From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49647) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d3GMU-0000em-7M for qemu-devel@nongnu.org; Wed, 26 Apr 2017 02:23:39 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d3GMS-0000Oq-Bj for qemu-devel@nongnu.org; Wed, 26 Apr 2017 02:23:38 -0400 From: "Emilio G. Cota" Date: Wed, 26 Apr 2017 02:23:22 -0400 Message-Id: <1493187803-4510-10-git-send-email-cota@braap.org> In-Reply-To: <1493187803-4510-1-git-send-email-cota@braap.org> References: <1493187803-4510-1-git-send-email-cota@braap.org> Subject: [Qemu-devel] [PATCH v3 09/10] target/i386: optimize indirect branches List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: Paolo Bonzini , Peter Crosthwaite , Richard Henderson , Peter Maydell , Eduardo Habkost , Andrzej Zaborowski , Aurelien Jarno , Alexander Graf , Stefan Weil , qemu-arm@nongnu.org, alex.bennee@linaro.org, Pranith Kumar Speed up indirect branches by jumping to the target if it is valid. Softmmu measurements (see later commit for user-mode numbers): Note: baseline (i.e. speedup == 1x) is QEMU v2.9.0. - SPECint06 (test set), x86_64-softmmu (Ubuntu 16.04 guest). Host: Intel i7-4790K @ 4.00GHz 2.4x +-+--------------------------------------------------------------------------------------------------------------+-+ | | | cross | 2.2x +cross+jr..........................................................................+++...........................+-+ | | | | +++ | | 2x +-+..............................................................................|..|............................+-+ | | | | | | | | 1.8x +-+..............................................................................|####...........................+-+ | |# |# | | **** |# | 1.6x +-+............................................................................*.|*.|#...........................+-+ | * |* |# | | * |* |# | 1.4x +-+.......................................................................+++..*.|*.|#...........................+-+ | ++++++ #### * |*++# +++ | | +++ | | #++# *++* # +++ | | 1.2x +-+......................###.....####....+++............|..|...........****..#.*..*..#....####...|.###.....####..+-+ | +++ **** # **** # #### ***### *++* # * * # #++# ****|# +++#++# | | ****### +++ *++* # *++* # ++# # #### *|* |# +++ * * # * * # *** # *| *|# **** # | 1x +-++-*++*++#++***###++*++*+#++*+-*++#+****++#++***++#+-*+*++#-+****##++*++*-+#+*++*-+#++*+*++#++*-+*+#++*++*++#-++-+ | * * # * * # * * # * * # * * # * * # *|* |# *++* # * * # * * # * * # * * # * * # | | * * # * * # * * # * * # * * # * * # *+*++# * * # * * # * * # * * # * * # * * # | 0.8x +-+--****###--***###--****##--****###-****###--***###--***###--****##--****###-****###--***###--****##--****###--+-+ astar bzip2 gcc gobmk h264ref hmmlibquantum mcf omnetpperlbench sjengxalancbmk hmean png: http://imgur.com/DU36YFU NB. 'cross' represents the previous commit. Signed-off-by: Emilio G. Cota --- target/i386/translate.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/target/i386/translate.c b/target/i386/translate.c index a065271..3bd80c0 100644 --- a/target/i386/translate.c +++ b/target/i386/translate.c @@ -4988,7 +4988,7 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, gen_push_v(s, cpu_T1); gen_op_jmp_v(cpu_T0); gen_bnd_jmp(s); - gen_eob(s); + gen_jr(s, cpu_T0); break; case 3: /* lcall Ev */ gen_op_ld_v(s, ot, cpu_T1, cpu_A0); @@ -5006,7 +5006,8 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, tcg_const_i32(dflag - 1), tcg_const_i32(s->pc - s->cs_base)); } - gen_eob(s); + tcg_gen_ld_tl(cpu_tmp4, cpu_env, offsetof(CPUX86State, eip)); + gen_jr(s, cpu_tmp4); break; case 4: /* jmp Ev */ if (dflag == MO_16) { @@ -5014,7 +5015,7 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, } gen_op_jmp_v(cpu_T0); gen_bnd_jmp(s); - gen_eob(s); + gen_jr(s, cpu_T0); break; case 5: /* ljmp Ev */ gen_op_ld_v(s, ot, cpu_T1, cpu_A0); @@ -5029,7 +5030,8 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, gen_op_movl_seg_T0_vm(R_CS); gen_op_jmp_v(cpu_T1); } - gen_eob(s); + tcg_gen_ld_tl(cpu_tmp4, cpu_env, offsetof(CPUX86State, eip)); + gen_jr(s, cpu_tmp4); break; case 6: /* push Ev */ gen_push_v(s, cpu_T0); @@ -6409,7 +6411,7 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, /* Note that gen_pop_T0 uses a zero-extending load. */ gen_op_jmp_v(cpu_T0); gen_bnd_jmp(s); - gen_eob(s); + gen_jr(s, cpu_T0); break; case 0xc3: /* ret */ ot = gen_pop_T0(s); @@ -6417,7 +6419,7 @@ static target_ulong disas_insn(CPUX86State *env, DisasContext *s, /* Note that gen_pop_T0 uses a zero-extending load. */ gen_op_jmp_v(cpu_T0); gen_bnd_jmp(s); - gen_eob(s); + gen_jr(s, cpu_T0); break; case 0xca: /* lret im */ val = cpu_ldsw_code(env, s->pc); -- 2.7.4