BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.riscv-europe.org//eu-summit-2026//speaker//JJFUJU
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-eu-summit-2026-BXQ73A@cfp.riscv-europe.org
DTSTART;TZID=CET:20260611T111000
DTEND;TZID=CET:20260611T112000
DESCRIPTION:This paper presents a standards-aligned microarchitectural exte
 nsion that leverages architecturally reserved RISC-V HINT encodings to ena
 ble lightweight Very Long Instruction Word (VLIW) execution while preservi
 ng full backward binary compatibility. Unlike conventional superscalar des
 igns that rely on dynamic scheduling\, speculative issue\, and complex haz
 ard detection\, our approach encodes static scheduling decisions in HINT i
 nstructions that execute as NOPs on unmodified cores. Modified implementat
 ions interpret these hints to form statically scheduled issue bundles\, ac
 hieving higher Instruction-Level Parallelism (ILP) without increasing ISA 
 surface area or compromising compliance.\n\nWe validate the proposal throu
 gh a full-stack methodology spanning ISA modeling\, RTL implementation\, a
 nd FPGA deployment. ISA semantics were prototyped using Google’s MPACT s
 imulator to evaluate bundle formation and decode behavior. We then extende
 d the OpenHW Group CVW (Wally) core to support 4-wide integer VLIW executi
 on via a widened multi-ported register file and parallel datapaths. The de
 sign was verified in Questa and Verilator and synthesized for FPGA-based c
 ycle-accurate measurement.\n\nEvaluation on representative DSP kernels (FF
 T\, FIR\, IIR\, and dot product) demonstrates substantial IPC and cycle-co
 unt improvements relative to scalar RV32I execution\, while maintaining bi
 nary compatibility and toolchain transparency. The proposed mechanism prov
 ides a path for energy-efficient ILP extraction in embedded and domain-spe
 cific systems\, illustrating how reserved ISA space can be systematically 
 exploited to deliver microarchitectural innovation without ecosystem fragm
 entation.
DTSTAMP:20260522T162846Z
LOCATION:Poster Island C
SUMMARY:STARBUG: RISC-V Hint Instructions for Lightweight VLIW Execution on
  Embedded DSP Workloads - Leo Marek
URL:https://cfp.riscv-europe.org/eu-summit-2026/talk/BXQ73A/
END:VEVENT
END:VCALENDAR