Re: [PATCH] watchdog: lenovo_se30g2_se60_wdt: Watchdog for Lenovo SE30G2 and SE60

Hi Guenter

On Tue, Jun 23, 2026, at 10:19 PM, Guenter Roeck wrote:
> On 6/22/26 08:50, Mark Pearson wrote:
>> On Mon, Jun 22, 2026, at 10:56 AM, [email protected] wrote:
>>> Thank you for your contribution! Sashiko AI review found 4 potential
>>> issue(s) to consider:
>>> - [High] The global `se_30g2_60_pdev` pointer is not cleared after
>>> being freed on the `platform_device_add()` error path, leading to a
>>> Use-After-Free and Double-Free during module exit.
>> 
>> I belive this is a valid issue - I'll fix in v2.
>> 
>>> - [High] SuperIO configuration mode is prematurely exited before
>>> reading `bram_base`, causing the read to fail and the driver to map an
>>> invalid I/O port address.
>> 
>> This one is I believe invalid and behaviour is correct. If someone wiser than me disagree's please let me know.
>> 
>
> How would anyone but you know ? Normally Super-IO access is disabled
> after the exit sequence is sent (as "lpc_write(0x2, 0x2);" in lpc_chip_id()
> seems to do). Maybe this EC is different and does not disable access to
> its IO space after "lpc_write(0x2, 0x2);". Or maybe that write does not
> disable Super-IO access at all. Again, who but you would know if the EC
> behaves differently than all the other Super-IO chips and embedded
> controllers out there ?
>
> If you say this is correct, we'll have to believe you.
>
> Thanks,
> Guenter

Yeah. I went and checked with the HW team on this one, and I've changed it.
Fix for this will be in v2 as well.

Mark