From dea7b7cb09a046e37593b9bf4577b8a482813d65 Mon Sep 17 00:00:00 2001
From: lkcl <lkcl@web>
Date: Sun, 12 Jun 2022 23:16:55 +0100
Subject: [PATCH]

---
 openpower/sv/mv.vec.mdwn | 57 +++++++++++++++++++++++++++++++++++++---
 1 file changed, 53 insertions(+), 4 deletions(-)

diff --git a/openpower/sv/mv.vec.mdwn b/openpower/sv/mv.vec.mdwn
index ec398bcb8..e13e4294c 100644
--- a/openpower/sv/mv.vec.mdwn
+++ b/openpower/sv/mv.vec.mdwn
@@ -69,7 +69,9 @@ There exists room within the `svshape` instruction of  [[sv/remap]]
 to request some alternative Matrix mappings, and there is also
 room within the reserved bits of `svremap` as well.
 
-# RM Mode Concept:
+# RM Pack/unpack
+
+Similar to [[sv/mv.swizzle]] 
 
 MVRM-2P-2S1D:
 
@@ -80,6 +82,53 @@ MVRM-2P-2S1D:
 | src_SUBVL    | `14:15`  | SUBVL for Source              |
 | MASK_SRC     | `16:18`  | Execution Mask for Source     |
 
-The inclusion of a separate src SUBVL would allow either
-`sv.mv RT.vecN RA.vecN` to mean contiguous sequential copy
-or it could mean zip/unzip (pack/unpack).
+The inclusion of a separate src SUBVL allows
+`sv.mv.swiz RT.vecN RA.vecN` to mean zip/unzip (pack/unpack).
+This is conceptually achieved by having both source and
+destination SUBVL be "outer" loops instead of inner loops.
+
+Illustrating a
+"normal" SVP64 operation with `SUBVL!=1:` (assuming no elwidth overrides):
+
+    def index():
+        for i in range(VL):
+            for j in range(SUBVL):
+                yield i*SUBVL+j
+
+    for idx in index():
+        operation_on(RA+idx)
+
+For a separate source/dest SUBVL (again, no elwidth overrides):
+
+    # yield an outer-SUBVL, inner VL loop with SRC SUBVL
+    def index_src():
+        for j in range(SRC_SUBVL):
+            for i in range(VL):
+                yield i+VL*j
+
+    # yield an outer-SUBVL, inner VL loop with DEST SUBVL
+    def index_dest():
+        for j in range(SUBVL):
+            for i in range(VL):
+                yield i+VL*j
+
+    # inner looping when SUBVLs are equal
+    if SRC_SUBVL == SUBVL:
+        for idx in index():
+            move_operation(RT+idx, RA+idx)
+    else:
+        # walk through both source and dest indices simultaneously
+        for src_idx, dst_idx in zip(index_src(), index_dst()):
+            move_operation(RT+dst_idx, RA+src_idx)
+
+"yield" from python is used here for simplicity and clarity.
+The two Finite State Machines for the generation of the source
+and destination element offsets progress incrementally in
+lock-step.
+
+Normal uaage, `SRC_SUBVL=1, SUBVL=2/3/4` gives
+a "pack" effect, and `SUBVL=1, SRC_SUBVL=2/3/4` gives an
+"unpack".  Setting both SUBVL and SRC_SUBVL to greater than
+1 will, unlike [[sv/mv.swizzle]], produce defined deterministic results,
+even if a little hard to understand.  Loops run through
+`MIN(SUBVL, SRC_SUBVL) * VL` elements.
-- 
2.30.2