linux-kernel - Re: [PATCH 2/2] memory: brcmstb_memc: Simplify compatible matching

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <86b6c9a8-5ecf-4aa5-a6cf-afee64d28efa@kernel.org>
Date: Thu, 5 Jun 2025 22:30:42 +0200
From: Krzysztof Kozlowski <krzk@...nel.org>
To: Florian Fainelli <florian.fainelli@...adcom.com>,
 Rob Herring <robh@...nel.org>
Cc: linux-kernel@...r.kernel.org, justin.chen@...adcom.com,
 Conor Dooley <conor+dt@...nel.org>,
 Broadcom internal kernel review list
 <bcm-kernel-feedback-list@...adcom.com>,
 "open list:OPEN FIRMWARE AND FLATTENED DEVICE TREE BINDINGS"
 <devicetree@...r.kernel.org>,
 "moderated list:BROADCOM BCM7XXX ARM ARCHITECTURE"
 <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH 2/2] memory: brcmstb_memc: Simplify compatible matching

On 05/06/2025 21:10, Florian Fainelli wrote:
> On 6/5/25 11:55, Rob Herring wrote:
>> On Fri, May 23, 2025 at 11:43:54AM -0700, Florian Fainelli wrote:
>>> Now that a "brcm,brcmstb-memc-ddr-rev-b.2.x" fallback compatible string
>>> has been defined, we can greatly simplify the matching within the driver
>>> to only look for that compatible string and nothing else.
>>>
>>> The fallback "brcm,brcmstb-memc-ddr" is also updated to assume the V21
>>> register layout since that is the most common nowadays.
>>>
>>> Signed-off-by: Florian Fainelli <florian.fainelli@...adcom.com>
>>> ---
>>>   drivers/memory/brcmstb_memc.c | 58 ++---------------------------------
>>>   1 file changed, 3 insertions(+), 55 deletions(-)
>>>
>>> diff --git a/drivers/memory/brcmstb_memc.c b/drivers/memory/brcmstb_memc.c
>>> index c87b37e2c1f0..ec4c198ddc49 100644
>>> --- a/drivers/memory/brcmstb_memc.c
>>> +++ b/drivers/memory/brcmstb_memc.c
>>> @@ -181,65 +181,13 @@ static const struct of_device_id brcmstb_memc_of_match[] = {
>>>   		.data = &brcmstb_memc_versions[BRCMSTB_MEMC_V20]
>>>   	},
>>>   	{
>>> -		.compatible = "brcm,brcmstb-memc-ddr-rev-b.2.1",
>>> +		.compatible = "brcm,brcmstb-memc-ddr-rev-b.2.x",
>>>   		.data = &brcmstb_memc_versions[BRCMSTB_MEMC_V21]
>>
>> This entry is pointless because the default will get V21.
>>
>> In fact, I don't think you need the new compatible string at all. It
>> doesn't work to add fallbacks after the fact.
> 
> I agree and would prefer to keep adding new compatible strings which is 

So you agree that adding such entries is pointless?

> what I initially did here:
> 
> https://lore.kernel.org/all/20241217194439.929040-2-florian.fainelli@broadcom.com/
> 
> but the feedback was that this should not be done, and hence this 
> attempt at defining a compatible string that would avoid needless churn.
> 
> So which way should I go now?

And the advice was to use v2.1 fallback, not replace v2.1 with something
else or keep adding pointless entries:
https://lore.kernel.org/all/2e33t7ft5ermsfr7c4ympxrn6l5sqdef3wml4hlbnhdupoouwj@gfjpbmowjadi/


Best regards,
Krzysztof