Re: [PATCH 2/2] hw/pxb-cxl: Support passthrough HDM Decoders unless overridden

[Jonathan Cameron]

On Thu, 26 Jan 2023 21:57:35 +0000
Fan Ni <fan.ni@samsung.com> wrote:

> On Wed, Jan 25, 2023 at 03:27:03PM +0000, Jonathan Cameron wrote:
> 
> > The CXL r3.0 specification allows for there to be no HDM decoders on CXL
> > Host Bridges if they have only a single root port. Instead, all accesses
> > directed to the host bridge (as specified in CXL Fixed Memory Windows)
> > are assumed to be routed to the single root port.
> > 
> > Linux currently assumes this implementation choice. So to simplify testing,
> > make QEMU emulation also default to no HDM decoders under these particular
> > circumstances, but provide a hdm_for_passthrough boolean option to have
> > HDM decoders as previously.
> > 
> > Technically this is breaking backwards compatibility, but given the only
> > known software stack used with the QEMU emulation is the Linux kernel
> > and this configuration did not work before this change, there are
> > unlikely to be any complaints that it now works. The option is retained
> > to allow testing of software that does allow for these HDM decoders to 
> > exist,
> > once someone writes it.
> > 
> > Reported-by: Fan Ni <fan.ni@samsung.com>
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

> >      pxb_dev_realize_common(dev, CXL, errp);
> > -    pxb_dev_reset(DEVICE(dev));
> > +    pxb_cxl_dev_reset(DEVICE(dev));
> >  }
> >  
> > +static Property pxb_cxl_dev_properties[] = {
> > +    /* Note: 0 is not a legal PXB bus number. */
> > +    DEFINE_PROP_UINT8("bus_nr", PXBDev, bus_nr, 0),
> > +    DEFINE_PROP_UINT16("numa_node", PXBDev, numa_node, 
> > NUMA_NODE_UNASSIGNED),
> > +    DEFINE_PROP_BOOL("bypass_iommu", PXBDev, bypass_iommu, false),
> > +    DEFINE_PROP_BOOL("hdm_for_passthrough", PXBDev, hdm_for_passthrough, 
> > false),  
> when setting hdm_for_passthrough to true at the qemu command line, we
> will see the segfault issue as before. I think this is expected as it
> is the logic in cxl_cfmws_find_device. Wondering if there will be
> following fixes to handle the case when hdm_for_passthrough is true.

Absolutely, I'd expect a kernel fix for that case, but it's probably not
high priority for anyone given we don't yet have any hardware that does that
(as far as I know anyway!)

I wanted to keep the control here to make that easy to test when we do
have the fix in place.

Jonathan