target-ppc: Altivec 2.07: Vector Polynomial Multiply Sum

This patch adds the Vectory Polynomial Multiply Sum instructions
introduced in Power ISA Version 2.07:

  - Vectory Polynomial Multiply Sum Byte (vpmsumb)
  - Vectory Polynomial Multiply Sum Halfword (vpmsumh)
  - Vectory Polynomial Multiply Sum Word (vpmsumw)
  - Vectory Polynomial Multiply Sum Doubleword (vpmsumd)

Signed-off-by: Tom Musta <tommusta@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>

3 files changed, 82 insertions, 0 deletions

diff --git a/target-ppc/helper.h b/target-ppc/helper.h
index aca712fa48..ca9eba58fa 100644
--- a/target-ppc/helper.h
+++ b/target-ppc/helper.h
@@ -311,6 +311,10 @@ DEF_HELPER_2(vpopcntw, void, avr, avr)
 DEF_HELPER_2(vpopcntd, void, avr, avr)
 DEF_HELPER_3(vbpermq, void, avr, avr, avr)
 DEF_HELPER_2(vgbbd, void, avr, avr)
+DEF_HELPER_3(vpmsumb, void, avr, avr, avr)
+DEF_HELPER_3(vpmsumh, void, avr, avr, avr)
+DEF_HELPER_3(vpmsumw, void, avr, avr, avr)
+DEF_HELPER_3(vpmsumd, void, avr, avr, avr)
 
 DEF_HELPER_2(xsadddp, void, env, i32)
 DEF_HELPER_2(xssubdp, void, env, i32)
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c
index cd0b88a8bc..4e8e507819 100644
--- a/target-ppc/int_helper.c
+++ b/target-ppc/int_helper.c
@@ -1351,6 +1351,76 @@ void helper_vgbbd(ppc_avr_t *r, ppc_avr_t *b)
     r->u64[1] = t[1];
 }
 
+#define PMSUM(name, srcfld, trgfld, trgtyp)                   \
+void helper_##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)  \
+{                                                             \
+    int i, j;                                                 \
+    trgtyp prod[sizeof(ppc_avr_t)/sizeof(a->srcfld[0])];      \
+                                                              \
+    VECTOR_FOR_INORDER_I(i, srcfld) {                         \
+        prod[i] = 0;                                          \
+        for (j = 0; j < sizeof(a->srcfld[0]) * 8; j++) {      \
+            if (a->srcfld[i] & (1ull<<j)) {                   \
+                prod[i] ^= ((trgtyp)b->srcfld[i] << j);       \
+            }                                                 \
+        }                                                     \
+    }                                                         \
+                                                              \
+    VECTOR_FOR_INORDER_I(i, trgfld) {                         \
+        r->trgfld[i] = prod[2*i] ^ prod[2*i+1];               \
+    }                                                         \
+}
+
+PMSUM(vpmsumb, u8, u16, uint16_t)
+PMSUM(vpmsumh, u16, u32, uint32_t)
+PMSUM(vpmsumw, u32, u64, uint64_t)
+
+void helper_vpmsumd(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b)
+{
+
+#ifdef CONFIG_INT128
+    int i, j;
+    __uint128_t prod[2];
+
+    VECTOR_FOR_INORDER_I(i, u64) {
+        prod[i] = 0;
+        for (j = 0; j < 64; j++) {
+            if (a->u64[i] & (1ull<<j)) {
+                prod[i] ^= (((__uint128_t)b->u64[i]) << j);
+            }
+        }
+    }
+
+    r->u128 = prod[0] ^ prod[1];
+
+#else
+    int i, j;
+    ppc_avr_t prod[2];
+
+    VECTOR_FOR_INORDER_I(i, u64) {
+        prod[i].u64[LO_IDX] = prod[i].u64[HI_IDX] = 0;
+        for (j = 0; j < 64; j++) {
+            if (a->u64[i] & (1ull<<j)) {
+                ppc_avr_t bshift;
+                if (j == 0) {
+                    bshift.u64[HI_IDX] = 0;
+                    bshift.u64[LO_IDX] = b->u64[i];
+                } else {
+                    bshift.u64[HI_IDX] = b->u64[i] >> (64-j);
+                    bshift.u64[LO_IDX] = b->u64[i] << j;
+                }
+                prod[i].u64[LO_IDX] ^= bshift.u64[LO_IDX];
+                prod[i].u64[HI_IDX] ^= bshift.u64[HI_IDX];
+            }
+        }
+    }
+
+    r->u64[LO_IDX] = prod[0].u64[LO_IDX] ^ prod[1].u64[LO_IDX];
+    r->u64[HI_IDX] = prod[0].u64[HI_IDX] ^ prod[1].u64[HI_IDX];
+#endif
+}
+
+
 #if defined(HOST_WORDS_BIGENDIAN)
 #define PKBIG 1
 #else
diff --git a/target-ppc/translate.c b/target-ppc/translate.c
index 75e7f95e7c..706bd5da92 100644
--- a/target-ppc/translate.c
+++ b/target-ppc/translate.c
@@ -7372,6 +7372,10 @@ GEN_VXFORM_DUAL(vclzd, PPC_NONE, PPC2_ALTIVEC_207, \
                 vpopcntd, PPC_NONE, PPC2_ALTIVEC_207)
 GEN_VXFORM(vbpermq, 6, 21);
 GEN_VXFORM_NOA(vgbbd, 6, 20);
+GEN_VXFORM(vpmsumb, 4, 16)
+GEN_VXFORM(vpmsumh, 4, 17)
+GEN_VXFORM(vpmsumw, 4, 18)
+GEN_VXFORM(vpmsumd, 4, 19)
 
 /***                           VSX extension                               ***/
 
@@ -10623,6 +10627,10 @@ GEN_VXFORM_DUAL(vclzd, vpopcntd, 1, 31, PPC_NONE, PPC2_ALTIVEC_207),
 
 GEN_VXFORM_207(vbpermq, 6, 21),
 GEN_VXFORM_207(vgbbd, 6, 20),
+GEN_VXFORM_207(vpmsumb, 4, 16),
+GEN_VXFORM_207(vpmsumh, 4, 17),
+GEN_VXFORM_207(vpmsumw, 4, 18),
+GEN_VXFORM_207(vpmsumd, 4, 19),
 
 GEN_HANDLER_E(lxsdx, 0x1F, 0x0C, 0x12, 0, PPC_NONE, PPC2_VSX),
 GEN_HANDLER_E(lxsiwax, 0x1F, 0x0C, 0x02, 0, PPC_NONE, PPC2_VSX207),