Philipp Rumpf wrote:
> On Wed, 14 Feb 2001, Grant Grundler wrote:
> > Having people look things up in the spec isn't very user friendly.
>
> Having the constants in some well-known header file should be sufficient,
> shouldn't it ?
I would hope anyone bothering to include the constants in a document would
spend a few minutes explaining them as well. Perhaps a bad assumption
on my part...
> It depends on the platform and maybe the exact PCI slot used, but I don't
> think it depends on the driver (unless MSI support is broken in which case
> you would want to fix it up in the driver).
correct.
> At least I can't find
> anything in the PCI 2.2 spec that would suggest we need to consult the
> driver before enabling MSIs with one message only.
I don't know how BIOS's treat this (if at all). Need to know this first.
If they manage this resource and pre-assign everything, ok.
That's how it goes.
But if generic PCI manages this, I prefer to avoid allocating resources
that may not get used. The host platform may have a limited pool of
usable MSI data values (think parisc EIRR bits) and some cards may want
to use more than one MSI.
grant
ps. seems this thread has gotten a bit far off from the original subject. :^/
Grant Grundler
parisc-linux {PCI|IOMMU|SMP} hacker
+1.408.447.7253
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Feb 15 2001 - 21:00:26 EST