Re: [Qemu-devel] [PATCH] block: fix big write

[Paolo Bonzini]

On 10/12/2014 16:47, Ming Lei wrote:
> > > Finally blkdev_issue_zeroout() can send WRITE SAME(10/16) directly
> > > and it can be from user space, fs, and block drivers.
> >
> > That is WRITE SAME without UNMAP, it is not used by mkfs, and Linux has
> > always honored max_write_same_blocks for it (defaulting to a 65535 block
> > limit for older devices that did not report a limit).
> 
> From QEMU view, blk_aio_write_zeroes() still need to handle
> case without UNMAP, and the default 65535 is just linux's current
> implementation, and even the recent patch tries to increase
> the default setting. Also the default limit might be bigger on other OS.

What is "another OS" that produces WRITE SAME with >2GB of data?
http://blogs.vmware.com/vsphere/2012/06/low-level-vaai-behaviour.html#more-3129
says ESX's default write same size is 1MB (2048 blocks).

Windows does not use WRITE SAME at all, according to Microsoft at
http://msdn.microsoft.com/en-us/library/windows/hardware/dn265487%28v=vs.85%29.aspx.

65535 is a sensible default for a host that doesn't know about
max_write_same_sectors.  Anything else would break physical drives as well.

Please produce a concrete case that is broken, with a released OS.

Paolo

> > So what *concrete* case would be fixed by adding extra little-used code
> > in QEMU to do the split?