Re: [PATCH 3/3] transport-pci: Improve queue msix vector register desc

["Michael S. Tsirkin" <mst@redhat.com>]

On Tue, Apr 04, 2023 at 06:14:20PM +0000, Parav Pandit wrote:
> 
> > From: Michael S. Tsirkin <mst@redhat.com>
> > Sent: Tuesday, April 4, 2023 3:21 AM
> 
> > Ok but "receiving" is confusing here. And the verb writes seems to ask for
> > direction, look at queue_desc as an example.
> > 
> What is the confusion in receiving?
> Driver some reason is configuring queue's vector even if it doesn't want to "receive" interrupts?
> Do you mean it is more verbose?

I mean driver does not even receive interrupts, APIC does.


> Word here "here" isn't necessary when describing the register itself, though it exists at other places.

if nothing else let's be consistent. without here it is not immediately clear
weather it's the register that is written to or the value specified
is written somewhere else.

> > Following that example:
> > 
> > 	The driver writes the MSI-X vector number used for virtqueue
> > interrupts here.
> > 
> Since a specific number is used here, article "an" is appropriate one.

No because it's the specific value used for interrupts.
Again, check other examples.

> > would you agree?