linux-kernel - Re: [PATCH v2] phy: exynos5-usbdrd: Fix broken USB on Exynos5422 (TYPEC dependency)

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] [day] [month] [year] [list]

Message-ID: <35d9c210-1ffb-4878-863c-912f389b18e5@kernel.org>
Date: Mon, 17 Feb 2025 10:03:03 +0100
From: Krzysztof Kozlowski <krzk@...nel.org>
To: André Draszik <andre.draszik@...aro.org>,
 Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
 Vinod Koul <vkoul@...nel.org>, Kishon Vijay Abraham I <kishon@...nel.org>,
 Alim Akhtar <alim.akhtar@...sung.com>,
 Peter Griffin <peter.griffin@...aro.org>, linux-phy@...ts.infradead.org,
 linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
 linux-samsung-soc@...r.kernel.org
Cc: Marek Szyprowski <m.szyprowski@...sung.com>
Subject: Re: [PATCH v2] phy: exynos5-usbdrd: Fix broken USB on Exynos5422
 (TYPEC dependency)

On 17/02/2025 06:59, André Draszik wrote:
> Hi Krzysztof,
> 
> On Sun, 2025-02-16 at 12:41 +0100, Krzysztof Kozlowski wrote:
>> Older Exynos designs, like Exynos5422, do not have USB Type-C and the
>> USB DRD PHY does not really depend on Type-C for these devices at all.
>> Incorrectly added optional dependency on CONFIG_TYPEC caused this driver
>> to be missing for exynos_defconfig and as result Exynos5422-based boards
>> like Hardkernel Odroid HC1 failed to probe USB.
>>
>> However ARM64 boards should have CONFIG_TYPEC enabled and in proper
>> state (e.g. =y when PHY_EXYNOS5_USBDRD is =y) for proper USB support on
>> Google GS101 boards.
>>
>> Add itermediate CONFIG_PHY_EXYNOS5_USBDRD_TYPEC symbol to skip TYPEC on
>> older boards, while still expressing optional dependency.
>>
>> Reported-by: Krzysztof Kozlowski <krzk@...nel.org>
>> Closes: https://krzk.eu/#/builders/21/builds/6139
>> Reported-by: Marek Szyprowski <m.szyprowski@...sung.com>
>> Closes: https://lore.kernel.org/all/3c0b77e6-357d-453e-8b63-4757c3231bde@samsung.com/
>> Fixes: 09dc674295a3 ("phy: exynos5-usbdrd: subscribe to orientation notifier if required")
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
>>
>> ---
>>
>> Patch for issue in linux-next
>>
>> Changes in v2:
>> 1. Add PHY_EXYNOS5_USBDRD_TYPEC, so arm64 defconfig will have both
>>    symbols in-sync
>> ---
>>  drivers/phy/samsung/Kconfig              | 12 +++++++++++-
>>  drivers/phy/samsung/phy-exynos5-usbdrd.c |  1 +
>>  2 files changed, 12 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/phy/samsung/Kconfig b/drivers/phy/samsung/Kconfig
>> index 7fba571c0e2b..b20ac6b75993 100644
>> --- a/drivers/phy/samsung/Kconfig
>> +++ b/drivers/phy/samsung/Kconfig
>> @@ -77,12 +77,22 @@ config PHY_S5PV210_USB2
>>  	  particular SoC is compiled in the driver. In case of S5PV210 two phys
>>  	  are available - device and host.
>>  
>> +# None of ARM32 Samsung boards use Type-C, however newer ARM64 do and
>> +# PHY_EXYNOS5_USBDRD driver needs TYPEC to be in a matching state to avoid link
>> +# failures (see optional dependencies in kconfig-language.rst).  Intermediate
>> +# PHY_EXYNOS5_USBDRD_TYPEC symbol allows to skip TYPEC on ARM32 boards.
>> +config PHY_EXYNOS5_USBDRD_TYPEC
>> +	tristate
>> +	depends on ARCH_EXYNOS
>> +	depends on ARM || TYPEC || !TYPEC
>> +	default y
>> +
>>  config PHY_EXYNOS5_USBDRD
>>  	tristate "Exynos5 SoC series USB DRD PHY driver"
>>  	depends on (ARCH_EXYNOS && OF) || COMPILE_TEST
>>  	depends on HAS_IOMEM
>> -	depends on TYPEC || (TYPEC=n && COMPILE_TEST)
> 
> Is it not possible to just drop the '&& COMPILE_TEST' part to make
> it
>   depends on TYPEC || TYPEC=n
> or
>   depends on TYPEC || !TYPEC
> (without any of the other changes)?
> 
> If not, why not, are there corner cases that doesn't catch? Some
> minimal testing seems to suggest that that works, too.

Yeah, I think I overcomplicated it. I wanted to make it really
independent from TYPEC on ARM, but that's not trivial and my patch fails
here.

Best regards,
Krzysztof