[libre-riscv-dev] sv.setvl encoding

[Luke Kenneth Casson Leighton]

this is sv.setvl strcncpy, with fail-on-first mode.

strncpy:
    mv a3, a0
    RegCSR[a3] = 8bit, a3, scalar
    RegCSR[a1] = 8bit, a1, scalar
    RegCSR[t0] = 8bit, t0, vector
    PredTb[t0] = ffirst, x0, inv
loop:
    SETVLI a2, t4, 8 # t4 and VL now 1..8
    ldb t0, (a1) # t0 fail first mode
    bne t0, x0, allnonzero # still ff
    # VL points to last nonzero
    GETVL t4       # from bne tests
    addi t4, t4, 1 # include zero
    SETVL t4       # set exactly to t4
    stb t0, (a3)   # store incl zero
    ret            # end subroutine
allnonzero:
    stb t0, (a3)    # VL legal range
    GETVL t4        # from bne tests
    add a1, a1, t4  # Bump src pointer
    sub a2, a2, t4  # Decrement count.
    add a3, a3, t4  # Bump dst pointer
    bnez a2, loop   # Anymore?
exit:
    ret

  i think... if sv.setvl actually made t4 *mirror* VL (i.e. it
*becomes* VL), the GETVLs and SETVLs can disappear.

strncpy:
    mv a3, a0
    RegCSR[a3] = 8bit, a3, scalar
    RegCSR[a1] = 8bit, a1, scalar
    RegCSR[t0] = 8bit, t0, vector
    PredTb[t0] = ffirst, x0, inv
loop:
    SETVLI a2, t4, 8 # t4 and VL now 1..8
    ldb t0, (a1) # t0 fail first mode
    bne t0, x0, allnonzero # still ff
    # VL (aka t4) points to last nonzero
    addi t4, t4, 1 # include zero
    stb t0, (a3)   # store incl zero
    ret            # end subroutine
allnonzero:
    stb t0, (a3)    # VL legal range
    add a1, a1, t4  # Bump src pointer
    sub a2, a2, t4  # Decrement count.
    add a3, a3, t4  # Bump dst pointer
    bnez a2, loop   # Anymore?
exit:
    ret

this saves one instruction in the loop and two in the exit path.

i thought about how it could be implemented, and the effect on the
Dependency Matrix: any write operation to the register would need to
stall the instruction issue phase, because the register's value has a
*direct* effect on the number of instructions that are issued.

 however that's just stalling the *issue* phase, something that SETVL
has to do anyway (i think).

l.