Update the inline-crypto engine DT binding to reflect that power-domain and
clock-names are now mandatory. Also update the maximum number of clocks
that can be specified to two. These new fields are mandatory because ICE
needs to vote on the power domain before it attempts to vote on the core
and iface clocks to avoid clock 'stuck' issues.
Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com>
---
.../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml
index c3408dcf5d20..1c2416117d4c 100644
--- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml
+++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml
@@ -28,12 +28,20 @@ properties:
maxItems: 1
clocks:
+ maxItems: 2
+
+ clock-names:
+ maxItems: 2
+
+ power-domains:
maxItems: 1
required:
- compatible
- reg
- clocks
+ - clock-names
+ - power-domains
additionalProperties: false
@@ -45,6 +53,10 @@ examples:
compatible = "qcom,sm8550-inline-crypto-engine",
"qcom,inline-crypto-engine";
reg = <0x01d88000 0x8000>;
- clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>;
+ clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>,
+ <&gcc GCC_UFS_PHY_AHB_CLK>;
+ clock-names = "ice_core_clk",
+ "iface_clk";
+ power-domains = <&gcc UFS_PHY_GDSC>;
};
...
--
2.34.1On 23/01/2026 08:11, Harshal Dev wrote: > Update the inline-crypto engine DT binding to reflect that power-domain and > clock-names are now mandatory. Also update the maximum number of clocks > that can be specified to two. These new fields are mandatory because ICE > needs to vote on the power domain before it attempts to vote on the core > and iface clocks to avoid clock 'stuck' issues. > > Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> > --- > .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- > 1 file changed, 13 insertions(+), 1 deletion(-) > > diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml > index c3408dcf5d20..1c2416117d4c 100644 > --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml > +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml > @@ -28,12 +28,20 @@ properties: > maxItems: 1 > > clocks: > + maxItems: 2 This is ABI break and your commit msg suggests things were not perfect, but it is not explicit - was this working or not? How is it that ICE was never tested? > + > + clock-names: > + maxItems: 2 Why "yellowpony" is a correct name? Please look at existing code to see how this is done. > + > + power-domains: > maxItems: 1 > > required: > - compatible > - reg > - clocks > + - clock-names > + - power-domains Another ABI break... Best regards, Krzysztof
Hi Krzysztof, On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: > On 23/01/2026 08:11, Harshal Dev wrote: >> Update the inline-crypto engine DT binding to reflect that power-domain and >> clock-names are now mandatory. Also update the maximum number of clocks >> that can be specified to two. These new fields are mandatory because ICE >> needs to vote on the power domain before it attempts to vote on the core >> and iface clocks to avoid clock 'stuck' issues. >> >> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >> --- >> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >> 1 file changed, 13 insertions(+), 1 deletion(-) >> >> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >> index c3408dcf5d20..1c2416117d4c 100644 >> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >> @@ -28,12 +28,20 @@ properties: >> maxItems: 1 >> >> clocks: >> + maxItems: 2 > > This is ABI break and your commit msg suggests things were not perfect, > but it is not explicit - was this working or not? How is it that ICE was > never tested? > I took some time to educate myself on the point of DT bindings stability being a strict requirement now, so I understand how these changes are breaking ABI, I'll send a better version of this again. As for your question of how it was working till now, it seems that things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS depends) initiates probe _before_ the unused clocks and power-domains are disabled by the kernel. And so, the un-clocked register access or clock 'stuck' isn't observed (since the clocks and power domains are already enabled). Perhaps I should write this scenario explicitly in the commit message? To maintain backward compatibility, let me introduce minItems and maxItems for clocks. When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as before. >> + >> + clock-names: >> + maxItems: 2 > > Why "yellowpony" is a correct name? Please look at existing code to see > how this is done. > Ack, I will try to list down the acceptable clock names here. > >> + >> + power-domains: >> maxItems: 1 >> >> required: >> - compatible >> - reg >> - clocks >> + - clock-names >> + - power-domains > > Another ABI break... Let me avoid adding clock-names and power-domains as required. Since for scenarios where a Linux distro uses CONFIG_SCSI_UFS_QCOM=y, this isn't mandatory as I explained above. > > Best regards, > Krzysztof Regards, Harshal
On 1/23/26 12:04 PM, Harshal Dev wrote: > Hi Krzysztof, > > On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >> On 23/01/2026 08:11, Harshal Dev wrote: >>> Update the inline-crypto engine DT binding to reflect that power-domain and >>> clock-names are now mandatory. Also update the maximum number of clocks >>> that can be specified to two. These new fields are mandatory because ICE >>> needs to vote on the power domain before it attempts to vote on the core >>> and iface clocks to avoid clock 'stuck' issues. >>> >>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>> --- >>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>> 1 file changed, 13 insertions(+), 1 deletion(-) >>> >>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>> index c3408dcf5d20..1c2416117d4c 100644 >>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>> @@ -28,12 +28,20 @@ properties: >>> maxItems: 1 >>> >>> clocks: >>> + maxItems: 2 >> >> This is ABI break and your commit msg suggests things were not perfect, >> but it is not explicit - was this working or not? How is it that ICE was >> never tested? >> > > I took some time to educate myself on the point of DT bindings stability being a > strict requirement now, so I understand how these changes are breaking ABI, I'll > send a better version of this again. > > As for your question of how it was working till now, it seems that > things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM > flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS > depends) initiates probe _before_ the unused clocks and power-domains are > disabled by the kernel. And so, the un-clocked register access or clock 'stuck' > isn't observed (since the clocks and power domains are already enabled). > Perhaps I should write this scenario explicitly in the commit message? > > To maintain backward compatibility, let me introduce minItems and maxItems for clocks. > When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as > before. You must not assume any particular kernel configuration clk_ignore_unused is a hack which leads to situations like this, since the bootloader doesn't clean up clocks it turned on, which leads to situations like this where someone who previously wrote this binding didn't care enough to **actually** test whether this device can operate with only the set of clocks it requires I believe in this case it absolutely makes sense to break things, but you must put the backstory in writing, in the commit message Konrad
Hi Krzysztof and Konrad, On 1/26/2026 3:59 PM, Konrad Dybcio wrote: > On 1/23/26 12:04 PM, Harshal Dev wrote: >> Hi Krzysztof, >> >> On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >>> On 23/01/2026 08:11, Harshal Dev wrote: >>>> Update the inline-crypto engine DT binding to reflect that power-domain and >>>> clock-names are now mandatory. Also update the maximum number of clocks >>>> that can be specified to two. These new fields are mandatory because ICE >>>> needs to vote on the power domain before it attempts to vote on the core >>>> and iface clocks to avoid clock 'stuck' issues. >>>> >>>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>>> --- >>>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>> >>>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>> index c3408dcf5d20..1c2416117d4c 100644 >>>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>> @@ -28,12 +28,20 @@ properties: >>>> maxItems: 1 >>>> >>>> clocks: >>>> + maxItems: 2 >>> >>> This is ABI break and your commit msg suggests things were not perfect, >>> but it is not explicit - was this working or not? How is it that ICE was >>> never tested? >>> >> >> I took some time to educate myself on the point of DT bindings stability being a >> strict requirement now, so I understand how these changes are breaking ABI, I'll >> send a better version of this again. >> >> As for your question of how it was working till now, it seems that >> things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM >> flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS >> depends) initiates probe _before_ the unused clocks and power-domains are >> disabled by the kernel. And so, the un-clocked register access or clock 'stuck' >> isn't observed (since the clocks and power domains are already enabled). >> Perhaps I should write this scenario explicitly in the commit message? >> >> To maintain backward compatibility, let me introduce minItems and maxItems for clocks. >> When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as >> before. > > You must not assume any particular kernel configuration > > clk_ignore_unused is a hack which leads to situations like this, since > the bootloader doesn't clean up clocks it turned on, which leads to > situations like this where someone who previously wrote this binding > didn't care enough to **actually** test whether this device can operate > with only the set of clocks it requires > > I believe in this case it absolutely makes sense to break things, but > you must put the backstory in writing, in the commit message > I took some more time to think this through, and I agree with you now Konrad. These DT bindings appear to be invalid from day-1. ICE being an independent and common IP for both UFS and SDCC, it cannot operate correctly without its power-domain and clocks being enabled first. Hence, it should be mandatory for them to be specified in the DT-node and the same should be reflected in the DT binding. The only reason I can think of for omitting the 'power-domain' and 'iface' clock in the original DT-binding for ICE is because we failed to test the driver on a production kernel where the 'clk_ignore_unused' flag is not passed on the cmdline. Or if we did test that way, we were just lucky to not run into a timing scenario where the probe for the driver is attempted _after_ the clocks are turned off by the kernel. Sending a new patch, which makes these two resources optional (to preserve the DT binding) would either imply that we are make this bug fix optional as well or asking the reporter to resort to some workaround such as overriding CONFIG_SCSI_UFS_QCOM to 'y'. Let us know your thoughts on this Krzysztof. Thanks, Harshal > Konrad
On 03/02/2026 10:26, Harshal Dev wrote: > Hi Krzysztof and Konrad, > > On 1/26/2026 3:59 PM, Konrad Dybcio wrote: >> On 1/23/26 12:04 PM, Harshal Dev wrote: >>> Hi Krzysztof, >>> >>> On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >>>> On 23/01/2026 08:11, Harshal Dev wrote: >>>>> Update the inline-crypto engine DT binding to reflect that power-domain and >>>>> clock-names are now mandatory. Also update the maximum number of clocks >>>>> that can be specified to two. These new fields are mandatory because ICE >>>>> needs to vote on the power domain before it attempts to vote on the core >>>>> and iface clocks to avoid clock 'stuck' issues. >>>>> >>>>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>>>> --- >>>>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>>> >>>>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>> index c3408dcf5d20..1c2416117d4c 100644 >>>>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>> @@ -28,12 +28,20 @@ properties: >>>>> maxItems: 1 >>>>> >>>>> clocks: >>>>> + maxItems: 2 >>>> >>>> This is ABI break and your commit msg suggests things were not perfect, >>>> but it is not explicit - was this working or not? How is it that ICE was >>>> never tested? >>>> >>> >>> I took some time to educate myself on the point of DT bindings stability being a >>> strict requirement now, so I understand how these changes are breaking ABI, I'll >>> send a better version of this again. >>> >>> As for your question of how it was working till now, it seems that >>> things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM >>> flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS >>> depends) initiates probe _before_ the unused clocks and power-domains are >>> disabled by the kernel. And so, the un-clocked register access or clock 'stuck' >>> isn't observed (since the clocks and power domains are already enabled). >>> Perhaps I should write this scenario explicitly in the commit message? >>> >>> To maintain backward compatibility, let me introduce minItems and maxItems for clocks. >>> When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as >>> before. >> >> You must not assume any particular kernel configuration >> >> clk_ignore_unused is a hack which leads to situations like this, since >> the bootloader doesn't clean up clocks it turned on, which leads to >> situations like this where someone who previously wrote this binding >> didn't care enough to **actually** test whether this device can operate >> with only the set of clocks it requires >> >> I believe in this case it absolutely makes sense to break things, but >> you must put the backstory in writing, in the commit message >> > > I took some more time to think this through, and I agree with you now Konrad. > > These DT bindings appear to be invalid from day-1. ICE being an independent > and common IP for both UFS and SDCC, it cannot operate correctly without its > power-domain and clocks being enabled first. Hence, it should be mandatory for > them to be specified in the DT-node and the same should be reflected in the DT > binding. > > The only reason I can think of for omitting the 'power-domain' and 'iface' clock > in the original DT-binding for ICE is because we failed to test the driver on > a production kernel where the 'clk_ignore_unused' flag is not passed on the cmdline. That's a reason to change ABI in the bindings, but not a reason to break in-kernel or out of tree DTS. > Or if we did test that way, we were just lucky to not run into a timing scenario > where the probe for the driver is attempted _after_ the clocks are turned off by the > kernel. > > Sending a new patch, which makes these two resources optional (to preserve the DT > binding) would either imply that we are make this bug fix optional as well or > asking the reporter to resort to some workaround such as overriding > CONFIG_SCSI_UFS_QCOM to 'y'. Either I do not understand the point or you still insist on breaking a working DTS on kernels with clk_ignore_unused, just because what exactly? You claim it did not work, but in fact it did work. So you claim it worked by luck, right? And what this patchset achieves? It breaks this "work by luck" into "100% not working and broken". I do not see how is this an improvement. My NAK for driver change stays. This is wrong approach - you cannot break working DTS. Best regards, Krzysztof
Hi Krzysztof, On 2/5/2026 4:47 PM, Krzysztof Kozlowski wrote: > On 03/02/2026 10:26, Harshal Dev wrote: >> Hi Krzysztof and Konrad, >> >> On 1/26/2026 3:59 PM, Konrad Dybcio wrote: >>> On 1/23/26 12:04 PM, Harshal Dev wrote: >>>> Hi Krzysztof, >>>> >>>> On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >>>>> On 23/01/2026 08:11, Harshal Dev wrote: >>>>>> Update the inline-crypto engine DT binding to reflect that power-domain and >>>>>> clock-names are now mandatory. Also update the maximum number of clocks >>>>>> that can be specified to two. These new fields are mandatory because ICE >>>>>> needs to vote on the power domain before it attempts to vote on the core >>>>>> and iface clocks to avoid clock 'stuck' issues. >>>>>> >>>>>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>>>>> --- >>>>>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>>>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>>>> >>>>>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>> index c3408dcf5d20..1c2416117d4c 100644 >>>>>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>> @@ -28,12 +28,20 @@ properties: >>>>>> maxItems: 1 >>>>>> >>>>>> clocks: >>>>>> + maxItems: 2 >>>>> >>>>> This is ABI break and your commit msg suggests things were not perfect, >>>>> but it is not explicit - was this working or not? How is it that ICE was >>>>> never tested? >>>>> >>>> >>>> I took some time to educate myself on the point of DT bindings stability being a >>>> strict requirement now, so I understand how these changes are breaking ABI, I'll >>>> send a better version of this again. >>>> >>>> As for your question of how it was working till now, it seems that >>>> things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM >>>> flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS >>>> depends) initiates probe _before_ the unused clocks and power-domains are >>>> disabled by the kernel. And so, the un-clocked register access or clock 'stuck' >>>> isn't observed (since the clocks and power domains are already enabled). >>>> Perhaps I should write this scenario explicitly in the commit message? >>>> >>>> To maintain backward compatibility, let me introduce minItems and maxItems for clocks. >>>> When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as >>>> before. >>> >>> You must not assume any particular kernel configuration >>> >>> clk_ignore_unused is a hack which leads to situations like this, since >>> the bootloader doesn't clean up clocks it turned on, which leads to >>> situations like this where someone who previously wrote this binding >>> didn't care enough to **actually** test whether this device can operate >>> with only the set of clocks it requires >>> >>> I believe in this case it absolutely makes sense to break things, but >>> you must put the backstory in writing, in the commit message >>> >> >> I took some more time to think this through, and I agree with you now Konrad. >> >> These DT bindings appear to be invalid from day-1. ICE being an independent >> and common IP for both UFS and SDCC, it cannot operate correctly without its >> power-domain and clocks being enabled first. Hence, it should be mandatory for >> them to be specified in the DT-node and the same should be reflected in the DT >> binding. >> >> The only reason I can think of for omitting the 'power-domain' and 'iface' clock >> in the original DT-binding for ICE is because we failed to test the driver on >> a production kernel where the 'clk_ignore_unused' flag is not passed on the cmdline. > > That's a reason to change ABI in the bindings, but not a reason to break > in-kernel or out of tree DTS. > >> Or if we did test that way, we were just lucky to not run into a timing scenario >> where the probe for the driver is attempted _after_ the clocks are turned off by the >> kernel. >> >> Sending a new patch, which makes these two resources optional (to preserve the DT >> binding) would either imply that we are make this bug fix optional as well or >> asking the reporter to resort to some workaround such as overriding >> CONFIG_SCSI_UFS_QCOM to 'y'. > > Either I do not understand the point or you still insist on breaking a > working DTS on kernels with clk_ignore_unused, just because what > exactly? You claim it did not work, but in fact it did work. So you > claim it worked by luck, right? And what this patchset achieves? It > breaks this "work by luck" into "100% not working and broken". I do not > see how is this an improvement. > My point is something more fundamental. It worked before and it will still continue to work if: 1. We pass the 'clk_ignore_unused' flag. or, 2. If the Linux distro is overriding CONFIG_SCSI_UFS_QCOM to 'y'. But that does not change the fact that the current DT binding does not fully describe all the resources required by the hardware block to function correctly. > My NAK for driver change stays. This is wrong approach - you cannot > break working DTS. > I agree that this patch in it's current form will break both the in-kernel and out of tree DTS written in accordance with the old binding. If this isn't acceptable at all then like you said we need to move forward in a way that preserves them. But I am trying to highlight that if we go forward this way, we are all agreeing that kernels with these old DTS can only work under the two conditions I described above. If the kernel doesn't obey either of those conditions, the kernel won't boot. If we are all aligned that preserving the older DTS is very important, since they still 'work' under certain conditions, then sure. I'm fine with sending a new patch that doesn't break the ABI as per my initial response. Let me know your thoughts. Cheers, Harshal > Best regards, > Krzysztof
On 06/02/2026 11:07, Harshal Dev wrote: > Hi Krzysztof, > > On 2/5/2026 4:47 PM, Krzysztof Kozlowski wrote: >> On 03/02/2026 10:26, Harshal Dev wrote: >>> Hi Krzysztof and Konrad, >>> >>> On 1/26/2026 3:59 PM, Konrad Dybcio wrote: >>>> On 1/23/26 12:04 PM, Harshal Dev wrote: >>>>> Hi Krzysztof, >>>>> >>>>> On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >>>>>> On 23/01/2026 08:11, Harshal Dev wrote: >>>>>>> Update the inline-crypto engine DT binding to reflect that power-domain and >>>>>>> clock-names are now mandatory. Also update the maximum number of clocks >>>>>>> that can be specified to two. These new fields are mandatory because ICE >>>>>>> needs to vote on the power domain before it attempts to vote on the core >>>>>>> and iface clocks to avoid clock 'stuck' issues. >>>>>>> >>>>>>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>>>>>> --- >>>>>>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>>>>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>>>>> >>>>>>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>> index c3408dcf5d20..1c2416117d4c 100644 >>>>>>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>> @@ -28,12 +28,20 @@ properties: >>>>>>> maxItems: 1 >>>>>>> >>>>>>> clocks: >>>>>>> + maxItems: 2 >>>>>> >>>>>> This is ABI break and your commit msg suggests things were not perfect, >>>>>> but it is not explicit - was this working or not? How is it that ICE was >>>>>> never tested? >>>>>> >>>>> >>>>> I took some time to educate myself on the point of DT bindings stability being a >>>>> strict requirement now, so I understand how these changes are breaking ABI, I'll >>>>> send a better version of this again. >>>>> >>>>> As for your question of how it was working till now, it seems that >>>>> things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM >>>>> flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS >>>>> depends) initiates probe _before_ the unused clocks and power-domains are >>>>> disabled by the kernel. And so, the un-clocked register access or clock 'stuck' >>>>> isn't observed (since the clocks and power domains are already enabled). >>>>> Perhaps I should write this scenario explicitly in the commit message? >>>>> >>>>> To maintain backward compatibility, let me introduce minItems and maxItems for clocks. >>>>> When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as >>>>> before. >>>> >>>> You must not assume any particular kernel configuration >>>> >>>> clk_ignore_unused is a hack which leads to situations like this, since >>>> the bootloader doesn't clean up clocks it turned on, which leads to >>>> situations like this where someone who previously wrote this binding >>>> didn't care enough to **actually** test whether this device can operate >>>> with only the set of clocks it requires >>>> >>>> I believe in this case it absolutely makes sense to break things, but >>>> you must put the backstory in writing, in the commit message >>>> >>> >>> I took some more time to think this through, and I agree with you now Konrad. >>> >>> These DT bindings appear to be invalid from day-1. ICE being an independent >>> and common IP for both UFS and SDCC, it cannot operate correctly without its >>> power-domain and clocks being enabled first. Hence, it should be mandatory for >>> them to be specified in the DT-node and the same should be reflected in the DT >>> binding. >>> >>> The only reason I can think of for omitting the 'power-domain' and 'iface' clock >>> in the original DT-binding for ICE is because we failed to test the driver on >>> a production kernel where the 'clk_ignore_unused' flag is not passed on the cmdline. >> >> That's a reason to change ABI in the bindings, but not a reason to break >> in-kernel or out of tree DTS. >> >>> Or if we did test that way, we were just lucky to not run into a timing scenario >>> where the probe for the driver is attempted _after_ the clocks are turned off by the >>> kernel. >>> >>> Sending a new patch, which makes these two resources optional (to preserve the DT >>> binding) would either imply that we are make this bug fix optional as well or >>> asking the reporter to resort to some workaround such as overriding >>> CONFIG_SCSI_UFS_QCOM to 'y'. >> >> Either I do not understand the point or you still insist on breaking a >> working DTS on kernels with clk_ignore_unused, just because what >> exactly? You claim it did not work, but in fact it did work. So you >> claim it worked by luck, right? And what this patchset achieves? It >> breaks this "work by luck" into "100% not working and broken". I do not >> see how is this an improvement. >> > > My point is something more fundamental. It worked before and it will still continue > to work if: > 1. We pass the 'clk_ignore_unused' flag. or, > 2. If the Linux distro is overriding CONFIG_SCSI_UFS_QCOM to 'y'. I do not agree with this. I already commented about your driver. If you do not believe me, apply your driver patch and show the test results of existing working device. > > But that does not change the fact that the current DT binding does not fully describe all > the resources required by the hardware block to function correctly. > >> My NAK for driver change stays. This is wrong approach - you cannot >> break working DTS. >> > > I agree that this patch in it's current form will break both the in-kernel and > out of tree DTS written in accordance with the old binding. If this isn't acceptable What? You just said few lines above: "it will still continue to work if:" So either this will continue to work or not. I don't understand this thread and honestly do not have patience for it. I gave you already reasoning what is wrong and why it is. Now it is just wasting my time. Best regards, Krzysztof
Hi Krzysztof, On 2/6/2026 4:20 PM, Krzysztof Kozlowski wrote: > On 06/02/2026 11:07, Harshal Dev wrote: >> Hi Krzysztof, >> >> On 2/5/2026 4:47 PM, Krzysztof Kozlowski wrote: >>> On 03/02/2026 10:26, Harshal Dev wrote: >>>> Hi Krzysztof and Konrad, >>>> >>>> On 1/26/2026 3:59 PM, Konrad Dybcio wrote: >>>>> On 1/23/26 12:04 PM, Harshal Dev wrote: >>>>>> Hi Krzysztof, >>>>>> >>>>>> On 1/23/2026 2:27 PM, Krzysztof Kozlowski wrote: >>>>>>> On 23/01/2026 08:11, Harshal Dev wrote: >>>>>>>> Update the inline-crypto engine DT binding to reflect that power-domain and >>>>>>>> clock-names are now mandatory. Also update the maximum number of clocks >>>>>>>> that can be specified to two. These new fields are mandatory because ICE >>>>>>>> needs to vote on the power domain before it attempts to vote on the core >>>>>>>> and iface clocks to avoid clock 'stuck' issues. >>>>>>>> >>>>>>>> Signed-off-by: Harshal Dev <harshal.dev@oss.qualcomm.com> >>>>>>>> --- >>>>>>>> .../bindings/crypto/qcom,inline-crypto-engine.yaml | 14 +++++++++++++- >>>>>>>> 1 file changed, 13 insertions(+), 1 deletion(-) >>>>>>>> >>>>>>>> diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>>> index c3408dcf5d20..1c2416117d4c 100644 >>>>>>>> --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>>> +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml >>>>>>>> @@ -28,12 +28,20 @@ properties: >>>>>>>> maxItems: 1 >>>>>>>> >>>>>>>> clocks: >>>>>>>> + maxItems: 2 >>>>>>> >>>>>>> This is ABI break and your commit msg suggests things were not perfect, >>>>>>> but it is not explicit - was this working or not? How is it that ICE was >>>>>>> never tested? >>>>>>> >>>>>> >>>>>> I took some time to educate myself on the point of DT bindings stability being a >>>>>> strict requirement now, so I understand how these changes are breaking ABI, I'll >>>>>> send a better version of this again. >>>>>> >>>>>> As for your question of how it was working till now, it seems that >>>>>> things were tested with the 'clk_ignore_unused' flag, or with CONFIG_SCSI_UFS_QCOM >>>>>> flag being override set to 'y'. When this is done, QCOM-ICE (on which QCOM-UFS >>>>>> depends) initiates probe _before_ the unused clocks and power-domains are >>>>>> disabled by the kernel. And so, the un-clocked register access or clock 'stuck' >>>>>> isn't observed (since the clocks and power domains are already enabled). >>>>>> Perhaps I should write this scenario explicitly in the commit message? >>>>>> >>>>>> To maintain backward compatibility, let me introduce minItems and maxItems for clocks. >>>>>> When the Linux distro uses CONFIG_SCSI_UFS_QCOM=y, we can do with just 1 clock as >>>>>> before. >>>>> >>>>> You must not assume any particular kernel configuration >>>>> >>>>> clk_ignore_unused is a hack which leads to situations like this, since >>>>> the bootloader doesn't clean up clocks it turned on, which leads to >>>>> situations like this where someone who previously wrote this binding >>>>> didn't care enough to **actually** test whether this device can operate >>>>> with only the set of clocks it requires >>>>> >>>>> I believe in this case it absolutely makes sense to break things, but >>>>> you must put the backstory in writing, in the commit message >>>>> >>>> >>>> I took some more time to think this through, and I agree with you now Konrad. >>>> >>>> These DT bindings appear to be invalid from day-1. ICE being an independent >>>> and common IP for both UFS and SDCC, it cannot operate correctly without its >>>> power-domain and clocks being enabled first. Hence, it should be mandatory for >>>> them to be specified in the DT-node and the same should be reflected in the DT >>>> binding. >>>> >>>> The only reason I can think of for omitting the 'power-domain' and 'iface' clock >>>> in the original DT-binding for ICE is because we failed to test the driver on >>>> a production kernel where the 'clk_ignore_unused' flag is not passed on the cmdline. >>> >>> That's a reason to change ABI in the bindings, but not a reason to break >>> in-kernel or out of tree DTS. >>> >>>> Or if we did test that way, we were just lucky to not run into a timing scenario >>>> where the probe for the driver is attempted _after_ the clocks are turned off by the >>>> kernel. >>>> >>>> Sending a new patch, which makes these two resources optional (to preserve the DT >>>> binding) would either imply that we are make this bug fix optional as well or >>>> asking the reporter to resort to some workaround such as overriding >>>> CONFIG_SCSI_UFS_QCOM to 'y'. >>> >>> Either I do not understand the point or you still insist on breaking a >>> working DTS on kernels with clk_ignore_unused, just because what >>> exactly? You claim it did not work, but in fact it did work. So you >>> claim it worked by luck, right? And what this patchset achieves? It >>> breaks this "work by luck" into "100% not working and broken". I do not >>> see how is this an improvement. >>> >> >> My point is something more fundamental. It worked before and it will still continue >> to work if: >> 1. We pass the 'clk_ignore_unused' flag. or, >> 2. If the Linux distro is overriding CONFIG_SCSI_UFS_QCOM to 'y'. > > I do not agree with this. I already commented about your driver. If you > do not believe me, apply your driver patch and show the test results of > existing working device. > > Apologies, it seems like I failed to explain correctly what I meant. Here I was talking about the existing in-tree ICE driver and not about this particular DT binding commit. This commit, as you rightly said and I mentioned below too, breaks backward compatibility for existing in-tree and out-of-tree DTS. >> >> But that does not change the fact that the current DT binding does not fully describe all >> the resources required by the hardware block to function correctly. >> >>> My NAK for driver change stays. This is wrong approach - you cannot >>> break working DTS. >>> >> >> I agree that this patch in it's current form will break both the in-kernel and >> out of tree DTS written in accordance with the old binding. If this isn't acceptable > > What? You just said few lines above: > "it will still continue to work if:" > I hope I am clear now, 'it' referred to the in-tree ICE driver and not to this particular DT schema commit. :) > So either this will continue to work or not. I don't understand this > thread and honestly do not have patience for it. I gave you already > reasoning what is wrong and why it is. Now it is just wasting my time. > Apologies again for the confusion. I totally agree, as replied previously too, that the updated DT binding breaks backward compatibility. Like I said, I will post another patch to preserve the correctness of existing in-tree and out-of-tree DTS. The only point I am trying to highlight for everyone's awareness is that as per this bug report https://lore.kernel.org/all/ZZYTYsaNUuWQg3tR@x1/ the kernel fails to boot with the existing DTS when the above two conditions aren't satisfied. Thank you, Harshal > Best regards, > Krzysztof
On 09/02/2026 06:43, Harshal Dev wrote: >>>> Either I do not understand the point or you still insist on breaking a >>>> working DTS on kernels with clk_ignore_unused, just because what >>>> exactly? You claim it did not work, but in fact it did work. So you >>>> claim it worked by luck, right? And what this patchset achieves? It >>>> breaks this "work by luck" into "100% not working and broken". I do not >>>> see how is this an improvement. >>>> >>> >>> My point is something more fundamental. It worked before and it will still continue >>> to work if: >>> 1. We pass the 'clk_ignore_unused' flag. or, >>> 2. If the Linux distro is overriding CONFIG_SCSI_UFS_QCOM to 'y'. >> >> I do not agree with this. I already commented about your driver. If you >> do not believe me, apply your driver patch and show the test results of >> existing working device. >> >> > > Apologies, it seems like I failed to explain correctly what I meant. > Here I was talking about the existing in-tree ICE driver and not about this particular DT > binding commit. This commit, as you rightly said and I mentioned below too, breaks backward > compatibility for existing in-tree and out-of-tree DTS. I was also talking about existing in-tree ICE driver, not about my commit. I do not believe anything will work fine with existing in-tree ICE driver. We shifted away from this binding in the discussion, imagine it does not matter. Just look at your driver code. Best regards, Krzysztof
© 2016 - 2026 Red Hat, Inc.