Re: [Qemu-ppc] Bug with REPORT LUNS in IBM VSCSI

[Nathan Whitehorn]

On 10/05/13 12:18, Nathan Whitehorn wrote:
> On 10/03/13 03:05, Paolo Bonzini wrote:
> > Il 02/10/2013 18:05, Nathan Whitehorn ha scritto:
> > > > Both Linux and Windows scan each channel separately.  Linux, in
> > > > addition, does not expect REPORT LUNS to use the logical unit 
> > > > addressing
> > > > method in the returned list of LUNs.  The driver converts the
> > > > channel/target/LUN tuple to that addressing method when it builds the
> > > > SRP commands.
> > > Right, I see that. And this is an approach that works, but is really
> > > really ugly since it is assumes a fixed range of possible LUNs. It also
> > > relies on some kind of inference about which LUNs actually exist since
> > > SRP has no mechanism to report a non-existant "target" back to the host
> > > (since there is only one target, it can't have non-existant targets, so
> > > there is no mechanism for this).
> > I can certainly change REPORT LUNS to report all subtargets, but I need
> > to be careful and not break Linux...
> >
> > Paolo
>
> I wrote this patch this morning. It makes VSCSI report all LUNs when a 
> REPORT_LUNS command is addressed to LUN 0 or the well-known "REPORT 
> LUNs" LUN (thus satisfying the standard and the FreeBSD driver) and 
> gives unchanged behavior for REPORT_LUNS commands addressed to other 
> LUNs (thus producing unchanged behavior under Linux and SLOF, which 
> use only LUN addressing and so never send anything to LUN 0). If this 
> looks reasonable to you, I'll submit it formally to qemu-devel.
>
> The behavior that makes Linux see no changes (reporting sub-LUNs as 
> before for REPORT_LUNS addressed to non-zero LUN values) seems to be 
> compliant with SPC-4. In particular, following table B.2, the behavior 
> of REPORT_LUNS addressed to non-zero non-well-known LUNs is "vendor 
> specific", so we can do whatever we like.
> -Nathan

There were a couple of minor issues with the patch, in particular a 
missing endianness flip for the REPORT_LUNS well-known LUN. I've 
attached a fixed version.
-Nathan

vscsi_luns.diff
Description: Text document