>> + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) >>> + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); >>> return ret; >>> + } >> >> I would prefer to have this encapsulated in a (inline) function. Maybe >> with better description then just the one line comment. >> >> Thanks, >> Matthias > > Ok, I will make the description more clear, do you means write the > function like below and used it on pwrap_write/pwrap_read. > > static inline void pwrap_leave_fsm_vldclr(struct pmic_wrapper *wrp) > { > if (pwrap_is_fsm_vldclr(wrp)) > pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); > } > I didn't remind of pwrap_is_fsm_vldclr neither, nice. Yes please use this function and add a descriptive comment on the top, so that in the future it is clear what is happening here. Thanks, Matthias > > Thanks, > Henry > >> >>> >>> pwrap_writel(wrp, (1 << 31) | ((adr >> 1) << 16) | wdata, >>> PWRAP_WACS2_CMD); >>> @@ -457,10 +463,16 @@ static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) >>> static int pwrap_read(struct pmic_wrapper *wrp, u32 adr, u32 *rdata) >>> { >>> int ret; >>> + u32 val; >>> >>> ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); >>> - if (ret) >>> + if (ret) { >>> + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ >>> + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); >>> + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) >>> + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); >>> return ret; >>> + } >>> >>> pwrap_writel(wrp, (adr >> 1) << 16, PWRAP_WACS2_CMD); >>> >>> > >
diff --git a/drivers/soc/mediatek/mtk-pmic-wrap.c b/drivers/soc/mediatek/mtk-pmic-wrap.c index 105597a..ccd5337 100644 --- a/drivers/soc/mediatek/mtk-pmic-wrap.c +++ b/drivers/soc/mediatek/mtk-pmic-wrap.c @@ -443,10 +443,16 @@ static int pwrap_wait_for_state(struct pmic_wrapper *wrp, static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) { int ret; + u32 val; ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); - if (ret) + if (ret) { + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); return ret; + } pwrap_writel(wrp, (1 << 31) | ((adr >> 1) << 16) | wdata, PWRAP_WACS2_CMD); @@ -457,10 +463,16 @@ static int pwrap_write(struct pmic_wrapper *wrp, u32 adr, u32 wdata) static int pwrap_read(struct pmic_wrapper *wrp, u32 adr, u32 *rdata) { int ret; + u32 val; ret = pwrap_wait_for_state(wrp, pwrap_is_fsm_idle); - if (ret) + if (ret) { + /* Clear vldclr bit if state is on the WACS_FSM_WFVLDCLR */ + val = pwrap_readl(wrp, PWRAP_WACS2_RDATA); + if (PWRAP_GET_WACS_FSM(val) == PWRAP_WACS_FSM_WFVLDCLR) + pwrap_writel(wrp, 1, PWRAP_WACS2_VLDCLR); return ret; + } pwrap_writel(wrp, (adr >> 1) << 16, PWRAP_WACS2_CMD);
Sometimes PMIC is too busy to send data in time to cause pmic wrap timeout, because pmic wrap is waiting for FSM_VLDCLR after finishing WACS2_CMD. It just return error when issue happened, so the state machine will stay on FSM_VLDCLR state when data send back later by PMIC and timeout again in next time because pmic wrap waiting for FSM_IDLE state at the begining of the read/write function. Clear the vldclr when timeout if state machine is stay on FSM_VLDCLR. Signed-off-by: Henry Chen <henryc.chen@mediatek.com> --- drivers/soc/mediatek/mtk-pmic-wrap.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-)