RE: [PATCH] of/irq: Fix msi-map calculation for nonzero rid-base

[Stuart Yoder]

> -----Original Message-----
> From: Marc Zyngier [mailto:marc.zyngier@xxxxxxx]
> Sent: Tuesday, February 09, 2016 6:06 AM
> To: Robin Murphy <robin.murphy@xxxxxxx>; robh+dt@xxxxxxxxxx; frowand.list@xxxxxxxxx;
> grant.likely@xxxxxxxxxx; devicetree@xxxxxxxxxxxxxxx
> Cc: mark.rutland@xxxxxxx; david.daney@xxxxxxxxxx; Stuart Yoder <stuart.yoder@xxxxxxx>;
> linux-arm-kernel@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx;
> stable@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH] of/irq: Fix msi-map calculation for nonzero rid-base
> 
> Hi Robin,
> 
> On 09/02/16 11:04, Robin Murphy wrote:
> > The existing msi-map code is fine for shifting the entire RID space
> > upwards, but attempting finer-grained remapping reveals a bug. It turns
> > out that we are mistakenly treating the msi-base part as an offset, not
> > as a new base to remap onto, so things get squiffy when rid-base is
> > nonzero. Fix this, and at the same time add a sanity check against
> > having msi-map-mask clash with a nonzero rid-base, as that's another
> > thing one can easily get wrong.
> >
> > CC: <stable@xxxxxxxxxxxxxxx>
> > Signed-off-by: Robin Murphy <robin.murphy@xxxxxxx>
> 
> Looks like Stuart and you both found the same bug at the same time:
> https://lkml.org/lkml/2016/2/8/1066
> 
> but yours seem more correct to me (the rid_base masking in Stuart's
> version seems odd).
> 
> > ---
> >  drivers/of/irq.c | 9 ++++++++-
> >  1 file changed, 8 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/of/irq.c b/drivers/of/irq.c
> > index 7ee21ae..e7bfc17 100644
> > --- a/drivers/of/irq.c
> > +++ b/drivers/of/irq.c
> > @@ -635,6 +635,13 @@ static u32 __of_msi_map_rid(struct device *dev, struct
> device_node **np,
> >  		msi_base = be32_to_cpup(msi_map + 2);
> >  		rid_len = be32_to_cpup(msi_map + 3);
> >
> > +		if (rid_base & ~map_mask) {
> > +			dev_err(parent_dev,
> > +				"Invalid msi-map translation - msi-map-mask (0x%x) ignores rid-
> base (0x%x)\n",
> > +				map_mask, rid_base);
> > +			return rid_out;
> > +		}
> > +
> >  		msi_controller_node = of_find_node_by_phandle(phandle);
> >
> >  		matched = (masked_rid >= rid_base &&
> > @@ -654,7 +661,7 @@ static u32 __of_msi_map_rid(struct device *dev, struct device_node
> **np,
> >  	if (!matched)
> >  		return rid_out;
> >
> > -	rid_out = masked_rid + msi_base;
> > +	rid_out = masked_rid - rid_base + msi_base;
> >  	dev_dbg(dev,
> >  		"msi-map at: %s, using mask %08x, rid-base: %08x, msi-base: %08x, length:
> %08x, rid: %08x -> %08x\n",
> >  		dev_name(parent_dev), map_mask, rid_base, msi_base,
> >

This computation:  masked_rid - rid_base

...doesn't seem right to me.  We are taking a rid that
has been already masked and subtracting a rid base that has
not been masked.   I don't see how you can combine masked
and unmasked values in the same calculation.

Say I have this msi mapping:

                   msi-map = <0x0100 &its 0x11 0x1>;
                   msi-map-mask = <0xff>;

masked_rid = 0x0
rid_base = 0x0100
msi_base = 0x11

masked_rid - rid_base is 0x0 - 0x0100...which does not
give the msi index/offset we want.

Correct final answer should be 0x11.

In my patch I masked the rid_base so it can be subtracted
from the masked_rid.

masked_rid_base = 0x00

   msi_base + (masked_rid - masked_rid_base) = 0x11


Stuart