On 6/16/25 10:57 AM, Akihiko Odaki wrote:

On 2025/06/14 16:27, Paolo Abeni wrote:

quoted

On 6/8/25 8:16 AM, Akihiko Odaki wrote:

quoted

On 2025/06/06 20:45, Paolo Abeni wrote:

quoted

+
+		/* Zero the trailing space provided by user-space, if any */
+		if (i < count && clear_user(argp, (count - i) * sizeof(u64)))

I think checking i < count is a premature optimization; it doesn't
matter even if we spend a bit longer because of the lack of the check.

FTR, the check is not an optimization. if `i` is greater than `count`,
`clear_user` is going to try to clear almost all the memory space (the
2nd argument is an unsigned one) and will likely return with error.

I think it's functionally needed.

The for loop tells i cannot be greater than count, so i < count will be 
false only when i == count. 

Right you are, /me goes grabbing more coffee...

Thanks,

Paolo