The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is specified much lower for the RK3588J variant. [1][2]
To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to its datasheet, [2] the RK3588J variant can actually run safely at higher CPU and GPU OPPs as well, but only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted specifically at industrial applications, is made (or binned) for.
Thus, only the CPU and GPU OPPs that are specified by the vendor to be safe throughout the entire RK3588J's extended temperature range may be provided, while anyone who actually can ensure that their RK3588J-based board is never going to run within the extended temperature range, may probably safely apply a DT overlay that adds the higher CPU and GPU OPPs. As we obviously can't know what will be the runtime temperature conditions for a particular board, we may provide only the always-safe OPPs.
With all this and the downstream RK3588(J) DT definitions [4][5] in mind, let's delete the RK3588J CPU and GPU OPPs that are not considered belonging to the normal operation mode for this SoC variant. To quote the RK3588J datasheet [2], "normal mode means the chipset works under safety voltage and frequency; for the industrial environment, highly recommend to keep in normal mode, the lifetime is reasonably guaranteed", while "overdrive mode brings higher frequency, and the voltage will increase accordingly; under the overdrive mode for a long time, the chipset may shorten the lifetime, especially in high temperature condition".
To sum up the RK3588J datasheet [2] and the vendor-provided DTs, [4][5] the maximum allowed CPU core and GPU frequencies are as follows:
IP core | Normal mode | Overdrive mode ------------+-------------+---------------- Cortex-A55 | 1,296 MHz | 1,704 MHz Cortex-A76 | 1,608 MHz | 2,016 MHz GPU | 700 MHz | 850 MHz
Unfortunately, when it comes to the actual voltages for the RK3588J CPU and GPU OPPs, there's a discrepancy between the RK3588J datasheet [2] and the downstream kernel code. [4][5] The RK3588J datasheet states that "the max. working voltage of CPU/GPU/NPU is 0.75 V under the normal mode", while the downstream kernel code actually allows voltage ranges that go up to 0.95 V, which is still within the voltage range allowed by the datasheet. However, the RK3588J datasheet also tells us to "strictly refer to the software configuration of SDK and the hardware reference design", so let's embrace the voltage ranges provided by the downstream kernel code, which also prevents the undesirable theoretical outcome of ending up with no usable OPPs on a particular board, as a result of the board's voltage regulator(s) being unable to deliver the exact voltages, for whatever reason.
The above-described voltage ranges for the RK3588J CPU OPPs remain taken from the downstream kernel code [4][5] by picking the highest, worst-bin values, which ensure that all RK3588J bins will work reliably. Yes, with some power inevitably wasted as unnecessarily generated heat, but the reliability is paramount, together with the longevity. This deficiency may be revisited separately at some point in the future.
The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been established earlier, [6] as a result of the "same-voltage, lower-frequency" OPPs being considered inefficient from the IPA governor's standpoint, which may also be revisited separately at some point in the future.
[1] https://wiki.friendlyelec.com/wiki/images/e/ee/Rockchip_RK3588_Datasheet_V1.... [2] https://wmsc.lcsc.com/wmsc/upload/file/pdf/v2/lcsc/2403201054_Rockchip-RK358... [3] https://lore.kernel.org/linux-rockchip/e55125ed-64fb-455e-b1e4-cebe2cf006e4@... [4] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [5] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [6] https://lore.kernel.org/all/20240229-rk-dts-additions-v3-5-6afe8473a631@gmai...
Fixes: 667885a68658 ("arm64: dts: rockchip: Add OPP data for CPU cores on RK3588j") Fixes: a7b2070505a2 ("arm64: dts: rockchip: Split GPU OPPs of RK3588 and RK3588j") Cc: stable@vger.kernel.org Cc: Heiko Stuebner heiko@sntech.de Cc: Alexey Charkov alchark@gmail.com Helped-by: Quentin Schulz quentin.schulz@cherry.de Signed-off-by: Dragan Simic dsimic@manjaro.org --- arch/arm64/boot/dts/rockchip/rk3588j.dtsi | 53 ++++++++--------------- 1 file changed, 17 insertions(+), 36 deletions(-)
diff --git a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi index bce72bac4503..3045cb3bd68c 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi @@ -11,74 +11,59 @@ cluster0_opp_table: opp-table-cluster0 { compatible = "operating-points-v2"; opp-shared;
- opp-1416000000 { - opp-hz = /bits/ 64 <1416000000>; + opp-1200000000 { + opp-hz = /bits/ 64 <1200000000>; opp-microvolt = <750000 750000 950000>; clock-latency-ns = <40000>; opp-suspend; }; - opp-1608000000 { - opp-hz = /bits/ 64 <1608000000>; - opp-microvolt = <887500 887500 950000>; - clock-latency-ns = <40000>; - }; - opp-1704000000 { - opp-hz = /bits/ 64 <1704000000>; - opp-microvolt = <937500 937500 950000>; + opp-1296000000 { + opp-hz = /bits/ 64 <1296000000>; + opp-microvolt = <775000 775000 950000>; clock-latency-ns = <40000>; }; };
cluster1_opp_table: opp-table-cluster1 { compatible = "operating-points-v2"; opp-shared;
+ opp-1200000000{ + opp-hz = /bits/ 64 <1200000000>; + opp-microvolt = <750000 750000 950000>; + clock-latency-ns = <40000>; + }; opp-1416000000 { opp-hz = /bits/ 64 <1416000000>; - opp-microvolt = <750000 750000 950000>; + opp-microvolt = <762500 762500 950000>; clock-latency-ns = <40000>; }; opp-1608000000 { opp-hz = /bits/ 64 <1608000000>; opp-microvolt = <787500 787500 950000>; clock-latency-ns = <40000>; }; - opp-1800000000 { - opp-hz = /bits/ 64 <1800000000>; - opp-microvolt = <875000 875000 950000>; - clock-latency-ns = <40000>; - }; - opp-2016000000 { - opp-hz = /bits/ 64 <2016000000>; - opp-microvolt = <950000 950000 950000>; - clock-latency-ns = <40000>; - }; };
cluster2_opp_table: opp-table-cluster2 { compatible = "operating-points-v2"; opp-shared;
+ opp-1200000000{ + opp-hz = /bits/ 64 <1200000000>; + opp-microvolt = <750000 750000 950000>; + clock-latency-ns = <40000>; + }; opp-1416000000 { opp-hz = /bits/ 64 <1416000000>; - opp-microvolt = <750000 750000 950000>; + opp-microvolt = <762500 762500 950000>; clock-latency-ns = <40000>; }; opp-1608000000 { opp-hz = /bits/ 64 <1608000000>; opp-microvolt = <787500 787500 950000>; clock-latency-ns = <40000>; }; - opp-1800000000 { - opp-hz = /bits/ 64 <1800000000>; - opp-microvolt = <875000 875000 950000>; - clock-latency-ns = <40000>; - }; - opp-2016000000 { - opp-hz = /bits/ 64 <2016000000>; - opp-microvolt = <950000 950000 950000>; - clock-latency-ns = <40000>; - }; };
gpu_opp_table: opp-table { @@ -104,10 +89,6 @@ opp-700000000 { opp-hz = /bits/ 64 <700000000>; opp-microvolt = <750000 750000 850000>; }; - opp-850000000 { - opp-hz = /bits/ 64 <800000000>; - opp-microvolt = <787500 787500 850000>; - }; }; };
Hi Dragan,
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
specified much lower for the RK3588J variant. [1][2]
To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to its datasheet, [2] the RK3588J variant can actually run safely at higher CPU and GPU OPPs as well, but only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted
"only when not enjoying the assumed extended temperature range" is extrapolated by me/us and not confirmed by Rockchip themselves. I've asked for a statement on what "industrial environment" they specify in the Normal Mode explanation means since it's the only time they use the term. I've yet to receive an answer. The only thing Rockchip in their datasheet is that the overdrive mode will shorten lifetime when used for a long time, especially in high temperature conditions. It's not clear whether we can use the overdrive mode even within the RK3588 typical range of operation.
specifically at industrial applications, is made (or binned) for.
Thus, only the CPU and GPU OPPs that are specified by the vendor to be safe throughout the entire RK3588J's extended temperature range may be provided, while anyone who actually can ensure that their RK3588J-based board is never going to run within the extended temperature range, may probably safely apply a DT overlay that adds the higher CPU and GPU OPPs. As we
Wouldn't say "safely" here, Rockchip still says that running overdrive mode for a long time may shorten lifetime... that followed by "especially in high temperature condition" doesn't mean that operating the RK3588J within the RK3588 typical range is safe to do.
obviously can't know what will be the runtime temperature conditions for a particular board, we may provide only the always-safe OPPs.
With all this and the downstream RK3588(J) DT definitions [4][5] in mind, let's delete the RK3588J CPU and GPU OPPs that are not considered belonging to the normal operation mode for this SoC variant. To quote the RK3588J datasheet [2], "normal mode means the chipset works under safety voltage and frequency; for the industrial environment, highly recommend to keep in normal mode, the lifetime is reasonably guaranteed", while "overdrive mode brings higher frequency, and the voltage will increase accordingly; under the overdrive mode for a long time, the chipset may shorten the lifetime, especially in high temperature condition".
To sum up the RK3588J datasheet [2] and the vendor-provided DTs, [4][5] the maximum allowed CPU core and GPU frequencies are as follows:
IP core | Normal mode | Overdrive mode------------+-------------+---------------- Cortex-A55 | 1,296 MHz | 1,704 MHz Cortex-A76 | 1,608 MHz | 2,016 MHz GPU | 700 MHz | 850 MHz
The NPU too is impacted by this, so maybe list it anyway here? Even if we don't support it right now and don't have OPPs for it.
Unfortunately, when it comes to the actual voltages for the RK3588J CPU and GPU OPPs, there's a discrepancy between the RK3588J datasheet [2] and the downstream kernel code. [4][5] The RK3588J datasheet states that "the max. working voltage of CPU/GPU/NPU is 0.75 V under the normal mode", while the downstream kernel code actually allows voltage ranges that go up to 0.95 V, which is still within the voltage range allowed by the datasheet. However, the RK3588J datasheet also tells us to "strictly refer to the software configuration of SDK and the hardware reference design", so let's embrace the voltage ranges provided by the downstream kernel code, which also prevents the undesirable theoretical outcome of ending up with no usable OPPs on a particular board, as a result of the board's voltage regulator(s) being unable to deliver the exact voltages, for whatever reason.
The above-described voltage ranges for the RK3588J CPU OPPs remain taken from the downstream kernel code [4][5] by picking the highest, worst-bin values, which ensure that all RK3588J bins will work reliably. Yes, with some power inevitably wasted as unnecessarily generated heat, but the reliability is paramount, together with the longevity. This deficiency may be revisited separately at some point in the future.
The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been
Interesting that we went for a different strategy for the GPU OPPs :)
established earlier, [6] as a result of the "same-voltage, lower-frequency" OPPs being considered inefficient from the IPA governor's standpoint, which may also be revisited separately at some point in the future.
[1] https://wiki.friendlyelec.com/wiki/images/e/ee/Rockchip_RK3588_Datasheet_V1.... [2] https://wmsc.lcsc.com/wmsc/upload/file/pdf/v2/lcsc/2403201054_Rockchip-RK358... [3] https://lore.kernel.org/linux-rockchip/e55125ed-64fb-455e-b1e4-cebe2cf006e4@... [4] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [5] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [6] https://lore.kernel.org/all/20240229-rk-dts-additions-v3-5-6afe8473a631@gmai...
Fixes: 667885a68658 ("arm64: dts: rockchip: Add OPP data for CPU cores on RK3588j") Fixes: a7b2070505a2 ("arm64: dts: rockchip: Split GPU OPPs of RK3588 and RK3588j") Cc: stable@vger.kernel.org Cc: Heiko Stuebner heiko@sntech.de Cc: Alexey Charkov alchark@gmail.com Helped-by: Quentin Schulz quentin.schulz@cherry.de
Reported-by/Suggested-by?
I don't see Helped-by in https://www.kernel.org/doc/html/latest/process/submitting-patches.html#using...
I see 2496b2aaacf137250f4ca449f465e2cadaabb0e8 got the Helped-by replaced by a Suggested-by for example, but I see other patches with Helped-by... if that is a standard trailer for kernel patches, then maybe we should add it to that doc?
Signed-off-by: Dragan Simic dsimic@manjaro.org
arch/arm64/boot/dts/rockchip/rk3588j.dtsi | 53 ++++++++--------------- 1 file changed, 17 insertions(+), 36 deletions(-)
diff --git a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi index bce72bac4503..3045cb3bd68c 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi @@ -11,74 +11,59 @@ cluster0_opp_table: opp-table-cluster0 { compatible = "operating-points-v2"; opp-shared;
opp-1416000000 {opp-hz = /bits/ 64 <1416000000>;
opp-1200000000 {opp-hz = /bits/ 64 <1200000000>; opp-microvolt = <750000 750000 950000>; clock-latency-ns = <40000>; opp-suspend;
opp-1608000000 {opp-hz = /bits/ 64 <1608000000>;opp-microvolt = <887500 887500 950000>;clock-latency-ns = <40000>;};opp-1704000000 {opp-hz = /bits/ 64 <1704000000>;opp-microvolt = <937500 937500 950000>;
opp-1296000000 {opp-hz = /bits/ 64 <1296000000>;opp-microvolt = <775000 775000 950000>;
Got tricked by this one.
In the Rockchip vendor kernel, the opp-microvolt is 750000 750000 950000, so the same as CPU OPP 1.2GHz. However, the opp-microvolt-L1 and L0 are higher than that. Only a couple of the OPPs in vendor kernel actually have opp-microvolt-L* higher than opp-microvolt, that is a noteworthy oddity for anyone reviewing this patch :)
Anyway, that is correct, we take the highest voltage among all defined opp-microvolt* properties.
I only have comments on the commit log, the diff is fine so:
Reviewed-by: Quentin Schulz quentin.schulz@cherry.de
Thanks! Quentin
Hello Quentin,
Thanks for your comments! Please see some responses below.
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
specified much lower for the RK3588J variant. [1][2]
To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to its datasheet, [2] the RK3588J variant can actually run safely at higher CPU and GPU OPPs as well, but only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted
"only when not enjoying the assumed extended temperature range" is extrapolated by me/us and not confirmed by Rockchip themselves. I've asked for a statement on what "industrial environment" they specify in the Normal Mode explanation means since it's the only time they use the term. I've yet to receive an answer. The only thing Rockchip in their datasheet is that the overdrive mode will shorten lifetime when used for a long time, especially in high temperature conditions. It's not clear whether we can use the overdrive mode even within the RK3588 typical range of operation.
True. I'll see to rephrase the patch description a bit in the v2, to avoid this kind of speculation. I mean, perhaps the speculation is right, but it hasn't been confirmed officially by Rockchip.
specifically at industrial applications, is made (or binned) for.
Thus, only the CPU and GPU OPPs that are specified by the vendor to be safe throughout the entire RK3588J's extended temperature range may be provided, while anyone who actually can ensure that their RK3588J-based board is never going to run within the extended temperature range, may probably safely apply a DT overlay that adds the higher CPU and GPU OPPs. As we
Wouldn't say "safely" here, Rockchip still says that running overdrive mode for a long time may shorten lifetime... that followed by "especially in high temperature condition" doesn't mean that operating the RK3588J within the RK3588 typical range is safe to do.
Also true, and I'll also try to address that in the v2.
obviously can't know what will be the runtime temperature conditions for a particular board, we may provide only the always-safe OPPs.
With all this and the downstream RK3588(J) DT definitions [4][5] in mind, let's delete the RK3588J CPU and GPU OPPs that are not considered belonging to the normal operation mode for this SoC variant. To quote the RK3588J datasheet [2], "normal mode means the chipset works under safety voltage and frequency; for the industrial environment, highly recommend to keep in normal mode, the lifetime is reasonably guaranteed", while "overdrive mode brings higher frequency, and the voltage will increase accordingly; under the overdrive mode for a long time, the chipset may shorten the lifetime, especially in high temperature condition".
To sum up the RK3588J datasheet [2] and the vendor-provided DTs, [4][5] the maximum allowed CPU core and GPU frequencies are as follows:
IP core | Normal mode | Overdrive mode------------+-------------+---------------- Cortex-A55 | 1,296 MHz | 1,704 MHz Cortex-A76 | 1,608 MHz | 2,016 MHz GPU | 700 MHz | 850 MHz
The NPU too is impacted by this, so maybe list it anyway here? Even if we don't support it right now and don't have OPPs for it.
Agreed, will add in the v2. Having all that information in a single place can only be helpful.
Unfortunately, when it comes to the actual voltages for the RK3588J CPU and GPU OPPs, there's a discrepancy between the RK3588J datasheet [2] and the downstream kernel code. [4][5] The RK3588J datasheet states that "the max. working voltage of CPU/GPU/NPU is 0.75 V under the normal mode", while the downstream kernel code actually allows voltage ranges that go up to 0.95 V, which is still within the voltage range allowed by the datasheet. However, the RK3588J datasheet also tells us to "strictly refer to the software configuration of SDK and the hardware reference design", so let's embrace the voltage ranges provided by the downstream kernel code, which also prevents the undesirable theoretical outcome of ending up with no usable OPPs on a particular board, as a result of the board's voltage regulator(s) being unable to deliver the exact voltages, for whatever reason.
The above-described voltage ranges for the RK3588J CPU OPPs remain taken from the downstream kernel code [4][5] by picking the highest, worst-bin values, which ensure that all RK3588J bins will work reliably. Yes, with some power inevitably wasted as unnecessarily generated heat, but the reliability is paramount, together with the longevity. This deficiency may be revisited separately at some point in the future.
The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been
Interesting that we went for a different strategy for the GPU OPPs :)
Good point, and I'm fully aware of that. :) Actually, I'm rather sure that omitting the additional CPU OPPs does no good to us, but I didn't want to argue about that when they were dropped originally, before I can have some hard numbers to prove it in a repeatable way.
established earlier, [6] as a result of the "same-voltage, lower-frequency" OPPs being considered inefficient from the IPA governor's standpoint, which may also be revisited separately at some point in the future.
[1] https://wiki.friendlyelec.com/wiki/images/e/ee/Rockchip_RK3588_Datasheet_V1.... [2] https://wmsc.lcsc.com/wmsc/upload/file/pdf/v2/lcsc/2403201054_Rockchip-RK358... [3] https://lore.kernel.org/linux-rockchip/e55125ed-64fb-455e-b1e4-cebe2cf006e4@... [4] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [5] https://raw.githubusercontent.com/rockchip-linux/kernel/604cec4004abe5a96c73... [6] https://lore.kernel.org/all/20240229-rk-dts-additions-v3-5-6afe8473a631@gmai...
Fixes: 667885a68658 ("arm64: dts: rockchip: Add OPP data for CPU cores on RK3588j") Fixes: a7b2070505a2 ("arm64: dts: rockchip: Split GPU OPPs of RK3588 and RK3588j") Cc: stable@vger.kernel.org Cc: Heiko Stuebner heiko@sntech.de Cc: Alexey Charkov alchark@gmail.com Helped-by: Quentin Schulz quentin.schulz@cherry.de
Reported-by/Suggested-by?
I don't see Helped-by in https://www.kernel.org/doc/html/latest/process/submitting-patches.html#using...
I see 2496b2aaacf137250f4ca449f465e2cadaabb0e8 got the Helped-by replaced by a Suggested-by for example, but I see other patches with Helped-by... if that is a standard trailer for kernel patches, then maybe we should add it to that doc?
Actually, I already tried to get the Helped-by tag added to the kernel documentation, by submitting a small patch series. [*] Unfortunately, it got rejected. :/
However, Heiko accepts Helped-by tags and nobody higher up the tree seems to complain, so we should be fine. :) It isn't the case with all maintainers, though.
[*] https://lore.kernel.org/all/cover.1730874296.git.dsimic@manjaro.org/T/#u
Signed-off-by: Dragan Simic dsimic@manjaro.org
arch/arm64/boot/dts/rockchip/rk3588j.dtsi | 53 ++++++++--------------- 1 file changed, 17 insertions(+), 36 deletions(-)
diff --git a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi index bce72bac4503..3045cb3bd68c 100644 --- a/arch/arm64/boot/dts/rockchip/rk3588j.dtsi +++ b/arch/arm64/boot/dts/rockchip/rk3588j.dtsi @@ -11,74 +11,59 @@ cluster0_opp_table: opp-table-cluster0 { compatible = "operating-points-v2"; opp-shared;
opp-1416000000 {opp-hz = /bits/ 64 <1416000000>;
opp-1200000000 {opp-hz = /bits/ 64 <1200000000>; opp-microvolt = <750000 750000 950000>; clock-latency-ns = <40000>; opp-suspend;
opp-1608000000 {opp-hz = /bits/ 64 <1608000000>;opp-microvolt = <887500 887500 950000>;clock-latency-ns = <40000>;};opp-1704000000 {opp-hz = /bits/ 64 <1704000000>;opp-microvolt = <937500 937500 950000>;
opp-1296000000 {opp-hz = /bits/ 64 <1296000000>;opp-microvolt = <775000 775000 950000>;Got tricked by this one.
In the Rockchip vendor kernel, the opp-microvolt is 750000 750000 950000, so the same as CPU OPP 1.2GHz. However, the opp-microvolt-L1 and L0 are higher than that. Only a couple of the OPPs in vendor kernel actually have opp-microvolt-L* higher than opp-microvolt, that is a noteworthy oddity for anyone reviewing this patch :)
Anyway, that is correct, we take the highest voltage among all defined opp-microvolt* properties.
Indeed, and I also almost got tricked by the same OPP. :)
I only have comments on the commit log, the diff is fine so:
Reviewed-by: Quentin Schulz quentin.schulz@cherry.de
Thanks once again for your detailed review! :)
Hi Dragan,
On 3/23/25 11:19 AM, Dragan Simic wrote:
Hello Quentin,
Thanks for your comments! Please see some responses below.
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
""" despite the two SoC variant datasheets specifying the same upper limit for the allowed temperature for both variants """
is incorrect. The whole range is different, yes it's only a 5°C difference for the upper limit, but they still are different.
specified much lower for the RK3588J variant. [1][2]
To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to its datasheet, [2] the RK3588J variant can actually run safely at higher CPU and GPU OPPs as well, but only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted
"only when not enjoying the assumed extended temperature range" is extrapolated by me/us and not confirmed by Rockchip themselves. I've asked for a statement on what "industrial environment" they specify in the Normal Mode explanation means since it's the only time they use the term. I've yet to receive an answer. The only thing Rockchip in their datasheet is that the overdrive mode will shorten lifetime when used for a long time, especially in high temperature conditions. It's not clear whether we can use the overdrive mode even within the RK3588 typical range of operation.
True. I'll see to rephrase the patch description a bit in the v2, to avoid this kind of speculation. I mean, perhaps the speculation is right, but it hasn't been confirmed officially by Rockchip.
Speculation is fine, but it should be worded as such.
[...]
The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been
Interesting that we went for a different strategy for the GPU OPPs :)
Good point, and I'm fully aware of that. :) Actually, I'm rather sure that omitting the additional CPU OPPs does no good to us, but I didn't want to argue about that when they were dropped originally, before I can have some hard numbers to prove it in a repeatable way.
I assume we'll have some patch in the future with those added and those hard numbers you're talking about, so looking forward to seeing it on the ML :)
[...]
Helped-by: Quentin Schulz quentin.schulz@cherry.de
Reported-by/Suggested-by?
I don't see Helped-by in https://eur02.safelinks.protection.outlook.com/? url=https%3A%2F%2Fwww.kernel.org%2Fdoc%2Fhtml%2Flatest%2Fprocess%2Fsubmitting-patches.html%23using-reported-by-tested-by-reviewed-by-suggested-by-and-fixes&data=05%7C02%7Cquentin.schulz%40cherry.de%7Cdc754791b6844506b11c08dd69f444a7%7C5e0e1b5221b54e7b83bb514ec460677e%7C0%7C0%7C638783220330058516%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=4bv9pUh6aSD0GVLJ4Zvuyvox1K0xxwf83KXX86QsvMo%3D&reserved=0
I see 2496b2aaacf137250f4ca449f465e2cadaabb0e8 got the Helped-by replaced by a Suggested-by for example, but I see other patches with Helped-by... if that is a standard trailer for kernel patches, then maybe we should add it to that doc?
Actually, I already tried to get the Helped-by tag added to the kernel documentation, by submitting a small patch series. [*] Unfortunately, it got rejected. :/
However, Heiko accepts Helped-by tags and nobody higher up the tree seems to complain, so we should be fine. :) It isn't the case with all maintainers, though.
[*] https://eur02.safelinks.protection.outlook.com/? url=https%3A%2F%2Flore.kernel.org%2Fall%2Fcover.1730874296.git.dsimic%40manjaro.org%2FT%2F%23u&data=05%7C02%7Cquentin.schulz%40cherry.de%7Cdc754791b6844506b11c08dd69f444a7%7C5e0e1b5221b54e7b83bb514ec460677e%7C0%7C0%7C638783220330070422%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=3dZgSG%2FBT6f%2Ffqs7D30HvEl18SzqYPwNeUGWBZfMAqM%3D&reserved=0
Are you trying to up the numbers of Helped-by in commit logs to make it a reasonable request to add the trailer in the documentation :) ?
Cheers, Quentin
Hello Quentin,
On 2025-03-24 10:23, Quentin Schulz wrote:
On 3/23/25 11:19 AM, Dragan Simic wrote:
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
""" despite the two SoC variant datasheets specifying the same upper limit for the allowed temperature for both variants """
is incorrect. The whole range is different, yes it's only a 5°C difference for the upper limit, but they still are different.
I just commented on this separately, with a couple of datasheet screenshots, before I saw your latest response. Please, have a look at that message.
specified much lower for the RK3588J variant. [1][2]
To be on the safe side and to ensure maximum longevity of the RK3588J SoCs, only the CPU and GPU OPPs that are declared by the vendor to be always safe for this SoC variant may be provided. As explained by the vendor [3] and according to its datasheet, [2] the RK3588J variant can actually run safely at higher CPU and GPU OPPs as well, but only when not enjoying the assumed extended temperature range that the RK3588J, as an SoC variant targeted
"only when not enjoying the assumed extended temperature range" is extrapolated by me/us and not confirmed by Rockchip themselves. I've asked for a statement on what "industrial environment" they specify in the Normal Mode explanation means since it's the only time they use the term. I've yet to receive an answer. The only thing Rockchip in their datasheet is that the overdrive mode will shorten lifetime when used for a long time, especially in high temperature conditions. It's not clear whether we can use the overdrive mode even within the RK3588 typical range of operation.
True. I'll see to rephrase the patch description a bit in the v2, to avoid this kind of speculation. I mean, perhaps the speculation is right, but it hasn't been confirmed officially by Rockchip.
Speculation is fine, but it should be worded as such.
Agreed, because that's our understanding so far, but it needs to be explained a bit better.
The provided RK3588J CPU OPPs follow the slightly debatable "provide only the highest-frequency OPP from the same-voltage group" approach that's been
Interesting that we went for a different strategy for the GPU OPPs :)
Good point, and I'm fully aware of that. :) Actually, I'm rather sure that omitting the additional CPU OPPs does no good to us, but I didn't want to argue about that when they were dropped originally, before I can have some hard numbers to prove it in a repeatable way.
I assume we'll have some patch in the future with those added and those hard numbers you're talking about, so looking forward to seeing it on the ML :)
Indeed, that's the plan, and there should be even more patches, which should remove the slightly annoying "xyz OPP is inefficient" warnings emitted by the IPA governor. :)
Helped-by: Quentin Schulz quentin.schulz@cherry.de
Reported-by/Suggested-by?
I don't see Helped-by in https://eur02.safelinks.protection.outlook.com/? url=https%3A%2F%2Fwww.kernel.org%2Fdoc%2Fhtml%2Flatest%2Fprocess%2Fsubmitting-patches.html%23using-reported-by-tested-by-reviewed-by-suggested-by-and-fixes&data=05%7C02%7Cquentin.schulz%40cherry.de%7Cdc754791b6844506b11c08dd69f444a7%7C5e0e1b5221b54e7b83bb514ec460677e%7C0%7C0%7C638783220330058516%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=4bv9pUh6aSD0GVLJ4Zvuyvox1K0xxwf83KXX86QsvMo%3D&reserved=0
I see 2496b2aaacf137250f4ca449f465e2cadaabb0e8 got the Helped-by replaced by a Suggested-by for example, but I see other patches with Helped-by... if that is a standard trailer for kernel patches, then maybe we should add it to that doc?
Actually, I already tried to get the Helped-by tag added to the kernel documentation, by submitting a small patch series. [*] Unfortunately, it got rejected. :/
However, Heiko accepts Helped-by tags and nobody higher up the tree seems to complain, so we should be fine. :) It isn't the case with all maintainers, though.
[*] https://eur02.safelinks.protection.outlook.com/? url=https%3A%2F%2Flore.kernel.org%2Fall%2Fcover.1730874296.git.dsimic%40manjaro.org%2FT%2F%23u&data=05%7C02%7Cquentin.schulz%40cherry.de%7Cdc754791b6844506b11c08dd69f444a7%7C5e0e1b5221b54e7b83bb514ec460677e%7C0%7C0%7C638783220330070422%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=3dZgSG%2FBT6f%2Ffqs7D30HvEl18SzqYPwNeUGWBZfMAqM%3D&reserved=0
Are you trying to up the numbers of Helped-by in commit logs to make it a reasonable request to add the trailer in the documentation :) ?
It's just that Helped-by is, to me, of a bit "higher value" than Suggested-by or Reported-by, because Helped-by means that the tagged person contributed more to the patch than just suggesting it or reporting a bug. In addition, having more Helped-by tags present in various commits can help a bit with, possibly, making it officially supported at some point in the future. :)
Hi Dragan,
On 3/24/25 10:53 AM, Dragan Simic wrote:
Hello Quentin,
On 2025-03-24 10:23, Quentin Schulz wrote:
On 3/23/25 11:19 AM, Dragan Simic wrote:
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
""" despite the two SoC variant datasheets specifying the same upper limit for the allowed temperature for both variants """
is incorrect. The whole range is different, yes it's only a 5°C difference for the upper limit, but they still are different.
I just commented on this separately, with a couple of datasheet screenshots, before I saw your latest response. Please, have a look at that message.
I see, I had a v1.3 datasheet opened:
https://github.com/FanX-Tek/rk3588-TRM-and-Datasheet/blob/master/Rockchip_RK...
Interestingly, it seems the RK3588S (still?) has a smaller operating range:
https://www.armboard.cn/download/Rockchip_RK3588S_Datasheet_V1.6-20240821.pd...
Cheers, Quentin
On 2025-03-24 11:20, Quentin Schulz wrote:
On 3/24/25 10:53 AM, Dragan Simic wrote:
On 2025-03-24 10:23, Quentin Schulz wrote:
On 3/23/25 11:19 AM, Dragan Simic wrote:
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
""" despite the two SoC variant datasheets specifying the same upper limit for the allowed temperature for both variants """
is incorrect. The whole range is different, yes it's only a 5°C difference for the upper limit, but they still are different.
I just commented on this separately, with a couple of datasheet screenshots, before I saw your latest response. Please, have a look at that message.
I see, I had a v1.3 datasheet opened:
https://github.com/FanX-Tek/rk3588-TRM-and-Datasheet/blob/master/Rockchip_RK...
Yup, the v1.6 of the RK3588 datasheet increased the upper ambient temperature limit from 80 to 85 oC.
Interestingly, it seems the RK3588S (still?) has a smaller operating range:
https://www.armboard.cn/download/Rockchip_RK3588S_Datasheet_V1.6-20240821.pd...
Oh, that's quite interesting, I had the v1.5 as the newest version. The v1.6 of the RK3588S datasheet actually lowered the upper ambient temperature limit from 85 down to 80 oC.
Hello Quentin,
On 2025-03-23 11:19, Dragan Simic wrote:
On 2025-03-21 10:53, Quentin Schulz wrote:
On 3/21/25 4:28 AM, Dragan Simic wrote:
The differences in the vendor-approved CPU and GPU OPPs for the standard Rockchip RK3588 variant [1] and the industrial Rockchip RK3588J variant [2] come from the latter, presumably, supporting an extended temperature range that's usually associated with industrial applications, despite the two SoC variant datasheets specifying the same upper limit for the allowed ambient temperature for both variants. However, the lower temperature limit is
RK3588 is rated for 0-80°C, RK3588J for -40-85°C, c.f. Recommended Operating Conditions, Table 3-2, Ambient Operating Temperature.
Indeed, which is why I specifically wrote "specifying the same upper limit", because having a lower negative temperature limit could hardly put the RK3588J in danger of overheating or running hotter. :)
Oh, now I see what you actually wrote above, which I misread a bit initially... In fact, the upper ambient temperature limit is the same for both RK3588J and RK3588, according to the datasheets; it's just that the lower temperature limit is much lower for the RK3588J, which the patch description says.
I'm not sure where did you find the 80 vs. 85 oC difference; please, see a couple of screenshots from the datasheets, linked below:
- RK3588 datasheet v1.6 (v1.7 is the same): https://0x0.st/8j1a.png - RK3588J datasheet v1.1: https://0x0.st/8j1m.png
The v2 of the patch is coming soon, with the patch description improved according to your suggestions.
linux-stable-mirror@lists.linaro.org
-
Dragan Simic -
Quentin Schulz