Re: [PATCH] 9pnet: allow making incomplete read requests

[Dominique Martinet]

Sergey Alirzaev wrote on Wed, Feb 05, 2020:
> A user doesn't necessarily want to wait for all the requested data to
> be available, since the waiting time for each request is unbounded.
> 
> The new method permits sending one read request at a time and getting
> the response ASAP, allowing to use 9pnet with synthetic file systems
> representing arbitrary data streams.

Much better, thanks!

> Signed-off-by: Sergey Alirzaev <l29ah@xxxxxxx>
> [...]
> +		if (n != count) {
> +			*err = -EFAULT;
> +			p9_tag_remove(clnt, req);
> +			return n;
>  		}
> -		p9_tag_remove(clnt, req);
> +	} else {
> +		iov_iter_advance(to, count);
> +		count;

Any reason for this stray 'count;' statement?

If you're ok with this I'll just take patch without that line, don't
bother resubmitting.
Will take a fair amount of time to make it to linux-next though, test
setup needs some love and I want to run tests even if this should be
straightforward...

Thanks,
-- 
Dominique