[PATCH V2 5/7] arm64/cpufeature: Add field details for ID_AA64DFR1_EL1 register

Mark Rutland mark.rutland at arm.com
Tue Dec 10 08:56:25 PST 2024


On Tue, Dec 10, 2024 at 04:41:44PM +0000, Will Deacon wrote:
> On Mon, Oct 28, 2024 at 11:04:24AM +0530, Anshuman Khandual wrote:
> > +static const struct arm64_ftr_bits ftr_id_aa64dfr1[] = {
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABL_CMPs_SHIFT, 8, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_DPFZS_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_EBEP_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ITE_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_ABLE_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_PMICNTR_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_SPMU_SHIFT, 4, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_CTX_CMPs_SHIFT, 8, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_WRPs_SHIFT, 8, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_NONSTRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_BRPs_SHIFT, 8, 0),
> > +	ARM64_FTR_BITS(FTR_HIDDEN, FTR_STRICT, FTR_LOWER_SAFE, ID_AA64DFR1_EL1_SYSPMUID_SHIFT, 8, 0),
> > +	ARM64_FTR_END,
> > +};
> 
> I think I mentioned this on an earlier series, but it would be useful to
> see some justification in the commit message as to why some of these
> features are considered STRICT vs NONSTRICT and why LOWER_SAFE is
> preferred over EXACT.
> 
> For example, why is EBEP strict whereas other PMU-related fields aren't?
> Why is the CTX_CMPs field treated differently to the same field in DFR0?
> 
> I'm not saying the above table is wrong, it just looks arbitrary without
> the justification.

FWIW, Anshuman and I discussed that on the v1 thread, after this v2
thread was posted. Anshuman promised to provide some rationale and make
some updates in the next version (i.e. v3):

  https://lore.kernel.org/linux-arm-kernel/8efe902c-8b9f-494a-b9da-430d8ced32ef@arm.com/

Mark.



More information about the linux-arm-kernel mailing list