From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35507) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cR8V8-0006VM-Lk for qemu-devel@nongnu.org; Tue, 10 Jan 2017 21:19:00 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cR8V7-0003Pq-KI for qemu-devel@nongnu.org; Tue, 10 Jan 2017 21:18:58 -0500 Received: from mail-pf0-x242.google.com ([2607:f8b0:400e:c00::242]:35075) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1cR8V7-0003PO-EY for qemu-devel@nongnu.org; Tue, 10 Jan 2017 21:18:57 -0500 Received: by mail-pf0-x242.google.com with SMTP id f144so12525977pfa.2 for ; Tue, 10 Jan 2017 18:18:57 -0800 (PST) Sender: Richard Henderson From: Richard Henderson Date: Tue, 10 Jan 2017 18:17:56 -0800 Message-Id: <20170111021820.24416-42-rth@twiddle.net> In-Reply-To: <20170111021820.24416-1-rth@twiddle.net> References: <20170111021820.24416-1-rth@twiddle.net> Subject: [Qemu-devel] [PULL 41/65] tcg/aarch64: Handle ctz and clz opcodes List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: peter.maydell@linaro.org Signed-off-by: Richard Henderson --- tcg/aarch64/tcg-target.h | 8 ++++---- tcg/aarch64/tcg-target.inc.c | 48 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 52 insertions(+), 4 deletions(-) diff --git a/tcg/aarch64/tcg-target.h b/tcg/aarch64/tcg-target.h index 976f493..9d6b00f 100644 --- a/tcg/aarch64/tcg-target.h +++ b/tcg/aarch64/tcg-target.h @@ -62,8 +62,8 @@ typedef enum { #define TCG_TARGET_HAS_eqv_i32 1 #define TCG_TARGET_HAS_nand_i32 0 #define TCG_TARGET_HAS_nor_i32 0 -#define TCG_TARGET_HAS_clz_i32 0 -#define TCG_TARGET_HAS_ctz_i32 0 +#define TCG_TARGET_HAS_clz_i32 1 +#define TCG_TARGET_HAS_ctz_i32 1 #define TCG_TARGET_HAS_deposit_i32 1 #define TCG_TARGET_HAS_extract_i32 1 #define TCG_TARGET_HAS_sextract_i32 1 @@ -96,8 +96,8 @@ typedef enum { #define TCG_TARGET_HAS_eqv_i64 1 #define TCG_TARGET_HAS_nand_i64 0 #define TCG_TARGET_HAS_nor_i64 0 -#define TCG_TARGET_HAS_clz_i64 0 -#define TCG_TARGET_HAS_ctz_i64 0 +#define TCG_TARGET_HAS_clz_i64 1 +#define TCG_TARGET_HAS_ctz_i64 1 #define TCG_TARGET_HAS_deposit_i64 1 #define TCG_TARGET_HAS_extract_i64 1 #define TCG_TARGET_HAS_sextract_i64 1 diff --git a/tcg/aarch64/tcg-target.inc.c b/tcg/aarch64/tcg-target.inc.c index 17c0b20..585b0d6 100644 --- a/tcg/aarch64/tcg-target.inc.c +++ b/tcg/aarch64/tcg-target.inc.c @@ -339,8 +339,12 @@ typedef enum { /* Conditional select instructions. */ I3506_CSEL = 0x1a800000, I3506_CSINC = 0x1a800400, + I3506_CSINV = 0x5a800000, + I3506_CSNEG = 0x5a800400, /* Data-processing (1 source) instructions. */ + I3507_CLZ = 0x5ac01000, + I3507_RBIT = 0x5ac00000, I3507_REV16 = 0x5ac00400, I3507_REV32 = 0x5ac00800, I3507_REV64 = 0x5ac00c00, @@ -993,6 +997,37 @@ static inline void tcg_out_mb(TCGContext *s, TCGArg a0) tcg_out32(s, sync[a0 & TCG_MO_ALL]); } +static void tcg_out_cltz(TCGContext *s, TCGType ext, TCGReg d, + TCGReg a0, TCGArg b, bool const_b, bool is_ctz) +{ + TCGReg a1 = a0; + if (is_ctz) { + a1 = TCG_REG_TMP; + tcg_out_insn(s, 3507, RBIT, ext, a1, a0); + } + if (const_b && b == (ext ? 64 : 32)) { + tcg_out_insn(s, 3507, CLZ, ext, d, a1); + } else { + AArch64Insn sel = I3506_CSEL; + + tcg_out_cmp(s, ext, a0, 0, 1); + tcg_out_insn(s, 3507, CLZ, ext, TCG_REG_TMP, a1); + + if (const_b) { + if (b == -1) { + b = TCG_REG_XZR; + sel = I3506_CSINV; + } else if (b == 0) { + b = TCG_REG_XZR; + } else { + tcg_out_movi(s, ext, d, b); + b = d; + } + } + tcg_out_insn_3506(s, sel, ext, d, TCG_REG_TMP, b, TCG_COND_NE); + } +} + #ifdef CONFIG_SOFTMMU /* helper signature: helper_ret_ld_mmu(CPUState *env, target_ulong addr, * TCGMemOpIdx oi, uintptr_t ra) @@ -1559,6 +1594,15 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, } break; + case INDEX_op_clz_i64: + case INDEX_op_clz_i32: + tcg_out_cltz(s, ext, a0, a1, a2, c2, false); + break; + case INDEX_op_ctz_i64: + case INDEX_op_ctz_i32: + tcg_out_cltz(s, ext, a0, a1, a2, c2, true); + break; + case INDEX_op_brcond_i32: a1 = (int32_t)a1; /* FALLTHRU */ @@ -1750,11 +1794,15 @@ static const TCGTargetOpDef aarch64_op_defs[] = { { INDEX_op_sar_i32, { "r", "r", "ri" } }, { INDEX_op_rotl_i32, { "r", "r", "ri" } }, { INDEX_op_rotr_i32, { "r", "r", "ri" } }, + { INDEX_op_clz_i32, { "r", "r", "rAL" } }, + { INDEX_op_ctz_i32, { "r", "r", "rAL" } }, { INDEX_op_shl_i64, { "r", "r", "ri" } }, { INDEX_op_shr_i64, { "r", "r", "ri" } }, { INDEX_op_sar_i64, { "r", "r", "ri" } }, { INDEX_op_rotl_i64, { "r", "r", "ri" } }, { INDEX_op_rotr_i64, { "r", "r", "ri" } }, + { INDEX_op_clz_i64, { "r", "r", "rAL" } }, + { INDEX_op_ctz_i64, { "r", "r", "rAL" } }, { INDEX_op_brcond_i32, { "r", "rA" } }, { INDEX_op_brcond_i64, { "r", "rA" } }, -- 2.9.3