In c9ec7088 "storage: extend preallocation flags support for qemu-img" the option to fallocate was added and meant to be active when (quote): "the XML described storage <allocation> matches its <capacity>" Up until recently 81a3042a12 "storage_util: fix qemu-img sparse allocation" the compared allocation size was an order of magnitude too small, but still it does use fallocate too often unless capacity>allocation. This change fixes the comparison to match the intended description of the feature. Fixes: c9ec7088c7a3f4cd26bb471f1f243931fff6f4f9 Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1759454 Fixes: https://bugs.launchpad.net/ubuntu/focal/+source/libvirt/+bug/1847105 Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com&gt; --- src/storage/storage_util.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

On Wed, Sep 2, 2020 at 6:49 PM Michal Privoznik <mprivozn(a)redhat.com&gt; wrote: > > On 9/2/20 3:58 PM, Christian Ehrhardt wrote: > > In c9ec7088 "storage: extend preallocation flags support for qemu-img" > > the option to fallocate was added and meant to be active when (quote): > > "the XML described storage <allocation> matches its <capacity>" > > > > Up until recently 81a3042a12 "storage_util: fix qemu-img sparse allocation" > > the compared allocation size was an order of magnitude too small, but still > > it does use fallocate too often unless capacity>allocation. > > > > This change fixes the comparison to match the intended description > > of the feature. > > > > Fixes: c9ec7088c7a3f4cd26bb471f1f243931fff6f4f9 > > Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1759454 > > Fixes: https://bugs.launchpad.net/ubuntu/focal/+source/libvirt/+bug/1847105 > > > > Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com&gt; > > --- > > src/storage/storage_util.c | 6 +++--- > > 1 file changed, 3 insertions(+), 3 deletions(-) > > Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com&gt; > > And sorry for making the mess earlier (~2 years ago). no problem - it turned out to be even more confusing. Due to some further testing and encouraged by feedback in the same direction by Richard Lager (on CC now) I realized that while the suggested change reads correct it will still not help my case :-/ Even if my fix lands, we are back to square one and would need virt-manager to submit a different XML. Remember: my target here would be to come back to pralloca=metadata as it was before for image creations from virt-manager. I've started that aspect of the discussion at the BZ [1] already. On the libvirt side allocation>capacity sounds like being wrong anyway.

And if that is so we have these possible conditions: - capacity==allocation now and before my change falloc - capacity>allocation now and before my change metadata - capacity<allocation before my change falloc, afterwards metadata (but this one seems invalid anyway) So I wonder are we really back at me asking Cole to let virt-manager request things differently which is how this started about a year ago? Or was I wrong trying to make the code to match the wording in the commit that added it and do we actually want it to behave differently (read no falloc) for the XMLs sent by virt-manager as of today?

On Thu, Sep 3, 2020 at 12:36 PM Daniel P. Berrangé <berrange(a)redhat.com&gt; wrote: > > On Thu, Sep 03, 2020 at 12:18:42PM +0200, Christian Ehrhardt wrote: > > On Wed, Sep 2, 2020 at 6:49 PM Michal Privoznik <mprivozn(a)redhat.com&gt; wrote: > > > > > > On 9/2/20 3:58 PM, Christian Ehrhardt wrote: > > > > In c9ec7088 "storage: extend preallocation flags support for qemu-img" > > > > the option to fallocate was added and meant to be active when (quote): > > > > "the XML described storage <allocation> matches its <capacity>" > > > > > > > > Up until recently 81a3042a12 "storage_util: fix qemu-img sparse allocation" > > > > the compared allocation size was an order of magnitude too small, but still > > > > it does use fallocate too often unless capacity>allocation. > > > > > > > > This change fixes the comparison to match the intended description > > > > of the feature. > > > > > > > > Fixes: c9ec7088c7a3f4cd26bb471f1f243931fff6f4f9 > > > > Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1759454 > > > > Fixes: https://bugs.launchpad.net/ubuntu/focal/+source/libvirt/+bug/1847105 > > > > > > > > Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com&gt; > > > > --- > > > > src/storage/storage_util.c | 6 +++--- > > > > 1 file changed, 3 insertions(+), 3 deletions(-) > > > > > > Reviewed-by: Michal Privoznik <mprivozn(a)redhat.com&gt; > > > > > > And sorry for making the mess earlier (~2 years ago). > > > > no problem - it turned out to be even more confusing. > > > > Due to some further testing and encouraged by feedback in the same > > direction by Richard Lager (on CC now) I realized that while the > > suggested change reads correct it will still not help my case :-/ > > > > Even if my fix lands, we are back to square one and would need > > virt-manager to submit a different XML. > > Remember: my target here would be to come back to pralloca=metadata as > > it was before for image creations from virt-manager. > > I've started that aspect of the discussion at the BZ [1] already. > > > > On the libvirt side allocation>capacity sounds like being wrong anyway. > > It is a bit wierd as an input XML from a mgmt app. It is to be expected > as an output XML from libvirt though. Some filesystems, notably XFS, > will sometimes speculatively over-allocate data extents in the belief > that further size-extending writes will probably arrive. So you can end > up with allocated blocks being greater than the current logical file > size. > > > And if that is so we have these possible conditions: > > - capacity==allocation now and before my change falloc > > - capacity>allocation now and before my change metadata > > - capacity<allocation before my change falloc, afterwards metadata > > (but this one seems invalid anyway) > > > > So I wonder are we really back at me asking Cole to let virt-manager > > request things differently which is how this started about a year ago? > > Or was I wrong trying to make the code to match the wording in the > > commit that added it and do we actually want it to behave differently > > (read no falloc) for the XMLs sent by virt-manager as of today? > > I think we should provide three flags > > VIR_STORAGE_VOL_CREATE_PREALLOC_METADATA > VIR_STORAGE_VOL_CREATE_PREALLOC_NONE > VIR_STORAGE_VOL_CREATE_PREALLOC_PAYLOAD > > as a way to get explicit behaviour, with those flags ignoring the > "allocation" field. Only look at "allocation" if none of the flags > are given for sake of back compat. Independent to the issue I initially wanted to solve and to the other discussion on why fallocate on ZFS makes barely any sense I like the suggestion Daniel.

On 9/3/20 5:18 AM, Christian Ehrhardt wrote: > Even if my fix lands, we are back to square one and would need > virt-manager to submit a different XML. > Remember: my target here would be to come back to pralloca=metadata as > it was before for image creations from virt-manager. Why is that your goal? If this is simply because OpenZFS doesn't support fallocate(mode=0),

that has (finally!) been resolved for the next release: https://github.com/openzfs/zfs/commit/f734301d2267cbb33eaffbca195fc93f1da...

ZFS will "fake" the fallocate() request. It'll check to make sure there's enough free space at the moment, which is about all it can do anyway. It can't reserve the space anyway, mostly because it is a copy-on-write filesystem. Even if the application writes zeros, ZFS will just throw them away anyway (assuming you are using compression, which everyone should be). > On the libvirt side allocation>capacity sounds like being wrong anyway. > And if that is so we have these possible conditions: > - capacity==allocation now and before my change falloc > - capacity>allocation now and before my change metadata > - capacity<allocation before my change falloc, afterwards metadata > (but this one seems invalid anyway) > > So I wonder are we really back at me asking Cole to let virt-manager > request things differently which is how this started about a year ago? Setting aside cases of semi-allocation (capacity > allocation != 0) and overprovisioning (allocation > capacity), I assume the common cases are thin provisioning (allocation == 0) and thick provisioning (capacity == allocation). virt-manager (at least in the way I use it) asks explicitly for the allocation and capacity. If virt-manager is properly conveying (and I'd assume it is) the user's capacity and allocation choices from the GUI to libvirt, then virt-manager is working correctly in my view and should be left alone. I believe the main goal for thick provisioning is to reserve the space as best as possible, because ENOSPC underneath a virtual machine is bad. Secondary goals would be allocating the space relatively contiguously for performance and accounting for the space immediately to help the administrator keep track of usage. If the filesystem supports fallocate(), using it accomplishes all of these goals in a very performant way. If the filesystem does not support fallocate(), then the application can either write zeros or do nothing. Writing zeros is slow, but achieves the goals to the extent possible. Not writing zeros is fast, but does not reserve/account for the space; though, depending on the filesystem, that might not be possible anyway. I think the question fundamentally comes down to: how strong do you take a "thick provisioning" request? Do you do everything in your power to achieve it (which would mean writing zeros*) or do you treat it as a hint that you'll only follow if it is fast to do so? If it's a demand, then try fallocate() but fall back to writing zeroes. (glibc's posix_fallocate() does exactly this.). If it's a hint, then only ever call fallocate(). I think it is reasonable to treat it as a demand and write zeros if fallocate() fails. If it is too slow, the admin will notice and can make the decision to (in the future) stop requesting thick provisioning and just request thin provisioning. In the ZFS case, why is the admin requesting thick provisioning anyway? * One could go further and defeat compression by writing random data. But that seems extreme, so I'm going to ignore that. -- Richard