Re: [Qemu-block] null-co undefined read behavior

[Kevin Wolf]

Am 13.06.2017 um 12:42 hat Stefan Hajnoczi geschrieben:
> On Mon, Jun 12, 2017 at 05:20:55PM -0400, John Snow wrote:
> > (3) Edit any tests to always use the zero option when using the null
> > driver, and then periodically keep updating this option in a losing
> > battle over time.
> 
> read-zeroes=on would have been a nicer default value because changing it
> could cause problems.
> 
> We can do the following:
> 
> Print a warning when null-* is instantiated with read-zeroes=off.  This
> will alert humans and it will make the missing option obvious in test
> output (hopefully we don't filter the warning out!).

I'm not a big fan of printing warnings everywhere, even more so if you
can't avoid getting them. If we overdo that, soon nobody will take them
serious any more.

We could possibly make read-zeroes a mandatory option in QAPI
(currently it seems it doesn't even exist - oops!) and if we really want
to print warnings for -drive, do it only if the default is used so that
passing an explicit read-zeroes=off removes the warning. Or we could
make it mandatory there, too.

These changes are incompatible, but I think the null-* use cases are
mostly about testing and debugging, so I feel we can be a bit laxer
about compatibility (like we already are with blkdebug).

Kevin

pgpPiPDITQQAv.pgp
Description: PGP signature