[REGRESSION] nvme: code command_id with a genctr for use-after-free validation crashes apple T2 SSD

[Orlando Chamberlain]

On 26/9/21 12:08, Keith Busch wrote:
> On Sat, Sep 25, 2021 at 11:40:19PM +0000, Orlando Chamberlain wrote:
>> I've managed to get it to boot by commenting out the counter increment, which might work
>> as a quirk:
> 
> Great, that's essentially what I proposed in a different thread (sorry, I see
> now you were not copied there):
> 
>   http://lists.infradead.org/pipermail/linux-nvme/2021-September/027665.html
> 
> The final patch may look a bit different, but it should be okay to test with
> the one in the link.
> 
I've tried that patch, it doesn't crash, but /dev/nvme0, /dev/nvme0n1 etc don't show up so
it doesn't progress past the initramfs. There weren't any errors other than preexisting
acpi table ones.

(I sent this a few minutes ago but I forgot to reply all so here it is again)