[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] Regression PCI passthrough from 4.5.5 to 4.6.0-rc1



 - Make QEMU setup the vectors when the table entries are unmasked,
   even if MSIX is not enabled.
 - Provide an hypercall so QEMU can unmask MSIX vectors on behalf of
   the guest. This would be used to unmask the entries if MSIX is
   enabled with table entries already unmasked.
Neither sounds right. As long as MSI-X is disabled (or the maskall
bit set), the contents of the table entries is meaningless. It is not
correct to assign any meaning to them in that state. And qemu
should not be unmasking the entries on behalf of the guest either.
It ought to be possible for Xen to know the state of the mask bits
at the time of binding the interrupts. After all, with a new hypercall
added like you propose it, there would still be a window in time
where neither setting (masked or unmasked) would be correct in
Xen's internal recording of state. Quite possibly the only solution
is for qemu to communicate the intended state of the mask bit
while binding the interrupt.

Hello Jan + Roger,

as someone who is not familiar with MSI-X stuff, I still have one obvious question: Why was all of this not a problem before the commit I bisected down? Everything did work before and to be honest there is a "business" side to my question: Since that patch it has actually been broken for at least 3 major Xen releaes (4.6-4.8). Does that make any sense? For the reputation of Xen?

Regards Andreas

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.