graphic: 0.903
virtual: 0.893
device: 0.857
semantic: 0.732
vnc: 0.710
network: 0.677
PID: 0.631
x86: 0.622
risc-v: 0.608
mistranslation: 0.580
files: 0.568
performance: 0.558
user-level: 0.552
socket: 0.517
arm: 0.515
register: 0.488
architecture: 0.485
i386: 0.483
boot: 0.447
VMM: 0.408
debug: 0.386
kernel: 0.380
ppc: 0.347
assembly: 0.282
permissions: 0.279
hypervisor: 0.235
peripherals: 0.218
TCG: 0.157
KVM: 0.019

 RFE: More info in qemu-img info/check

Originally filed in Fedora bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=861375

"""
qemu-img info currently give me info like this:

image: /home/alex/.local/share/gnome-boxes/images/Fedora 16
file format: qcow2
virtual size: 11G (11794287616 bytes)
disk size: 4.5G
cluster_size: 65536

In order to figure out the "health" of an image there is some more information I would like:

in-use disk size - I.e the subset of disk size that is not marked as unused due to e.g. TRIM operations

amount of compressed clusters. I.e. "is it useful to re-compress the image".

Fragmentation estimation.

This would be useful to both sysadmins in general and for automated things like
what we want to do in gnome-boxes:
https://bugzilla.gnome.org/show_bug.cgi?id=685032
"""

As mentioned in the original report, qemu-img check currently has fragmentation stats, but only for QED.

qemu-img check has reported allocated clusters, compressed clusters and fragmentation for qcow2 images since February 2013 (QEMU 1.5).