On Sun Oct 11, 2026 at 11:51 PM JST, haider wrote:
> GSP-RM checksums a command queue element by XORing it in 64-bit words,
> so for an element whose size is not a multiple of 8 it also reads up to
> 7 bytes past its end, and expects them to be zero: its _checkSum32()
> "assumes that the data is padded out with zeros to the next 8-byte
> alignment". NVIDIA's kernel driver zeroes these bytes before computing
> the checksum, and nouveau builds each element in a zeroed buffer.
>
> allocate_command() hands out the queue slots as they are, so these bytes
> still hold whatever an earlier, longer element left there. Once the
> queue has wrapped, a command of such a size can then fail GSP-RM's
> checksum: GSP-RM never answers it, and soon stops answering at all.
>
> Nova does not send enough commands for the queue to wrap yet, as it is
> zeroed at probe, but this will break as soon as it does.
>
> Clear each element and its padding before writing it.
The checksum code is imminently going away with the switch to the r000
firmware version, so I don't think there is much value in fixing it now.
>
> Fixes: 75f6b1de8133 ("gpu: nova-core: gsp: Add GSP command queue bindings and
> handling")
> Assisted-by: LLM
> Signed-off-by: haider <[email protected]>
For future contributions, please use as real name as per the
contribution guidelines [1].
Also as a general comment, both the commit log and code comment are
unsufferably long and tedious. When working with a LLM, please
edit/rewrite its output to something concise and to the point.
[1]
https://www.kernel.org/doc/html/v4.17/process/submitting-patches.html#developer-s-certificate-of-origin-1-1
