From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:52540) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bozr6-0002hv-9m for qemu-devel@nongnu.org; Tue, 27 Sep 2016 17:24:01 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bozr3-0008UO-7C for qemu-devel@nongnu.org; Tue, 27 Sep 2016 17:23:59 -0400 Received: from mail-pf0-x241.google.com ([2607:f8b0:400e:c00::241]:34423) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bozr2-0008TP-Sw for qemu-devel@nongnu.org; Tue, 27 Sep 2016 17:23:57 -0400 Received: by mail-pf0-x241.google.com with SMTP id 21so1212580pfy.1 for ; Tue, 27 Sep 2016 14:23:56 -0700 (PDT) Sender: Richard Henderson From: Richard Henderson Date: Tue, 27 Sep 2016 14:23:52 -0700 Message-Id: <1475011433-24456-2-git-send-email-rth@twiddle.net> In-Reply-To: <1475011433-24456-1-git-send-email-rth@twiddle.net> References: <1475011433-24456-1-git-send-email-rth@twiddle.net> Subject: [Qemu-devel] [PATCH 1/2] tcg: Add tcg_gen_mulsu2_{i32,i64,tl} List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Sagar Karandikar This multiply has one signed input and one unsigned input, producing the full double-width result. Signed-off-by: Richard Henderson --- tcg/tcg-op.c | 43 +++++++++++++++++++++++++++++++++++++++++++ tcg/tcg-op.h | 4 ++++ 2 files changed, 47 insertions(+) diff --git a/tcg/tcg-op.c b/tcg/tcg-op.c index 291d50b..d92a6cd 100644 --- a/tcg/tcg-op.c +++ b/tcg/tcg-op.c @@ -688,6 +688,33 @@ void tcg_gen_muls2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) } } +void tcg_gen_mulsu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2) +{ + if (TCG_TARGET_REG_BITS == 32) { + TCGv_i32 t0 = tcg_temp_new_i32(); + TCGv_i32 t1 = tcg_temp_new_i32(); + TCGv_i32 t2 = tcg_temp_new_i32(); + tcg_gen_mulu2_i32(t0, t1, arg1, arg2); + /* Adjust for negative input for the signed arg1. */ + tcg_gen_sari_i32(t2, arg1, 31); + tcg_gen_and_i32(t2, t2, arg2); + tcg_gen_sub_i32(rh, t1, t2); + tcg_gen_mov_i32(rl, t0); + tcg_temp_free_i32(t0); + tcg_temp_free_i32(t1); + tcg_temp_free_i32(t2); + } else { + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + tcg_gen_ext_i32_i64(t0, arg1); + tcg_gen_extu_i32_i64(t1, arg2); + tcg_gen_mul_i64(t0, t0, t1); + tcg_gen_extr_i64_i32(rl, rh, t0); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + } +} + void tcg_gen_ext8s_i32(TCGv_i32 ret, TCGv_i32 arg) { if (TCG_TARGET_HAS_ext8s_i32) { @@ -1758,6 +1785,22 @@ void tcg_gen_muls2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) } } +void tcg_gen_mulsu2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2) +{ + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); + TCGv_i64 t2 = tcg_temp_new_i64(); + tcg_gen_mulu2_i64(t0, t1, arg1, arg2); + /* Adjust for negative input for the signed arg1. */ + tcg_gen_sari_i64(t2, arg1, 63); + tcg_gen_and_i64(t2, t2, arg2); + tcg_gen_sub_i64(rh, t1, t2); + tcg_gen_mov_i64(rl, t0); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + tcg_temp_free_i64(t2); +} + /* Size changing operations. */ void tcg_gen_extrl_i64_i32(TCGv_i32 ret, TCGv_i64 arg) diff --git a/tcg/tcg-op.h b/tcg/tcg-op.h index 02cb376..56407d2 100644 --- a/tcg/tcg-op.h +++ b/tcg/tcg-op.h @@ -306,6 +306,7 @@ void tcg_gen_sub2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 al, TCGv_i32 ah, TCGv_i32 bl, TCGv_i32 bh); void tcg_gen_mulu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2); void tcg_gen_muls2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2); +void tcg_gen_mulsu2_i32(TCGv_i32 rl, TCGv_i32 rh, TCGv_i32 arg1, TCGv_i32 arg2); void tcg_gen_ext8s_i32(TCGv_i32 ret, TCGv_i32 arg); void tcg_gen_ext16s_i32(TCGv_i32 ret, TCGv_i32 arg); void tcg_gen_ext8u_i32(TCGv_i32 ret, TCGv_i32 arg); @@ -482,6 +483,7 @@ void tcg_gen_sub2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 al, TCGv_i64 ah, TCGv_i64 bl, TCGv_i64 bh); void tcg_gen_mulu2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2); void tcg_gen_muls2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2); +void tcg_gen_mulsu2_i64(TCGv_i64 rl, TCGv_i64 rh, TCGv_i64 arg1, TCGv_i64 arg2); void tcg_gen_not_i64(TCGv_i64 ret, TCGv_i64 arg); void tcg_gen_ext8s_i64(TCGv_i64 ret, TCGv_i64 arg); void tcg_gen_ext16s_i64(TCGv_i64 ret, TCGv_i64 arg); @@ -932,6 +934,7 @@ static inline void tcg_gen_qemu_st64(TCGv_i64 arg, TCGv addr, int mem_index) #define tcg_gen_sub2_tl tcg_gen_sub2_i64 #define tcg_gen_mulu2_tl tcg_gen_mulu2_i64 #define tcg_gen_muls2_tl tcg_gen_muls2_i64 +#define tcg_gen_mulsu2_tl tcg_gen_mulsu2_i64 #else #define tcg_gen_movi_tl tcg_gen_movi_i32 #define tcg_gen_mov_tl tcg_gen_mov_i32 @@ -1009,6 +1012,7 @@ static inline void tcg_gen_qemu_st64(TCGv_i64 arg, TCGv addr, int mem_index) #define tcg_gen_sub2_tl tcg_gen_sub2_i32 #define tcg_gen_mulu2_tl tcg_gen_mulu2_i32 #define tcg_gen_muls2_tl tcg_gen_muls2_i32 +#define tcg_gen_mulsu2_tl tcg_gen_mulsu2_i32 #endif #if UINTPTR_MAX == UINT32_MAX -- 2.5.5