mbox series

[jammy,xilinx-zynqmp,0/1] Fix backported kria device tree changes

Message ID 20240220043059.276750-1-portia.stephens@canonical.com
Headers show
Series Fix backported kria device tree changes | expand

Message

Portia Stephens Feb. 20, 2024, 4:30 a.m. UTC
[ Impact ]

* Kria device tree's were backported from Xilinx's 6.1 tree in order to add
support for the KD240 platform
(https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
* Testing had previously been done a development branch with non-upstreamable
patches. 3 issues were introduced to the KD240 image that was not present on
the development branch.
* Since all Xilinx device tree's are so interdependent all Kria and ZCU device
trees were updated including certified platforms.

[ Test Plan ]

* QA will run certification testing on the KD240 platform
* Normal certification testing will be run on all other certified platforms

[ Where problems could occur ]

* This impacts the device tree for certified Xilinx platforms which could break
any of the device touched by the change.

[ Other Info ]

* Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366

* Private launchpad bugs that contain the regressions failure:
https://bugs.launchpad.net/limerick/+bug/2051228
https://bugs.launchpad.net/limerick/+bug/2051224
https://bugs.launchpad.net/limerick/+bug/2051201

Portia Stephens (1):
  UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes

 .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
 arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
 2 files changed, 25 insertions(+), 19 deletions(-)

Comments

Manuel Diewald Feb. 20, 2024, 6:39 p.m. UTC | #1
On Tue, Feb 20, 2024 at 02:30:58PM +1000, Portia Stephens wrote:
> [ Impact ]
> 
> * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> support for the KD240 platform
> (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> * Testing had previously been done a development branch with non-upstreamable
> patches. 3 issues were introduced to the KD240 image that was not present on
> the development branch.
> * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> trees were updated including certified platforms.
> 
> [ Test Plan ]
> 
> * QA will run certification testing on the KD240 platform
> * Normal certification testing will be run on all other certified platforms
> 
> [ Where problems could occur ]
> 
> * This impacts the device tree for certified Xilinx platforms which could break
> any of the device touched by the change.
> 
> [ Other Info ]
> 
> * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> 
> * Private launchpad bugs that contain the regressions failure:
> https://bugs.launchpad.net/limerick/+bug/2051228
> https://bugs.launchpad.net/limerick/+bug/2051224
> https://bugs.launchpad.net/limerick/+bug/2051201
> 
> Portia Stephens (1):
>   UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> 
>  .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
>  arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
>  2 files changed, 25 insertions(+), 19 deletions(-)
> 
> -- 
> 2.34.1
> 
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team

The Launchpad bug should target series 'Jammy'. A brief [Fix] section
could also be helpful, to give some context to the changes maybe. Either
way, LGTM.

Acked-by: Manuel Diewald <manuel.diewald@canonical.com>
Wen-chien Jesse Sung Feb. 21, 2024, 2:06 a.m. UTC | #2
Acked-by: Wen-chien Jesse Sung <jesse.sung@canonical.com>
Portia Stephens Feb. 21, 2024, 2:38 a.m. UTC | #3
On Wed, Feb 21, 2024 at 4:39 AM Manuel Diewald
<manuel.diewald@canonical.com> wrote:
>
> On Tue, Feb 20, 2024 at 02:30:58PM +1000, Portia Stephens wrote:
> > [ Impact ]
> >
> > * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> > support for the KD240 platform
> > (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> > * Testing had previously been done a development branch with non-upstreamable
> > patches. 3 issues were introduced to the KD240 image that was not present on
> > the development branch.
> > * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> > trees were updated including certified platforms.
> >
> > [ Test Plan ]
> >
> > * QA will run certification testing on the KD240 platform
> > * Normal certification testing will be run on all other certified platforms
> >
> > [ Where problems could occur ]
> >
> > * This impacts the device tree for certified Xilinx platforms which could break
> > any of the device touched by the change.
> >
> > [ Other Info ]
> >
> > * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> >
> > * Private launchpad bugs that contain the regressions failure:
> > https://bugs.launchpad.net/limerick/+bug/2051228
> > https://bugs.launchpad.net/limerick/+bug/2051224
> > https://bugs.launchpad.net/limerick/+bug/2051201
> >
> > Portia Stephens (1):
> >   UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> >
> >  .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
> >  arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
> >  2 files changed, 25 insertions(+), 19 deletions(-)
> >
> > --
> > 2.34.1
> >
> >
> > --
> > kernel-team mailing list
> > kernel-team@lists.ubuntu.com
> > https://lists.ubuntu.com/mailman/listinfo/kernel-team
>
> The Launchpad bug should target series 'Jammy'. A brief [Fix] section
> could also be helpful, to give some context to the changes maybe. Either
> way, LGTM.
>

I've updated the Launchpad bug target series and added a Fixes section
 that describes the changes a bit more.

> Acked-by: Manuel Diewald <manuel.diewald@canonical.com>
>
> --
>  Manuel
Roxana Nicolescu Feb. 21, 2024, 8:03 a.m. UTC | #4
On 20/02/2024 05:30, Portia Stephens wrote:
> [ Impact ]
>
> * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> support for the KD240 platform
> (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> * Testing had previously been done a development branch with non-upstreamable
> patches. 3 issues were introduced to the KD240 image that was not present on
> the development branch.
> * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> trees were updated including certified platforms.
>
> [ Test Plan ]
>
> * QA will run certification testing on the KD240 platform
> * Normal certification testing will be run on all other certified platforms
>
> [ Where problems could occur ]
>
> * This impacts the device tree for certified Xilinx platforms which could break
> any of the device touched by the change.
>
> [ Other Info ]
>
> * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
>
> * Private launchpad bugs that contain the regressions failure:
> https://bugs.launchpad.net/limerick/+bug/2051228
> https://bugs.launchpad.net/limerick/+bug/2051224
> https://bugs.launchpad.net/limerick/+bug/2051201
>
> Portia Stephens (1):
>    UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
>
>   .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
>   arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
>   2 files changed, 25 insertions(+), 19 deletions(-)
>
Hi,

Any reason the subject does not include SRU? It messes up my filters.
The mailing list receives other type of emails, not only patches, and 
this is
what I use to filter patches. 
https://wiki.ubuntu.com/Kernel/Dev/StablePatchFormat

Roxana
Manuel Diewald Feb. 21, 2024, 8:29 a.m. UTC | #5
On Wed, Feb 21, 2024 at 09:03:25AM +0100, Roxana Nicolescu wrote:
> 
> On 20/02/2024 05:30, Portia Stephens wrote:
> > [ Impact ]
> > 
> > * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> > support for the KD240 platform
> > (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> > * Testing had previously been done a development branch with non-upstreamable
> > patches. 3 issues were introduced to the KD240 image that was not present on
> > the development branch.
> > * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> > trees were updated including certified platforms.
> > 
> > [ Test Plan ]
> > 
> > * QA will run certification testing on the KD240 platform
> > * Normal certification testing will be run on all other certified platforms
> > 
> > [ Where problems could occur ]
> > 
> > * This impacts the device tree for certified Xilinx platforms which could break
> > any of the device touched by the change.
> > 
> > [ Other Info ]
> > 
> > * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> > 
> > * Private launchpad bugs that contain the regressions failure:
> > https://bugs.launchpad.net/limerick/+bug/2051228
> > https://bugs.launchpad.net/limerick/+bug/2051224
> > https://bugs.launchpad.net/limerick/+bug/2051201
> > 
> > Portia Stephens (1):
> >    UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> > 
> >   .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
> >   arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
> >   2 files changed, 25 insertions(+), 19 deletions(-)
> > 
> Hi,
> 
> Any reason the subject does not include SRU? It messes up my filters.
> The mailing list receives other type of emails, not only patches, and this
> is
> what I use to filter patches.
> https://wiki.ubuntu.com/Kernel/Dev/StablePatchFormat
> 
> Roxana
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team

I think the SRU tag is not mandatory since the kernel is not a stable
kernel yet.
Andrei Gherzan Feb. 21, 2024, 10:48 a.m. UTC | #6
On 24/02/20 02:30PM, Portia Stephens wrote:
> [ Impact ]
> 
> * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> support for the KD240 platform
> (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> * Testing had previously been done a development branch with non-upstreamable
> patches. 3 issues were introduced to the KD240 image that was not present on
> the development branch.
> * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> trees were updated including certified platforms.
> 
> [ Test Plan ]
> 
> * QA will run certification testing on the KD240 platform
> * Normal certification testing will be run on all other certified platforms
> 
> [ Where problems could occur ]
> 
> * This impacts the device tree for certified Xilinx platforms which could break
> any of the device touched by the change.
> 
> [ Other Info ]
> 
> * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> 
> * Private launchpad bugs that contain the regressions failure:
> https://bugs.launchpad.net/limerick/+bug/2051228
> https://bugs.launchpad.net/limerick/+bug/2051224
> https://bugs.launchpad.net/limerick/+bug/2051201
> 
> Portia Stephens (1):
>   UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> 
>  .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
>  arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++------

Acked-by: Andrei Gherzan <andrei.gherzan@canonical.com>
Roxana Nicolescu Feb. 21, 2024, 11:30 a.m. UTC | #7
On 21/02/2024 09:29, Manuel Diewald wrote:
> On Wed, Feb 21, 2024 at 09:03:25AM +0100, Roxana Nicolescu wrote:
>> On 20/02/2024 05:30, Portia Stephens wrote:
>>> [ Impact ]
>>>
>>> * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
>>> support for the KD240 platform
>>> (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
>>> * Testing had previously been done a development branch with non-upstreamable
>>> patches. 3 issues were introduced to the KD240 image that was not present on
>>> the development branch.
>>> * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
>>> trees were updated including certified platforms.
>>>
>>> [ Test Plan ]
>>>
>>> * QA will run certification testing on the KD240 platform
>>> * Normal certification testing will be run on all other certified platforms
>>>
>>> [ Where problems could occur ]
>>>
>>> * This impacts the device tree for certified Xilinx platforms which could break
>>> any of the device touched by the change.
>>>
>>> [ Other Info ]
>>>
>>> * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
>>>
>>> * Private launchpad bugs that contain the regressions failure:
>>> https://bugs.launchpad.net/limerick/+bug/2051228
>>> https://bugs.launchpad.net/limerick/+bug/2051224
>>> https://bugs.launchpad.net/limerick/+bug/2051201
>>>
>>> Portia Stephens (1):
>>>     UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
>>>
>>>    .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
>>>    arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
>>>    2 files changed, 25 insertions(+), 19 deletions(-)
>>>
>> Hi,
>>
>> Any reason the subject does not include SRU? It messes up my filters.
>> The mailing list receives other type of emails, not only patches, and this
>> is
>> what I use to filter patches.
>> https://wiki.ubuntu.com/Kernel/Dev/StablePatchFormat
>>
>> Roxana
>>
>> -- 
>> kernel-team mailing list
>> kernel-team@lists.ubuntu.com
>> https://lists.ubuntu.com/mailman/listinfo/kernel-team
> I think the SRU tag is not mandatory since the kernel is not a stable
> kernel yet.
>
Good point indeed. But I think at least PATCH should be used to show
it is a patch.

Roxana
Manuel Diewald Feb. 21, 2024, 1:20 p.m. UTC | #8
On Wed, Feb 21, 2024 at 12:30:43PM +0100, Roxana Nicolescu wrote:
> 
> On 21/02/2024 09:29, Manuel Diewald wrote:
> > On Wed, Feb 21, 2024 at 09:03:25AM +0100, Roxana Nicolescu wrote:
> > > On 20/02/2024 05:30, Portia Stephens wrote:
> > > > [ Impact ]
> > > > 
> > > > * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> > > > support for the KD240 platform
> > > > (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> > > > * Testing had previously been done a development branch with non-upstreamable
> > > > patches. 3 issues were introduced to the KD240 image that was not present on
> > > > the development branch.
> > > > * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> > > > trees were updated including certified platforms.
> > > > 
> > > > [ Test Plan ]
> > > > 
> > > > * QA will run certification testing on the KD240 platform
> > > > * Normal certification testing will be run on all other certified platforms
> > > > 
> > > > [ Where problems could occur ]
> > > > 
> > > > * This impacts the device tree for certified Xilinx platforms which could break
> > > > any of the device touched by the change.
> > > > 
> > > > [ Other Info ]
> > > > 
> > > > * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> > > > 
> > > > * Private launchpad bugs that contain the regressions failure:
> > > > https://bugs.launchpad.net/limerick/+bug/2051228
> > > > https://bugs.launchpad.net/limerick/+bug/2051224
> > > > https://bugs.launchpad.net/limerick/+bug/2051201
> > > > 
> > > > Portia Stephens (1):
> > > >     UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> > > > 
> > > >    .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
> > > >    arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
> > > >    2 files changed, 25 insertions(+), 19 deletions(-)
> > > > 
> > > Hi,
> > > 
> > > Any reason the subject does not include SRU? It messes up my filters.
> > > The mailing list receives other type of emails, not only patches, and this
> > > is
> > > what I use to filter patches.
> > > https://wiki.ubuntu.com/Kernel/Dev/StablePatchFormat
> > > 
> > > Roxana
> > > 
> > > -- 
> > > kernel-team mailing list
> > > kernel-team@lists.ubuntu.com
> > > https://lists.ubuntu.com/mailman/listinfo/kernel-team
> > I think the SRU tag is not mandatory since the kernel is not a stable
> > kernel yet.
> > 
> Good point indeed. But I think at least PATCH should be used to show
> it is a patch.
> 
> Roxana
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team

I agree, if something like PATCH was mandatory, it would be possible to
filter for things that require reviews trivially - which is nice to
have.
Portia Stephens Feb. 21, 2024, 11:06 p.m. UTC | #9
On Wed, Feb 21, 2024 at 02:20:09PM +0100, Manuel Diewald wrote:
> On Wed, Feb 21, 2024 at 12:30:43PM +0100, Roxana Nicolescu wrote:
> > 
> > On 21/02/2024 09:29, Manuel Diewald wrote:
> > > On Wed, Feb 21, 2024 at 09:03:25AM +0100, Roxana Nicolescu wrote:
> > > > On 20/02/2024 05:30, Portia Stephens wrote:
> > > > > [ Impact ]
> > > > > 
> > > > > * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> > > > > support for the KD240 platform
> > > > > (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> > > > > * Testing had previously been done a development branch with non-upstreamable
> > > > > patches. 3 issues were introduced to the KD240 image that was not present on
> > > > > the development branch.
> > > > > * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> > > > > trees were updated including certified platforms.
> > > > > 
> > > > > [ Test Plan ]
> > > > > 
> > > > > * QA will run certification testing on the KD240 platform
> > > > > * Normal certification testing will be run on all other certified platforms
> > > > > 
> > > > > [ Where problems could occur ]
> > > > > 
> > > > > * This impacts the device tree for certified Xilinx platforms which could break
> > > > > any of the device touched by the change.
> > > > > 
> > > > > [ Other Info ]
> > > > > 
> > > > > * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
> > > > > 
> > > > > * Private launchpad bugs that contain the regressions failure:
> > > > > https://bugs.launchpad.net/limerick/+bug/2051228
> > > > > https://bugs.launchpad.net/limerick/+bug/2051224
> > > > > https://bugs.launchpad.net/limerick/+bug/2051201
> > > > > 
> > > > > Portia Stephens (1):
> > > > >     UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
> > > > > 
> > > > >    .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
> > > > >    arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
> > > > >    2 files changed, 25 insertions(+), 19 deletions(-)
> > > > > 
> > > > Hi,
> > > > 
> > > > Any reason the subject does not include SRU? It messes up my filters.
> > > > The mailing list receives other type of emails, not only patches, and this
> > > > is
> > > > what I use to filter patches.
> > > > https://wiki.ubuntu.com/Kernel/Dev/StablePatchFormat
> > > > 
> > > > Roxana
> > > > 
> > > > -- 
> > > > kernel-team mailing list
> > > > kernel-team@lists.ubuntu.com
> > > > https://lists.ubuntu.com/mailman/listinfo/kernel-team
> > > I think the SRU tag is not mandatory since the kernel is not a stable
> > > kernel yet.

I'll add the SRU flag in the future, I am essentially treating it as a
SRU'd kernel anyways. Requiring 2 reviews, submitting the SRU
justification, etc. For context, it is still a development kernel
because we have one remaining enablement project we are working on now
and we (PM's, customer, me) were concerned about the impact of limtiing
ourselves to the SRU cycle on this enablment. Once that is wrapped up I will
move it out of development.

> > > 
> > Good point indeed. But I think at least PATCH should be used to show
> > it is a patch.
> > 
> > Roxana
> > 
> > -- 
> > kernel-team mailing list
> > kernel-team@lists.ubuntu.com
> > https://lists.ubuntu.com/mailman/listinfo/kernel-team
> 
> I agree, if something like PATCH was mandatory, it would be possible to
> filter for things that require reviews trivially - which is nice to
> have.
> 
> -- 
>  Manuel
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team
Portia Stephens Feb. 26, 2024, 3:19 a.m. UTC | #10
On Tue, Feb 20, 2024 at 2:31 PM Portia Stephens
<portia.stephens@canonical.com> wrote:
>
> [ Impact ]
>
> * Kria device tree's were backported from Xilinx's 6.1 tree in order to add
> support for the KD240 platform
> (https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2046280) .
> * Testing had previously been done a development branch with non-upstreamable
> patches. 3 issues were introduced to the KD240 image that was not present on
> the development branch.
> * Since all Xilinx device tree's are so interdependent all Kria and ZCU device
> trees were updated including certified platforms.
>
> [ Test Plan ]
>
> * QA will run certification testing on the KD240 platform
> * Normal certification testing will be run on all other certified platforms
>
> [ Where problems could occur ]
>
> * This impacts the device tree for certified Xilinx platforms which could break
> any of the device touched by the change.
>
> [ Other Info ]
>
> * Buglink: https://bugs.launchpad.net/ubuntu/+source/linux-xilinx-zynqmp/+bug/2054366
>
> * Private launchpad bugs that contain the regressions failure:
> https://bugs.launchpad.net/limerick/+bug/2051228
> https://bugs.launchpad.net/limerick/+bug/2051224
> https://bugs.launchpad.net/limerick/+bug/2051201
>
> Portia Stephens (1):
>   UBUNTU: SAUCE: zynqmp.dtsi fix incorrectly backported changes
>
>  .../arm64/boot/dts/xilinx/zynqmp-clk-ccf.dtsi |  2 +-
>  arch/arm64/boot/dts/xilinx/zynqmp.dtsi        | 42 +++++++++++--------
>  2 files changed, 25 insertions(+), 19 deletions(-)
>
> --
> 2.34.1
>