On Tue, 24 Mar 2026 12:23:54 +0100
"Danilo Krummrich" <[email protected]> wrote:
> On Tue Mar 24, 2026 at 1:18 AM CET, Deborah Brouwer wrote:
> > + register! {
> > + /// GPU identification register.
> > + pub(crate) GPU_ID(u32) @ 0x0 {
> > + /// Status of the GPU release.
> > + 3:0 ver_status;
> > + /// Minor release version number.
> > + 11:4 ver_minor;
> > + /// Major release version number.
> > + 15:12 ver_major;
> > + /// Product identifier.
> > + 19:16 prod_major;
> > + /// Architecture patch revision.
> > + 23:20 arch_rev;
> > + /// Architecture minor revision.
> > + 27:24 arch_minor;
> > + /// Architecture major revision.
> > + 31:28 arch_major;
> > + }
>
> Are you sure that this is the field order you want to choose for Tyr?
> nova-core
> had this order in the past (we're changing this currently), as it came from
> OpenRM headers, but it is pretty unusual for datasheets and TRMs, which is
> also
> why the register!() macro examples use the typical and recommended order, i.e.
>
> pub(crate) GPU_ID(u32) @ 0x0 {
> /// Architecture major revision.
> 31:28 arch_major;
> /// Architecture minor revision.
> 27:24 arch_minor;
> /// Architecture patch revision.
> 23:20 arch_rev;
> /// Product identifier.
> 19:16 prod_major;
> /// Major release version number.
> 15:12 ver_major;
> /// Minor release version number.
> 11:4 ver_minor;
> /// Status of the GPU release.
> 3:0 ver_status;
> }
It's defined in ascending bit order in the datasheets we have, so if
we're ever going to auto-generate those from the xml, we'd likely have
the same definitions Deborah came up with, unless the script re-orders
things in descending bit order.
