DYNAMIC_FTRACE and I-pipe patch

DYNAMIC_FTRACE and I-pipe patch

[Pierre FICHEUX]

Hi,

One of my customers uses a 5.4.70 kernel with an iMX8M board and a
custom I-pipe patch (with Xenomai 3.1). It looks like all the I-pipe
patches disable the DYNAMIC_FTRACE option :

diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index f3f2fc8ad81a..98d7c64b7b8f 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -532,6 +532,7 @@ config DYNAMIC_FTRACE
  bool "enable/disable function tracing dynamically"
  depends on FUNCTION_TRACER
  depends on HAVE_DYNAMIC_FTRACE
+ depends on !IPIPE
  default y
  help
   This option will modify all the calls to function tracing

That's not the case for a recent Dovetail based kernel.

Could you give me some information about such a limitation ?

Thanks in advance.

regards

-- 

Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
                             http://www.smile.fr
                             https://smile.eu/fr/offres/embarque-iot
I would love to change the world, but they won't give me the source code

RE: DYNAMIC_FTRACE and I-pipe patch

[Chen, Hongzhan]

>-----Original Message-----
>From: Pierre FICHEUX <pierre.ficheux@smile.fr>
>Sent: Tuesday, March 5, 2024 7:44 AM
>To: xenomai <xenomai@xenomai.org>
>Subject: DYNAMIC_FTRACE and I-pipe patch
>
>Hi,
>
>One of my customers uses a 5.4.70 kernel with an iMX8M board and a
>custom I-pipe patch (with Xenomai 3.1). It looks like all the I-pipe
>patches disable the DYNAMIC_FTRACE option :
>
>diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
>index f3f2fc8ad81a..98d7c64b7b8f 100644
>--- a/kernel/trace/Kconfig
>+++ b/kernel/trace/Kconfig
>@@ -532,6 +532,7 @@ config DYNAMIC_FTRACE
>  bool "enable/disable function tracing dynamically"
>  depends on FUNCTION_TRACER
>  depends on HAVE_DYNAMIC_FTRACE
>+ depends on !IPIPE
>  default y
>  help
>   This option will modify all the calls to function tracing
>
>That's not the case for a recent Dovetail based kernel.
>
>Could you give me some information about such a limitation ?

Maybe you can refer to https://source.denx.de/Xenomai/xenomai/-/wikis/Dovetail.
It mentioned " The I-pipe provides its own function tracer based on the Ftrace
engine, but still duplicating some of its features while
restricting others." but  I do not know if it includes such limitation IPIPE was restricting but
dovetail improved.

Regards

Hongzhan Chen
>
>Thanks in advance.
>
>regards
>
>--
>
>Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
>                             http://www.smile.fr
>                             https://smile.eu/fr/offres/embarque-iot
>I would love to change the world, but they won't give me the source code

Re: DYNAMIC_FTRACE and I-pipe patch

[Pierre FICHEUX]

Hi,

You talk about the I-pipe tracer?

--
Pierre

> Le 5 mars 2024 à 02:13, Chen, Hongzhan <hongzhan.chen@intel.com> a écrit :
> 
> 
> 
>> -----Original Message-----
>> From: Pierre FICHEUX <pierre.ficheux@smile.fr>
>> Sent: Tuesday, March 5, 2024 7:44 AM
>> To: xenomai <xenomai@xenomai.org>
>> Subject: DYNAMIC_FTRACE and I-pipe patch
>> 
>> Hi,
>> 
>> One of my customers uses a 5.4.70 kernel with an iMX8M board and a
>> custom I-pipe patch (with Xenomai 3.1). It looks like all the I-pipe
>> patches disable the DYNAMIC_FTRACE option :
>> 
>> diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
>> index f3f2fc8ad81a..98d7c64b7b8f 100644
>> --- a/kernel/trace/Kconfig
>> +++ b/kernel/trace/Kconfig
>> @@ -532,6 +532,7 @@ config DYNAMIC_FTRACE
>> bool "enable/disable function tracing dynamically"
>> depends on FUNCTION_TRACER
>> depends on HAVE_DYNAMIC_FTRACE
>> + depends on !IPIPE
>> default y
>> help
>>  This option will modify all the calls to function tracing
>> 
>> That's not the case for a recent Dovetail based kernel.
>> 
>> Could you give me some information about such a limitation ?
> 
> Maybe you can refer to https://www.google.com/url?q=https://source.denx.de/Xenomai/xenomai/-/wikis/Dovetail&source=gmail-imap&ust=1710206004000000&usg=AOvVaw0IXHqzeoruHeu6lnhYcPQw.
> It mentioned " The I-pipe provides its own function tracer based on the Ftrace
> engine, but still duplicating some of its features while
> restricting others." but  I do not know if it includes such limitation IPIPE was restricting but
> dovetail improved.
> 
> Regards
> 
> Hongzhan Chen
>> 
>> Thanks in advance.
>> 
>> regards
>> 
>> --
>> 
>> Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
>>                            https://www.google.com/url?q=http://www.smile.fr&source=gmail-imap&ust=1710206004000000&usg=AOvVaw2X5QyoSLrMfOonkGBSzX5c
>>                            https://www.google.com/url?q=https://smile.eu/fr/offres/embarque-iot&source=gmail-imap&ust=1710206004000000&usg=AOvVaw3N1t1B1QlZt68DA6MYbiQJ
>> I would love to change the world, but they won't give me the source code
> 

RE: DYNAMIC_FTRACE and I-pipe patch

[Chen, Hongzhan]

>-----Original Message-----
>From: Pierre FICHEUX <pierre.ficheux@smile.fr>
>Sent: Tuesday, March 5, 2024 12:39 PM
>To: Chen, Hongzhan <hongzhan.chen@intel.com>
>Cc: xenomai <xenomai@xenomai.org>
>Subject: Re: DYNAMIC_FTRACE and I-pipe patch
>
>Hi,
>
>You talk about the I-pipe tracer?

I am talking about the issues that dovetail want to address as Philippe described in the link. The DYNAMIC_FTRACE limitation you mentioned with I-PIPE maybe one of the issues that dovetail already fixed. That is why it is not the case for dovetail. 

Regards

Hongzhan Chen

>
>--
>Pierre
>
>> Le 5 mars 2024 à 02:13, Chen, Hongzhan <hongzhan.chen@intel.com> a
>écrit :
>>
>> 
>>
>>> -----Original Message-----
>>> From: Pierre FICHEUX <pierre.ficheux@smile.fr>
>>> Sent: Tuesday, March 5, 2024 7:44 AM
>>> To: xenomai <xenomai@xenomai.org>
>>> Subject: DYNAMIC_FTRACE and I-pipe patch
>>>
>>> Hi,
>>>
>>> One of my customers uses a 5.4.70 kernel with an iMX8M board and a
>>> custom I-pipe patch (with Xenomai 3.1). It looks like all the I-pipe
>>> patches disable the DYNAMIC_FTRACE option :
>>>
>>> diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
>>> index f3f2fc8ad81a..98d7c64b7b8f 100644
>>> --- a/kernel/trace/Kconfig
>>> +++ b/kernel/trace/Kconfig
>>> @@ -532,6 +532,7 @@ config DYNAMIC_FTRACE
>>> bool "enable/disable function tracing dynamically"
>>> depends on FUNCTION_TRACER
>>> depends on HAVE_DYNAMIC_FTRACE
>>> + depends on !IPIPE
>>> default y
>>> help
>>>  This option will modify all the calls to function tracing
>>>
>>> That's not the case for a recent Dovetail based kernel.
>>>
>>> Could you give me some information about such a limitation ?
>>
>> Maybe you can refer to
>https://www.google.com/url?q=https://source.denx.de/Xenomai/xenomai/-
>/wikis/Dovetail&source=gmail-
>imap&ust=1710206004000000&usg=AOvVaw0IXHqzeoruHeu6lnhYcPQw.
>> It mentioned " The I-pipe provides its own function tracer based on the
>Ftrace
>> engine, but still duplicating some of its features while
>> restricting others." but  I do not know if it includes such limitation IPIPE was
>restricting but
>> dovetail improved.
>>
>> Regards
>>
>> Hongzhan Chen
>>>
>>> Thanks in advance.
>>>
>>> regards
>>>
>>> --
>>>
>>> Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
>>>
>https://www.google.com/url?q=http://www.smile.fr&source=gmail-
>imap&ust=1710206004000000&usg=AOvVaw2X5QyoSLrMfOonkGBSzX5c
>>>
>https://www.google.com/url?q=https://smile.eu/fr/offres/embarque-
>iot&source=gmail-
>imap&ust=1710206004000000&usg=AOvVaw3N1t1B1QlZt68DA6MYbiQJ
>>> I would love to change the world, but they won't give me the source code
>>

Re: DYNAMIC_FTRACE and I-pipe patch

[Pierre FICHEUX]

So a piece of advice could be to switch to a Dovetail compatible kernel ?

The soc is an imx8m mini (designed by Compulab) using an adapted
version of linux-imx. Is there any porting guide?

https://mediawiki.compulab.com/w/index.php?title=UCM-iMX8M-Mini_NXP_iMX8M-Mini_Yocto_Linux

Le mar. 5 mars 2024 à 06:03, Chen, Hongzhan <hongzhan.chen@intel.com> a écrit :
>
>
>
> >-----Original Message-----
> >From: Pierre FICHEUX <pierre.ficheux@smile.fr>
> >Sent: Tuesday, March 5, 2024 12:39 PM
> >To: Chen, Hongzhan <hongzhan.chen@intel.com>
> >Cc: xenomai <xenomai@xenomai.org>
> >Subject: Re: DYNAMIC_FTRACE and I-pipe patch
> >
> >Hi,
> >
> >You talk about the I-pipe tracer?
>
> I am talking about the issues that dovetail want to address as Philippe described in the link. The DYNAMIC_FTRACE limitation you mentioned with I-PIPE maybe one of the issues that dovetail already fixed. That is why it is not the case for dovetail.
>
> Regards
>
> Hongzhan Chen
>
> >
> >--
> >Pierre
> >
> >> Le 5 mars 2024 à 02:13, Chen, Hongzhan <hongzhan.chen@intel.com> a
> >écrit :
> >>
> >>
> >>
> >>> -----Original Message-----
> >>> From: Pierre FICHEUX <pierre.ficheux@smile.fr>
> >>> Sent: Tuesday, March 5, 2024 7:44 AM
> >>> To: xenomai <xenomai@xenomai.org>
> >>> Subject: DYNAMIC_FTRACE and I-pipe patch
> >>>
> >>> Hi,
> >>>
> >>> One of my customers uses a 5.4.70 kernel with an iMX8M board and a
> >>> custom I-pipe patch (with Xenomai 3.1). It looks like all the I-pipe
> >>> patches disable the DYNAMIC_FTRACE option :
> >>>
> >>> diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
> >>> index f3f2fc8ad81a..98d7c64b7b8f 100644
> >>> --- a/kernel/trace/Kconfig
> >>> +++ b/kernel/trace/Kconfig
> >>> @@ -532,6 +532,7 @@ config DYNAMIC_FTRACE
> >>> bool "enable/disable function tracing dynamically"
> >>> depends on FUNCTION_TRACER
> >>> depends on HAVE_DYNAMIC_FTRACE
> >>> + depends on !IPIPE
> >>> default y
> >>> help
> >>>  This option will modify all the calls to function tracing
> >>>
> >>> That's not the case for a recent Dovetail based kernel.
> >>>
> >>> Could you give me some information about such a limitation ?
> >>
> >> Maybe you can refer to
> >https://www.google.com/url?q=https://source.denx.de/Xenomai/xenomai/-
> >/wikis/Dovetail&source=gmail-
> >imap&ust=1710206004000000&usg=AOvVaw0IXHqzeoruHeu6lnhYcPQw.
> >> It mentioned " The I-pipe provides its own function tracer based on the
> >Ftrace
> >> engine, but still duplicating some of its features while
> >> restricting others." but  I do not know if it includes such limitation IPIPE was
> >restricting but
> >> dovetail improved.
> >>
> >> Regards
> >>
> >> Hongzhan Chen
> >>>
> >>> Thanks in advance.
> >>>
> >>> regards
> >>>
> >>> --
> >>>
> >>> Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
> >>>
> >https://www.google.com/url?q=http://www.smile.fr&source=gmail-
> >imap&ust=1710206004000000&usg=AOvVaw2X5QyoSLrMfOonkGBSzX5c
> >>>
> >https://www.google.com/url?q=https://smile.eu/fr/offres/embarque-
> >iot&source=gmail-
> >imap&ust=1710206004000000&usg=AOvVaw3N1t1B1QlZt68DA6MYbiQJ
> >>> I would love to change the world, but they won't give me the source code
> >>



--

Pierre FICHEUX -/- CTO Smile ECS, France -\- pierre.ficheux@smile.fr
                             http://www.smile.fr
                             https://smile.eu/fr/offres/embarque-iot

I would love to change the world, but they won't give me the source code