diff mbox series

[2/3] arm64: dts: qcom: sa8775p: Add CPU OPP tables to scale DDR/L3

Message ID 20241017-sa8775p-cpufreq-l3-ddr-scaling-v1-2-074e0fb80b33@quicinc.com
State New
Headers show
Series Add support to scale DDR and L3 on SA8775P | expand

Commit Message

Jagadeesh Kona Oct. 17, 2024, 9:28 a.m. UTC
From: Shivnandan Kumar <quic_kshivnan@quicinc.com>

Add OPP tables required to scale DDR and L3 per freq-domain
on SA8775P platform.

Signed-off-by: Shivnandan Kumar <quic_kshivnan@quicinc.com>
Signed-off-by: Jagadeesh Kona <quic_jkona@quicinc.com>
---
 arch/arm64/boot/dts/qcom/sa8775p.dtsi | 178 ++++++++++++++++++++++++++++++++++
 1 file changed, 178 insertions(+)

Comments

Jagadeesh Kona Nov. 11, 2024, 1:09 p.m. UTC | #1
On 10/17/2024 9:12 PM, Brian Masney wrote:
> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
>> +	cpu0_opp_table: opp-table-cpu0 {
>> +		compatible = "operating-points-v2";
>> +		opp-shared;
>> +
>> +		cpu0_opp_1267mhz: opp-1267200000 {
>> +			opp-hz = /bits/ 64 <1267200000>;
>> +			opp-peak-kBps = <6220800 29491200>;
>> +		};
>> +
>> +		cpu0_opp_1363mhz: opp-1363200000 {
>> +			opp-hz = /bits/ 64 <1363200000>;
>> +			opp-peak-kBps = <6220800 29491200>;
>> +		};
> 
> [snip]
> 
>> +	cpu4_opp_table: opp-table-cpu4 {
>> +		compatible = "operating-points-v2";
>> +		opp-shared;
>> +
>> +		cpu4_opp_1267mhz: opp-1267200000 {
>> +			opp-hz = /bits/ 64 <1267200000>;
>> +			opp-peak-kBps = <6220800 29491200>;
>> +		};
>> +
>> +		cpu4_opp_1363mhz: opp-1363200000 {
>> +			opp-hz = /bits/ 64 <1363200000>;
>> +			opp-peak-kBps = <6220800 29491200>;
>> +		};
> 
> There's no functional differences in the cpu0 and cpu4 opp tables. Can
> a single table be used?
> 
> This aligns with my recollection that this particular SoC only has the
> gold cores.
> 
> Brian
> 

Thanks Brian for your review. Sorry for the delayed response.

We require separate OPP tables for CPU0 and CPU4 to allow independent
scaling of DDR and L3 frequencies for each CPU domain, with the final
DDR and L3 frequencies being an aggregate of both.

If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
won't be invoked for CPU4. As a result both CPU devices will end up in sharing
the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
votes of other.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1588

Thanks,
Jagadeesh
Konrad Dybcio Nov. 30, 2024, 2:32 p.m. UTC | #2
On 14.11.2024 11:48 PM, Dmitry Baryshkov wrote:
> On Mon, Nov 11, 2024 at 06:39:48PM +0530, Jagadeesh Kona wrote:
>>
>>
>> On 10/17/2024 9:12 PM, Brian Masney wrote:
>>> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
>>>> +	cpu0_opp_table: opp-table-cpu0 {
>>>> +		compatible = "operating-points-v2";
>>>> +		opp-shared;
>>>> +
>>>> +		cpu0_opp_1267mhz: opp-1267200000 {
>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>> +
>>>> +		cpu0_opp_1363mhz: opp-1363200000 {
>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>
>>> [snip]
>>>
>>>> +	cpu4_opp_table: opp-table-cpu4 {
>>>> +		compatible = "operating-points-v2";
>>>> +		opp-shared;
>>>> +
>>>> +		cpu4_opp_1267mhz: opp-1267200000 {
>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>> +
>>>> +		cpu4_opp_1363mhz: opp-1363200000 {
>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>
>>> There's no functional differences in the cpu0 and cpu4 opp tables. Can
>>> a single table be used?
>>>
>>> This aligns with my recollection that this particular SoC only has the
>>> gold cores.
>>>
>>> Brian
>>>
>>
>> Thanks Brian for your review. Sorry for the delayed response.
>>
>> We require separate OPP tables for CPU0 and CPU4 to allow independent
>> scaling of DDR and L3 frequencies for each CPU domain, with the final
>> DDR and L3 frequencies being an aggregate of both.
>>
>> If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
>> won't be invoked for CPU4. As a result both CPU devices will end up in sharing
>> the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
>> votes of other.

Oh that's a fun find.. clocks get the same treatment.. very bad,
but may explain some schroedingerbugs.

Taking a peek at some code paths, wouldn't dropping opp-shared
solve our issues? dev_pm_opp_set_sharing_cpus() overrides it

Konrad
Jagadeesh Kona Dec. 3, 2024, 3:03 p.m. UTC | #3
On 11/15/2024 4:18 AM, Dmitry Baryshkov wrote:
> On Mon, Nov 11, 2024 at 06:39:48PM +0530, Jagadeesh Kona wrote:
>>
>>
>> On 10/17/2024 9:12 PM, Brian Masney wrote:
>>> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
>>>> +	cpu0_opp_table: opp-table-cpu0 {
>>>> +		compatible = "operating-points-v2";
>>>> +		opp-shared;
>>>> +
>>>> +		cpu0_opp_1267mhz: opp-1267200000 {
>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>> +
>>>> +		cpu0_opp_1363mhz: opp-1363200000 {
>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>
>>> [snip]
>>>
>>>> +	cpu4_opp_table: opp-table-cpu4 {
>>>> +		compatible = "operating-points-v2";
>>>> +		opp-shared;
>>>> +
>>>> +		cpu4_opp_1267mhz: opp-1267200000 {
>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>> +
>>>> +		cpu4_opp_1363mhz: opp-1363200000 {
>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>> +		};
>>>
>>> There's no functional differences in the cpu0 and cpu4 opp tables. Can
>>> a single table be used?
>>>
>>> This aligns with my recollection that this particular SoC only has the
>>> gold cores.
>>>
>>> Brian
>>>
>>
>> Thanks Brian for your review. Sorry for the delayed response.
>>
>> We require separate OPP tables for CPU0 and CPU4 to allow independent
>> scaling of DDR and L3 frequencies for each CPU domain, with the final
>> DDR and L3 frequencies being an aggregate of both.
>>
>> If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
>> won't be invoked for CPU4. As a result both CPU devices will end up in sharing
>> the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
>> votes of other.
> 
> All of this should be a part of the commit message.
> 

Thanks Dmitry for your review. Will include this in commit text while posting the next series.

Thanks,
Jagadeesh

>>
>> [1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1588
>>
>> Thanks,
>> Jagadeesh
>>  
>
Dmitry Baryshkov Dec. 4, 2024, 3:13 a.m. UTC | #4
On Tue, Dec 03, 2024 at 08:33:46PM +0530, Jagadeesh Kona wrote:
> 
> 
> On 11/30/2024 8:02 PM, Konrad Dybcio wrote:
> > On 14.11.2024 11:48 PM, Dmitry Baryshkov wrote:
> >> On Mon, Nov 11, 2024 at 06:39:48PM +0530, Jagadeesh Kona wrote:
> >>>
> >>>
> >>> On 10/17/2024 9:12 PM, Brian Masney wrote:
> >>>> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
> >>>>> +	cpu0_opp_table: opp-table-cpu0 {
> >>>>> +		compatible = "operating-points-v2";
> >>>>> +		opp-shared;
> >>>>> +
> >>>>> +		cpu0_opp_1267mhz: opp-1267200000 {
> >>>>> +			opp-hz = /bits/ 64 <1267200000>;
> >>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>> +		};
> >>>>> +
> >>>>> +		cpu0_opp_1363mhz: opp-1363200000 {
> >>>>> +			opp-hz = /bits/ 64 <1363200000>;
> >>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>> +		};
> >>>>
> >>>> [snip]
> >>>>
> >>>>> +	cpu4_opp_table: opp-table-cpu4 {
> >>>>> +		compatible = "operating-points-v2";
> >>>>> +		opp-shared;
> >>>>> +
> >>>>> +		cpu4_opp_1267mhz: opp-1267200000 {
> >>>>> +			opp-hz = /bits/ 64 <1267200000>;
> >>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>> +		};
> >>>>> +
> >>>>> +		cpu4_opp_1363mhz: opp-1363200000 {
> >>>>> +			opp-hz = /bits/ 64 <1363200000>;
> >>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>> +		};
> >>>>
> >>>> There's no functional differences in the cpu0 and cpu4 opp tables. Can
> >>>> a single table be used?
> >>>>
> >>>> This aligns with my recollection that this particular SoC only has the
> >>>> gold cores.
> >>>>
> >>>> Brian
> >>>>
> >>>
> >>> Thanks Brian for your review. Sorry for the delayed response.
> >>>
> >>> We require separate OPP tables for CPU0 and CPU4 to allow independent
> >>> scaling of DDR and L3 frequencies for each CPU domain, with the final
> >>> DDR and L3 frequencies being an aggregate of both.
> >>>
> >>> If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
> >>> won't be invoked for CPU4. As a result both CPU devices will end up in sharing
> >>> the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
> >>> votes of other.
> > 
> > Oh that's a fun find.. clocks get the same treatment.. very bad,
> > but may explain some schroedingerbugs.
> > 
> > Taking a peek at some code paths, wouldn't dropping opp-shared
> > solve our issues? dev_pm_opp_set_sharing_cpus() overrides it
> > 
> > Konrad
> 
> Thanks Konrad for your review.
> 
> Yes, correct. I tried dropping opp-shared but it is again getting set due to
> dev_pm_opp_set_sharing_cpus().

It should be set, but then it should get the limited CPU mask rather
than the full CPU set. Isn't that enough for your case?
Jagadeesh Kona Dec. 4, 2024, 8:45 a.m. UTC | #5
On 12/4/2024 8:43 AM, Dmitry Baryshkov wrote:
> On Tue, Dec 03, 2024 at 08:33:46PM +0530, Jagadeesh Kona wrote:
>>
>>
>> On 11/30/2024 8:02 PM, Konrad Dybcio wrote:
>>> On 14.11.2024 11:48 PM, Dmitry Baryshkov wrote:
>>>> On Mon, Nov 11, 2024 at 06:39:48PM +0530, Jagadeesh Kona wrote:
>>>>>
>>>>>
>>>>> On 10/17/2024 9:12 PM, Brian Masney wrote:
>>>>>> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
>>>>>>> +	cpu0_opp_table: opp-table-cpu0 {
>>>>>>> +		compatible = "operating-points-v2";
>>>>>>> +		opp-shared;
>>>>>>> +
>>>>>>> +		cpu0_opp_1267mhz: opp-1267200000 {
>>>>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>>>>> +		};
>>>>>>> +
>>>>>>> +		cpu0_opp_1363mhz: opp-1363200000 {
>>>>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>>>>> +		};
>>>>>>
>>>>>> [snip]
>>>>>>
>>>>>>> +	cpu4_opp_table: opp-table-cpu4 {
>>>>>>> +		compatible = "operating-points-v2";
>>>>>>> +		opp-shared;
>>>>>>> +
>>>>>>> +		cpu4_opp_1267mhz: opp-1267200000 {
>>>>>>> +			opp-hz = /bits/ 64 <1267200000>;
>>>>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>>>>> +		};
>>>>>>> +
>>>>>>> +		cpu4_opp_1363mhz: opp-1363200000 {
>>>>>>> +			opp-hz = /bits/ 64 <1363200000>;
>>>>>>> +			opp-peak-kBps = <6220800 29491200>;
>>>>>>> +		};
>>>>>>
>>>>>> There's no functional differences in the cpu0 and cpu4 opp tables. Can
>>>>>> a single table be used?
>>>>>>
>>>>>> This aligns with my recollection that this particular SoC only has the
>>>>>> gold cores.
>>>>>>
>>>>>> Brian
>>>>>>
>>>>>
>>>>> Thanks Brian for your review. Sorry for the delayed response.
>>>>>
>>>>> We require separate OPP tables for CPU0 and CPU4 to allow independent
>>>>> scaling of DDR and L3 frequencies for each CPU domain, with the final
>>>>> DDR and L3 frequencies being an aggregate of both.
>>>>>
>>>>> If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
>>>>> won't be invoked for CPU4. As a result both CPU devices will end up in sharing
>>>>> the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
>>>>> votes of other.
>>>
>>> Oh that's a fun find.. clocks get the same treatment.. very bad,
>>> but may explain some schroedingerbugs.
>>>
>>> Taking a peek at some code paths, wouldn't dropping opp-shared
>>> solve our issues? dev_pm_opp_set_sharing_cpus() overrides it
>>>
>>> Konrad
>>
>> Thanks Konrad for your review.
>>
>> Yes, correct. I tried dropping opp-shared but it is again getting set due to
>> dev_pm_opp_set_sharing_cpus().
> 
> It should be set, but then it should get the limited CPU mask rather
> than the full CPU set. Isn't that enough for your case?
> 

Even if we call dev_pm_opp_set_sharing_cpus() with the limited CPU mask, it adds
OPP_TABLE_ACCESS_SHARED flag to the OPP table. Due to this flag being set, if this
same opp table is used for another CPU domain(CPU4-7) also in DT, then _managed_opp[1]
which gets called inside from dev_pm_opp_of_add_table() for CPU4 will return the same
CPU0 OPP table. 

Due to above, _allocate_opp_table() [2] won't be invoked for CPU4 but instead CPU4 will be
added as device under the CPU0 OPP table [3]. Due to this, dev_pm_opp_of_find_icc_paths() [4]
won't be invoked for CPU4 device and hence CPU4 won't be able to independently scale it's
interconnects. Both CPU0 and CPU4 devices will scale the same ICC path which can lead to one
device overwriting the BW vote placed by other device. So we need two separate OPP tables for
both domains.

[1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1600
[2] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1613
[3] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1606
[4] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1484

Thanks,
Jagadeesh
Dmitry Baryshkov Dec. 4, 2024, 11 a.m. UTC | #6
On Wed, Dec 04, 2024 at 02:15:09PM +0530, Jagadeesh Kona wrote:
> 
> 
> On 12/4/2024 8:43 AM, Dmitry Baryshkov wrote:
> > On Tue, Dec 03, 2024 at 08:33:46PM +0530, Jagadeesh Kona wrote:
> >>
> >>
> >> On 11/30/2024 8:02 PM, Konrad Dybcio wrote:
> >>> On 14.11.2024 11:48 PM, Dmitry Baryshkov wrote:
> >>>> On Mon, Nov 11, 2024 at 06:39:48PM +0530, Jagadeesh Kona wrote:
> >>>>>
> >>>>>
> >>>>> On 10/17/2024 9:12 PM, Brian Masney wrote:
> >>>>>> On Thu, Oct 17, 2024 at 02:58:31PM +0530, Jagadeesh Kona wrote:
> >>>>>>> +	cpu0_opp_table: opp-table-cpu0 {
> >>>>>>> +		compatible = "operating-points-v2";
> >>>>>>> +		opp-shared;
> >>>>>>> +
> >>>>>>> +		cpu0_opp_1267mhz: opp-1267200000 {
> >>>>>>> +			opp-hz = /bits/ 64 <1267200000>;
> >>>>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>>>> +		};
> >>>>>>> +
> >>>>>>> +		cpu0_opp_1363mhz: opp-1363200000 {
> >>>>>>> +			opp-hz = /bits/ 64 <1363200000>;
> >>>>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>>>> +		};
> >>>>>>
> >>>>>> [snip]
> >>>>>>
> >>>>>>> +	cpu4_opp_table: opp-table-cpu4 {
> >>>>>>> +		compatible = "operating-points-v2";
> >>>>>>> +		opp-shared;
> >>>>>>> +
> >>>>>>> +		cpu4_opp_1267mhz: opp-1267200000 {
> >>>>>>> +			opp-hz = /bits/ 64 <1267200000>;
> >>>>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>>>> +		};
> >>>>>>> +
> >>>>>>> +		cpu4_opp_1363mhz: opp-1363200000 {
> >>>>>>> +			opp-hz = /bits/ 64 <1363200000>;
> >>>>>>> +			opp-peak-kBps = <6220800 29491200>;
> >>>>>>> +		};
> >>>>>>
> >>>>>> There's no functional differences in the cpu0 and cpu4 opp tables. Can
> >>>>>> a single table be used?
> >>>>>>
> >>>>>> This aligns with my recollection that this particular SoC only has the
> >>>>>> gold cores.
> >>>>>>
> >>>>>> Brian
> >>>>>>
> >>>>>
> >>>>> Thanks Brian for your review. Sorry for the delayed response.
> >>>>>
> >>>>> We require separate OPP tables for CPU0 and CPU4 to allow independent
> >>>>> scaling of DDR and L3 frequencies for each CPU domain, with the final
> >>>>> DDR and L3 frequencies being an aggregate of both.
> >>>>>
> >>>>> If we use a single OPP table for both CPU domains, then _allocate_opp_table() [1]
> >>>>> won't be invoked for CPU4. As a result both CPU devices will end up in sharing
> >>>>> the same ICC path handle, which could lead to one CPU device overwriting the bandwidth
> >>>>> votes of other.
> >>>
> >>> Oh that's a fun find.. clocks get the same treatment.. very bad,
> >>> but may explain some schroedingerbugs.
> >>>
> >>> Taking a peek at some code paths, wouldn't dropping opp-shared
> >>> solve our issues? dev_pm_opp_set_sharing_cpus() overrides it
> >>>
> >>> Konrad
> >>
> >> Thanks Konrad for your review.
> >>
> >> Yes, correct. I tried dropping opp-shared but it is again getting set due to
> >> dev_pm_opp_set_sharing_cpus().
> > 
> > It should be set, but then it should get the limited CPU mask rather
> > than the full CPU set. Isn't that enough for your case?
> > 
> 
> Even if we call dev_pm_opp_set_sharing_cpus() with the limited CPU mask, it adds
> OPP_TABLE_ACCESS_SHARED flag to the OPP table. Due to this flag being set, if this
> same opp table is used for another CPU domain(CPU4-7) also in DT, then _managed_opp[1]
> which gets called inside from dev_pm_opp_of_add_table() for CPU4 will return the same
> CPU0 OPP table. 
> 
> Due to above, _allocate_opp_table() [2] won't be invoked for CPU4 but instead CPU4 will be
> added as device under the CPU0 OPP table [3]. Due to this, dev_pm_opp_of_find_icc_paths() [4]
> won't be invoked for CPU4 device and hence CPU4 won't be able to independently scale it's
> interconnects. Both CPU0 and CPU4 devices will scale the same ICC path which can lead to one
> device overwriting the BW vote placed by other device. So we need two separate OPP tables for
> both domains.

Ack, that makes sense. Thanks for the explanation!

> 
> [1] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1600
> [2] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1613
> [3] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1606
> [4] https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/drivers/opp/core.c#n1484
> 
> Thanks,
> Jagadeesh
diff mbox series

Patch

diff --git a/arch/arm64/boot/dts/qcom/sa8775p.dtsi b/arch/arm64/boot/dts/qcom/sa8775p.dtsi
index d8b90bd4b1f05604185f015929a1f296799ad6a4..47eca50b30ffa38a652706014d35ef9e833003ec 100644
--- a/arch/arm64/boot/dts/qcom/sa8775p.dtsi
+++ b/arch/arm64/boot/dts/qcom/sa8775p.dtsi
@@ -48,6 +48,7 @@  CPU0: cpu@0 {
 			next-level-cache = <&L2_0>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu0_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl0 MASTER_EPSS_L3_APPS
@@ -74,6 +75,7 @@  CPU1: cpu@100 {
 			next-level-cache = <&L2_1>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu0_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl0 MASTER_EPSS_L3_APPS
@@ -95,6 +97,7 @@  CPU2: cpu@200 {
 			next-level-cache = <&L2_2>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu0_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl0 MASTER_EPSS_L3_APPS
@@ -116,6 +119,7 @@  CPU3: cpu@300 {
 			next-level-cache = <&L2_3>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu0_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl0 MASTER_EPSS_L3_APPS
@@ -137,6 +141,7 @@  CPU4: cpu@10000 {
 			next-level-cache = <&L2_4>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu4_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl1 MASTER_EPSS_L3_APPS
@@ -164,6 +169,7 @@  CPU5: cpu@10100 {
 			next-level-cache = <&L2_5>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu4_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl1 MASTER_EPSS_L3_APPS
@@ -185,6 +191,7 @@  CPU6: cpu@10200 {
 			next-level-cache = <&L2_6>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu4_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl1 MASTER_EPSS_L3_APPS
@@ -206,6 +213,7 @@  CPU7: cpu@10300 {
 			next-level-cache = <&L2_7>;
 			capacity-dmips-mhz = <1024>;
 			dynamic-power-coefficient = <100>;
+			operating-points-v2 = <&cpu4_opp_table>;
 			interconnects = <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY
 					&mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ACTIVE_ONLY>,
 					<&epss_l3_cl1 MASTER_EPSS_L3_APPS
@@ -299,6 +307,176 @@  CLUSTER_SLEEP_APSS_RSC_PC: cluster-sleep-1 {
 		};
 	};
 
+	cpu0_opp_table: opp-table-cpu0 {
+		compatible = "operating-points-v2";
+		opp-shared;
+
+		cpu0_opp_1267mhz: opp-1267200000 {
+			opp-hz = /bits/ 64 <1267200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu0_opp_1363mhz: opp-1363200000 {
+			opp-hz = /bits/ 64 <1363200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu0_opp_1459mhz: opp-1459200000 {
+			opp-hz = /bits/ 64 <1459200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu0_opp_1536mhz: opp-1536000000 {
+			opp-hz = /bits/ 64 <1536000000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu0_opp_1632mhz: opp-1632000000 {
+			opp-hz = /bits/ 64 <1632000000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_1708mhz: opp-1708800000 {
+			opp-hz = /bits/ 64 <1708800000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_1785mhz: opp-1785600000 {
+			opp-hz = /bits/ 64 <1785600000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_1862mhz: opp-1862400000 {
+			opp-hz = /bits/ 64 <1862400000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_1939mhz: opp-1939200000 {
+			opp-hz = /bits/ 64 <1939200000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_2016mhz: opp-2016000000 {
+			opp-hz = /bits/ 64 <2016000000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu0_opp_2112mhz: opp-2112000000 {
+			opp-hz = /bits/ 64 <2112000000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu0_opp_2188mhz: opp-2188800000 {
+			opp-hz = /bits/ 64 <2188800000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu0_opp_2265mhz: opp-2265600000 {
+			opp-hz = /bits/ 64 <2265600000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu0_opp_2361mhz: opp-2361600000 {
+			opp-hz = /bits/ 64 <2361600000>;
+			opp-peak-kBps = <12787200 51609600>;
+		};
+
+		cpu0_opp_2457mhz: opp-2457600000 {
+			opp-hz = /bits/ 64 <2457600000>;
+			opp-peak-kBps = <12787200 51609600>;
+		};
+
+		cpu0_opp_2553mhz: opp-2553600000 {
+			opp-hz = /bits/ 64 <2553600000>;
+			opp-peak-kBps = <12787200 54681600>;
+		};
+	};
+
+	cpu4_opp_table: opp-table-cpu4 {
+		compatible = "operating-points-v2";
+		opp-shared;
+
+		cpu4_opp_1267mhz: opp-1267200000 {
+			opp-hz = /bits/ 64 <1267200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu4_opp_1363mhz: opp-1363200000 {
+			opp-hz = /bits/ 64 <1363200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu4_opp_1459mhz: opp-1459200000 {
+			opp-hz = /bits/ 64 <1459200000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu4_opp_1536mhz: opp-1536000000 {
+			opp-hz = /bits/ 64 <1536000000>;
+			opp-peak-kBps = <6220800 29491200>;
+		};
+
+		cpu4_opp_1632mhz: opp-1632000000 {
+			opp-hz = /bits/ 64 <1632000000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_1708mhz: opp-1708800000 {
+			opp-hz = /bits/ 64 <1708800000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_1785mhz: opp-1785600000 {
+			opp-hz = /bits/ 64 <1785600000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_1862mhz: opp-1862400000 {
+			opp-hz = /bits/ 64 <1862400000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_1939mhz: opp-1939200000 {
+			opp-hz = /bits/ 64 <1939200000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_2016mhz: opp-2016000000 {
+			opp-hz = /bits/ 64 <2016000000>;
+			opp-peak-kBps = <6835200 39321600>;
+		};
+
+		cpu4_opp_2112mhz: opp-2112000000 {
+			opp-hz = /bits/ 64 <2112000000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu4_opp_2188mhz: opp-2188800000 {
+			opp-hz = /bits/ 64 <2188800000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu4_opp_2265mhz: opp-2265600000 {
+			opp-hz = /bits/ 64 <2265600000>;
+			opp-peak-kBps = <8371200 49766400>;
+		};
+
+		cpu4_opp_2361mhz: opp-2361600000 {
+			opp-hz = /bits/ 64 <2361600000>;
+			opp-peak-kBps = <12787200 51609600>;
+		};
+
+		cpu4_opp_2457mhz: opp-2457600000 {
+			opp-hz = /bits/ 64 <2457600000>;
+			opp-peak-kBps = <12787200 51609600>;
+		};
+
+		cpu4_opp_2553mhz: opp-2553600000 {
+			opp-hz = /bits/ 64 <2553600000>;
+			opp-peak-kBps = <12787200 54681600>;
+		};
+	};
+
 	dummy-sink {
 		compatible = "arm,coresight-dummy-sink";