Message ID | 20220307065527.156132-3-npiggin@gmail.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | Fix PVR matching, add AIL cap compatibility | expand |
On Mon, Mar 07, 2022 at 04:55:25PM +1000, Nicholas Piggin wrote: > The POWER9 DD2.0 PVR does not follow the same format as the other > POWER9/10 PVRs, it includes a non-zero value in the "chip type" field. I'm unclear whether you're describing the hardware PVR here, or the value in qemu. > This does not cause problems because the pvr check is masks it out and > matches against the base, but it's a small inconsistency. Zero the > field. I assume this is making the qemu model match the hardware, but that's not entirely clear to me from the commit message. > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> > --- > target/ppc/cpu-models.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/target/ppc/cpu-models.h b/target/ppc/cpu-models.h > index 76775a74a9..b42f2ab162 100644 > --- a/target/ppc/cpu-models.h > +++ b/target/ppc/cpu-models.h > @@ -349,7 +349,7 @@ enum { > CPU_POWERPC_POWER8NVL_v10 = 0x004C0100, > CPU_POWERPC_POWER9_BASE = 0x004E0000, > CPU_POWERPC_POWER9_DD1 = 0x004E0100, > - CPU_POWERPC_POWER9_DD20 = 0x004E1200, > + CPU_POWERPC_POWER9_DD20 = 0x004E0200, > CPU_POWERPC_POWER10_BASE = 0x00800000, > CPU_POWERPC_POWER10_DD1 = 0x00800100, > CPU_POWERPC_POWER10_DD20 = 0x00800200,
diff --git a/target/ppc/cpu-models.h b/target/ppc/cpu-models.h index 76775a74a9..b42f2ab162 100644 --- a/target/ppc/cpu-models.h +++ b/target/ppc/cpu-models.h @@ -349,7 +349,7 @@ enum { CPU_POWERPC_POWER8NVL_v10 = 0x004C0100, CPU_POWERPC_POWER9_BASE = 0x004E0000, CPU_POWERPC_POWER9_DD1 = 0x004E0100, - CPU_POWERPC_POWER9_DD20 = 0x004E1200, + CPU_POWERPC_POWER9_DD20 = 0x004E0200, CPU_POWERPC_POWER10_BASE = 0x00800000, CPU_POWERPC_POWER10_DD1 = 0x00800100, CPU_POWERPC_POWER10_DD20 = 0x00800200,
The POWER9 DD2.0 PVR does not follow the same format as the other POWER9/10 PVRs, it includes a non-zero value in the "chip type" field. This does not cause problems because the pvr check is masks it out and matches against the base, but it's a small inconsistency. Zero the field. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> --- target/ppc/cpu-models.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)