target/riscv: vector widening integer add and subtract

Signed-off-by: LIU Zhiwei <zhiwei_liu@c-sky.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-Id: <20200701152549.1218-12-zhiwei_liu@c-sky.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

1 files changed, 111 insertions, 0 deletions

diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c
index 26ddc2dc6f..06435ed82a 100644
--- a/target/riscv/vector_helper.c
+++ b/target/riscv/vector_helper.c
@@ -1017,3 +1017,114 @@ void HELPER(vec_rsubs64)(void *d, void *a, uint64_t b, uint32_t desc)
         *(uint64_t *)(d + i) = b - *(uint64_t *)(a + i);
     }
 }
+
+/* Vector Widening Integer Add/Subtract */
+#define WOP_UUU_B uint16_t, uint8_t, uint8_t, uint16_t, uint16_t
+#define WOP_UUU_H uint32_t, uint16_t, uint16_t, uint32_t, uint32_t
+#define WOP_UUU_W uint64_t, uint32_t, uint32_t, uint64_t, uint64_t
+#define WOP_SSS_B int16_t, int8_t, int8_t, int16_t, int16_t
+#define WOP_SSS_H int32_t, int16_t, int16_t, int32_t, int32_t
+#define WOP_SSS_W int64_t, int32_t, int32_t, int64_t, int64_t
+#define WOP_WUUU_B  uint16_t, uint8_t, uint16_t, uint16_t, uint16_t
+#define WOP_WUUU_H  uint32_t, uint16_t, uint32_t, uint32_t, uint32_t
+#define WOP_WUUU_W  uint64_t, uint32_t, uint64_t, uint64_t, uint64_t
+#define WOP_WSSS_B  int16_t, int8_t, int16_t, int16_t, int16_t
+#define WOP_WSSS_H  int32_t, int16_t, int32_t, int32_t, int32_t
+#define WOP_WSSS_W  int64_t, int32_t, int64_t, int64_t, int64_t
+RVVCALL(OPIVV2, vwaddu_vv_b, WOP_UUU_B, H2, H1, H1, DO_ADD)
+RVVCALL(OPIVV2, vwaddu_vv_h, WOP_UUU_H, H4, H2, H2, DO_ADD)
+RVVCALL(OPIVV2, vwaddu_vv_w, WOP_UUU_W, H8, H4, H4, DO_ADD)
+RVVCALL(OPIVV2, vwsubu_vv_b, WOP_UUU_B, H2, H1, H1, DO_SUB)
+RVVCALL(OPIVV2, vwsubu_vv_h, WOP_UUU_H, H4, H2, H2, DO_SUB)
+RVVCALL(OPIVV2, vwsubu_vv_w, WOP_UUU_W, H8, H4, H4, DO_SUB)
+RVVCALL(OPIVV2, vwadd_vv_b, WOP_SSS_B, H2, H1, H1, DO_ADD)
+RVVCALL(OPIVV2, vwadd_vv_h, WOP_SSS_H, H4, H2, H2, DO_ADD)
+RVVCALL(OPIVV2, vwadd_vv_w, WOP_SSS_W, H8, H4, H4, DO_ADD)
+RVVCALL(OPIVV2, vwsub_vv_b, WOP_SSS_B, H2, H1, H1, DO_SUB)
+RVVCALL(OPIVV2, vwsub_vv_h, WOP_SSS_H, H4, H2, H2, DO_SUB)
+RVVCALL(OPIVV2, vwsub_vv_w, WOP_SSS_W, H8, H4, H4, DO_SUB)
+RVVCALL(OPIVV2, vwaddu_wv_b, WOP_WUUU_B, H2, H1, H1, DO_ADD)
+RVVCALL(OPIVV2, vwaddu_wv_h, WOP_WUUU_H, H4, H2, H2, DO_ADD)
+RVVCALL(OPIVV2, vwaddu_wv_w, WOP_WUUU_W, H8, H4, H4, DO_ADD)
+RVVCALL(OPIVV2, vwsubu_wv_b, WOP_WUUU_B, H2, H1, H1, DO_SUB)
+RVVCALL(OPIVV2, vwsubu_wv_h, WOP_WUUU_H, H4, H2, H2, DO_SUB)
+RVVCALL(OPIVV2, vwsubu_wv_w, WOP_WUUU_W, H8, H4, H4, DO_SUB)
+RVVCALL(OPIVV2, vwadd_wv_b, WOP_WSSS_B, H2, H1, H1, DO_ADD)
+RVVCALL(OPIVV2, vwadd_wv_h, WOP_WSSS_H, H4, H2, H2, DO_ADD)
+RVVCALL(OPIVV2, vwadd_wv_w, WOP_WSSS_W, H8, H4, H4, DO_ADD)
+RVVCALL(OPIVV2, vwsub_wv_b, WOP_WSSS_B, H2, H1, H1, DO_SUB)
+RVVCALL(OPIVV2, vwsub_wv_h, WOP_WSSS_H, H4, H2, H2, DO_SUB)
+RVVCALL(OPIVV2, vwsub_wv_w, WOP_WSSS_W, H8, H4, H4, DO_SUB)
+GEN_VEXT_VV(vwaddu_vv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwaddu_vv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwaddu_vv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwsubu_vv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwsubu_vv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwsubu_vv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwadd_vv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwadd_vv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwadd_vv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwsub_vv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwsub_vv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwsub_vv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwaddu_wv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwaddu_wv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwaddu_wv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwsubu_wv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwsubu_wv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwsubu_wv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwadd_wv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwadd_wv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwadd_wv_w, 4, 8, clearq)
+GEN_VEXT_VV(vwsub_wv_b, 1, 2, clearh)
+GEN_VEXT_VV(vwsub_wv_h, 2, 4, clearl)
+GEN_VEXT_VV(vwsub_wv_w, 4, 8, clearq)
+
+RVVCALL(OPIVX2, vwaddu_vx_b, WOP_UUU_B, H2, H1, DO_ADD)
+RVVCALL(OPIVX2, vwaddu_vx_h, WOP_UUU_H, H4, H2, DO_ADD)
+RVVCALL(OPIVX2, vwaddu_vx_w, WOP_UUU_W, H8, H4, DO_ADD)
+RVVCALL(OPIVX2, vwsubu_vx_b, WOP_UUU_B, H2, H1, DO_SUB)
+RVVCALL(OPIVX2, vwsubu_vx_h, WOP_UUU_H, H4, H2, DO_SUB)
+RVVCALL(OPIVX2, vwsubu_vx_w, WOP_UUU_W, H8, H4, DO_SUB)
+RVVCALL(OPIVX2, vwadd_vx_b, WOP_SSS_B, H2, H1, DO_ADD)
+RVVCALL(OPIVX2, vwadd_vx_h, WOP_SSS_H, H4, H2, DO_ADD)
+RVVCALL(OPIVX2, vwadd_vx_w, WOP_SSS_W, H8, H4, DO_ADD)
+RVVCALL(OPIVX2, vwsub_vx_b, WOP_SSS_B, H2, H1, DO_SUB)
+RVVCALL(OPIVX2, vwsub_vx_h, WOP_SSS_H, H4, H2, DO_SUB)
+RVVCALL(OPIVX2, vwsub_vx_w, WOP_SSS_W, H8, H4, DO_SUB)
+RVVCALL(OPIVX2, vwaddu_wx_b, WOP_WUUU_B, H2, H1, DO_ADD)
+RVVCALL(OPIVX2, vwaddu_wx_h, WOP_WUUU_H, H4, H2, DO_ADD)
+RVVCALL(OPIVX2, vwaddu_wx_w, WOP_WUUU_W, H8, H4, DO_ADD)
+RVVCALL(OPIVX2, vwsubu_wx_b, WOP_WUUU_B, H2, H1, DO_SUB)
+RVVCALL(OPIVX2, vwsubu_wx_h, WOP_WUUU_H, H4, H2, DO_SUB)
+RVVCALL(OPIVX2, vwsubu_wx_w, WOP_WUUU_W, H8, H4, DO_SUB)
+RVVCALL(OPIVX2, vwadd_wx_b, WOP_WSSS_B, H2, H1, DO_ADD)
+RVVCALL(OPIVX2, vwadd_wx_h, WOP_WSSS_H, H4, H2, DO_ADD)
+RVVCALL(OPIVX2, vwadd_wx_w, WOP_WSSS_W, H8, H4, DO_ADD)
+RVVCALL(OPIVX2, vwsub_wx_b, WOP_WSSS_B, H2, H1, DO_SUB)
+RVVCALL(OPIVX2, vwsub_wx_h, WOP_WSSS_H, H4, H2, DO_SUB)
+RVVCALL(OPIVX2, vwsub_wx_w, WOP_WSSS_W, H8, H4, DO_SUB)
+GEN_VEXT_VX(vwaddu_vx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwaddu_vx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwaddu_vx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwsubu_vx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwsubu_vx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwsubu_vx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwadd_vx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwadd_vx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwadd_vx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwsub_vx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwsub_vx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwsub_vx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwaddu_wx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwaddu_wx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwaddu_wx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwsubu_wx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwsubu_wx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwsubu_wx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwadd_wx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwadd_wx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwadd_wx_w, 4, 8, clearq)
+GEN_VEXT_VX(vwsub_wx_b, 1, 2, clearh)
+GEN_VEXT_VX(vwsub_wx_h, 2, 4, clearl)
+GEN_VEXT_VX(vwsub_wx_w, 4, 8, clearq)