| Message ID | 20221114-narmstrong-sm8550-upstream-i2c-master-hub-v3-0-f6a20dc9996e@linaro.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | soc: qcom: add support for the I2C Master Hub | expand |
On Tue, Nov 29, 2022 at 03:47:00PM +0100, Neil Armstrong wrote: > The I2C Master Hub is a stripped down version of the GENI Serial Engine > QUP Wrapper Controller but only supporting I2C serial engines without > DMA support. > > The I2C Master Hub only supports a variant of the I2C serial engine with: > - a separate "core" clock > - no DMA support > - non discoverable fixed FIFO size > > Since DMA isn't supported, the wrapper doesn't need the Master AHB clock > and the iommus property neither. > > This patchset adds the bindings changes to the QUPv3 wrapper and I2C serial > element bindings to reflect the different resources requirements. > > In order to reuse the QUPv3 wrapper and I2C serial element driver support, > the I2C Master Hub requirements are expressed in new desc structs passed > as device match data. Is everyone fine if I take all this via the I2C tree?
On Thu, Dec 01, 2022 at 11:49:21PM +0100, Wolfram Sang wrote: > On Tue, Nov 29, 2022 at 03:47:00PM +0100, Neil Armstrong wrote: > > The I2C Master Hub is a stripped down version of the GENI Serial Engine > > QUP Wrapper Controller but only supporting I2C serial engines without > > DMA support. > > > > The I2C Master Hub only supports a variant of the I2C serial engine with: > > - a separate "core" clock > > - no DMA support > > - non discoverable fixed FIFO size > > > > Since DMA isn't supported, the wrapper doesn't need the Master AHB clock > > and the iommus property neither. > > > > This patchset adds the bindings changes to the QUPv3 wrapper and I2C serial > > element bindings to reflect the different resources requirements. > > > > In order to reuse the QUPv3 wrapper and I2C serial element driver support, > > the I2C Master Hub requirements are expressed in new desc structs passed > > as device match data. > > Is everyone fine if I take all this via the I2C tree? I did this now. All applied to for-next, thanks!
On 05/12/2022 09:31, Wolfram Sang wrote: > On Thu, Dec 01, 2022 at 11:49:21PM +0100, Wolfram Sang wrote: >> On Tue, Nov 29, 2022 at 03:47:00PM +0100, Neil Armstrong wrote: >>> The I2C Master Hub is a stripped down version of the GENI Serial Engine >>> QUP Wrapper Controller but only supporting I2C serial engines without >>> DMA support. >>> >>> The I2C Master Hub only supports a variant of the I2C serial engine with: >>> - a separate "core" clock >>> - no DMA support >>> - non discoverable fixed FIFO size >>> >>> Since DMA isn't supported, the wrapper doesn't need the Master AHB clock >>> and the iommus property neither. >>> >>> This patchset adds the bindings changes to the QUPv3 wrapper and I2C serial >>> element bindings to reflect the different resources requirements. >>> >>> In order to reuse the QUPv3 wrapper and I2C serial element driver support, >>> the I2C Master Hub requirements are expressed in new desc structs passed >>> as device match data. >> >> Is everyone fine if I take all this via the I2C tree? > > I did this now. > > All applied to for-next, thanks! > Thanks ! Neil
The I2C Master Hub is a stripped down version of the GENI Serial Engine QUP Wrapper Controller but only supporting I2C serial engines without DMA support. The I2C Master Hub only supports a variant of the I2C serial engine with: - a separate "core" clock - no DMA support - non discoverable fixed FIFO size Since DMA isn't supported, the wrapper doesn't need the Master AHB clock and the iommus property neither. This patchset adds the bindings changes to the QUPv3 wrapper and I2C serial element bindings to reflect the different resources requirements. In order to reuse the QUPv3 wrapper and I2C serial element driver support, the I2C Master Hub requirements are expressed in new desc structs passed as device match data. To: Andy Gross <agross@kernel.org> To: Bjorn Andersson <andersson@kernel.org> Cc: Konrad Dybcio <konrad.dybcio@linaro.org> To: Rob Herring <robh+dt@kernel.org> To: Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org> Cc: linux-arm-msm@vger.kernel.org Cc: devicetree@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-i2c@vger.kernel.org Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org> --- Changes in v3: - Fixed check of DT clocks count on geni-se.c with Kondar help - Added Krzysztof's Reviewed-by on patches 1 & 2 - Link to v2: https://lore.kernel.org/r/20221114-narmstrong-sm8550-upstream-i2c-master-hub-v2-0-aadaa6997b28@linaro.org Changes in v2: - Fixed all commits messages to remove "This" and fix grammar - Fixed the bindings by moving the if in allOf:if - Fixed the bindings by adding minItems: & maxItems: instead of true - Added a warning about clock count in patch 3 - Added Reviewed-by from Konrad on patches 3, 4 & 5 - Link to v1: https://lore.kernel.org/r/20221114-narmstrong-sm8550-upstream-i2c-master-hub-v1-0-64449106a148@linaro.org --- Neil Armstrong (6): dt-bindings: qcom: geni-se: document I2C Master Hub wrapper variant dt-bindings: i2c: qcom-geni: document I2C Master Hub serial I2C engine soc: qcom: geni-se: add desc struct to specify clocks from device match data soc: qcom: geni-se: add support for I2C Master Hub wrapper variant i2c: qcom-geni: add desc struct to prepare support for I2C Master Hub variant i2c: qcom-geni: add support for I2C Master Hub variant .../bindings/i2c/qcom,i2c-geni-qcom.yaml | 64 +++++++++++++++--- .../devicetree/bindings/soc/qcom/qcom,geni-se.yaml | 44 ++++++++++-- drivers/i2c/busses/i2c-qcom-geni.c | 58 +++++++++++++++- drivers/soc/qcom/qcom-geni-se.c | 79 ++++++++++++++++++---- 4 files changed, 212 insertions(+), 33 deletions(-) --- base-commit: 094226ad94f471a9f19e8f8e7140a09c2625abaa change-id: 20221114-narmstrong-sm8550-upstream-i2c-master-hub-44a7fb19475e Best regards,