qemu/target
Emilio G. Cota b4aa297781 target/i386: optimize indirect branches
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.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <1493263764-18657-11-git-send-email-cota@braap.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
2017-06-05 09:25:42 -07:00
..
alpha migration: Remove unneeded includes of migration/vmstate.h 2017-06-01 18:49:22 +02:00
arm target/arm: optimize indirect branches 2017-06-05 09:25:42 -07:00
cris qom/cpu: move tlb_flush to cpu_common_reset 2017-01-13 14:24:31 +00:00
hppa migration: Remove unneeded includes of migration/vmstate.h 2017-06-01 18:49:22 +02:00
i386 target/i386: optimize indirect branches 2017-06-05 09:25:42 -07:00
lm32 qom/cpu: move tlb_flush to cpu_common_reset 2017-01-13 14:24:31 +00:00
m68k This is the same as the v3 posted except a re-base and a few extra signoffs 2017-01-16 18:23:02 +00:00
microblaze cputlb: drop flush_global flag from tlb_flush 2017-01-13 14:24:37 +00:00
mips target/mips: fix delay slot detection in gen_msa_branch() 2017-03-20 11:19:14 +00:00
moxie qom/cpu: move tlb_flush to cpu_common_reset 2017-01-13 14:24:31 +00:00
nios2 target/nios2: Fix 64-bit ilp32 compilation 2017-06-05 09:25:42 -07:00
openrisc target/openrisc: Support non-busy idle state using PMR SPR 2017-05-04 09:39:14 +09:00
ppc target/ppc: reset reservation in do_rfi() 2017-05-24 11:39:52 +10:00
s390x migration: Remove unneeded includes of migration/vmstate.h 2017-06-01 18:49:22 +02:00
sh4 target/sh4: fix RTE instruction delay slot 2017-05-30 21:00:56 +02:00
sparc shutdown: Add source information to SHUTDOWN and RESET 2017-05-23 13:28:17 +02:00
tilegx migration: Remove unneeded includes of migration/vmstate.h 2017-06-01 18:49:22 +02:00
tricore qom/cpu: move tlb_flush to cpu_common_reset 2017-01-13 14:24:31 +00:00
unicore32 cputlb: drop flush_global flag from tlb_flush 2017-01-13 14:24:37 +00:00
xtensa target/xtensa fixes for 2.9: 2017-03-18 17:24:49 +00:00