OPF ISA WG External RFC ls005 v1: XLEN

  • RFC Author: Luke Kenneth Casson Leighton.
  • RFC Contributors/Ideas: Jacob Lifshay, Toshaan Bharvani
  • Funded by NLnet under the NGI Zero Entrust EU Horizon Europe Grant 101069594

URLs:

Severity: Major

Status: New

Date: 22 Dec 2022

Target v3.2B

Books and Section affected:

    Everything (in a consistent, regular and systematic fashion)

Summary

    Exactly as is already done in RISC-V, convert the entire use of 64-bit hard-coding to "XLEN".
    Exactly as is in RISC-V, options then include PowerISA-32, PowerISA-64 and PowerISA-128.
    Unlike in RISC-V, the concept of PowerISA-16 and PowerISA-8 is also floated, for Embedded,
    AI, Edge, Processing-in-Memory, Distributed Computing and other purposes.

Submitter: Luke Leighton (Libre-SOC)

Requester: Libre-SOC

Impact on processor:

    Entirely new processors, entirely new markets.

Impact on software:

    Massive but regular, consistent, and systematic.

Keywords:

    XLEN

Motivation

The Power ISA is far too massive, making it wholly unsuited for Embedded markets and adversely impacting its reach and potential. The RISC paradigm it is based on has gone too far into PackedSIMD (128-bit). Fixing this is relatively and conceptually straightforward: allow 32-bit and even 16-bit and 8-bit implementations, and use the opportunity to allow future Scalar 128-bit implementations in the exact same strategic way that RISC-V has RV128.

Register files are redefined to XLEN width but are permitted to "group" registers together to create 16-bit, 32-bit and 64-bit addresses. In this way, the limitations of what would otherwise restrict the usefulness of a severely-targetted application-specific processor may be overcome in order to make it still possible to (at reduced performance) still run general-purpose applications. AI application-specific processing or other Processing-In-Memory or other specialist design therefore may for example focus a balance of raw computing power heavily onto 8-bit or 16-bit computation, but still gain the benefit of the Power ISA and everything it brings. Contrast this with the more "normal" approach of creating heavily-focussed specialist "AI" Engines incapable of Turing-completeness and the benefits are clear.

Note: SVP64 requires this change as a 100% critical dependency. SIMD back-end ALUs process Vectors of "Elements" at 8, 16 and 32-bit (and 64-bit), read from, processed, and returned to, the standard Scalar Register Files, with byte-level write-enable lines.

Changes

For all pseudocode right across the board in all Scalar operations, replace hard-coded "64" with "XLEN". This work is already underway as sponsored by NLnet in the Libre-SOC Power ISA Pseudocode. The default is obviously recommended to be "XLEN=64" in order to create zero disruption.

Definitions of the Register File(s) for GPR and FPR are then changed to be "XLEN" wide. However, for Embedded purposes (XLEN=32/16/8), an SPR controls whether (and how many) sequentially-grouped registers are taken together to create 16-bit, 32-bit and 64-bit addresses (depending on application need). GPR is obvious, FPR is quirky. SVP64 redefines FP ops (those not ending in "s") to be "full width" and all ops ending in "s" to be "half of the full width".

  • XLEN=64 keeps FPR "full width" exactly as presently defined, and "half width" exactly as presently defined.
  • XLEN=32 overrides FPR "full width" operations to full BFP32, and "half width" to be "BFP16 stored in an BFP32"
  • XLEN=16 redefines FPR "full width" operations to full IEEE BFP16 and leaves "half width" RESERVED (there is no IEEE version of FP8).
  • XLEN=8 redefines FPR "full width" operations to bfloat16 and leaves "half width" RESERVED.

\newpage{}