Message ID | 1698835699-28550-1-git-send-email-quic_ziqichen@quicinc.com |
---|---|
State | Superseded |
Headers | show |
Series | dt-bindings: ufs: Add msi-parent for UFS MCQ | expand |
On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: > The Message Signaled Interrupts (MSI) has been introduced > to UFS driver since the MCQ be enabled. Not really relevant when a driver supported MSI, but the when the h/w did. Has UFS always supported MSI? It was added in some version of the spec? > Hence in UFS DT > node we need to give the msi-parent property that point > to the hardware entity which serves as the MSI controller > for this UFS controller. > > Signed-off-by: Ziqi Chen <quic_ziqichen@quicinc.com> > --- > Documentation/devicetree/bindings/ufs/ufs-common.yaml | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/Documentation/devicetree/bindings/ufs/ufs-common.yaml b/Documentation/devicetree/bindings/ufs/ufs-common.yaml > index bbaee4f5..42309bb 100644 > --- a/Documentation/devicetree/bindings/ufs/ufs-common.yaml > +++ b/Documentation/devicetree/bindings/ufs/ufs-common.yaml > @@ -73,6 +73,8 @@ properties: > description: > Specifies max. load that can be drawn from VCCQ2 supply. > > + msi-parent: true > + > dependencies: > freq-table-hz: [ clocks ] > > -- > 2.7.4 >
On 11/6/23 06:48, Rob Herring wrote: > On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: >> The Message Signaled Interrupts (MSI) has been introduced >> to UFS driver since the MCQ be enabled. > > Not really relevant when a driver supported MSI, but the when the h/w > did. Has UFS always supported MSI? It was added in some version of the > spec? MSI support has been introduced in UFSHCI version 4.0 and I think that the controller vendor can decide whether or not to implement MSI. Does this mean that the patch needs to be improved? Thanks, Bart.
On 11/7/2023 1:56 AM, Bart Van Assche wrote: > On 11/6/23 06:48, Rob Herring wrote: >> On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: >>> The Message Signaled Interrupts (MSI) has been introduced >>> to UFS driver since the MCQ be enabled. >> >> Not really relevant when a driver supported MSI, but the when the h/w >> did. Has UFS always supported MSI? It was added in some version of the >> spec? > > MSI support has been introduced in UFSHCI version 4.0 and I think that > the controller vendor can decide whether or not to implement MSI. Does > this mean that the patch needs to be improved? > > Thanks, > > Bart. > Hi Rob, in my previous patch " dt-bindings: ufs: qcom: Add msi-parent for UFS MCQ", Manivannan Sadhasivam also think instead of qcom,ufs.yaml, this should be part of ufs-common.yaml. please let me know if there is any others I need to improve for this patch. Thanks a lot. Best Regards, Ziqi
On Mon, Nov 6, 2023 at 11:56 AM Bart Van Assche <bvanassche@acm.org> wrote: > > On 11/6/23 06:48, Rob Herring wrote: > > On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: > >> The Message Signaled Interrupts (MSI) has been introduced > >> to UFS driver since the MCQ be enabled. > > > > Not really relevant when a driver supported MSI, but the when the h/w > > did. Has UFS always supported MSI? It was added in some version of the > > spec? > > MSI support has been introduced in UFSHCI version 4.0 and I think that > the controller vendor can decide whether or not to implement MSI. Does > this mean that the patch needs to be improved? Yes, this information is what should be in the commit msg rather than driver details. Rob
On Thu, Nov 09, 2023 at 07:55:14AM -0600, Rob Herring wrote: > On Mon, Nov 6, 2023 at 11:56 AM Bart Van Assche <bvanassche@acm.org> wrote: > > > > On 11/6/23 06:48, Rob Herring wrote: > > > On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: > > >> The Message Signaled Interrupts (MSI) has been introduced > > >> to UFS driver since the MCQ be enabled. > > > > > > Not really relevant when a driver supported MSI, but the when the h/w > > > did. Has UFS always supported MSI? It was added in some version of the > > > spec? > > > > MSI support has been introduced in UFSHCI version 4.0 and I think that > > the controller vendor can decide whether or not to implement MSI. Does > > this mean that the patch needs to be improved? > > Yes, this information is what should be in the commit msg rather than > driver details. > Yes, agreed. Ziqi, please update the commit message to incorporate the hw details about when MCQ/MSI got introduced. Devicetree binding should describe the hw, not the driver. - Mani > Rob
On 11/10/2023 12:16 AM, Manivannan Sadhasivam wrote: > On Thu, Nov 09, 2023 at 07:55:14AM -0600, Rob Herring wrote: >> On Mon, Nov 6, 2023 at 11:56 AM Bart Van Assche <bvanassche@acm.org> wrote: >>> >>> On 11/6/23 06:48, Rob Herring wrote: >>>> On Wed, Nov 01, 2023 at 06:48:13PM +0800, Ziqi Chen wrote: >>>>> The Message Signaled Interrupts (MSI) has been introduced >>>>> to UFS driver since the MCQ be enabled. >>>> >>>> Not really relevant when a driver supported MSI, but the when the h/w >>>> did. Has UFS always supported MSI? It was added in some version of the >>>> spec? >>> >>> MSI support has been introduced in UFSHCI version 4.0 and I think that >>> the controller vendor can decide whether or not to implement MSI. Does >>> this mean that the patch needs to be improved? >> >> Yes, this information is what should be in the commit msg rather than >> driver details. >> > > Yes, agreed. Ziqi, please update the commit message to incorporate the hw > details about when MCQ/MSI got introduced. Devicetree binding should describe > the hw, not the driver. > Thank you all, I will update the commit message in next version. - Ziqi > - Mani > >> Rob >
On 01/11/2023 11:48, Ziqi Chen wrote: > The Message Signaled Interrupts (MSI) has been introduced > to UFS driver since the MCQ be enabled. Hence in UFS DT > node we need to give the msi-parent property that point > to the hardware entity which serves as the MSI controller > for this UFS controller. > > Signed-off-by: Ziqi Chen <quic_ziqichen@quicinc.com> Which tree or next did you use as base for this patch? It does not apply, neither on October's next nor on current. It does not apply on v6.7-rc1 either... Best regards, Krzysztof
diff --git a/Documentation/devicetree/bindings/ufs/ufs-common.yaml b/Documentation/devicetree/bindings/ufs/ufs-common.yaml index bbaee4f5..42309bb 100644 --- a/Documentation/devicetree/bindings/ufs/ufs-common.yaml +++ b/Documentation/devicetree/bindings/ufs/ufs-common.yaml @@ -73,6 +73,8 @@ properties: description: Specifies max. load that can be drawn from VCCQ2 supply. + msi-parent: true + dependencies: freq-table-hz: [ clocks ]
The Message Signaled Interrupts (MSI) has been introduced to UFS driver since the MCQ be enabled. Hence in UFS DT node we need to give the msi-parent property that point to the hardware entity which serves as the MSI controller for this UFS controller. Signed-off-by: Ziqi Chen <quic_ziqichen@quicinc.com> --- Documentation/devicetree/bindings/ufs/ufs-common.yaml | 2 ++ 1 file changed, 2 insertions(+)