SVP64 mp3 assembler almost correct

lkcl@fizzy:~/src/libresoc/openpower-isa/media$ diff -u /tmp/x1 /tmp/x2
--- /tmp/x1     2021-06-15 15:51:06.016530558 +0100
+++ /tmp/x2     2021-06-15 15:51:16.816549772 +0100
@@ -2,7 +2,7 @@
 00000010  91 38 c2 aa b0 65 00 2a  ec 4d ea 2a c6 99 f3 ab  |.8...e.*.M.*....|
 00000020  5e 86 5c 2b c7 79 ce aa  05 b1 30 ab c2 b7 35 2c  |^.\+.y....0...5,|
 00000030  b2 48 35 ab 00 d8 91 2b  26 7e 6c 2b 52 c9 ac ac  |.H5....+&~l+R...|
-00000040  00 00 00 00 fc 57 cf 2c  61 ee ce ab 00 3a ff ab  |.....W.,a....:..|
+00000040  00 00 00 80 fc 57 cf 2c  61 ee ce ab 00 3a ff ab  |.....W.,a....:..|
 00000050  42 b6 f1 2b d2 49 08 ad  46 dd 5c 2c b9 f3 0d 2c  |B..+.I..F.\,...,|
 00000060  2c 2d b3 ac 8d 26 55 2d  ec 4d ea ac 0b 6c 08 ac  |,-...&U-.M...l..|
 00000070  2c a3 e6 2c 52 97 db ac  a8 7c 03 2d 1d b3 58 ac  |,..,R....|.-..X.|

diff --git a/media/audio/mp3/mp3_0_apply_window_float_basicsv.s b/media/audio/mp3/mp3_0_apply_window_float_basicsv.s
index 9ae5f721ed92d2a1fc5d5e8b6badbc97a50ae328..c75cb574822e4e881888ab07a70a7940da0899f2 100644 (file)
--- a/media/audio/mp3/mp3_0_apply_window_float_basicsv.s
+++ b/media/audio/mp3/mp3_0_apply_window_float_basicsv.s
@@ -50,6 +50,10 @@ ff_mpadsp_apply_window_float_sv:
        mulli 0, incr, 31
        add out2, out, 0
 
+       # set predicate mask (for taking out one add in mapreduce)
+       addi 30, 0, 127 # 127 = 0b0111_1111 - 7 bits
+
+       # set Vector Length
        setvl 0, 0, 7, 1, 1 # setvli MVL=8, VL=8
        # sv.addi win2, win, 124
 
@@ -59,7 +63,9 @@ ff_mpadsp_apply_window_float_sv:
        # SUM8(MACS, sum, w, p)
        # sv.lfs/els fv0.v, 256(win)
        # sv.lfs/els fv1.v, 256(p)
-       # sv.fmuls/mr sum, fv0.v, fv1.v
+       # TODO sv.fmadds/mr sum, fv0.v, fv1.v
+       # sv.fmuls fv0.v, fv0.v, fv1.v
+       # sv.fadds/mr/m=r30 sum, fv0.v, sum
 
        lfiwax tmpsum, 0, 9 # zero it
        addi p, buf, 192
@@ -67,7 +73,9 @@ ff_mpadsp_apply_window_float_sv:
        # SUM8(MLSS, sum, w + 32, p)
        # sv.lfs/els fv0.v, 256(win)
        # sv.lfs/els fv1.v, 256(p)
-       # sv.fmuls/mr tmpsum, fv0.v, fv1.v
+       # TODO sv.fmadds/mr tmpsum, fv0.v, fv1.v
+       # sv.fmuls fv0.v, fv0.v, fv1.v
+       # sv.fadds/mr/m=r30 tmpsum, fv0.v, tmpsum
        fsubs sum, sum, tmpsum
        subi win, win, 128
 
@@ -89,8 +97,12 @@ ff_mpadsp_apply_window_float_sv:
                # sv.lfs/els fv0.v, 256(p)
                # sv.lfs/els fv1.v, 256(win)
                # sv.lfs/els fv2.v, 256(win2)
-               # sv.fmuls/mr sum, fv0.v, fv1.v
-               # sv.fmuls/mr sum2, fv0.v, fv2.v
+               # TODO sv.fmadds/mr sum, fv0.v, fv1.v
+               # sv.fmuls fv1.v, fv0.v, fv1.v
+               # sv.fadds/mr/m=r30 sum, fv1.v, sum
+               # TODO sv.fmadds/mr sum2, fv0.v, fv2.v
+               # sv.fmuls fv0.v, fv0.v, fv2.v
+               # sv.fadds/mr/m=r30 sum2, fv0.v, sum2
                fneg sum2, sum2
 
                addi p, buf, 192
@@ -104,8 +116,12 @@ ff_mpadsp_apply_window_float_sv:
                # sv.lfs/els fv0.v, 256(p)
                # sv.lfs/els fv1.v, 256(win)
                # sv.lfs/els fv2.v, 256(win2)
-               # sv.fmuls/mr tmpsum, fv0.v, fv1.v
-               # sv.fmuls/mr tmpsum2, fv0.v, fv2.v
+               # TODO sv.fmadds/mr tmpsum, fv0.v, fv1.v
+               # sv.fmuls fv1.v, fv0.v, fv1.v
+               # sv.fadds/mr/m=r30 tmpsum, fv1.v, tmpsum
+               # TODO sv.fmadds/mr tmpsum2, fv0.v, fv2.v
+               # sv.fmuls fv0.v, fv0.v, fv2.v
+               # sv.fadds/mr/m=r30 tmpsum2, fv0.v, tmpsum2
                fsubs sum, sum, tmpsum
                fsubs sum2, sum2, tmpsum2
 
@@ -128,7 +144,9 @@ ff_mpadsp_apply_window_float_sv:
        # SUM8(MLSS, sum, w + 32, p)
        # sv.lfs/els fv0.v, 256(win)
        # sv.lfs/els fv1.v, 256(p)
-       # sv.fmuls/mr sum, fv0.v, fv1.v
+       # TODO sv.fmadds/mr sum, fv0.v, fv1.v
+       # sv.fmuls fv0.v, fv0.v, fv1.v
+       # sv.fadds/mr/m=r30 sum, fv0.v, sum
        fneg sum, sum
 
        stfs sum, 0(out)