Re: [RFC PATCH spice 1/2] QXL interface: add functions to identify monitors in the guest

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

 



Hi Gerd,

On Wed, 2018-10-10 at 12:37 +0200, Gerd Hoffmann wrote:
>   Hi,
> 
> > > + * Sets the hardware (e.g. PCI) path of the graphics device
> > > represented by this QXL interface instance.
> > 
> > So, this comment suggests that the caller might be able to provide a
> > path that is not a PCI path. But the implementation below (mostly the
> > debug output, I suppose...) assumes a PCI path. Do we need to support
> > non-PCI paths?
> 
> Certainly not for the initial revision, maybe never.
> 
> But thanks to the "pci/" prefix we should be able to add support for
> other paths later in case it turns out we need it.
> 
> > > + * Returns: The actual SPICE server monitor_id associated with the
> > > display
> > 
> > So, if I remember correctly, Gerd recommended returning this value from
> > the function. But I think it needs more explanation here. What exactly
> > is a "monitor_id" supposed to represent? It is not used in your follow-
> > up qemu patch so it's hard to tell.
> 
> IIRC the plan was to ditch the global monior_id idea and use the
> (channel_id, display_id) tuple everywhere ...

Not sure what exactly you mean here by "global monitor_id". Not sure
about "use the (channel_id, display_id)" either, it doesn't seem quite
correct.

The plan was (and still is) to limit the use cases to the following
two:

* Legacy QXL on linux with multiple monitors per display channel, but
only this single display channel. Multiple display channels are not
supported in this case, so no streaming etc.

* Limit the number of monitors per display channel to one for all other
cases.

With these limitations, the display_id = channel_id + monitor_id
formula that is used on the client remains unique. With one more
condition, that I think I should add, and that is that monitor_id is
always 0 for the multiple display channel case. It seems it may come up
that the monitor_id could be non-zero, e.g. for the virtio-gpu case...

So the IDs used are:

monitors_config server -> client:
(channel_id, monitor_id)

monitors_config client -> server and possibly server -> vd_agent:
display_id

I hope it's clear like this :)

Cheers,
Lukas

> cheers,
>   Gerd
> 
_______________________________________________
Spice-devel mailing list
Spice-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/spice-devel




[Index of Archives]     [Linux Virtualization]     [Linux Virtualization]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]     [Monitors]