RISC-V: Adding T-Head FMemIdx extension

This patch adds support for the T-Head FMemIdx instructions.
The patch uses the T-Head specific decoder and translation.

Co-developed-by: LIU Zhiwei <zhiwei_liu@linux.alibaba.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Message-Id: <20230131202013.2541053-11-christoph.muellner@vrull.eu>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

target/riscv/cpu.c

@@ -114,6 +114,7 @@ static const struct isa_ext_data isa_edata_arr[] = {
     ISA_EXT_DATA_ENTRY(xtheadbs, true, PRIV_VERSION_1_11_0, ext_xtheadbs),
     ISA_EXT_DATA_ENTRY(xtheadcmo, true, PRIV_VERSION_1_11_0, ext_xtheadcmo),
     ISA_EXT_DATA_ENTRY(xtheadcondmov, true, PRIV_VERSION_1_11_0, ext_xtheadcondmov),
+    ISA_EXT_DATA_ENTRY(xtheadfmemidx, true, PRIV_VERSION_1_11_0, ext_xtheadfmemidx),
     ISA_EXT_DATA_ENTRY(xtheadmac, true, PRIV_VERSION_1_11_0, ext_xtheadmac),
     ISA_EXT_DATA_ENTRY(xtheadmemidx, true, PRIV_VERSION_1_11_0, ext_xtheadmemidx),
     ISA_EXT_DATA_ENTRY(xtheadmempair, true, PRIV_VERSION_1_11_0, ext_xtheadmempair),
@@ -1102,6 +1103,7 @@ static Property riscv_cpu_extensions[] = {
     DEFINE_PROP_BOOL("xtheadbs", RISCVCPU, cfg.ext_xtheadbs, false),
     DEFINE_PROP_BOOL("xtheadcmo", RISCVCPU, cfg.ext_xtheadcmo, false),
     DEFINE_PROP_BOOL("xtheadcondmov", RISCVCPU, cfg.ext_xtheadcondmov, false),
+    DEFINE_PROP_BOOL("xtheadfmemidx", RISCVCPU, cfg.ext_xtheadfmemidx, false),
     DEFINE_PROP_BOOL("xtheadmac", RISCVCPU, cfg.ext_xtheadmac, false),
     DEFINE_PROP_BOOL("xtheadmemidx", RISCVCPU, cfg.ext_xtheadmemidx, false),
     DEFINE_PROP_BOOL("xtheadmempair", RISCVCPU, cfg.ext_xtheadmempair, false),

target/riscv/cpu.h

@@ -478,6 +478,7 @@ struct RISCVCPUConfig {
     bool ext_xtheadbs;
     bool ext_xtheadcmo;
     bool ext_xtheadcondmov;
+    bool ext_xtheadfmemidx;
     bool ext_xtheadmac;
     bool ext_xtheadmemidx;
     bool ext_xtheadmempair;

target/riscv/insn_trans/trans_xthead.c.inc

@@ -46,6 +46,12 @@
     }                                            \
 } while (0)
 
+#define REQUIRE_XTHEADFMEMIDX(ctx) do {          \
+    if (!ctx->cfg_ptr->ext_xtheadfmemidx) {      \
+        return false;                            \
+    }                                            \
+} while (0)
+
 #define REQUIRE_XTHEADMAC(ctx) do {              \
     if (!ctx->cfg_ptr->ext_xtheadmac) {          \
         return false;                            \
@@ -341,6 +347,108 @@ static bool trans_th_mvnez(DisasContext *ctx, arg_th_mveqz *a)
     return gen_th_condmove(ctx, a, TCG_COND_NE);
 }
 
+/* XTheadFMem */
+
+/*
+ * Load 64-bit float from indexed address.
+ * If !zext_offs, then address is rs1 + (rs2 << imm2).
+ * If  zext_offs, then address is rs1 + (zext(rs2[31:0]) << imm2).
+ */
+static bool gen_fload_idx(DisasContext *ctx, arg_th_memidx *a, MemOp memop,
+                          bool zext_offs)
+{
+    TCGv_i64 rd = cpu_fpr[a->rd];
+    TCGv addr = get_th_address_indexed(ctx, a->rs1, a->rs2, a->imm2, zext_offs);
+
+    tcg_gen_qemu_ld_i64(rd, addr, ctx->mem_idx, memop);
+    if ((memop & MO_SIZE) == MO_32) {
+        gen_nanbox_s(rd, rd);
+    }
+
+    mark_fs_dirty(ctx);
+    return true;
+}
+
+/*
+ * Store 64-bit float to indexed address.
+ * If !zext_offs, then address is rs1 + (rs2 << imm2).
+ * If  zext_offs, then address is rs1 + (zext(rs2[31:0]) << imm2).
+ */
+static bool gen_fstore_idx(DisasContext *ctx, arg_th_memidx *a, MemOp memop,
+                           bool zext_offs)
+{
+    TCGv_i64 rd = cpu_fpr[a->rd];
+    TCGv addr = get_th_address_indexed(ctx, a->rs1, a->rs2, a->imm2, zext_offs);
+
+    tcg_gen_qemu_st_i64(rd, addr, ctx->mem_idx, memop);
+
+    return true;
+}
+
+static bool trans_th_flrd(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVD);
+    return gen_fload_idx(ctx, a, MO_TEUQ, false);
+}
+
+static bool trans_th_flrw(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVF);
+    return gen_fload_idx(ctx, a, MO_TEUL, false);
+}
+
+static bool trans_th_flurd(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVD);
+    return gen_fload_idx(ctx, a, MO_TEUQ, true);
+}
+
+static bool trans_th_flurw(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVF);
+    return gen_fload_idx(ctx, a, MO_TEUL, true);
+}
+
+static bool trans_th_fsrd(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVD);
+    return gen_fstore_idx(ctx, a, MO_TEUQ, false);
+}
+
+static bool trans_th_fsrw(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVF);
+    return gen_fstore_idx(ctx, a, MO_TEUL, false);
+}
+
+static bool trans_th_fsurd(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVD);
+    return gen_fstore_idx(ctx, a, MO_TEUQ, true);
+}
+
+static bool trans_th_fsurw(DisasContext *ctx, arg_th_memidx *a)
+{
+    REQUIRE_XTHEADFMEMIDX(ctx);
+    REQUIRE_FPU;
+    REQUIRE_EXT(ctx, RVF);
+    return gen_fstore_idx(ctx, a, MO_TEUL, true);
+}
+
 /* XTheadMac */
 
 static bool gen_th_mac(DisasContext *ctx, arg_r *a,

target/riscv/translate.c

@@ -134,7 +134,8 @@ static bool has_xthead_p(DisasContext *ctx  __attribute__((__unused__)))
 {
     return ctx->cfg_ptr->ext_xtheadba || ctx->cfg_ptr->ext_xtheadbb ||
            ctx->cfg_ptr->ext_xtheadbs || ctx->cfg_ptr->ext_xtheadcmo ||
-           ctx->cfg_ptr->ext_xtheadcondmov || ctx->cfg_ptr->ext_xtheadmac ||
+           ctx->cfg_ptr->ext_xtheadcondmov ||
+           ctx->cfg_ptr->ext_xtheadfmemidx || ctx->cfg_ptr->ext_xtheadmac ||
            ctx->cfg_ptr->ext_xtheadmemidx || ctx->cfg_ptr->ext_xtheadmempair ||
            ctx->cfg_ptr->ext_xtheadsync;
 }

target/riscv/xthead.decode

@@ -100,6 +100,16 @@ th_l2cache_iall  0000000 10110 00000 000 00000 0001011
 th_mveqz         0100000 ..... .....  001 ..... 0001011 @r
 th_mvnez         0100001 ..... .....  001 ..... 0001011 @r
 
+# XTheadFMemIdx
+th_flrd          01100 .. ..... ..... 110 ..... 0001011 @th_memidx
+th_flrw          01000 .. ..... ..... 110 ..... 0001011 @th_memidx
+th_flurd         01110 .. ..... ..... 110 ..... 0001011 @th_memidx
+th_flurw         01010 .. ..... ..... 110 ..... 0001011 @th_memidx
+th_fsrd          01100 .. ..... ..... 111 ..... 0001011 @th_memidx
+th_fsrw          01000 .. ..... ..... 111 ..... 0001011 @th_memidx
+th_fsurd         01110 .. ..... ..... 111 ..... 0001011 @th_memidx
+th_fsurw         01010 .. ..... ..... 111 ..... 0001011 @th_memidx
+
 # XTheadMac
 th_mula          00100 00 ..... ..... 001 ..... 0001011 @r
 th_mulah         00101 00 ..... ..... 001 ..... 0001011 @r