linux-kernel - Re: [PATCH] ARM: dts: omap3-gta04: Drop superfluous omap36xx compatible

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <6B245439-F9FE-4931-A0DE-81F825BB5FE5@goldelico.com>
Date:   Wed, 4 Oct 2023 14:50:30 +0200
From:   "H. Nikolaus Schaller" <hns@...delico.com>
To:     Andreas Kemnade <andreas@...nade.info>
Cc:     bcousson@...libre.com, Tony Lindgren <tony@...mide.com>,
        robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org,
        conor+dt@...nel.org, linux-omap@...r.kernel.org,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ARM: dts: omap3-gta04: Drop superfluous omap36xx
 compatible



> Am 04.10.2023 um 14:42 schrieb H. Nikolaus Schaller <hns@...delico.com>:
> 
> 
> 
>> Am 04.10.2023 um 13:54 schrieb Andreas Kemnade <andreas@...nade.info>:
>> 
>> On Wed, 4 Oct 2023 13:39:03 +0200
>> "H. Nikolaus Schaller" <hns@...delico.com> wrote:
>> 
>>>> Am 04.10.2023 um 13:03 schrieb Andreas Kemnade <andreas@...nade.info>:
>>>> 
>>>> On Wed, 4 Oct 2023 12:50:16 +0200
>>>> "H. Nikolaus Schaller" <hns@...delico.com> wrote:
>>>> 
>>>>> Hi Andreas,
>>>>> 
>>>>>> Am 04.10.2023 um 08:53 schrieb Andreas Kemnade <andreas@...nade.info>:
>>>>>> 
>>>>>> Drop omap36xx compatible as done in other omap3630 devices.
>>>>>> This has apparently fallen through the lattice.
>>>>>> 
>>>>>> Signed-off-by: Andreas Kemnade <andreas@...nade.info>
>>>>>> ---
>>>>>> arch/arm/boot/dts/ti/omap/omap3-gta04.dtsi | 2 +-
>>>>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>>>> 
>>>>>> diff --git a/arch/arm/boot/dts/ti/omap/omap3-gta04.dtsi b/arch/arm/boot/dts/ti/omap/omap3-gta04.dtsi
>>>>>> index b6b27e93857f..3661340009e7 100644
>>>>>> --- a/arch/arm/boot/dts/ti/omap/omap3-gta04.dtsi
>>>>>> +++ b/arch/arm/boot/dts/ti/omap/omap3-gta04.dtsi
>>>>>> @@ -11,7 +11,7 @@
>>>>>> 
>>>>>> / {
>>>>>> 	model = "OMAP3 GTA04";
>>>>>> -	compatible = "goldelico,gta04", "ti,omap3630", "ti,omap36xx", "ti,omap3";    
>>>>> 
>>>>> there seem to be some more references to ti,omap36xx:
>>>>> 
>>>>> arch/arm/boot/dts/ti/omap/omap3-lilly-a83x.dtsi:	compatible = "incostartec,omap3-lilly-a83x", "ti,omap3630", "ti,omap36xx", "ti,omap3";  
>>>> 
>>>> apperently all the dtsi are fallen through the lattice when handling the dts.
>>>> 
>>>> 
>>>>> arch/arm/mach-omap2/board-generic.c:	"ti,omap36xx",
>>>>> drivers/clk/ti/dpll.c:	     of_machine_is_compatible("ti,omap36xx")) &&
>>>>> drivers/cpufreq/ti-cpufreq.c:	{ .compatible = "ti,omap36xx", .data = &omap36xx_soc_data, },
>>>>> 
>>>>> So are you sure that we can remove it without replacement or code fixes in dpll and cpufreq (board-generic is probably no issue)?
>>>>> 
>>>> see discussion of:
>>>> 
>>>> commit e341f338180c84cd98af3016cf5bcfde45a041fb
>>>> Author: Andrew Davis <afd@...com>
>>>> Date:   Thu Feb 16 09:33:38 2023 -0600
>>>> 
>>>>  ARM: dts: omap: Drop ti,omap36xx compatible  
>>> 
>>> Ah, I wasn't aware of this.
>>> 
>>>> 
>>>> all the places also basically check for omap36xx || omap3630.  
>>> 
>>> 
>>> Yes, I have checked but drivers/cpufreq/ti-cpufreq.c seems to be an
>>> exception (unless I am missing some other patch).
>>> 
>> No:
>>       { .compatible = "ti,omap36xx", .data = &omap36xx_soc_data, },
>>       { .compatible = "ti,omap3630", .data = &omap36xx_soc_data, },
> 
> Well, in v6.6-rc4 I see:
> 
> static const struct of_device_id ti_cpufreq_of_match[] = {
> 	{ .compatible = "ti,am33xx", .data = &am3x_soc_data, },
> 	{ .compatible = "ti,am3517", .data = &am3517_soc_data, },
> 	{ .compatible = "ti,am43", .data = &am4x_soc_data, },
> 	{ .compatible = "ti,dra7", .data = &dra7_soc_data },
> 	{ .compatible = "ti,omap34xx", .data = &omap34xx_soc_data, },
> 	{ .compatible = "ti,omap36xx", .data = &omap36xx_soc_data, },
> 	{ .compatible = "ti,am625", .data = &am625_soc_data, },
> 	{ .compatible = "ti,am62a7", .data = &am625_soc_data, },
> 	/* legacy */
> 	{ .compatible = "ti,omap3430", .data = &omap34xx_soc_data, },
> 	{ .compatible = "ti,omap3630", .data = &omap36xx_soc_data, },
> 	{},
> };
> 
> Either the "/* legacy */" or the "ti,omap36xx" seems as if it should be removed.
> But it seems to break the systematic approach of this table.
> 
>> The bindings also only specify omap3630.
> 
> What I think is that the background was (before bindings documentation
> was invented) that there are drivers covering all variants of omap36xx
> (incl. am37xx and dm37xx) and some parts specific to a single version.

Ah, there is indeed an issue with removing omap36xx and replacing by
omap3630.

What about the PVR/SGX driver. This needs a compatible that can distinguish
between the DM3725 and DM3730. The first is w/o SGX and the second one
with. Having all summarized as omap3630 does not allow to load the
PVR/SGX driver based on the board specific compatible entry.

AFAIR this was the original idea behind 

	compatible = "goldelico,gta04", "ti,omap3630", "ti,omap36xx", "ti,omap3"; 

Only with this we can make the SGX driver depend on "ti,omap3630" and all
other general omap36xx stuff on "ti,omap36xx".

So it seems as if there was a reason to have both, and the omap36xx is not
completely superflous to be dropped.

BR,
Nikolaus