riscv: support new "riscv,isa-extensions" string-array
ClosedPublic
Actions

Authored by br on Thu, Oct 1, 6:13 PM.

Details

Reviewers

mhorne
bnovkov

Group Reviewers

riscv

Commits

rG3e01ebda6829: riscv: support the new "riscv,isa-extensions" string-array.

Summary

Support the new "riscv,isa-extensions" property on RISC-V hart nodes in FDT.

The "riscv,isa" property is deprecated, but cannot be removed because doing so would break compatibility with existing DTBs. The new properties replace it: "riscv,isa-base" describes the base ISA and "riscv,isa-extensions" is a string array containing the supported ISA extensions.

See
https://github.com/torvalds/linux/commit/aeb71e42caae2031ec849a858080d81462cacca9
https://github.com/torvalds/linux/blob/master/Documentation/devicetree/bindings/riscv/extensions.yaml

The "riscv,isa-extensions" property can be relatively large; on the Spacemit K3 it is approximately 300 bytes.

The FreeBSD OFW interface does not provide access to the underlying FDT property data without copying it, and memory allocation is not possible this early. So allocate a static buffer for the property instead.

Reuse the existing parse_riscv_isa() implementation to parse both the new extension property and the legacy "riscv,isa" property.

Test Plan

Tested in QEMU and Spacemit K3.

Diff Detail

Lint

Lint Skipped

Unit

Tests Skipped

Event Timeline

br requested review of this revision.Thu, Oct 1, 6:13 PM

br created this revision.

Thanks! I have a half-baked implementation of this and I encountered the same issue with OFW interface. There are useful functions implemented by libfdt which we cannot use, e.g. fdt_stringlist_get(). I wondered about what a change to the OFW interface would look like to accommodate something like this...

(In the end, it is not so important 😆 )

sys/riscv/riscv/identcpu.c
409	You could use `isa_extensions[]` here too to hold the result. It is harmless either way.
470–474

This revision is now accepted and ready to land.Fri, Oct 2, 2:07 PM

LGTM, one minor inline comment aside.

sys/riscv/riscv/identcpu.c
415	I know that this was carried over from the previous implementation, but I think this should be turned into an explicit panic. Same thing goes for the `KASSERT` in `parse_isa_extensions`.