On Thu, Feb 17, 2022 at 04:44:56PM -0600, Eric Blake wrote:
On Thu, Feb 17, 2022 at 10:00:52PM +0000, Richard W.M. Jones wrote:
> On Thu, Feb 17, 2022 at 03:58:07PM -0600, Eric Blake wrote:
> > The NBD protocol spec recommends EINVAL (rather than EIO) for requests
> > that violate block size constraints. Do we want to hardcode EINVAL,
> > or else tweak the error policy to allow the user to specify which
> > errno to provide?
>
> I err on not overthinking this. I'll change it to EINVAL, and if ever
> we need to choose another errno then we can add a knob for it.
Just noticed this XXX comment that we should now be able to test with
blocksize-policy of error:
https://gitlab.com/nbdkit/libnbd/-/blob/master/tests/errors.c#L464
Wow, that test is huge! Looking at the history it seems to have
accreted new tests over several years.
I've partially split it into smaller, more mangeable tests. There's
more to be done before I can add that check. I may have another go at
this later.
https://gitlab.com/nbdkit/libnbd/-/commit/3f57d7a37a7c5a1d8495f3bc1880e22...
Rich.
--
Richard Jones, Virtualization Group, Red Hat
http://people.redhat.com/~rjones
Read my programming and virtualization blog:
http://rwmj.wordpress.com
virt-df lists disk usage of guests without needing to install any
software inside the virtual machine. Supports Linux and Windows.
http://people.redhat.com/~rjones/virt-df/