linux-kernel - Re: [Patch V1 1/4] memory: tegra: Add support for mc interrupts

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]

Date:   Wed, 19 Jan 2022 14:17:00 +0530
From:   Ashish Mhetre <amhetre@...dia.com>
To:     Dmitry Osipenko <digetx@...il.com>, <thierry.reding@...il.com>,
        <jonathanh@...dia.com>, <linux-tegra@...r.kernel.org>,
        <krzysztof.kozlowski@...onical.com>, <linux-kernel@...r.kernel.org>
CC:     <Snikam@...dia.com>, <vdumpa@...dia.com>
Subject: Re: [Patch V1 1/4] memory: tegra: Add support for mc interrupts



On 1/12/2022 1:43 PM, Dmitry Osipenko wrote:
> External email: Use caution opening links or attachments
> 
> 
> 11.01.2022 21:45, Ashish Mhetre пишет:
>>
>> @@ -765,16 +768,21 @@ static int tegra_mc_probe(struct platform_device *pdev)
>>                        return err;
>>        }
>>
>> -     if (mc->soc->ops && mc->soc->ops->handle_irq) {
>> +     if (mc->soc->interrupt_ops && mc->soc->interrupt_ops->handle_irq) {
>>                mc->irq = platform_get_irq(pdev, 0);
>>                if (mc->irq < 0)
>>                        return mc->irq;
>>
>>                WARN(!mc->soc->client_id_mask, "missing client ID mask for this SoC\n");
>>
>> +             /* clear any mc-errs that occurred before. */
> 
> s/mc-errs/Memory Controller errors/
> 
Sure, I'll update these in next version.
>> +             if (mc->soc->interrupt_ops->clear_interrupt)
>> +                     mc->soc->interrupt_ops->clear_interrupt(mc);
> 
> There is no explanation of this change neither in the code, nor in the
> commit message. Please always provide detailed descriptions for a
> non-trivial changes.
> 
> Interrupts aren't cleared intentionally by the driver, otherwise you'll
> never know about early-boot memory faults which happened before the
> probe. Hence this change is incorrect.
That's true, we should be logging early-boot memory faults as well.
Ideally there shouldn't be any early-boot faults as all clients will
be up after MC, right? But I agree that we should be checking and
logging if any interrupt is present.