Message ID | 1585718145-29537-1-git-send-email-sanm@codeaurora.org |
---|---|
Headers | show |
Series | ADD interconnect support for Qualcomm DWC3 driver | expand |
Quoting Sandeep Maheswaram (2020-03-31 22:15:44) > Populate USB DT nodes with interconnect properties. > > Signed-off-by: Sandeep Maheswaram <sanm@codeaurora.org> > --- Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Quoting Sandeep Maheswaram (2020-03-31 22:15:45) > Populate USB DT nodes with interconnect properties. > > Signed-off-by: Sandeep Maheswaram <sanm@codeaurora.org> > --- Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Hi Felipe, all patches of this series have been reviewed and there are no outstanding comments, so I guess it should be ready to land? Thanks Matthias On Wed, Apr 01, 2020 at 10:45:41AM +0530, Sandeep Maheswaram wrote: > This path series aims to add interconnect support in > dwc3-qcom driver on SDM845 and SC7180 SoCs. > > Changes from v6 -> v7 > > [PATCH 2/4] Fixed review comments from Matthias in DWC3 driver. > > Other patches remain unchanged. > > Changes from v5 -> v6 > > [PATCH 1/4] Addressed comments from Rob. > > [PATCH 2/4] Fixed review comments from Matthias in DWC3 driver. > > [PATCH 3/4] Ignoring 80 char limit in defining interconnect paths. > > Added [PATCH 4/4] in this series. Adding interconnect nodes for SC7180. > Depends on patch https://patchwork.kernel.org/patch/11417989/. > > Changes from v4 -> v5 > > [PATCH 1/3] Added the interconnect properties in yaml. This patch depends > on series https://patchwork.kernel.org/cover/11372641/. > > [PATCH 2/3] Fixed review comments from Matthias in DWC3 driver. > > [PATCH 3/3] Modified as per the new interconnect nodes in sdm845. Depends > on series https://patchwork.kernel.org/cover/11372211/. > > > Changes from v3 -> v4 > > Fixed review comments from Matthias > > [PATCH 1/3] and [PATCH 3/3] remains unchanged > > Changes from v2 -> v3 > > Fixed review comments from Matthias and Manu > > changed the functions prefix from usb_* to dwc3_qcom_* > > Changes since V1: > > Comments by Georgi Djakov on "[PATCH 2/3]" addressed > > [PATCH 1/3] and [PATCH 3/3] remains unchanged > > > Sandeep Maheswaram (4): > dt-bindings: usb: qcom,dwc3: Introduce interconnect properties for > Qualcomm DWC3 driver > usb: dwc3: qcom: Add interconnect support in dwc3 driver > arm64: dts: qcom: sdm845: Add interconnect properties for USB > arm64: dts: qcom: sc7180: Add interconnect properties for USB > > .../devicetree/bindings/usb/qcom,dwc3.yaml | 8 ++ > arch/arm64/boot/dts/qcom/sc7180.dtsi | 4 + > arch/arm64/boot/dts/qcom/sdm845.dtsi | 8 ++ > drivers/usb/dwc3/dwc3-qcom.c | 128 ++++++++++++++++++++- > 4 files changed, 146 insertions(+), 2 deletions(-) > > -- > QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member > of Code Aurora Forum, hosted by The Linux Foundation >
Hi Felipe, Any update about landing this series. Regards Sandeep On 4/30/2020 12:05 AM, Matthias Kaehlcke wrote: > Hi Felipe, > > all patches of this series have been reviewed and there are no outstanding > comments, so I guess it should be ready to land? > > Thanks > > Matthias > > On Wed, Apr 01, 2020 at 10:45:41AM +0530, Sandeep Maheswaram wrote: >> This path series aims to add interconnect support in >> dwc3-qcom driver on SDM845 and SC7180 SoCs. >> >> Changes from v6 -> v7 >> > [PATCH 2/4] Fixed review comments from Matthias in DWC3 driver. >> > Other patches remain unchanged. >> >> Changes from v5 -> v6 >> > [PATCH 1/4] Addressed comments from Rob. >> > [PATCH 2/4] Fixed review comments from Matthias in DWC3 driver. >> > [PATCH 3/4] Ignoring 80 char limit in defining interconnect paths. >> > Added [PATCH 4/4] in this series. Adding interconnect nodes for SC7180. >> Depends on patch https://patchwork.kernel.org/patch/11417989/. >> >> Changes from v4 -> v5 >> > [PATCH 1/3] Added the interconnect properties in yaml. This patch depends >> on series https://patchwork.kernel.org/cover/11372641/. >> > [PATCH 2/3] Fixed review comments from Matthias in DWC3 driver. >> > [PATCH 3/3] Modified as per the new interconnect nodes in sdm845. Depends >> on series https://patchwork.kernel.org/cover/11372211/. >> >> >> Changes from v3 -> v4 >> > Fixed review comments from Matthias >> > [PATCH 1/3] and [PATCH 3/3] remains unchanged >> >> Changes from v2 -> v3 >> > Fixed review comments from Matthias and Manu >> > changed the functions prefix from usb_* to dwc3_qcom_* >> >> Changes since V1: >> > Comments by Georgi Djakov on "[PATCH 2/3]" addressed >> > [PATCH 1/3] and [PATCH 3/3] remains unchanged >> >> >> Sandeep Maheswaram (4): >> dt-bindings: usb: qcom,dwc3: Introduce interconnect properties for >> Qualcomm DWC3 driver >> usb: dwc3: qcom: Add interconnect support in dwc3 driver >> arm64: dts: qcom: sdm845: Add interconnect properties for USB >> arm64: dts: qcom: sc7180: Add interconnect properties for USB >> >> .../devicetree/bindings/usb/qcom,dwc3.yaml | 8 ++ >> arch/arm64/boot/dts/qcom/sc7180.dtsi | 4 + >> arch/arm64/boot/dts/qcom/sdm845.dtsi | 8 ++ >> drivers/usb/dwc3/dwc3-qcom.c | 128 ++++++++++++++++++++- >> 4 files changed, 146 insertions(+), 2 deletions(-) >> >> -- >> QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member >> of Code Aurora Forum, hosted by The Linux Foundation >>
"Sandeep Maheswaram (Temp)" <sanm@codeaurora.org> writes: > Hi Felipe, > > Any update about landing this series. in my tree now
Hi, Sandeep Maheswaram <sanm@codeaurora.org> writes: > +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > +{ > + struct device *dev = qcom->dev; > + int ret; > + > + if (!device_is_bound(&qcom->dwc3->dev)) > + return -EPROBE_DEFER; this breaks allmodconfig. I'm dropping this series from my queue for this merge window.
Felipe Balbi <balbi@kernel.org> writes: > Hi, > > Sandeep Maheswaram <sanm@codeaurora.org> writes: >> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >> +{ >> + struct device *dev = qcom->dev; >> + int ret; >> + >> + if (!device_is_bound(&qcom->dwc3->dev)) >> + return -EPROBE_DEFER; > > this breaks allmodconfig. I'm dropping this series from my queue for > this merge window. Sorry, I meant this patch ;-)
On Thu, May 14, 2020 at 02:30:28PM +0300, Felipe Balbi wrote: > Felipe Balbi <balbi@kernel.org> writes: > > > Hi, > > > > Sandeep Maheswaram <sanm@codeaurora.org> writes: > >> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > >> +{ > >> + struct device *dev = qcom->dev; > >> + int ret; > >> + > >> + if (!device_is_bound(&qcom->dwc3->dev)) > >> + return -EPROBE_DEFER; > > > > this breaks allmodconfig. I'm dropping this series from my queue for > > this merge window. > > Sorry, I meant this patch ;-) I guess that's due to INTERCONNECT being a module. There is currently a discussion about this with Viresh and Georgi in response to another automated build failure. Viresh suggests changing CONFIG_INTERCONNECT from tristate to bool, which seems sensible to me given that interconnect is a core subsystem. Let's hold back with this patch/series then until that is sorted out.
On Fri, 15 May 2020 at 02:33, Georgi Djakov <georgi.djakov@linaro.org> wrote: > ---8<--- > diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig > index 206caa0ea1c6..6661788b1a76 100644 > --- a/drivers/usb/dwc3/Kconfig > +++ b/drivers/usb/dwc3/Kconfig > @@ -129,6 +129,7 @@ config USB_DWC3_QCOM > tristate "Qualcomm Platform" > depends on ARCH_QCOM || COMPILE_TEST > depends on EXTCON || !EXTCON > + depends on INTERCONNECT || !INTERCONNECT Again, as I said yesterday. This looks incorrect and may not fix the problem.. With this we will be able to select USB_DWC3_QCOM even when INTERCONNECT=m. What we perhaps need here is: depends on INTERCONNECT != m -- viresh
Hi, On 5/15/2020 9:27 AM, Viresh Kumar wrote: > On Fri, 15 May 2020 at 02:33, Georgi Djakov <georgi.djakov@linaro.org> wrote: > >> ---8<--- >> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig >> index 206caa0ea1c6..6661788b1a76 100644 >> --- a/drivers/usb/dwc3/Kconfig >> +++ b/drivers/usb/dwc3/Kconfig >> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM >> tristate "Qualcomm Platform" >> depends on ARCH_QCOM || COMPILE_TEST >> depends on EXTCON || !EXTCON >> + depends on INTERCONNECT || !INTERCONNECT > Again, as I said yesterday. This looks incorrect and may not fix the problem.. > > With this we will be able to select USB_DWC3_QCOM even when INTERCONNECT=m. DWC3_QCOM in that case would be 'm' if INTERCONNECT =m and that should be fine? > What we perhaps need here is: > depends on INTERCONNECT != m > > -- > viresh
On 15-05-20, 09:53, Manu Gautam wrote: > Hi, > > On 5/15/2020 9:27 AM, Viresh Kumar wrote: > > On Fri, 15 May 2020 at 02:33, Georgi Djakov <georgi.djakov@linaro.org> wrote: > > > >> ---8<--- > >> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig > >> index 206caa0ea1c6..6661788b1a76 100644 > >> --- a/drivers/usb/dwc3/Kconfig > >> +++ b/drivers/usb/dwc3/Kconfig > >> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM > >> tristate "Qualcomm Platform" > >> depends on ARCH_QCOM || COMPILE_TEST > >> depends on EXTCON || !EXTCON > >> + depends on INTERCONNECT || !INTERCONNECT > > Again, as I said yesterday. This looks incorrect and may not fix the problem.. > > > > With this we will be able to select USB_DWC3_QCOM even when INTERCONNECT=m. > > DWC3_QCOM in that case would be 'm' if INTERCONNECT =m and > that should be fine? Ahh, I was speaking in context of PM_OPP where it is a bool. Perhaps this works just fine with DWC3_QCOM, sorry for the confusion.
Hi, Georgi Djakov <georgi.djakov@linaro.org> writes: > On 5/14/20 20:13, Matthias Kaehlcke wrote: >> On Thu, May 14, 2020 at 02:30:28PM +0300, Felipe Balbi wrote: >>> Felipe Balbi <balbi@kernel.org> writes: >>> >>>> Hi, >>>> >>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: >>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>> +{ >>>>> + struct device *dev = qcom->dev; >>>>> + int ret; >>>>> + >>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>> + return -EPROBE_DEFER; >>>> >>>> this breaks allmodconfig. I'm dropping this series from my queue for >>>> this merge window. >>> >>> Sorry, I meant this patch ;-) >> >> I guess that's due to INTERCONNECT being a module. There is currently a > > I believe it's because of this: > ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] undefined! > >> discussion about this with Viresh and Georgi in response to another >> automated build failure. Viresh suggests changing CONFIG_INTERCONNECT >> from tristate to bool, which seems sensible to me given that interconnect >> is a core subsystem. > > The problem you are talking about would arise when INTERCONNECT=m and > USB_DWC3_QCOM=y and it definitely exists here and could be triggered with > randconfig build. So i suggest to squash also the diff below. > > Thanks, > Georgi > > ---8<--- > diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig > index 206caa0ea1c6..6661788b1a76 100644 > --- a/drivers/usb/dwc3/Kconfig > +++ b/drivers/usb/dwc3/Kconfig > @@ -129,6 +129,7 @@ config USB_DWC3_QCOM > tristate "Qualcomm Platform" > depends on ARCH_QCOM || COMPILE_TEST > depends on EXTCON || !EXTCON > + depends on INTERCONNECT || !INTERCONNECT I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to device_is_bound()
Hi, On 5/15/20 08:54, Felipe Balbi wrote: > > Hi, > > Georgi Djakov <georgi.djakov@linaro.org> writes: >> On 5/14/20 20:13, Matthias Kaehlcke wrote: >>> On Thu, May 14, 2020 at 02:30:28PM +0300, Felipe Balbi wrote: >>>> Felipe Balbi <balbi@kernel.org> writes: >>>> >>>>> Hi, >>>>> >>>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: >>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>>> +{ >>>>>> + struct device *dev = qcom->dev; >>>>>> + int ret; >>>>>> + >>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>>> + return -EPROBE_DEFER; >>>>> >>>>> this breaks allmodconfig. I'm dropping this series from my queue for >>>>> this merge window. >>>> >>>> Sorry, I meant this patch ;-) >>> >>> I guess that's due to INTERCONNECT being a module. There is currently a >> >> I believe it's because of this: >> ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] undefined! >> >>> discussion about this with Viresh and Georgi in response to another >>> automated build failure. Viresh suggests changing CONFIG_INTERCONNECT >>> from tristate to bool, which seems sensible to me given that interconnect >>> is a core subsystem. >> >> The problem you are talking about would arise when INTERCONNECT=m and >> USB_DWC3_QCOM=y and it definitely exists here and could be triggered with >> randconfig build. So i suggest to squash also the diff below. >> >> Thanks, >> Georgi >> >> ---8<--- >> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig >> index 206caa0ea1c6..6661788b1a76 100644 >> --- a/drivers/usb/dwc3/Kconfig >> +++ b/drivers/usb/dwc3/Kconfig >> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM >> tristate "Qualcomm Platform" >> depends on ARCH_QCOM || COMPILE_TEST >> depends on EXTCON || !EXTCON >> + depends on INTERCONNECT || !INTERCONNECT > > I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to device_is_bound() Agree, but just to clarify, that these are two separate issues that need to be fixed. The device_is_bound() is the first one and USB_DWC3_QCOM=y combined with INTERCONNECT=m is the second one. Thanks, Georgi
Hi, Georgi Djakov <georgi.djakov@linaro.org> writes: >>>>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: >>>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>>>> +{ >>>>>>> + struct device *dev = qcom->dev; >>>>>>> + int ret; >>>>>>> + >>>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>>>> + return -EPROBE_DEFER; >>>>>> >>>>>> this breaks allmodconfig. I'm dropping this series from my queue for >>>>>> this merge window. >>>>> >>>>> Sorry, I meant this patch ;-) >>>> >>>> I guess that's due to INTERCONNECT being a module. There is currently a >>> >>> I believe it's because of this: >>> ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] undefined! >>> >>>> discussion about this with Viresh and Georgi in response to another >>>> automated build failure. Viresh suggests changing CONFIG_INTERCONNECT >>>> from tristate to bool, which seems sensible to me given that interconnect >>>> is a core subsystem. >>> >>> The problem you are talking about would arise when INTERCONNECT=m and >>> USB_DWC3_QCOM=y and it definitely exists here and could be triggered with >>> randconfig build. So i suggest to squash also the diff below. >>> >>> Thanks, >>> Georgi >>> >>> ---8<--- >>> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig >>> index 206caa0ea1c6..6661788b1a76 100644 >>> --- a/drivers/usb/dwc3/Kconfig >>> +++ b/drivers/usb/dwc3/Kconfig >>> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM >>> tristate "Qualcomm Platform" >>> depends on ARCH_QCOM || COMPILE_TEST >>> depends on EXTCON || !EXTCON >>> + depends on INTERCONNECT || !INTERCONNECT >> >> I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to device_is_bound() > > Agree, but just to clarify, that these are two separate issues that need to > be fixed. The device_is_bound() is the first one and USB_DWC3_QCOM=y combined > with INTERCONNECT=m is the second one. If INTERCONNECT=m, QCOM3 shouldn't be y. I think the following is enough: depends on INTERCONNECT=y || INTERCONNECT=USB_DWC3_QCOM
On Thu 14 May 23:29 PDT 2020, Felipe Balbi wrote: > > Hi, > > Georgi Djakov <georgi.djakov@linaro.org> writes: > >>>>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: > >>>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > >>>>>>> +{ > >>>>>>> + struct device *dev = qcom->dev; > >>>>>>> + int ret; > >>>>>>> + > >>>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) > >>>>>>> + return -EPROBE_DEFER; > >>>>>> > >>>>>> this breaks allmodconfig. I'm dropping this series from my queue for > >>>>>> this merge window. > >>>>> > >>>>> Sorry, I meant this patch ;-) > >>>> > >>>> I guess that's due to INTERCONNECT being a module. There is currently a > >>> > >>> I believe it's because of this: > >>> ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] undefined! > >>> > >>>> discussion about this with Viresh and Georgi in response to another > >>>> automated build failure. Viresh suggests changing CONFIG_INTERCONNECT > >>>> from tristate to bool, which seems sensible to me given that interconnect > >>>> is a core subsystem. > >>> > >>> The problem you are talking about would arise when INTERCONNECT=m and > >>> USB_DWC3_QCOM=y and it definitely exists here and could be triggered with > >>> randconfig build. So i suggest to squash also the diff below. > >>> > >>> Thanks, > >>> Georgi > >>> > >>> ---8<--- > >>> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig > >>> index 206caa0ea1c6..6661788b1a76 100644 > >>> --- a/drivers/usb/dwc3/Kconfig > >>> +++ b/drivers/usb/dwc3/Kconfig > >>> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM > >>> tristate "Qualcomm Platform" > >>> depends on ARCH_QCOM || COMPILE_TEST > >>> depends on EXTCON || !EXTCON > >>> + depends on INTERCONNECT || !INTERCONNECT > >> > >> I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to device_is_bound() > > > > Agree, but just to clarify, that these are two separate issues that need to > > be fixed. The device_is_bound() is the first one and USB_DWC3_QCOM=y combined > > with INTERCONNECT=m is the second one. > > If INTERCONNECT=m, QCOM3 shouldn't be y. I think the following is > enough: > > depends on INTERCONNECT=y || INTERCONNECT=USB_DWC3_QCOM > This misses the case where INTERCONNECT=n and USB_DWC3_QCOM=[ym] which I don't see a reason for breaking. But if only INTERCONNECT where a bool, then we don't need to specify a depends on, because it will either be there, or the stubs will. We've come to this conclusion in a lot of different frameworks and I don't see why we should do this differently with INTERCONNECT. Regards, Bjorn
On 18-05-20, 11:35, Bjorn Andersson wrote: > This misses the case where INTERCONNECT=n and USB_DWC3_QCOM=[ym] which > I don't see a reason for breaking. > > But if only INTERCONNECT where a bool, then we don't need to specify a > depends on, because it will either be there, or the stubs will. > We've come to this conclusion in a lot of different frameworks and I > don't see why we should do this differently with INTERCONNECT. INTERCONNECT is a bool now and the patch has been pushed to linux-next already.
Hi Felipe, Please let me know how to go forward with this patch Regards Sandeep On 5/19/2020 12:05 AM, Bjorn Andersson wrote: > On Thu 14 May 23:29 PDT 2020, Felipe Balbi wrote: > >> Hi, >> >> Georgi Djakov <georgi.djakov@linaro.org> writes: >>>>>>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: >>>>>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>>>>>> +{ >>>>>>>>> + struct device *dev = qcom->dev; >>>>>>>>> + int ret; >>>>>>>>> + >>>>>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>>>>>> + return -EPROBE_DEFER; >>>>>>>> this breaks allmodconfig. I'm dropping this series from my queue for >>>>>>>> this merge window. >>>>>>> Sorry, I meant this patch ;-) >>>>>> I guess that's due to INTERCONNECT being a module. There is currently a >>>>> I believe it's because of this: >>>>> ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] undefined! >>>>> >>>>>> discussion about this with Viresh and Georgi in response to another >>>>>> automated build failure. Viresh suggests changing CONFIG_INTERCONNECT >>>>>> from tristate to bool, which seems sensible to me given that interconnect >>>>>> is a core subsystem. >>>>> The problem you are talking about would arise when INTERCONNECT=m and >>>>> USB_DWC3_QCOM=y and it definitely exists here and could be triggered with >>>>> randconfig build. So i suggest to squash also the diff below. >>>>> >>>>> Thanks, >>>>> Georgi >>>>> >>>>> ---8<--- >>>>> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig >>>>> index 206caa0ea1c6..6661788b1a76 100644 >>>>> --- a/drivers/usb/dwc3/Kconfig >>>>> +++ b/drivers/usb/dwc3/Kconfig >>>>> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM >>>>> tristate "Qualcomm Platform" >>>>> depends on ARCH_QCOM || COMPILE_TEST >>>>> depends on EXTCON || !EXTCON >>>>> + depends on INTERCONNECT || !INTERCONNECT >>>> I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to device_is_bound() >>> Agree, but just to clarify, that these are two separate issues that need to >>> be fixed. The device_is_bound() is the first one and USB_DWC3_QCOM=y combined >>> with INTERCONNECT=m is the second one. >> If INTERCONNECT=m, QCOM3 shouldn't be y. I think the following is >> enough: >> >> depends on INTERCONNECT=y || INTERCONNECT=USB_DWC3_QCOM >> > This misses the case where INTERCONNECT=n and USB_DWC3_QCOM=[ym] which > I don't see a reason for breaking. > > But if only INTERCONNECT where a bool, then we don't need to specify a > depends on, because it will either be there, or the stubs will. > We've come to this conclusion in a lot of different frameworks and I > don't see why we should do this differently with INTERCONNECT. > > Regards, > Bjorn
On 26.05.20 14:04, Sandeep Maheswaram (Temp) wrote: > Hi Felipe, > > Please let me know how to go forward with this patch > Hi Sandeep, Please just add a patch to fix the allmodconfig error. Felipe has suggested to introduce a separate patch which exports the device_is_bound() function. This export should precede the addition of interconnect support. Also regarding the "depends on INTERCONNECT || !INTERCONNECT" change, no "depends on" would be needed, as we just made the interconnect framework bool. Thanks, Georgi > Regards > > Sandeep > > On 5/19/2020 12:05 AM, Bjorn Andersson wrote: >> On Thu 14 May 23:29 PDT 2020, Felipe Balbi wrote: >> >>> Hi, >>> >>> Georgi Djakov <georgi.djakov@linaro.org> writes: >>>>>>>>> Sandeep Maheswaram <sanm@codeaurora.org> writes: >>>>>>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>>>>>>> +{ >>>>>>>>>> + struct device *dev = qcom->dev; >>>>>>>>>> + int ret; >>>>>>>>>> + >>>>>>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>>>>>>> + return -EPROBE_DEFER; >>>>>>>>> this breaks allmodconfig. I'm dropping this series from my >>>>>>>>> queue for >>>>>>>>> this merge window. >>>>>>>> Sorry, I meant this patch ;-) >>>>>>> I guess that's due to INTERCONNECT being a module. There is >>>>>>> currently a >>>>>> I believe it's because of this: >>>>>> ERROR: modpost: "device_is_bound" [drivers/usb/dwc3/dwc3-qcom.ko] >>>>>> undefined! >>>>>> >>>>>>> discussion about this with Viresh and Georgi in response to another >>>>>>> automated build failure. Viresh suggests changing >>>>>>> CONFIG_INTERCONNECT >>>>>>> from tristate to bool, which seems sensible to me given that >>>>>>> interconnect >>>>>>> is a core subsystem. >>>>>> The problem you are talking about would arise when INTERCONNECT=m and >>>>>> USB_DWC3_QCOM=y and it definitely exists here and could be >>>>>> triggered with >>>>>> randconfig build. So i suggest to squash also the diff below. >>>>>> >>>>>> Thanks, >>>>>> Georgi >>>>>> >>>>>> ---8<--- >>>>>> diff --git a/drivers/usb/dwc3/Kconfig b/drivers/usb/dwc3/Kconfig >>>>>> index 206caa0ea1c6..6661788b1a76 100644 >>>>>> --- a/drivers/usb/dwc3/Kconfig >>>>>> +++ b/drivers/usb/dwc3/Kconfig >>>>>> @@ -129,6 +129,7 @@ config USB_DWC3_QCOM >>>>>> tristate "Qualcomm Platform" >>>>>> depends on ARCH_QCOM || COMPILE_TEST >>>>>> depends on EXTCON || !EXTCON >>>>>> + depends on INTERCONNECT || !INTERCONNECT >>>>> I would prefer to see a patch adding EXPORT_SYMBOL_GPL() to >>>>> device_is_bound() >>>> Agree, but just to clarify, that these are two separate issues that >>>> need to >>>> be fixed. The device_is_bound() is the first one and USB_DWC3_QCOM=y >>>> combined >>>> with INTERCONNECT=m is the second one. >>> If INTERCONNECT=m, QCOM3 shouldn't be y. I think the following is >>> enough: >>> >>> depends on INTERCONNECT=y || INTERCONNECT=USB_DWC3_QCOM >>> >> This misses the case where INTERCONNECT=n and USB_DWC3_QCOM=[ym] which >> I don't see a reason for breaking. >> >> But if only INTERCONNECT where a bool, then we don't need to specify a >> depends on, because it will either be there, or the stubs will. >> We've come to this conclusion in a lot of different frameworks and I >> don't see why we should do this differently with INTERCONNECT. >> >> Regards, >> Bjorn >
Hi, Georgi Djakov <georgi.djakov@linaro.org> writes: > On 26.05.20 14:04, Sandeep Maheswaram (Temp) wrote: >> Hi Felipe, >> >> Please let me know how to go forward with this patch (don't top-post!) > Please just add a patch to fix the allmodconfig error. Felipe has > suggested to introduce a separate patch which exports the > device_is_bound() function. This export should precede the addition > of interconnect support. > > Also regarding the "depends on INTERCONNECT || !INTERCONNECT" change, > no "depends on" would be needed, as we just made the interconnect > framework bool. y'all have lost the current merge window, I guess. I'm not sure Greg will take last minute changes to drivers base and I have already sent him my pull request for v5.8. On the plus side, this gives you the chance to run hundreds of randbuilds with your patches.
On 5/26/2020 5:13 PM, Felipe Balbi wrote: > Hi, > > Georgi Djakov <georgi.djakov@linaro.org> writes: >> On 26.05.20 14:04, Sandeep Maheswaram (Temp) wrote: >>> Hi Felipe, >>> >>> Please let me know how to go forward with this patch > (don't top-post!) > >> Please just add a patch to fix the allmodconfig error. Felipe has >> suggested to introduce a separate patch which exports the >> device_is_bound() function. This export should precede the addition >> of interconnect support. >> >> Also regarding the "depends on INTERCONNECT || !INTERCONNECT" change, >> no "depends on" would be needed, as we just made the interconnect >> framework bool. > y'all have lost the current merge window, I guess. I'm not sure Greg > will take last minute changes to drivers base and I have already sent > him my pull request for v5.8. On the plus side, this gives you the > chance to run hundreds of randbuilds with your patches. Hi Georgi, I am assuming that the patch which exports the device_is_bound() function will solve the allmodconfig error. Or do i need to change anything in dwc3 driver?
Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > index 1dfd024..d33ae86 100644 > --- a/drivers/usb/dwc3/dwc3-qcom.c > +++ b/drivers/usb/dwc3/dwc3-qcom.c > @@ -76,8 +85,13 @@ struct dwc3_qcom { > enum usb_dr_mode mode; > bool is_suspended; > bool pm_suspended; > + struct icc_path *usb_ddr_icc_path; > + struct icc_path *apps_usb_icc_path; > }; > > +static int dwc3_qcom_interconnect_enable(struct dwc3_qcom *qcom); > +static int dwc3_qcom_interconnect_disable(struct dwc3_qcom *qcom); Please get rid of these. We shouldn't need forward declarations. > + > static inline void dwc3_qcom_setbits(void __iomem *base, u32 offset, u32 val) > { > u32 reg; > @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > return 0; > } > > + > +/** > + * dwc3_qcom_interconnect_init() - Get interconnect path handles > + * @qcom: Pointer to the concerned usb core. > + * > + */ > +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > +{ > + struct device *dev = qcom->dev; > + int ret; > + > + if (!device_is_bound(&qcom->dwc3->dev)) > + return -EPROBE_DEFER; How is this supposed to work? I see that this was added in an earlier revision of this patch series but there isn't any mention of why device_is_bound() is used here. It would be great if there was a comment detailing why this is necessary. It sounds like maximum_speed is important? Furthermore, dwc3_qcom_interconnect_init() is called by dwc3_qcom_probe() which is the function that registers the device for qcom->dwc3->dev. If that device doesn't probe between the time it is registered by dwc3_qcom_probe() and this function is called then we'll fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the qcom->dwc3->dev device from the platform bus because we call of_platform_depopulate() on the error path of dwc3_qcom_probe(). So isn't this whole thing racy and can potentially lead us to a driver probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing and we're trying to time it just right so that driver for dwc3 binds before we setup interconnects? I don't know if dwc3 can communicate to the wrapper but that would be more of a direct way to do this. Or maybe the wrapper should try to read the DT property for maximum speed and fallback to a worst case high bandwidth value if it can't figure it out itself without help from dwc3 core.
On 6/3/2020 11:06 PM, Stephen Boyd wrote: > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) >> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c >> index 1dfd024..d33ae86 100644 >> --- a/drivers/usb/dwc3/dwc3-qcom.c >> +++ b/drivers/usb/dwc3/dwc3-qcom.c >> @@ -76,8 +85,13 @@ struct dwc3_qcom { >> enum usb_dr_mode mode; >> bool is_suspended; >> bool pm_suspended; >> + struct icc_path *usb_ddr_icc_path; >> + struct icc_path *apps_usb_icc_path; >> }; >> >> +static int dwc3_qcom_interconnect_enable(struct dwc3_qcom *qcom); >> +static int dwc3_qcom_interconnect_disable(struct dwc3_qcom *qcom); > Please get rid of these. We shouldn't need forward declarations. Will do in next version. > >> + >> static inline void dwc3_qcom_setbits(void __iomem *base, u32 offset, u32 val) >> { >> u32 reg; >> @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) >> return 0; >> } >> >> + >> +/** >> + * dwc3_qcom_interconnect_init() - Get interconnect path handles >> + * @qcom: Pointer to the concerned usb core. >> + * >> + */ >> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >> +{ >> + struct device *dev = qcom->dev; >> + int ret; >> + >> + if (!device_is_bound(&qcom->dwc3->dev)) >> + return -EPROBE_DEFER; > How is this supposed to work? I see that this was added in an earlier > revision of this patch series but there isn't any mention of why > device_is_bound() is used here. It would be great if there was a comment > detailing why this is necessary. It sounds like maximum_speed is > important? > > Furthermore, dwc3_qcom_interconnect_init() is called by > dwc3_qcom_probe() which is the function that registers the device for > qcom->dwc3->dev. If that device doesn't probe between the time it is > registered by dwc3_qcom_probe() and this function is called then we'll > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > qcom->dwc3->dev device from the platform bus because we call > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > So isn't this whole thing racy and can potentially lead us to a driver > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > and we're trying to time it just right so that driver for dwc3 binds > before we setup interconnects? I don't know if dwc3 can communicate to > the wrapper but that would be more of a direct way to do this. Or maybe > the wrapper should try to read the DT property for maximum speed and > fallback to a worst case high bandwidth value if it can't figure it out > itself without help from dwc3 core. > This was added in V4 to address comments from Matthias in V3 https://patchwork.kernel.org/patch/11148587/
Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) > > On 6/3/2020 11:06 PM, Stephen Boyd wrote: > > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > >> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > >> index 1dfd024..d33ae86 100644 > >> --- a/drivers/usb/dwc3/dwc3-qcom.c > >> +++ b/drivers/usb/dwc3/dwc3-qcom.c > >> @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > >> return 0; > >> } > >> > >> + > >> +/** > >> + * dwc3_qcom_interconnect_init() - Get interconnect path handles > >> + * @qcom: Pointer to the concerned usb core. > >> + * > >> + */ > >> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > >> +{ > >> + struct device *dev = qcom->dev; > >> + int ret; > >> + > >> + if (!device_is_bound(&qcom->dwc3->dev)) > >> + return -EPROBE_DEFER; > > How is this supposed to work? I see that this was added in an earlier > > revision of this patch series but there isn't any mention of why > > device_is_bound() is used here. It would be great if there was a comment > > detailing why this is necessary. It sounds like maximum_speed is > > important? > > > > Furthermore, dwc3_qcom_interconnect_init() is called by > > dwc3_qcom_probe() which is the function that registers the device for > > qcom->dwc3->dev. If that device doesn't probe between the time it is > > registered by dwc3_qcom_probe() and this function is called then we'll > > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > > qcom->dwc3->dev device from the platform bus because we call > > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > > > So isn't this whole thing racy and can potentially lead us to a driver > > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > > and we're trying to time it just right so that driver for dwc3 binds > > before we setup interconnects? I don't know if dwc3 can communicate to > > the wrapper but that would be more of a direct way to do this. Or maybe > > the wrapper should try to read the DT property for maximum speed and > > fallback to a worst case high bandwidth value if it can't figure it out > > itself without help from dwc3 core. > > > This was added in V4 to address comments from Matthias in V3 > > https://patchwork.kernel.org/patch/11148587/ > Yes, that why I said: "I see that this was added in an earlier revision of this patch series but there isn't any mention of why device_is_bound() is used here. It would be great if there was a comment detailing why this is necessary. It sounds like maximum_speed is important?" Can you please respond to the rest of my email?
On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: > Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) > > > > On 6/3/2020 11:06 PM, Stephen Boyd wrote: > > > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > > >> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > > >> index 1dfd024..d33ae86 100644 > > >> --- a/drivers/usb/dwc3/dwc3-qcom.c > > >> +++ b/drivers/usb/dwc3/dwc3-qcom.c > > >> @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > > >> return 0; > > >> } > > >> > > >> + > > >> +/** > > >> + * dwc3_qcom_interconnect_init() - Get interconnect path handles > > >> + * @qcom: Pointer to the concerned usb core. > > >> + * > > >> + */ > > >> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > > >> +{ > > >> + struct device *dev = qcom->dev; > > >> + int ret; > > >> + > > >> + if (!device_is_bound(&qcom->dwc3->dev)) > > >> + return -EPROBE_DEFER; > > > How is this supposed to work? I see that this was added in an earlier > > > revision of this patch series but there isn't any mention of why > > > device_is_bound() is used here. It would be great if there was a comment > > > detailing why this is necessary. It sounds like maximum_speed is > > > important? > > > > > > Furthermore, dwc3_qcom_interconnect_init() is called by > > > dwc3_qcom_probe() which is the function that registers the device for > > > qcom->dwc3->dev. If that device doesn't probe between the time it is > > > registered by dwc3_qcom_probe() and this function is called then we'll > > > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > > > qcom->dwc3->dev device from the platform bus because we call > > > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > > > > > So isn't this whole thing racy and can potentially lead us to a driver > > > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > > > and we're trying to time it just right so that driver for dwc3 binds > > > before we setup interconnects? I don't know if dwc3 can communicate to > > > the wrapper but that would be more of a direct way to do this. Or maybe > > > the wrapper should try to read the DT property for maximum speed and > > > fallback to a worst case high bandwidth value if it can't figure it out > > > itself without help from dwc3 core. > > > > > This was added in V4 to address comments from Matthias in V3 > > > > https://patchwork.kernel.org/patch/11148587/ > > > > Yes, that why I said: > > "I see that this was added in an earlier > revision of this patch series but there isn't any mention of why > device_is_bound() is used here. It would be great if there was a comment > detailing why this is necessary. It sounds like maximum_speed is > important?" > > Can you please respond to the rest of my email? I agree with Stephen that using device_is_bound() isn't a good option in this case, when I suggested it I wasn't looking at the big picture of how probing the core driver is triggered, sorry about that. Reading the speed from the DT with usb_get_maximum_speed() as Stephen suggests would be an option, the inconvenient is that we then essentially require the property to be defined, while the core driver gets a suitable value from hardware registers. Not sure if the wrapper driver could read from the same registers. One option could be to poll device_is_bound() for 100 ms (or so), with sleeps between polls. It's not elegant but would probably work if we don't find a better solution.
On 6/16/2020 1:12 AM, Matthias Kaehlcke wrote: > On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: >> Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) >>> On 6/3/2020 11:06 PM, Stephen Boyd wrote: >>>> Quoting Sandeep Maheswaram (2020-03-31 22:15:43) >>>>> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c >>>>> index 1dfd024..d33ae86 100644 >>>>> --- a/drivers/usb/dwc3/dwc3-qcom.c >>>>> +++ b/drivers/usb/dwc3/dwc3-qcom.c >>>>> @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) >>>>> return 0; >>>>> } >>>>> >>>>> + >>>>> +/** >>>>> + * dwc3_qcom_interconnect_init() - Get interconnect path handles >>>>> + * @qcom: Pointer to the concerned usb core. >>>>> + * >>>>> + */ >>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>> +{ >>>>> + struct device *dev = qcom->dev; >>>>> + int ret; >>>>> + >>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>> + return -EPROBE_DEFER; >>>> How is this supposed to work? I see that this was added in an earlier >>>> revision of this patch series but there isn't any mention of why >>>> device_is_bound() is used here. It would be great if there was a comment >>>> detailing why this is necessary. It sounds like maximum_speed is >>>> important? >>>> >>>> Furthermore, dwc3_qcom_interconnect_init() is called by >>>> dwc3_qcom_probe() which is the function that registers the device for >>>> qcom->dwc3->dev. If that device doesn't probe between the time it is >>>> registered by dwc3_qcom_probe() and this function is called then we'll >>>> fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the >>>> qcom->dwc3->dev device from the platform bus because we call >>>> of_platform_depopulate() on the error path of dwc3_qcom_probe(). >>>> >>>> So isn't this whole thing racy and can potentially lead us to a driver >>>> probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing >>>> and we're trying to time it just right so that driver for dwc3 binds >>>> before we setup interconnects? I don't know if dwc3 can communicate to >>>> the wrapper but that would be more of a direct way to do this. Or maybe >>>> the wrapper should try to read the DT property for maximum speed and >>>> fallback to a worst case high bandwidth value if it can't figure it out >>>> itself without help from dwc3 core. >>>> >>> This was added in V4 to address comments from Matthias in V3 >>> >>> https://patchwork.kernel.org/patch/11148587/ >>> >> Yes, that why I said: >> >> "I see that this was added in an earlier >> revision of this patch series but there isn't any mention of why >> device_is_bound() is used here. It would be great if there was a comment >> detailing why this is necessary. It sounds like maximum_speed is >> important?" >> >> Can you please respond to the rest of my email? > I agree with Stephen that using device_is_bound() isn't a good option > in this case, when I suggested it I wasn't looking at the big picture > of how probing the core driver is triggered, sorry about that. > > Reading the speed from the DT with usb_get_maximum_speed() as Stephen > suggests would be an option, the inconvenient is that we then > essentially require the property to be defined, while the core driver > gets a suitable value from hardware registers. Not sure if the wrapper > driver could read from the same registers. > > One option could be to poll device_is_bound() for 100 ms (or so), with > sleeps between polls. It's not elegant but would probably work if we > don't find a better solution. if (np) ret = dwc3_qcom_of_register_core(pdev); else ret = dwc3_qcom_acpi_register_core(pdev); if (ret) { dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); goto depopulate; } ret = dwc3_qcom_interconnect_init(qcom); if (ret) goto depopulate; qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev); Before calling dwc3_qcom_interconnect_init we are checking if (ret) { dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); goto depopulate; } Doesn't this condition confirm the core driver is probed?
On Tue, Jun 16, 2020 at 10:22:47AM +0530, Sandeep Maheswaram (Temp) wrote: > > On 6/16/2020 1:12 AM, Matthias Kaehlcke wrote: > > On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: > > > Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) > > > > On 6/3/2020 11:06 PM, Stephen Boyd wrote: > > > > > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > > > > > > diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > index 1dfd024..d33ae86 100644 > > > > > > --- a/drivers/usb/dwc3/dwc3-qcom.c > > > > > > +++ b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > > > > > > return 0; > > > > > > } > > > > > > + > > > > > > +/** > > > > > > + * dwc3_qcom_interconnect_init() - Get interconnect path handles > > > > > > + * @qcom: Pointer to the concerned usb core. > > > > > > + * > > > > > > + */ > > > > > > +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > > > > > > +{ > > > > > > + struct device *dev = qcom->dev; > > > > > > + int ret; > > > > > > + > > > > > > + if (!device_is_bound(&qcom->dwc3->dev)) > > > > > > + return -EPROBE_DEFER; > > > > > How is this supposed to work? I see that this was added in an earlier > > > > > revision of this patch series but there isn't any mention of why > > > > > device_is_bound() is used here. It would be great if there was a comment > > > > > detailing why this is necessary. It sounds like maximum_speed is > > > > > important? > > > > > > > > > > Furthermore, dwc3_qcom_interconnect_init() is called by > > > > > dwc3_qcom_probe() which is the function that registers the device for > > > > > qcom->dwc3->dev. If that device doesn't probe between the time it is > > > > > registered by dwc3_qcom_probe() and this function is called then we'll > > > > > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > > > > > qcom->dwc3->dev device from the platform bus because we call > > > > > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > > > > > > > > > So isn't this whole thing racy and can potentially lead us to a driver > > > > > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > > > > > and we're trying to time it just right so that driver for dwc3 binds > > > > > before we setup interconnects? I don't know if dwc3 can communicate to > > > > > the wrapper but that would be more of a direct way to do this. Or maybe > > > > > the wrapper should try to read the DT property for maximum speed and > > > > > fallback to a worst case high bandwidth value if it can't figure it out > > > > > itself without help from dwc3 core. > > > > > > > > > This was added in V4 to address comments from Matthias in V3 > > > > > > > > https://patchwork.kernel.org/patch/11148587/ > > > > > > > Yes, that why I said: > > > > > > "I see that this was added in an earlier > > > revision of this patch series but there isn't any mention of why > > > device_is_bound() is used here. It would be great if there was a comment > > > detailing why this is necessary. It sounds like maximum_speed is > > > important?" > > > > > > Can you please respond to the rest of my email? > > I agree with Stephen that using device_is_bound() isn't a good option > > in this case, when I suggested it I wasn't looking at the big picture > > of how probing the core driver is triggered, sorry about that. > > > > Reading the speed from the DT with usb_get_maximum_speed() as Stephen > > suggests would be an option, the inconvenient is that we then > > essentially require the property to be defined, while the core driver > > gets a suitable value from hardware registers. Not sure if the wrapper > > driver could read from the same registers. > > > > One option could be to poll device_is_bound() for 100 ms (or so), with > > sleeps between polls. It's not elegant but would probably work if we > > don't find a better solution. > if (np) > ret = dwc3_qcom_of_register_core(pdev); > else > ret = dwc3_qcom_acpi_register_core(pdev); > > if (ret) { > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > goto depopulate; > } > > ret = dwc3_qcom_interconnect_init(qcom); > if (ret) > goto depopulate; > > qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev); > > Before calling dwc3_qcom_interconnect_init we are checking > > if (ret) { > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > goto depopulate; > } > > Doesn't this condition confirm the core driver is probed? Not really: // called under the hood by of_platform_populate() static int really_probe(struct device *dev, struct device_driver *drv) { ... if (dev->bus->probe) { ret = dev->bus->probe(dev); if (ret) goto probe_failed; } else if (drv->probe) { ret = drv->probe(dev); if (ret) goto probe_failed; } ... probe_failed: ... /* * Ignore errors returned by ->probe so that the next driver can try * its luck. */ ret = 0; ... return ret; } As a result of_platform_populate() in dwc3_qcom_of_register_core() returns 0 even when probing the device failed: [ 0.244339] dwc3-qcom a6f8800.usb: DBG: populate [ 0.244772] dwc3 a600000.dwc3: DBG: dwc3_probe [ 0.245237] dwc3 a600000.dwc3: DBG: dwc3_probe err: -517 [ 0.245264] dwc3-qcom a6f8800.usb: DBG: populate (done) [ 0.245317] dwc3-qcom a6f8800.usb: DBG: dwc3_qcom_interconnect_init() failed: -517 Probe fails because the interconnect stuff isn't ready yet, otherwise it could access invalid data. A later _populate() is successful and the probing of the core is done synchronously, i.e. after _populate() the core driver is fully initialized: [ 3.898106] dwc3-qcom a6f8800.usb: DBG: populate [ 3.908356] dwc3 a600000.dwc3: DBG: dwc3_probe [ 4.205104] dwc3 a600000.dwc3: DBG: dwc3_probe (done) [ 4.210305] dwc3-qcom a6f8800.usb: DBG: populate (done) The synchronous probing in _populate() suggests that using device_is_bound() would actually be a valid option, either the core device was successfully probed or not, there should be no race. I sent a patch that adds this check to dwc3_qcom_of_register_core(), which is less confusing and makes clear that the core device is valid unless this function returns an error: https://lore.kernel.org/patchwork/patch/1257279/ It might make sense to add your "driver core:Export the symbol device_is_bound" patch, mine and this one to a single series. Thanks Matthias
On Tue, Jun 16, 2020 at 01:38:49PM -0700, Matthias Kaehlcke wrote: > On Tue, Jun 16, 2020 at 10:22:47AM +0530, Sandeep Maheswaram (Temp) wrote: > > > > On 6/16/2020 1:12 AM, Matthias Kaehlcke wrote: > > > On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: > > > > Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) > > > > > On 6/3/2020 11:06 PM, Stephen Boyd wrote: > > > > > > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > > > > > > > diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > index 1dfd024..d33ae86 100644 > > > > > > > --- a/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > +++ b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > > > > > > > return 0; > > > > > > > } > > > > > > > + > > > > > > > +/** > > > > > > > + * dwc3_qcom_interconnect_init() - Get interconnect path handles > > > > > > > + * @qcom: Pointer to the concerned usb core. > > > > > > > + * > > > > > > > + */ > > > > > > > +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > > > > > > > +{ > > > > > > > + struct device *dev = qcom->dev; > > > > > > > + int ret; > > > > > > > + > > > > > > > + if (!device_is_bound(&qcom->dwc3->dev)) > > > > > > > + return -EPROBE_DEFER; > > > > > > How is this supposed to work? I see that this was added in an earlier > > > > > > revision of this patch series but there isn't any mention of why > > > > > > device_is_bound() is used here. It would be great if there was a comment > > > > > > detailing why this is necessary. It sounds like maximum_speed is > > > > > > important? > > > > > > > > > > > > Furthermore, dwc3_qcom_interconnect_init() is called by > > > > > > dwc3_qcom_probe() which is the function that registers the device for > > > > > > qcom->dwc3->dev. If that device doesn't probe between the time it is > > > > > > registered by dwc3_qcom_probe() and this function is called then we'll > > > > > > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > > > > > > qcom->dwc3->dev device from the platform bus because we call > > > > > > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > > > > > > > > > > > So isn't this whole thing racy and can potentially lead us to a driver > > > > > > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > > > > > > and we're trying to time it just right so that driver for dwc3 binds > > > > > > before we setup interconnects? I don't know if dwc3 can communicate to > > > > > > the wrapper but that would be more of a direct way to do this. Or maybe > > > > > > the wrapper should try to read the DT property for maximum speed and > > > > > > fallback to a worst case high bandwidth value if it can't figure it out > > > > > > itself without help from dwc3 core. > > > > > > > > > > > This was added in V4 to address comments from Matthias in V3 > > > > > > > > > > https://patchwork.kernel.org/patch/11148587/ > > > > > > > > > Yes, that why I said: > > > > > > > > "I see that this was added in an earlier > > > > revision of this patch series but there isn't any mention of why > > > > device_is_bound() is used here. It would be great if there was a comment > > > > detailing why this is necessary. It sounds like maximum_speed is > > > > important?" > > > > > > > > Can you please respond to the rest of my email? > > > I agree with Stephen that using device_is_bound() isn't a good option > > > in this case, when I suggested it I wasn't looking at the big picture > > > of how probing the core driver is triggered, sorry about that. > > > > > > Reading the speed from the DT with usb_get_maximum_speed() as Stephen > > > suggests would be an option, the inconvenient is that we then > > > essentially require the property to be defined, while the core driver > > > gets a suitable value from hardware registers. Not sure if the wrapper > > > driver could read from the same registers. > > > > > > One option could be to poll device_is_bound() for 100 ms (or so), with > > > sleeps between polls. It's not elegant but would probably work if we > > > don't find a better solution. > > if (np) > > ret = dwc3_qcom_of_register_core(pdev); > > else > > ret = dwc3_qcom_acpi_register_core(pdev); > > > > if (ret) { > > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > > goto depopulate; > > } > > > > ret = dwc3_qcom_interconnect_init(qcom); > > if (ret) > > goto depopulate; > > > > qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev); > > > > Before calling dwc3_qcom_interconnect_init we are checking > > > > if (ret) { > > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > > goto depopulate; > > } > > > > Doesn't this condition confirm the core driver is probed? > > Not really: > > // called under the hood by of_platform_populate() > static int really_probe(struct device *dev, struct device_driver *drv) > { > ... > > if (dev->bus->probe) { > ret = dev->bus->probe(dev); > if (ret) > goto probe_failed; > } else if (drv->probe) { > ret = drv->probe(dev); > if (ret) > goto probe_failed; > } > > ... > > probe_failed: > ... > > /* > * Ignore errors returned by ->probe so that the next driver can try > * its luck. > */ > ret = 0; > > ... > > return ret; > } > > As a result of_platform_populate() in dwc3_qcom_of_register_core() > returns 0 even when probing the device failed: > > [ 0.244339] dwc3-qcom a6f8800.usb: DBG: populate > [ 0.244772] dwc3 a600000.dwc3: DBG: dwc3_probe > [ 0.245237] dwc3 a600000.dwc3: DBG: dwc3_probe err: -517 > [ 0.245264] dwc3-qcom a6f8800.usb: DBG: populate (done) > [ 0.245317] dwc3-qcom a6f8800.usb: DBG: dwc3_qcom_interconnect_init() failed: -517 > > Probe fails because the interconnect stuff isn't ready yet, otherwise > it could access invalid data. > > A later _populate() is successful and the probing of the core is done > synchronously, i.e. after _populate() the core driver is fully > initialized: > > [ 3.898106] dwc3-qcom a6f8800.usb: DBG: populate > [ 3.908356] dwc3 a600000.dwc3: DBG: dwc3_probe > [ 4.205104] dwc3 a600000.dwc3: DBG: dwc3_probe (done) > [ 4.210305] dwc3-qcom a6f8800.usb: DBG: populate (done) > > The synchronous probing in _populate() suggests that using device_is_bound() > would actually be a valid option, either the core device was successfully > probed or not, there should be no race. > > I sent a patch that adds this check to dwc3_qcom_of_register_core(), which > is less confusing and makes clear that the core device is valid unless > this function returns an error: > > https://lore.kernel.org/patchwork/patch/1257279/ > > It might make sense to add your "driver core:Export the symbol > device_is_bound" patch, mine and this one to a single series. From the discussion on "driver core:Export the symbol device_is_bound" (https://patchwork.kernel.org/patch/11584225/) it is clear that this won't fly. The split dwc3 driver is considered a broken design. This is what Rob Herring said: We never should have had this split either in the DT binding nor driver(s) as if the SoC wrapper crap and licensed IP block are independent things. The thing to do here is either make the DWC3 code a library which drivers call (e.g. SDHCI) or add hooks into the DWC3 driver for platform specifics (e.g. Designware PCI). Neither is a simple solution though. That seems to be the desirable solution in the longer term, but it doesn't seem reasonable to me to expect you to fix this design issue to add interconnect support. Some possible options to move forward: - try to determine the max speed without involving the core device - select a reasonable default when 'maximum-speed' is not specified - use the core device to determine the max speed and pray
On 7/1/2020 4:12 AM, Matthias Kaehlcke wrote: > On Tue, Jun 16, 2020 at 01:38:49PM -0700, Matthias Kaehlcke wrote: >> On Tue, Jun 16, 2020 at 10:22:47AM +0530, Sandeep Maheswaram (Temp) wrote: >>> On 6/16/2020 1:12 AM, Matthias Kaehlcke wrote: >>>> On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: >>>>> Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) >>>>>> On 6/3/2020 11:06 PM, Stephen Boyd wrote: >>>>>>> Quoting Sandeep Maheswaram (2020-03-31 22:15:43) >>>>>>>> diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c >>>>>>>> index 1dfd024..d33ae86 100644 >>>>>>>> --- a/drivers/usb/dwc3/dwc3-qcom.c >>>>>>>> +++ b/drivers/usb/dwc3/dwc3-qcom.c >>>>>>>> @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) >>>>>>>> return 0; >>>>>>>> } >>>>>>>> + >>>>>>>> +/** >>>>>>>> + * dwc3_qcom_interconnect_init() - Get interconnect path handles >>>>>>>> + * @qcom: Pointer to the concerned usb core. >>>>>>>> + * >>>>>>>> + */ >>>>>>>> +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) >>>>>>>> +{ >>>>>>>> + struct device *dev = qcom->dev; >>>>>>>> + int ret; >>>>>>>> + >>>>>>>> + if (!device_is_bound(&qcom->dwc3->dev)) >>>>>>>> + return -EPROBE_DEFER; >>>>>>> How is this supposed to work? I see that this was added in an earlier >>>>>>> revision of this patch series but there isn't any mention of why >>>>>>> device_is_bound() is used here. It would be great if there was a comment >>>>>>> detailing why this is necessary. It sounds like maximum_speed is >>>>>>> important? >>>>>>> >>>>>>> Furthermore, dwc3_qcom_interconnect_init() is called by >>>>>>> dwc3_qcom_probe() which is the function that registers the device for >>>>>>> qcom->dwc3->dev. If that device doesn't probe between the time it is >>>>>>> registered by dwc3_qcom_probe() and this function is called then we'll >>>>>>> fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the >>>>>>> qcom->dwc3->dev device from the platform bus because we call >>>>>>> of_platform_depopulate() on the error path of dwc3_qcom_probe(). >>>>>>> >>>>>>> So isn't this whole thing racy and can potentially lead us to a driver >>>>>>> probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing >>>>>>> and we're trying to time it just right so that driver for dwc3 binds >>>>>>> before we setup interconnects? I don't know if dwc3 can communicate to >>>>>>> the wrapper but that would be more of a direct way to do this. Or maybe >>>>>>> the wrapper should try to read the DT property for maximum speed and >>>>>>> fallback to a worst case high bandwidth value if it can't figure it out >>>>>>> itself without help from dwc3 core. >>>>>>> >>>>>> This was added in V4 to address comments from Matthias in V3 >>>>>> >>>>>> https://patchwork.kernel.org/patch/11148587/ >>>>>> >>>>> Yes, that why I said: >>>>> >>>>> "I see that this was added in an earlier >>>>> revision of this patch series but there isn't any mention of why >>>>> device_is_bound() is used here. It would be great if there was a comment >>>>> detailing why this is necessary. It sounds like maximum_speed is >>>>> important?" >>>>> >>>>> Can you please respond to the rest of my email? >>>> I agree with Stephen that using device_is_bound() isn't a good option >>>> in this case, when I suggested it I wasn't looking at the big picture >>>> of how probing the core driver is triggered, sorry about that. >>>> >>>> Reading the speed from the DT with usb_get_maximum_speed() as Stephen >>>> suggests would be an option, the inconvenient is that we then >>>> essentially require the property to be defined, while the core driver >>>> gets a suitable value from hardware registers. Not sure if the wrapper >>>> driver could read from the same registers. >>>> >>>> One option could be to poll device_is_bound() for 100 ms (or so), with >>>> sleeps between polls. It's not elegant but would probably work if we >>>> don't find a better solution. >>> if (np) >>> ret = dwc3_qcom_of_register_core(pdev); >>> else >>> ret = dwc3_qcom_acpi_register_core(pdev); >>> >>> if (ret) { >>> dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); >>> goto depopulate; >>> } >>> >>> ret = dwc3_qcom_interconnect_init(qcom); >>> if (ret) >>> goto depopulate; >>> >>> qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev); >>> >>> Before calling dwc3_qcom_interconnect_init we are checking >>> >>> if (ret) { >>> dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); >>> goto depopulate; >>> } >>> >>> Doesn't this condition confirm the core driver is probed? >> Not really: >> >> // called under the hood by of_platform_populate() >> static int really_probe(struct device *dev, struct device_driver *drv) >> { >> ... >> >> if (dev->bus->probe) { >> ret = dev->bus->probe(dev); >> if (ret) >> goto probe_failed; >> } else if (drv->probe) { >> ret = drv->probe(dev); >> if (ret) >> goto probe_failed; >> } >> >> ... >> >> probe_failed: >> ... >> >> /* >> * Ignore errors returned by ->probe so that the next driver can try >> * its luck. >> */ >> ret = 0; >> >> ... >> >> return ret; >> } >> >> As a result of_platform_populate() in dwc3_qcom_of_register_core() >> returns 0 even when probing the device failed: >> >> [ 0.244339] dwc3-qcom a6f8800.usb: DBG: populate >> [ 0.244772] dwc3 a600000.dwc3: DBG: dwc3_probe >> [ 0.245237] dwc3 a600000.dwc3: DBG: dwc3_probe err: -517 >> [ 0.245264] dwc3-qcom a6f8800.usb: DBG: populate (done) >> [ 0.245317] dwc3-qcom a6f8800.usb: DBG: dwc3_qcom_interconnect_init() failed: -517 >> >> Probe fails because the interconnect stuff isn't ready yet, otherwise >> it could access invalid data. >> >> A later _populate() is successful and the probing of the core is done >> synchronously, i.e. after _populate() the core driver is fully >> initialized: >> >> [ 3.898106] dwc3-qcom a6f8800.usb: DBG: populate >> [ 3.908356] dwc3 a600000.dwc3: DBG: dwc3_probe >> [ 4.205104] dwc3 a600000.dwc3: DBG: dwc3_probe (done) >> [ 4.210305] dwc3-qcom a6f8800.usb: DBG: populate (done) >> >> The synchronous probing in _populate() suggests that using device_is_bound() >> would actually be a valid option, either the core device was successfully >> probed or not, there should be no race. >> >> I sent a patch that adds this check to dwc3_qcom_of_register_core(), which >> is less confusing and makes clear that the core device is valid unless >> this function returns an error: >> >> https://lore.kernel.org/patchwork/patch/1257279/ >> >> It might make sense to add your "driver core:Export the symbol >> device_is_bound" patch, mine and this one to a single series. > From the discussion on "driver core:Export the symbol device_is_bound" > (https://patchwork.kernel.org/patch/11584225/) it is clear that > this won't fly. The split dwc3 driver is considered a broken > design. > > This is what Rob Herring said: > > We never should have had this split either in the DT binding nor > driver(s) as if the SoC wrapper crap and licensed IP block are > independent things. The thing to do here is either make the DWC3 code > a library which drivers call (e.g. SDHCI) or add hooks into the DWC3 > driver for platform specifics (e.g. Designware PCI). Neither is a > simple solution though. > > That seems to be the desirable solution in the longer term, but it > doesn't seem reasonable to me to expect you to fix this design issue > to add interconnect support. > > Some possible options to move forward: > > - try to determine the max speed without involving the core device > - select a reasonable default when 'maximum-speed' is not specified > - use the core device to determine the max speed and pray Can we do as below to get speed qcom->max_speed = usb_get_maximum_speed(&qcom->dwc3->dev); as they were doing similarly in below code to get mode qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev);
On Tue, Jul 07, 2020 at 10:41:24AM +0530, Sandeep Maheswaram (Temp) wrote: > > On 7/1/2020 4:12 AM, Matthias Kaehlcke wrote: > > On Tue, Jun 16, 2020 at 01:38:49PM -0700, Matthias Kaehlcke wrote: > > > On Tue, Jun 16, 2020 at 10:22:47AM +0530, Sandeep Maheswaram (Temp) wrote: > > > > On 6/16/2020 1:12 AM, Matthias Kaehlcke wrote: > > > > > On Thu, Jun 04, 2020 at 04:16:31AM -0700, Stephen Boyd wrote: > > > > > > Quoting Sandeep Maheswaram (Temp) (2020-06-04 02:43:09) > > > > > > > On 6/3/2020 11:06 PM, Stephen Boyd wrote: > > > > > > > > Quoting Sandeep Maheswaram (2020-03-31 22:15:43) > > > > > > > > > diff --git a/drivers/usb/dwc3/dwc3-qcom.c b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > > > index 1dfd024..d33ae86 100644 > > > > > > > > > --- a/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > > > +++ b/drivers/usb/dwc3/dwc3-qcom.c > > > > > > > > > @@ -285,6 +307,101 @@ static int dwc3_qcom_resume(struct dwc3_qcom *qcom) > > > > > > > > > return 0; > > > > > > > > > } > > > > > > > > > + > > > > > > > > > +/** > > > > > > > > > + * dwc3_qcom_interconnect_init() - Get interconnect path handles > > > > > > > > > + * @qcom: Pointer to the concerned usb core. > > > > > > > > > + * > > > > > > > > > + */ > > > > > > > > > +static int dwc3_qcom_interconnect_init(struct dwc3_qcom *qcom) > > > > > > > > > +{ > > > > > > > > > + struct device *dev = qcom->dev; > > > > > > > > > + int ret; > > > > > > > > > + > > > > > > > > > + if (!device_is_bound(&qcom->dwc3->dev)) > > > > > > > > > + return -EPROBE_DEFER; > > > > > > > > How is this supposed to work? I see that this was added in an earlier > > > > > > > > revision of this patch series but there isn't any mention of why > > > > > > > > device_is_bound() is used here. It would be great if there was a comment > > > > > > > > detailing why this is necessary. It sounds like maximum_speed is > > > > > > > > important? > > > > > > > > > > > > > > > > Furthermore, dwc3_qcom_interconnect_init() is called by > > > > > > > > dwc3_qcom_probe() which is the function that registers the device for > > > > > > > > qcom->dwc3->dev. If that device doesn't probe between the time it is > > > > > > > > registered by dwc3_qcom_probe() and this function is called then we'll > > > > > > > > fail dwc3_qcom_probe() with -EPROBE_DEFER. And that will remove the > > > > > > > > qcom->dwc3->dev device from the platform bus because we call > > > > > > > > of_platform_depopulate() on the error path of dwc3_qcom_probe(). > > > > > > > > > > > > > > > > So isn't this whole thing racy and can potentially lead us to a driver > > > > > > > > probe loop where the wrapper (dwc3_qcom) and the core (dwc3) are probing > > > > > > > > and we're trying to time it just right so that driver for dwc3 binds > > > > > > > > before we setup interconnects? I don't know if dwc3 can communicate to > > > > > > > > the wrapper but that would be more of a direct way to do this. Or maybe > > > > > > > > the wrapper should try to read the DT property for maximum speed and > > > > > > > > fallback to a worst case high bandwidth value if it can't figure it out > > > > > > > > itself without help from dwc3 core. > > > > > > > > > > > > > > > This was added in V4 to address comments from Matthias in V3 > > > > > > > > > > > > > > https://patchwork.kernel.org/patch/11148587/ > > > > > > > > > > > > > Yes, that why I said: > > > > > > > > > > > > "I see that this was added in an earlier > > > > > > revision of this patch series but there isn't any mention of why > > > > > > device_is_bound() is used here. It would be great if there was a comment > > > > > > detailing why this is necessary. It sounds like maximum_speed is > > > > > > important?" > > > > > > > > > > > > Can you please respond to the rest of my email? > > > > > I agree with Stephen that using device_is_bound() isn't a good option > > > > > in this case, when I suggested it I wasn't looking at the big picture > > > > > of how probing the core driver is triggered, sorry about that. > > > > > > > > > > Reading the speed from the DT with usb_get_maximum_speed() as Stephen > > > > > suggests would be an option, the inconvenient is that we then > > > > > essentially require the property to be defined, while the core driver > > > > > gets a suitable value from hardware registers. Not sure if the wrapper > > > > > driver could read from the same registers. > > > > > > > > > > One option could be to poll device_is_bound() for 100 ms (or so), with > > > > > sleeps between polls. It's not elegant but would probably work if we > > > > > don't find a better solution. > > > > if (np) > > > > ret = dwc3_qcom_of_register_core(pdev); > > > > else > > > > ret = dwc3_qcom_acpi_register_core(pdev); > > > > > > > > if (ret) { > > > > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > > > > goto depopulate; > > > > } > > > > > > > > ret = dwc3_qcom_interconnect_init(qcom); > > > > if (ret) > > > > goto depopulate; > > > > > > > > qcom->mode = usb_get_dr_mode(&qcom->dwc3->dev); > > > > > > > > Before calling dwc3_qcom_interconnect_init we are checking > > > > > > > > if (ret) { > > > > dev_err(dev, "failed to register DWC3 Core, err=%d\n", ret); > > > > goto depopulate; > > > > } > > > > > > > > Doesn't this condition confirm the core driver is probed? > > > Not really: > > > > > > // called under the hood by of_platform_populate() > > > static int really_probe(struct device *dev, struct device_driver *drv) > > > { > > > ... > > > > > > if (dev->bus->probe) { > > > ret = dev->bus->probe(dev); > > > if (ret) > > > goto probe_failed; > > > } else if (drv->probe) { > > > ret = drv->probe(dev); > > > if (ret) > > > goto probe_failed; > > > } > > > > > > ... > > > > > > probe_failed: > > > ... > > > > > > /* > > > * Ignore errors returned by ->probe so that the next driver can try > > > * its luck. > > > */ > > > ret = 0; > > > > > > ... > > > > > > return ret; > > > } > > > > > > As a result of_platform_populate() in dwc3_qcom_of_register_core() > > > returns 0 even when probing the device failed: > > > > > > [ 0.244339] dwc3-qcom a6f8800.usb: DBG: populate > > > [ 0.244772] dwc3 a600000.dwc3: DBG: dwc3_probe > > > [ 0.245237] dwc3 a600000.dwc3: DBG: dwc3_probe err: -517 > > > [ 0.245264] dwc3-qcom a6f8800.usb: DBG: populate (done) > > > [ 0.245317] dwc3-qcom a6f8800.usb: DBG: dwc3_qcom_interconnect_init() failed: -517 > > > > > > Probe fails because the interconnect stuff isn't ready yet, otherwise > > > it could access invalid data. > > > > > > A later _populate() is successful and the probing of the core is done > > > synchronously, i.e. after _populate() the core driver is fully > > > initialized: > > > > > > [ 3.898106] dwc3-qcom a6f8800.usb: DBG: populate > > > [ 3.908356] dwc3 a600000.dwc3: DBG: dwc3_probe > > > [ 4.205104] dwc3 a600000.dwc3: DBG: dwc3_probe (done) > > > [ 4.210305] dwc3-qcom a6f8800.usb: DBG: populate (done) > > > > > > The synchronous probing in _populate() suggests that using device_is_bound() > > > would actually be a valid option, either the core device was successfully > > > probed or not, there should be no race. > > > > > > I sent a patch that adds this check to dwc3_qcom_of_register_core(), which > > > is less confusing and makes clear that the core device is valid unless > > > this function returns an error: > > > > > > https://lore.kernel.org/patchwork/patch/1257279/ > > > > > > It might make sense to add your "driver core:Export the symbol > > > device_is_bound" patch, mine and this one to a single series. > > From the discussion on "driver core:Export the symbol device_is_bound" > > (https://patchwork.kernel.org/patch/11584225/) it is clear that > > this won't fly. The split dwc3 driver is considered a broken > > design. > > > > This is what Rob Herring said: > > > > We never should have had this split either in the DT binding nor > > driver(s) as if the SoC wrapper crap and licensed IP block are > > independent things. The thing to do here is either make the DWC3 code > > a library which drivers call (e.g. SDHCI) or add hooks into the DWC3 > > driver for platform specifics (e.g. Designware PCI). Neither is a > > simple solution though. > > > > That seems to be the desirable solution in the longer term, but it > > doesn't seem reasonable to me to expect you to fix this design issue > > to add interconnect support. > > > > Some possible options to move forward: > > > > - try to determine the max speed without involving the core device > > - select a reasonable default when 'maximum-speed' is not specified > > - use the core device to determine the max speed and pray > > Can we do as below to get speed > > qcom->max_speed = usb_get_maximum_speed(&qcom->dwc3->dev); Yes, that would get the maximum speed from the DT if it is specified. In case of USB_SPEED_UNKNOWN you probably want to assume it's USB_SPEED_SUPER, which in the worst case would result in the ICC running at a higher speed than needed.