[isar-cip-core,4/5] .reproducible-check-ci.yml: Add BBB target

Commit Message

Venkata Pyla Sept. 11, 2023, 1:31 p.m. UTC

From: venkata pyla <[e-mail]>

Signed-off-by: venkata pyla <[e-mail]>
---
 .reproducible-check-ci.yml | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

Comments

Jan Kiszka Sept. 12, 2023, 3:08 a.m. UTC | #1

On 11.09.23 19:01, [e-mail] wrote:
> From: venkata pyla <[e-mail]>
> 

Which additional aspects of reproducibility does this target test?

Jan

> Signed-off-by: venkata pyla <[e-mail]>
> ---
>  .reproducible-check-ci.yml | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
> 
> diff --git a/.reproducible-check-ci.yml b/.reproducible-check-ci.yml
> index dec28e2..6bcc215 100644
> --- a/.reproducible-check-ci.yml
> +++ b/.reproducible-check-ci.yml
> @@ -87,6 +87,13 @@ build:qemu-arm-repro-build:
>    variables:
>      target: qemu-arm
>  
> +build:bbb-repro-build:
> +  extends:
> +    - .repro-build
> +  variables:
> +    target: bbb
> +    extension: "none"
> +
>  # repro build test
>  test:qemu-amd64-repro-test:
>    extends:
> @@ -111,3 +118,12 @@ test:qemu-arm-repro-test:
>      target: qemu-arm
>    dependencies:
>      - build:qemu-arm-repro-build
> +
> +test:bbb-repro-test:
> +  extends:
> +    - .repro-test
> +  variables:
> +    target: bbb
> +    extension: "none"
> +  dependencies:
> +    - build:bbb-repro-build

Venkata Pyla Sept. 12, 2023, 6:52 a.m. UTC | #2

>-----Original Message-----
>From: Jan Kiszka <[e-mail]>
>Sent: Tuesday, September 12, 2023 8:39 AM
>To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>tsip.com>; [e-mail]
>Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
><[e-mail]>
>Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
>
>On 11.09.23 19:01, [e-mail] wrote:
>> From: venkata pyla <[e-mail]>
>>
>
>Which additional aspects of reproducibility does this target test?

It will try to verify additionally beagle bone specific files like MLO, u-boot and dtb files.


>
>Jan
>
>> Signed-off-by: venkata pyla <[e-mail]>
>> ---
>>  .reproducible-check-ci.yml | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)
>>
>> diff --git a/.reproducible-check-ci.yml b/.reproducible-check-ci.yml
>> index dec28e2..6bcc215 100644
>> --- a/.reproducible-check-ci.yml
>> +++ b/.reproducible-check-ci.yml
>> @@ -87,6 +87,13 @@ build:qemu-arm-repro-build:
>>    variables:
>>      target: qemu-arm
>>
>> +build:bbb-repro-build:
>> +  extends:
>> +    - .repro-build
>> +  variables:
>> +    target: bbb
>> +    extension: "none"
>> +
>>  # repro build test
>>  test:qemu-amd64-repro-test:
>>    extends:
>> @@ -111,3 +118,12 @@ test:qemu-arm-repro-test:
>>      target: qemu-arm
>>    dependencies:
>>      - build:qemu-arm-repro-build
>> +
>> +test:bbb-repro-test:
>> +  extends:
>> +    - .repro-test
>> +  variables:
>> +    target: bbb
>> +    extension: "none"
>> +  dependencies:
>> +    - build:bbb-repro-build
>
>--
>Siemens AG, Technology
>Linux Expert Center

Jan Kiszka Sept. 12, 2023, 7:50 a.m. UTC | #3

On 12.09.23 12:22, [e-mail] wrote:
> 
> 
>> -----Original Message-----
>> From: Jan Kiszka <[e-mail]>
>> Sent: Tuesday, September 12, 2023 8:39 AM
>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>> tsip.com>; [e-mail]
>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>> <[e-mail]>
>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
>>
>> On 11.09.23 19:01, [e-mail] wrote:
>>> From: venkata pyla <[e-mail]>
>>>
>>
>> Which additional aspects of reproducibility does this target test?
> 
> It will try to verify additionally beagle bone specific files like MLO, u-boot and dtb files.
> 

OK, so this includes firmware generation, understood. Does it also
include something that is now checked twice by other targets? I.e. can
this supersede some other target?

Jan

Venkata Pyla Sept. 13, 2023, 4:07 a.m. UTC | #4

>-----Original Message-----
>From: Jan Kiszka <[e-mail]>
>Sent: Tuesday, September 12, 2023 1:20 PM
>To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>tsip.com>; [e-mail]
>Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
><[e-mail]>
>Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
>
>On 12.09.23 12:22, [e-mail] wrote:
>>
>>
>>> -----Original Message-----
>>> From: Jan Kiszka <[e-mail]>
>>> Sent: Tuesday, September 12, 2023 8:39 AM
>>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>>> tsip.com>; [e-mail]
>>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>>> <[e-mail]>
>>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB
>>> target
>>>
>>> On 11.09.23 19:01, [e-mail] wrote:
>>>> From: venkata pyla <[e-mail]>
>>>>
>>>
>>> Which additional aspects of reproducibility does this target test?
>>
>> It will try to verify additionally beagle bone specific files like MLO, u-boot and
>dtb files.
>>
>
>OK, so this includes firmware generation, understood. Does it also include
>something that is now checked twice by other targets? I.e. can this supersede
>some other target?

No, the kernel is different because it uses different configs, rootfs contents also different as it installs additional packages like u-boot-script and u-boot-tools.
So the bbb target image contents are not verified in previous targets.

>
>Jan
>
>--
>Siemens AG, Technology
>Linux Expert Center

Jan Kiszka Sept. 14, 2023, 5:52 a.m. UTC | #5

On 13.09.23 09:37, [e-mail] wrote:
> 
> 
>> -----Original Message-----
>> From: Jan Kiszka <[e-mail]>
>> Sent: Tuesday, September 12, 2023 1:20 PM
>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>> tsip.com>; [e-mail]
>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>> <[e-mail]>
>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
>>
>> On 12.09.23 12:22, [e-mail] wrote:
>>>
>>>
>>>> -----Original Message-----
>>>> From: Jan Kiszka <[e-mail]>
>>>> Sent: Tuesday, September 12, 2023 8:39 AM
>>>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>>>> tsip.com>; [e-mail]
>>>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>>>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>>>> <[e-mail]>
>>>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB
>>>> target
>>>>
>>>> On 11.09.23 19:01, [e-mail] wrote:
>>>>> From: venkata pyla <[e-mail]>
>>>>>
>>>>
>>>> Which additional aspects of reproducibility does this target test?
>>>
>>> It will try to verify additionally beagle bone specific files like MLO, u-boot and
>> dtb files.
>>>
>>
>> OK, so this includes firmware generation, understood. Does it also include
>> something that is now checked twice by other targets? I.e. can this supersede
>> some other target?
> 
> No, the kernel is different because it uses different configs, rootfs contents also different as it installs additional packages like u-boot-script and u-boot-tools.
> So the bbb target image contents are not verified in previous targets.

So far, we are not looking for testing each artifact we could produce
for reproducibility, just like we do not check each configuration
combination a user can select. Instead, we are testing representative
combinations. So my question would be if the BBB can be sufficiently
representative for qemu-arm specifically.

Jan

Venkata Pyla Sept. 14, 2023, 6:14 a.m. UTC | #6

> -----Original Message-----
> From: Jan Kiszka <[e-mail]>
> Sent: Thursday, September 14, 2023 11:23 AM
> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
> tsip.com>; [e-mail]
> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
> <[e-mail]>
> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
> 
> On 13.09.23 09:37, [e-mail] wrote:
> >
> >
> >> -----Original Message-----
> >> From: Jan Kiszka <[e-mail]>
> >> Sent: Tuesday, September 12, 2023 1:20 PM
> >> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
> >> tsip.com>; [e-mail]
> >> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
> >> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
> >> <[e-mail]>
> >> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB
> >> target
> >>
> >> On 12.09.23 12:22, [e-mail] wrote:
> >>>
> >>>
> >>>> -----Original Message-----
> >>>> From: Jan Kiszka <[e-mail]>
> >>>> Sent: Tuesday, September 12, 2023 8:39 AM
> >>>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
> >>>> tsip.com>; [e-mail]
> >>>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting)
> <dinesh.kumar@toshiba-
> >>>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
> >>>> <[e-mail]>
> >>>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add
> >>>> BBB target
> >>>>
> >>>> On 11.09.23 19:01, [e-mail] wrote:
> >>>>> From: venkata pyla <[e-mail]>
> >>>>>
> >>>>
> >>>> Which additional aspects of reproducibility does this target test?
> >>>
> >>> It will try to verify additionally beagle bone specific files like
> >>> MLO, u-boot and
> >> dtb files.
> >>>
> >>
> >> OK, so this includes firmware generation, understood. Does it also
> >> include something that is now checked twice by other targets? I.e.
> >> can this supersede some other target?
> >
> > No, the kernel is different because it uses different configs, rootfs contents
> also different as it installs additional packages like u-boot-script and u-boot-
> tools.
> > So the bbb target image contents are not verified in previous targets.
> 
> So far, we are not looking for testing each artifact we could produce for
> reproducibility, just like we do not check each configuration combination a
> user can select. Instead, we are testing representative combinations. So my
> question would be if the BBB can be sufficiently representative for qemu-arm
> specifically.

Ha understood now, you mean if BBB can verify all the contents that qemu-arm can produce then we can drop the qemu-arm target and use only BBB in reproducible check CI?

If this is the question, then the answer is "No" BBB can not represent completely as qemu-arm because it has additional features like secure boot and encryptions are not yet supported in BBB.

> 
> Jan
> 
> --
> Siemens AG, Technology
> Linux Expert Center

Jan Kiszka Sept. 14, 2023, 6:18 a.m. UTC | #7

On 14.09.23 11:44, [e-mail] wrote:
> 
> 
>> -----Original Message-----
>> From: Jan Kiszka <[e-mail]>
>> Sent: Thursday, September 14, 2023 11:23 AM
>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>> tsip.com>; [e-mail]
>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>> <[e-mail]>
>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB target
>>
>> On 13.09.23 09:37, [e-mail] wrote:
>>>
>>>
>>>> -----Original Message-----
>>>> From: Jan Kiszka <[e-mail]>
>>>> Sent: Tuesday, September 12, 2023 1:20 PM
>>>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>>>> tsip.com>; [e-mail]
>>>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting) <dinesh.kumar@toshiba-
>>>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>>>> <[e-mail]>
>>>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add BBB
>>>> target
>>>>
>>>> On 12.09.23 12:22, [e-mail] wrote:
>>>>>
>>>>>
>>>>>> -----Original Message-----
>>>>>> From: Jan Kiszka <[e-mail]>
>>>>>> Sent: Tuesday, September 12, 2023 8:39 AM
>>>>>> To: pyla venkata(ＴＳＩＰ TMIEC ODG Porting) <Venkata.Pyla@toshiba-
>>>>>> tsip.com>; [e-mail]
>>>>>> Cc: dinesh kumar(ＴＳＩＰ TMIEC ODG Porting)
>> <dinesh.kumar@toshiba-
>>>>>> tsip.com>; hayashi kazuhiro(林 和宏 ＤＭＥ ○ＤＩＧ□ＭＰＳ○ＭＰ４)
>>>>>> <[e-mail]>
>>>>>> Subject: Re: [isar-cip-core 4/5] .reproducible-check-ci.yml: Add
>>>>>> BBB target
>>>>>>
>>>>>> On 11.09.23 19:01, [e-mail] wrote:
>>>>>>> From: venkata pyla <[e-mail]>
>>>>>>>
>>>>>>
>>>>>> Which additional aspects of reproducibility does this target test?
>>>>>
>>>>> It will try to verify additionally beagle bone specific files like
>>>>> MLO, u-boot and
>>>> dtb files.
>>>>>
>>>>
>>>> OK, so this includes firmware generation, understood. Does it also
>>>> include something that is now checked twice by other targets? I.e.
>>>> can this supersede some other target?
>>>
>>> No, the kernel is different because it uses different configs, rootfs contents
>> also different as it installs additional packages like u-boot-script and u-boot-
>> tools.
>>> So the bbb target image contents are not verified in previous targets.
>>
>> So far, we are not looking for testing each artifact we could produce for
>> reproducibility, just like we do not check each configuration combination a
>> user can select. Instead, we are testing representative combinations. So my
>> question would be if the BBB can be sufficiently representative for qemu-arm
>> specifically.
> 
> Ha understood now, you mean if BBB can verify all the contents that qemu-arm can produce then we can drop the qemu-arm target and use only BBB in reproducible check CI?
> 
> If this is the question, then the answer is "No" BBB can not represent completely as qemu-arm because it has additional features like secure boot and encryptions are not yet supported in BBB.
> 

Good point. Merging this as is.

Once we are fully reproducible, I suppose we can drop these side-checks
and just add them to ever CI run. But that is not yet in reach,
unfortunately.

Thanks,
Jan

Patch

diff --git a/.reproducible-check-ci.yml b/.reproducible-check-ci.yml
index dec28e2..6bcc215 100644
--- a/.reproducible-check-ci.yml
+++ b/.reproducible-check-ci.yml
@@ -87,6 +87,13 @@  build:qemu-arm-repro-build:
   variables:
     target: qemu-arm
 
+build:bbb-repro-build:
+  extends:
+    - .repro-build
+  variables:
+    target: bbb
+    extension: "none"
+
 # repro build test
 test:qemu-amd64-repro-test:
   extends:
@@ -111,3 +118,12 @@  test:qemu-arm-repro-test:
     target: qemu-arm
   dependencies:
     - build:qemu-arm-repro-build
+
+test:bbb-repro-test:
+  extends:
+    - .repro-test
+  variables:
+    target: bbb
+    extension: "none"
+  dependencies:
+    - build:bbb-repro-build