linux-kernel - Re: [Patch v7 2/4] memory: tegra: Add MC error logging on tegra186 onward

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:   Fri, 15 Apr 2022 00:14:47 +0300
From:   Dmitry Osipenko <dmitry.osipenko@...labora.com>
To:     Ashish Mhetre <amhetre@...dia.com>, digetx@...il.com,
        krzysztof.kozlowski@...aro.org, thierry.reding@...il.com,
        jonathanh@...dia.com, robh+dt@...nel.org,
        linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        linux-tegra@...r.kernel.org, krzysztof.kozlowski+dt@...aro.org
Cc:     vdumpa@...dia.com, Snikam@...dia.com
Subject: Re: [Patch v7 2/4] memory: tegra: Add MC error logging on tegra186
 onward

On 4/14/22 08:31, Ashish Mhetre wrote:
> 
> 
> On 4/14/2022 2:43 AM, Dmitry Osipenko wrote:
>> External email: Use caution opening links or attachments
>>
>>
>> On 4/13/22 12:40, Ashish Mhetre wrote:
>>> +irqreturn_t tegra30_mc_handle_irq(int irq, void *data)
>>>   {
>>>        struct tegra_mc *mc = data;
>>> +     unsigned int bit, channel;
>>>        unsigned long status;
>>> -     unsigned int bit;
>>>
>>> -     /* mask all interrupts to avoid flooding */
>>> -     status = mc_readl(mc, MC_INTSTATUS) & mc->soc->intmask;
>>> +     if (mc->soc->num_channels) {
>>> +             u32 global_status;
>>> +             int err;
>>> +
>>> +             global_status = mc_ch_readl(mc, MC_BROADCAST_CHANNEL,
>>> MC_GLOBAL_INTSTATUS);
>>
>> This will crash if mc->bcast_ch_regs = ERR_PTR(-EINVAL) for older dtbs.
> 
> Actually interrupts won't occur till we write MC_INTMASK register from
> broadcast channel with appropriate intmask value. I have added check in
> tegra_mc_probe() while registering irq which will write MC_INTMASK from
> broadcast only when mc->bcast_ch_regs is initialized i.e.
> !IS_ERR(mc->bcast_ch_regs).
> So interrupt handler won't be triggered at all if
> mc->bcast_ch_regs = ERR_PTR(-EINVAL).
> 

Should be cleaner to set mc->bcast_ch_regs to NULL anyways. The ERR_PTR
doesn't add much value and only makes code less readable.