diff options
author | Richard Henderson <richard.henderson@linaro.org> | 2022-11-07 10:42:56 +1100 |
---|---|---|
committer | Richard Henderson <richard.henderson@linaro.org> | 2023-05-16 16:30:25 -0700 |
commit | 12fde9bcdb52118495d10c32ed375679f23e323c (patch) | |
tree | af2fc03fc6bf402cad84ec11be6799fd0a373f8f /tcg | |
parent | 7b8801071951c55dc506c1fca8b40ba292a28d6e (diff) | |
download | qemu-12fde9bcdb52118495d10c32ed375679f23e323c.zip qemu-12fde9bcdb52118495d10c32ed375679f23e323c.tar.gz qemu-12fde9bcdb52118495d10c32ed375679f23e323c.tar.bz2 |
tcg: Add INDEX_op_qemu_{ld,st}_i128
Add opcodes for backend support for 128-bit memory operations.
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Diffstat (limited to 'tcg')
-rw-r--r-- | tcg/aarch64/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/arm/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/i386/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/loongarch64/tcg-target.h | 1 | ||||
-rw-r--r-- | tcg/mips/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/optimize.c | 2 | ||||
-rw-r--r-- | tcg/ppc/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/riscv/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/s390x/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/sparc64/tcg-target.h | 2 | ||||
-rw-r--r-- | tcg/tcg-op.c | 69 | ||||
-rw-r--r-- | tcg/tcg.c | 10 | ||||
-rw-r--r-- | tcg/tci/tcg-target.h | 2 |
13 files changed, 93 insertions, 7 deletions
diff --git a/tcg/aarch64/tcg-target.h b/tcg/aarch64/tcg-target.h index 378e01d..74ee2ed 100644 --- a/tcg/aarch64/tcg-target.h +++ b/tcg/aarch64/tcg-target.h @@ -129,6 +129,8 @@ extern bool have_lse2; #define TCG_TARGET_HAS_muluh_i64 1 #define TCG_TARGET_HAS_mulsh_i64 1 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_TARGET_HAS_v64 1 #define TCG_TARGET_HAS_v128 1 #define TCG_TARGET_HAS_v256 0 diff --git a/tcg/arm/tcg-target.h b/tcg/arm/tcg-target.h index 4c2d333..65efc53 100644 --- a/tcg/arm/tcg-target.h +++ b/tcg/arm/tcg-target.h @@ -125,6 +125,8 @@ extern bool use_neon_instructions; #define TCG_TARGET_HAS_rem_i32 0 #define TCG_TARGET_HAS_qemu_st8_i32 0 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_TARGET_HAS_v64 use_neon_instructions #define TCG_TARGET_HAS_v128 use_neon_instructions #define TCG_TARGET_HAS_v256 0 diff --git a/tcg/i386/tcg-target.h b/tcg/i386/tcg-target.h index 8fe6958..943af67 100644 --- a/tcg/i386/tcg-target.h +++ b/tcg/i386/tcg-target.h @@ -194,6 +194,8 @@ extern bool have_atomic16; #define TCG_TARGET_HAS_qemu_st8_i32 1 #endif +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + /* We do not support older SSE systems, only beginning with AVX1. */ #define TCG_TARGET_HAS_v64 have_avx1 #define TCG_TARGET_HAS_v128 have_avx1 diff --git a/tcg/loongarch64/tcg-target.h b/tcg/loongarch64/tcg-target.h index 75c3d80..482901a 100644 --- a/tcg/loongarch64/tcg-target.h +++ b/tcg/loongarch64/tcg-target.h @@ -168,6 +168,7 @@ typedef enum { #define TCG_TARGET_HAS_muls2_i64 0 #define TCG_TARGET_HAS_muluh_i64 1 #define TCG_TARGET_HAS_mulsh_i64 1 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 #define TCG_TARGET_DEFAULT_MO (0) diff --git a/tcg/mips/tcg-target.h b/tcg/mips/tcg-target.h index 47088af..7277a11 100644 --- a/tcg/mips/tcg-target.h +++ b/tcg/mips/tcg-target.h @@ -204,6 +204,8 @@ extern bool use_mips32r2_instructions; #define TCG_TARGET_HAS_ext16u_i64 0 /* andi rt, rs, 0xffff */ #endif +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_TARGET_DEFAULT_MO 0 #define TCG_TARGET_NEED_LDST_LABELS diff --git a/tcg/optimize.c b/tcg/optimize.c index 9614fa3..da400b9 100644 --- a/tcg/optimize.c +++ b/tcg/optimize.c @@ -2186,11 +2186,13 @@ void tcg_optimize(TCGContext *s) break; case INDEX_op_qemu_ld_i32: case INDEX_op_qemu_ld_i64: + case INDEX_op_qemu_ld_i128: done = fold_qemu_ld(&ctx, op); break; case INDEX_op_qemu_st_i32: case INDEX_op_qemu_st8_i32: case INDEX_op_qemu_st_i64: + case INDEX_op_qemu_st_i128: done = fold_qemu_st(&ctx, op); break; CASE_OP_32_64(rem): diff --git a/tcg/ppc/tcg-target.h b/tcg/ppc/tcg-target.h index d55f026..0914380 100644 --- a/tcg/ppc/tcg-target.h +++ b/tcg/ppc/tcg-target.h @@ -149,6 +149,8 @@ extern bool have_vsx; #define TCG_TARGET_HAS_mulsh_i64 1 #endif +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + /* * While technically Altivec could support V64, it has no 64-bit store * instruction and substituting two 32-bit stores makes the generated diff --git a/tcg/riscv/tcg-target.h b/tcg/riscv/tcg-target.h index dece3b3..494c986 100644 --- a/tcg/riscv/tcg-target.h +++ b/tcg/riscv/tcg-target.h @@ -163,6 +163,8 @@ typedef enum { #define TCG_TARGET_HAS_muluh_i64 1 #define TCG_TARGET_HAS_mulsh_i64 1 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_TARGET_DEFAULT_MO (0) #define TCG_TARGET_NEED_LDST_LABELS diff --git a/tcg/s390x/tcg-target.h b/tcg/s390x/tcg-target.h index fe05680..170007b 100644 --- a/tcg/s390x/tcg-target.h +++ b/tcg/s390x/tcg-target.h @@ -140,6 +140,8 @@ extern uint64_t s390_facilities[3]; #define TCG_TARGET_HAS_muluh_i64 0 #define TCG_TARGET_HAS_mulsh_i64 0 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_TARGET_HAS_v64 HAVE_FACILITY(VECTOR) #define TCG_TARGET_HAS_v128 HAVE_FACILITY(VECTOR) #define TCG_TARGET_HAS_v256 0 diff --git a/tcg/sparc64/tcg-target.h b/tcg/sparc64/tcg-target.h index f6cd869..31c5537 100644 --- a/tcg/sparc64/tcg-target.h +++ b/tcg/sparc64/tcg-target.h @@ -151,6 +151,8 @@ extern bool use_vis3_instructions; #define TCG_TARGET_HAS_muluh_i64 use_vis3_instructions #define TCG_TARGET_HAS_mulsh_i64 0 +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + #define TCG_AREG0 TCG_REG_I0 #define TCG_TARGET_DEFAULT_MO (0) diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c index b13ded1..c419228 100644 --- a/tcg/tcg-op.c +++ b/tcg/tcg-op.c @@ -3205,7 +3205,7 @@ static void canonicalize_memop_i128_as_i64(MemOp ret[2], MemOp orig) void tcg_gen_qemu_ld_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) { - MemOpIdx oi = make_memop_idx(memop, idx); + const MemOpIdx oi = make_memop_idx(memop, idx); tcg_debug_assert((memop & MO_SIZE) == MO_128); tcg_debug_assert((memop & MO_SIGN) == 0); @@ -3213,9 +3213,36 @@ void tcg_gen_qemu_ld_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) tcg_gen_req_mo(TCG_MO_LD_LD | TCG_MO_ST_LD); addr = plugin_prep_mem_callbacks(addr); - /* TODO: allow the tcg backend to see the whole operation. */ + /* TODO: For now, force 32-bit hosts to use the helper. */ + if (TCG_TARGET_HAS_qemu_ldst_i128 && TCG_TARGET_REG_BITS == 64) { + TCGv_i64 lo, hi; + TCGArg addr_arg; + MemOpIdx adj_oi; + bool need_bswap = false; + + if ((memop & MO_BSWAP) && !tcg_target_has_memory_bswap(memop)) { + lo = TCGV128_HIGH(val); + hi = TCGV128_LOW(val); + adj_oi = make_memop_idx(memop & ~MO_BSWAP, idx); + need_bswap = true; + } else { + lo = TCGV128_LOW(val); + hi = TCGV128_HIGH(val); + adj_oi = oi; + } + +#if TARGET_LONG_BITS == 32 + addr_arg = tcgv_i32_arg(addr); +#else + addr_arg = tcgv_i64_arg(addr); +#endif + tcg_gen_op4ii_i64(INDEX_op_qemu_ld_i128, lo, hi, addr_arg, adj_oi); - if (use_two_i64_for_i128(memop)) { + if (need_bswap) { + tcg_gen_bswap64_i64(lo, lo); + tcg_gen_bswap64_i64(hi, hi); + } + } else if (use_two_i64_for_i128(memop)) { MemOp mop[2]; TCGv addr_p8; TCGv_i64 x, y; @@ -3258,7 +3285,7 @@ void tcg_gen_qemu_ld_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) void tcg_gen_qemu_st_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) { - MemOpIdx oi = make_memop_idx(memop, idx); + const MemOpIdx oi = make_memop_idx(memop, idx); tcg_debug_assert((memop & MO_SIZE) == MO_128); tcg_debug_assert((memop & MO_SIGN) == 0); @@ -3266,9 +3293,39 @@ void tcg_gen_qemu_st_i128(TCGv_i128 val, TCGv addr, TCGArg idx, MemOp memop) tcg_gen_req_mo(TCG_MO_ST_LD | TCG_MO_ST_ST); addr = plugin_prep_mem_callbacks(addr); - /* TODO: allow the tcg backend to see the whole operation. */ + /* TODO: For now, force 32-bit hosts to use the helper. */ + + if (TCG_TARGET_HAS_qemu_ldst_i128 && TCG_TARGET_REG_BITS == 64) { + TCGv_i64 lo, hi; + TCGArg addr_arg; + MemOpIdx adj_oi; + bool need_bswap = false; + + if ((memop & MO_BSWAP) && !tcg_target_has_memory_bswap(memop)) { + lo = tcg_temp_new_i64(); + hi = tcg_temp_new_i64(); + tcg_gen_bswap64_i64(lo, TCGV128_HIGH(val)); + tcg_gen_bswap64_i64(hi, TCGV128_LOW(val)); + adj_oi = make_memop_idx(memop & ~MO_BSWAP, idx); + need_bswap = true; + } else { + lo = TCGV128_LOW(val); + hi = TCGV128_HIGH(val); + adj_oi = oi; + } + +#if TARGET_LONG_BITS == 32 + addr_arg = tcgv_i32_arg(addr); +#else + addr_arg = tcgv_i64_arg(addr); +#endif + tcg_gen_op4ii_i64(INDEX_op_qemu_st_i128, lo, hi, addr_arg, adj_oi); - if (use_two_i64_for_i128(memop)) { + if (need_bswap) { + tcg_temp_free_i64(lo); + tcg_temp_free_i64(hi); + } + } else if (use_two_i64_for_i128(memop)) { MemOp mop[2]; TCGv addr_p8; TCGv_i64 x, y; @@ -1735,6 +1735,10 @@ bool tcg_op_supported(TCGOpcode op) case INDEX_op_qemu_st8_i32: return TCG_TARGET_HAS_qemu_st8_i32; + case INDEX_op_qemu_ld_i128: + case INDEX_op_qemu_st_i128: + return TCG_TARGET_HAS_qemu_ldst_i128; + case INDEX_op_mov_i32: case INDEX_op_setcond_i32: case INDEX_op_brcond_i32: @@ -2187,7 +2191,7 @@ static const char * const cond_name[] = [TCG_COND_GTU] = "gtu" }; -static const char * const ldst_name[] = +static const char * const ldst_name[(MO_BSWAP | MO_SSIZE) + 1] = { [MO_UB] = "ub", [MO_SB] = "sb", @@ -2201,6 +2205,8 @@ static const char * const ldst_name[] = [MO_BEUL] = "beul", [MO_BESL] = "besl", [MO_BEUQ] = "beq", + [MO_128 + MO_BE] = "beo", + [MO_128 + MO_LE] = "leo", }; static const char * const alignment_name[(MO_AMASK >> MO_ASHIFT) + 1] = { @@ -2357,6 +2363,8 @@ static void tcg_dump_ops(TCGContext *s, FILE *f, bool have_prefs) case INDEX_op_qemu_st8_i32: case INDEX_op_qemu_ld_i64: case INDEX_op_qemu_st_i64: + case INDEX_op_qemu_ld_i128: + case INDEX_op_qemu_st_i128: { const char *s_al, *s_op, *s_at; MemOpIdx oi = op->args[k++]; diff --git a/tcg/tci/tcg-target.h b/tcg/tci/tcg-target.h index 364012e..28dc6d5 100644 --- a/tcg/tci/tcg-target.h +++ b/tcg/tci/tcg-target.h @@ -127,6 +127,8 @@ #define TCG_TARGET_HAS_mulu2_i32 1 #endif /* TCG_TARGET_REG_BITS == 64 */ +#define TCG_TARGET_HAS_qemu_ldst_i128 0 + /* Number of registers available. */ #define TCG_TARGET_NB_REGS 16 |