On 05/05/2017 05:34 PM, Max Reitz wrote: > On 29.04.2017 21:14, Eric Blake wrote: >> Use blkdebug's new geometry constraints to emulate setups that >> have needed past regression fixes: write zeroes asserting >> when running through a loopback block device with max-transfer >> smaller than cluster size, and discard rounding away portions >> of requests not aligned to preferred boundaries. Also, add >> coverage that the block layer is honoring max transfer limits. >>
>> +function verify_io()
>> +{
>> + if ($QEMU_IMG info -f "$IMGFMT" "$TEST_IMG" |
>> +
>> +verify_io | $QEMU_IO "$TEST_IMG" | _filter_qemu_io
>
> This conflicts with Fam's image locking series that has been introduced
> in the meantime (and unfortunately I'm the one who has to base his block
> queue on Kevin's...). I suppose it's because the qemu_io process is
> launched before the qemu_img info process.
Indeed. My test was heavily modeled after 046; there, our solution
(currently commit 64ace79 on Kevin's tree) was to add -U, not -r.
>
> Simply adding an -r to the qemu_io command fixes this, however. I'll do
> so in my branch, assuming you're OK with that. :-)
I'd lean towards -U, but yes, I'm fine if you make whatever one-line
tweak is needed to obey the rules.
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3266
Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature
