| Message ID | 20240112-opp_support-v6-0-77bbf7d0cc37@quicinc.com |
|---|---|
| Headers | show |
| Series | PCI: qcom: Add support for OPP | expand |
On 12/01/2024 14:22, Krishna chaitanya chundru wrote: > CPU-PCIe path consits for registers PCIe BAR space, config space. > As there is less access on this path compared to pcie to mem path > add minimum vote i.e GEN1x1 bandwidth always. > > In suspend remove the cpu vote after register space access is done. > > Fixes: c4860af88d0c ("PCI: qcom: Add basic interconnect support") If this patch is a Fixes then don't you need the accompanying dts change as a parallel Fixes too ? i.e. without the dts update - you won't have the nodes in the dts to consume => applying this code to the stable kernel absent the dts will result in no functional change and therefore no bugfix. I'm not sure if you are asked to put a Fixes here but it seems to be it should either be dropped or require a parallel Fixes: tag for the dts and yaml changes. What is the bug this change fixes in the backport ? > cc: stable@vger.kernel.org > Signed-off-by: Krishna chaitanya chundru <quic_krichai@quicinc.com> --- bod
On Fri, Jan 12, 2024 at 07:52:00PM +0530, Krishna chaitanya chundru wrote: > Add the interconnects path as required property for sm8450 platform. There's no explaination here as to why you need two different compatibles for the instances on this device. Please add one. Thanks, Conor. > > Signed-off-by: Krishna chaitanya chundru <quic_krichai@quicinc.com> > --- > Documentation/devicetree/bindings/pci/qcom,pcie.yaml | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml > index eadba38171e1..bc28669f6fa0 100644 > --- a/Documentation/devicetree/bindings/pci/qcom,pcie.yaml > +++ b/Documentation/devicetree/bindings/pci/qcom,pcie.yaml > @@ -777,6 +777,8 @@ allOf: > - qcom,pcie-sa8540p > - qcom,pcie-sa8775p > - qcom,pcie-sc8280xp > + - qcom,pcie-sm8450-pcie0 > + - qcom,pcie-sm8450-pcie1 > then: > required: > - interconnects > > -- > 2.42.0 >
On 12.01.2024 16:17, Bryan O'Donoghue wrote: > On 12/01/2024 14:22, Krishna chaitanya chundru wrote: >> CPU-PCIe path consits for registers PCIe BAR space, config space. >> As there is less access on this path compared to pcie to mem path >> add minimum vote i.e GEN1x1 bandwidth always. >> >> In suspend remove the cpu vote after register space access is done. >> >> Fixes: c4860af88d0c ("PCI: qcom: Add basic interconnect support") > > If this patch is a Fixes then don't you need the accompanying dts change as a parallel Fixes too ? > > i.e. without the dts update - you won't have the nodes in the dts to consume => applying this code to the stable kernel absent the dts will result in no functional change and therefore no bugfix. The Fixes tag denotes a bug fix, its use for backport autosel is just a nice "coincidence". Fixing a lack of a required icc path and having to rely on BL leftovers / keepalive bus settings is definitely worth this tag in my eyes. Konrad
On 1/12/2024 8:47 PM, Bryan O'Donoghue wrote: > On 12/01/2024 14:22, Krishna chaitanya chundru wrote: >> CPU-PCIe path consits for registers PCIe BAR space, config space. >> As there is less access on this path compared to pcie to mem path >> add minimum vote i.e GEN1x1 bandwidth always. >> >> In suspend remove the cpu vote after register space access is done. >> >> Fixes: c4860af88d0c ("PCI: qcom: Add basic interconnect support") > > If this patch is a Fixes then don't you need the accompanying dts change > as a parallel Fixes too ? > > i.e. without the dts update - you won't have the nodes in the dts to > consume => applying this code to the stable kernel absent the dts will > result in no functional change and therefore no bugfix. > > I'm not sure if you are asked to put a Fixes here but it seems to be it > should either be dropped or require a parallel Fixes: tag for the dts > and yaml changes. > > What is the bug this change fixes in the backport ? > There is no change required in the dts because the cpu-pcie path is already present in the dts. So till now driver is ignoring that path, that's why we tagged with fixed. -Krishna Chaitanya >> cc: stable@vger.kernel.org >> Signed-off-by: Krishna chaitanya chundru <quic_krichai@quicinc.com> > > --- > bod
On 16/01/2024 04:52, Krishna Chaitanya Chundru wrote: >> > There is no change required in the dts because the cpu-pcie path is > already present in the dts. Not at c4860af88d0cb1bb006df12615c5515ae509f73b its not, those dts entries get added later. But anyway re-reading your commit log "vote for minimum bandwidth as at c4860af88d0cb1bb006df12615c5515ae509f73b" makes sense to me. Reviewed-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
On Fri, Jan 12, 2024 at 11:33:15PM +0100, Konrad Dybcio wrote: > On 12.01.2024 16:17, Bryan O'Donoghue wrote: > > On 12/01/2024 14:22, Krishna chaitanya chundru wrote: > >> CPU-PCIe path consits for registers PCIe BAR space, config space. > >> As there is less access on this path compared to pcie to mem path > >> add minimum vote i.e GEN1x1 bandwidth always. > >> > >> In suspend remove the cpu vote after register space access is done. > >> > >> Fixes: c4860af88d0c ("PCI: qcom: Add basic interconnect support") > > > > If this patch is a Fixes then don't you need the accompanying dts change as a parallel Fixes too ? > > > > i.e. without the dts update - you won't have the nodes in the dts to consume => applying this code to the stable kernel absent the dts will result in no functional change and therefore no bugfix. > > The Fixes tag denotes a bug fix, its use for backport autosel is just > a nice "coincidence". > > Fixing a lack of a required icc path and having to rely on BL leftovers > / keepalive bus settings is definitely worth this tag in my eyes. An incomplete implementation can sometimes be considered a bug, but not always. If this is needed to enable a new use case, then it's hard to argue that the original omission was a bug. And as I just mentioned to Krishna, the above Fixes tag is not correct as that commit did not *introduce* any issue. It solved the bit that was strictly needed for sc8280xp, but now it seems you may need something more for an even newer platform (even if no details and motivation was included in the commit message as it should have been). Johan
On 1/16/24 11:52, Johan Hovold wrote: > On Fri, Jan 12, 2024 at 11:33:15PM +0100, Konrad Dybcio wrote: >> On 12.01.2024 16:17, Bryan O'Donoghue wrote: >>> On 12/01/2024 14:22, Krishna chaitanya chundru wrote: >>>> CPU-PCIe path consits for registers PCIe BAR space, config space. >>>> As there is less access on this path compared to pcie to mem path >>>> add minimum vote i.e GEN1x1 bandwidth always. >>>> >>>> In suspend remove the cpu vote after register space access is done. >>>> >>>> Fixes: c4860af88d0c ("PCI: qcom: Add basic interconnect support") >>> >>> If this patch is a Fixes then don't you need the accompanying dts change as a parallel Fixes too ? >>> >>> i.e. without the dts update - you won't have the nodes in the dts to consume => applying this code to the stable kernel absent the dts will result in no functional change and therefore no bugfix. >> >> The Fixes tag denotes a bug fix, its use for backport autosel is just >> a nice "coincidence". >> >> Fixing a lack of a required icc path and having to rely on BL leftovers >> / keepalive bus settings is definitely worth this tag in my eyes. > > An incomplete implementation can sometimes be considered a bug, but not > always. If this is needed to enable a new use case, then it's hard to > argue that the original omission was a bug. > > And as I just mentioned to Krishna, the above Fixes tag is not correct > as that commit did not *introduce* any issue. It solved the bit that was > strictly needed for sc8280xp, but now it seems you may need something > more for an even newer platform (even if no details and motivation was > included in the commit message as it should have been). The PCIe hardware seems to be piggybacking off of others' bus bandwidth requests and I think it's just been pure luck that it didn't simply refuse to work on previous generations. So indeed, the commit message seems incomplete in explaining where the problem lies Konrad
This patch adds support for OPP to vote for the performance state of RPMH power domain based upon GEN speed it PCIe got enumerated. QCOM Resource Power Manager-hardened (RPMh) is a hardware block which maintains hardware state of a regulator by performing max aggregation of the requests made by all of the processors. PCIe controller can operate on different RPMh performance state of power domain based up on the speed of the link. And this performance state varies from target to target. It is manadate to scale the performance state based up on the PCIe speed link operates so that SoC can run under optimum power conditions. Add Operating Performance Points(OPP) support to vote for RPMh state based upon GEN speed link is operating. Before link up PCIe driver will vote for the maximum performance state. As now we are adding ICC BW vote in OPP, the ICC BW voting depends both GEN speed and link width using opp-level to indicate the opp entry table will be difficult. In PCIe certain gen speeds like GEN1x2 & GEN2X1 or GEN3x2 & GEN4x1 use same icc bw if we use freq in the opp table to represent the PCIe Gen speed number of PCIe entries can reduced. So going back to use freq in the opp table instead of level. Signed-off-by: Krishna chaitanya chundru <quic_krichai@quicinc.com> --- Changes frm v5: - Add ICC BW voting as part of OPP, rebase the latest kernel, and only - either OPP or ICC BW voting will supported we removed the patch to - return eror for icc opp update patch. - As we added the icc bw voting in opp table I am not including reviewed - by tags given in previous patch. - Use opp freq to find opp entries as now we need to include pcie link - also in to considerations. - Add CPU-PCIe BW voting which is not present till now. - Drop PCI: qcom: Return error from 'qcom_pcie_icc_update' as either opp or icc bw - only one executes and there is no need to fail if opp or icc update fails. - Link for v5: https://lore.kernel.org/linux-arm-msm/20231101063323.GH2897@thinkpad/T/ Changes from v4: - Added a separate patch for returning error from the qcom_pcie_upadate and moved opp update logic to icc_update and used a bool variable to update the opp. - Addressed comments made by pavan. changes from v3: - Removing the opp vote on suspend when the link is not up and link is not up and add debug prints as suggested by pavan. - Added dev_pm_opp_find_level_floor API to find the highest opp to vote. changes from v2: - Instead of using the freq based opp search use level based as suggested by Dmitry Baryshkov. Changes from v1: - Addressed comments from Krzysztof Kozlowski. - Added the rpmhpd_opp_xxx phandle as suggested by pavan. - Added dev_pm_opp_set_opp API call which was missed on previous patch. --- Krishna chaitanya chundru (6): dt-bindings: PCI: qcom: Add interconnects path as required property arm64: dts: qcom: sm8450: Add interconnect path to PCIe node PCI: qcom: Add missing icc bandwidth vote for cpu to PCIe path dt-bindings: pci: qcom: Add opp table arm64: dts: qcom: sm8450: Add opp table support to PCIe PCI: qcom: Add OPP support to scale performance state of power domain .../devicetree/bindings/pci/qcom,pcie.yaml | 6 ++ arch/arm64/boot/dts/qcom/sm8450.dtsi | 82 +++++++++++++++ drivers/pci/controller/dwc/pcie-qcom.c | 114 ++++++++++++++++++--- 3 files changed, 187 insertions(+), 15 deletions(-) --- base-commit: 70d201a40823acba23899342d62bc2644051ad2e change-id: 20240112-opp_support-3a1839c6a171 Best regards,