[Libre-soc-isa] [Bug 1071] add parallel prefix sum remap mode

bugzilla-daemon at libre-soc.org bugzilla-daemon at libre-soc.org
Fri Apr 28 10:04:04 BST 2023


--- Comment #16 from Jacob Lifshay <programmerjake at gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #15)
> drat, you're right.


> of course. okaaay sigh so this involves creating a new Form, with Parallel
> Reduction similar to svshape2 "carving out" its own niche... urrr...

as mentioned on irc, I think we should use one more XO bit and support a GPR
for X (not Y or Z) for all svshape modes.

SV Shape Dynamic:
svshaped RA, SVyd, SVzd, SVrm, vf

if RA = 0 then x_len <- VL
else x_len <- (RA)
VL <- remapped_len(x_len, ...)
# since we don't want to set MAXVL from a GPR:
MAXVL <- remapped_len(x_len=MAXVL, ...)

You are receiving this mail because:
You are on the CC list for the bug.

More information about the Libre-SOC-ISA mailing list