[PATCH v5 00/11] RISC-V: Probe DT extension support using riscv,isa-extensions & riscv,isa-base

* [PATCH v5 00/11] RISC-V: Probe DT extension support using riscv,isa-extensions & riscv,isa-base
@ 2023-07-13 12:10 Conor Dooley
  2023-07-13 12:10 ` [PATCH v5 01/11] RISC-V: Provide a more helpful error message on invalid ISA strings Conor Dooley
                   ` (11 more replies)
  0 siblings, 12 replies; 13+ messages in thread
From: Conor Dooley @ 2023-07-13 12:10 UTC (permalink / raw)
  To: palmer
  Cc: conor, conor.dooley, Rob Herring, Krzysztof Kozlowski,
	Paul Walmsley, Albert Ou, Jonathan Corbet, Andrew Jones,
	Heiko Stuebner, Evan Green, Sunil V L, linux-doc, linux-riscv,
	devicetree, linux-kernel

Hey,

Based on my latest iteration of deprecating riscv,isa [1], here's an
implementation of the new properties for Linux. The first few patches,
up to "RISC-V: split riscv_fill_hwcap() in 3", are all prep work that
further tames some of the extension related code, on top of my already
applied series that cleans up the ISA string parser.
Perhaps "RISC-V: shunt isa_ext_arr to cpufeature.c" is a bit gratuitous,
but I figured a bit of coalescing of extension related data structures
would be a good idea. Note that riscv,isa will still be used in the
absence of the new properties. Palmer suggested adding a Kconfig option
to turn off the fallback for DT, which I have gone and done. It's locked
behind the NONPORTABLE option for good reason.

In v2, I've also come up with a more reasonable name for the new
function I added & fixed up various comments from Drew and Evan.

In v3, there's the new commandline option that Drew suggested. I have
Also picked up a patch from Palmer that adds more helpful prints where
harts fail the checks in riscv_early_of_processor_id(), and I've
sprinkled a few more of those prints in my new additions to the
function.

v4 just rebases on v6.5-rc1 and fixes the nommu build issue due to a
missing __init.

In v5, I've fixed issues spotted by myself & Evan. I'm not the worlds
biggest fan of the strlen() calls inside the macro - but that's going to
go away again almost immediately if the scalar crypto stuff gets merged.
I also spotted an issue with a rebase I did at some point, where the
dedicated properties did not use isainfo->isa & created a bitmap for
each cpu, which would've caused the per-hart extension tracking to
break.

Cheers,
Conor.

[1] (it's in v6.5-rc1 now)

CC: Rob Herring <robh+dt@kernel.org>
CC: Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>
CC: Paul Walmsley <paul.walmsley@sifive.com>
CC: Palmer Dabbelt <palmer@dabbelt.com>
CC: Albert Ou <aou@eecs.berkeley.edu>
CC: Jonathan Corbet <corbet@lwn.net>
CC: Andrew Jones <ajones@ventanamicro.com>
CC: Heiko Stuebner <heiko.stuebner@vrull.eu>
CC: Evan Green <evan@rivosinc.com>
CC: Sunil V L <sunilvl@ventanamicro.com>
CC: linux-doc@vger.kernel.org
CC: linux-riscv@lists.infradead.org
CC: devicetree@vger.kernel.org
CC: linux-kernel@vger.kernel.org

Conor Dooley (9):
  RISC-V: drop a needless check in print_isa_ext()
  RISC-V: shunt isa_ext_arr to cpufeature.c
  RISC-V: repurpose riscv_isa_ext array in riscv_fill_hwcap()
  RISC-V: add missing single letter extension definitions
  RISC-V: add single letter extensions to riscv_isa_ext
  RISC-V: split riscv_fill_hwcap() in 3
  RISC-V: enable extension detection from dedicated properties
  RISC-V: try new extension properties in of_early_processor_hartid()
  RISC-V: provide Kconfig & commandline options to control parsing
    "riscv,isa"

Heiko Stuebner (1):
  RISC-V: don't parse dt/acpi isa string to get rv32/rv64

Palmer Dabbelt (1):
  RISC-V: Provide a more helpful error message on invalid ISA strings

 .../admin-guide/kernel-parameters.txt         |   7 +
 arch/riscv/Kconfig                            |  18 +
 arch/riscv/include/asm/hwcap.h                |  17 +-
 arch/riscv/kernel/cpu.c                       | 179 +++---
 arch/riscv/kernel/cpufeature.c                | 521 ++++++++++++------
 5 files changed, 439 insertions(+), 303 deletions(-)

-- 
2.40.1

^ permalink raw reply	[flat|nested] 13+ messages in thread