On 2/20/20 10:34 AM, Alberto Garcia wrote:
On Thu 20 Feb 2020 04:28:07 PM CET, Eric Blake wrote:
Images without subclusters are treated as if they had exactly one,
with subcluster_size = cluster_size.
The qcow2 spec changes earlier in the series made it sound like your
choices are exactly 1 or 32,
+#define QCOW_MAX_SUBCLUSTERS_PER_CLUSTER 32
+
...but this name sounds like other values (2, 4, 8, 16) might be
possible?
I guess I didn't want to call it QCOW_SUBCLUSTERS_PER_CLUSTER because
there's already BDRVQcow2State.subclusters_per_cluster. And that one can
have two possible values (1 and 32) so 32 would be the maximum.
I get your point, however, and I'm open to suggestions.
Maybe QCOW_EXTL2_SUBCLUSTERS_PER_CLUSTER
since it is a hard-coded property of the EXTL2 feature.
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3226
Virtualization: qemu.org | libvirt.org
