target/loongarch: Implement xvadda
This patch includes:
- XVADDA.{B/H/W/D}.
Signed-off-by: Song Gao <gaosong@loongson.cn>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20230914022645.1151356-24-gaosong@loongson.cn>
diff --git a/target/loongarch/disas.c b/target/loongarch/disas.c
index bbe7ad8..51fbd78 100644
--- a/target/loongarch/disas.c
+++ b/target/loongarch/disas.c
@@ -1851,6 +1851,11 @@
INSN_LASX(xvabsd_wu, vvv)
INSN_LASX(xvabsd_du, vvv)
+INSN_LASX(xvadda_b, vvv)
+INSN_LASX(xvadda_h, vvv)
+INSN_LASX(xvadda_w, vvv)
+INSN_LASX(xvadda_d, vvv)
+
INSN_LASX(xvreplgr2vr_b, vr)
INSN_LASX(xvreplgr2vr_h, vr)
INSN_LASX(xvreplgr2vr_w, vr)
diff --git a/target/loongarch/insn_trans/trans_vec.c.inc b/target/loongarch/insn_trans/trans_vec.c.inc
index 37b8420..c8d3567 100644
--- a/target/loongarch/insn_trans/trans_vec.c.inc
+++ b/target/loongarch/insn_trans/trans_vec.c.inc
@@ -1672,6 +1672,10 @@
TRANS(vadda_h, LSX, gvec_vvv, MO_16, do_vadda)
TRANS(vadda_w, LSX, gvec_vvv, MO_32, do_vadda)
TRANS(vadda_d, LSX, gvec_vvv, MO_64, do_vadda)
+TRANS(xvadda_b, LASX, gvec_xxx, MO_8, do_vadda)
+TRANS(xvadda_h, LASX, gvec_xxx, MO_16, do_vadda)
+TRANS(xvadda_w, LASX, gvec_xxx, MO_32, do_vadda)
+TRANS(xvadda_d, LASX, gvec_xxx, MO_64, do_vadda)
TRANS(vmax_b, LSX, gvec_vvv, MO_8, tcg_gen_gvec_smax)
TRANS(vmax_h, LSX, gvec_vvv, MO_16, tcg_gen_gvec_smax)
diff --git a/target/loongarch/insns.decode b/target/loongarch/insns.decode
index c086ee9..f3722e3 100644
--- a/target/loongarch/insns.decode
+++ b/target/loongarch/insns.decode
@@ -1432,6 +1432,11 @@
xvabsd_wu 0111 01000110 00110 ..... ..... ..... @vvv
xvabsd_du 0111 01000110 00111 ..... ..... ..... @vvv
+xvadda_b 0111 01000101 11000 ..... ..... ..... @vvv
+xvadda_h 0111 01000101 11001 ..... ..... ..... @vvv
+xvadda_w 0111 01000101 11010 ..... ..... ..... @vvv
+xvadda_d 0111 01000101 11011 ..... ..... ..... @vvv
+
xvreplgr2vr_b 0111 01101001 11110 00000 ..... ..... @vr
xvreplgr2vr_h 0111 01101001 11110 00001 ..... ..... @vr
xvreplgr2vr_w 0111 01101001 11110 00010 ..... ..... @vr
diff --git a/target/loongarch/vec_helper.c b/target/loongarch/vec_helper.c
index 35b207a..ec6d86c 100644
--- a/target/loongarch/vec_helper.c
+++ b/target/loongarch/vec_helper.c
@@ -394,22 +394,24 @@
#define DO_VABS(a) ((a < 0) ? (-a) : (a))
-#define DO_VADDA(NAME, BIT, E, DO_OP) \
-void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t v) \
-{ \
- int i; \
- VReg *Vd = (VReg *)vd; \
- VReg *Vj = (VReg *)vj; \
- VReg *Vk = (VReg *)vk; \
- for (i = 0; i < LSX_LEN/BIT; i++) { \
- Vd->E(i) = DO_OP(Vj->E(i)) + DO_OP(Vk->E(i)); \
- } \
+#define DO_VADDA(NAME, BIT, E) \
+void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t desc) \
+{ \
+ int i; \
+ VReg *Vd = (VReg *)vd; \
+ VReg *Vj = (VReg *)vj; \
+ VReg *Vk = (VReg *)vk; \
+ int oprsz = simd_oprsz(desc); \
+ \
+ for (i = 0; i < oprsz / (BIT / 8); i++) { \
+ Vd->E(i) = DO_VABS(Vj->E(i)) + DO_VABS(Vk->E(i)); \
+ } \
}
-DO_VADDA(vadda_b, 8, B, DO_VABS)
-DO_VADDA(vadda_h, 16, H, DO_VABS)
-DO_VADDA(vadda_w, 32, W, DO_VABS)
-DO_VADDA(vadda_d, 64, D, DO_VABS)
+DO_VADDA(vadda_b, 8, B)
+DO_VADDA(vadda_h, 16, H)
+DO_VADDA(vadda_w, 32, W)
+DO_VADDA(vadda_d, 64, D)
#define DO_MIN(a, b) (a < b ? a : b)
#define DO_MAX(a, b) (a > b ? a : b)
