linux-kernel - Re: fsl,p2020-esdhc sdhci quirks

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: <ec06a3c3-93dd-3d3f-e436-5d7cc3714531@intel.com>
Date:   Tue, 3 Dec 2019 13:15:59 +0200
From:   Adrian Hunter <adrian.hunter@...el.com>
To:     Rasmus Villemoes <linux@...musvillemoes.dk>,
        Yinbo Zhu <yinbo.zhu@....com>
Cc:     Ulf Hansson <ulf.hansson@...aro.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: fsl,p2020-esdhc sdhci quirks

On 3/12/19 12:49 pm, Rasmus Villemoes wrote:
> On 03/12/2019 11.15, Rasmus Villemoes wrote:
>> Hi
>>
>> Commits
>>
>> 05cb6b2a66fa - mmc: sdhci-of-esdhc: add erratum eSDHC-A001 and A-008358
>> support
>> a46e42712596 - mmc: sdhci-of-esdhc: add erratum eSDHC5 support
>>
>> seem a bit odd, in that they set bits from the SDHCI_* namespace in the
>> ->quirks2 member:
>>
>>                 host->quirks2 |= SDHCI_QUIRK_RESET_AFTER_REQUEST;
>>                 host->quirks2 |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;
> 
> FWIW, with something like the patch below, sparse (make C=1) would complain
> 
> drivers/mmc/host/sdhci-of-esdhc.c:1306:31: warning: invalid assignment: |=
> drivers/mmc/host/sdhci-of-esdhc.c:1306:31:    left side has type
> restricted sdhci_quirk2_t
> drivers/mmc/host/sdhci-of-esdhc.c:1306:31:    right side has type
> restricted sdhci_quirk_t
> drivers/mmc/host/sdhci-of-esdhc.c:1307:31: warning: invalid assignment: |=
> drivers/mmc/host/sdhci-of-esdhc.c:1307:31:    left side has type
> restricted sdhci_quirk2_t
> drivers/mmc/host/sdhci-of-esdhc.c:1307:31:    right side has type
> restricted sdhci_quirk_t
> 
> But maybe that's too much churn, dunno.

Presumably testing would have caught this, so I think it would be better if
people tested their patches.