yuzu/unicorn
1
0
Fork 0
mirror of https://github.com/yuzu-emu/unicorn.git synced 2025-01-11 08:15:40 +00:00
Code Issues Packages Projects Releases Wiki Activity
unicorn/qemu
History
Emilio G. Cota 7d0440dec4
tb-hash: improve tb_jmp_cache hash function in user mode 
Optimizations to cross-page chaining and indirect branches make
performance more sensitive to the hit rate of tb_jmp_cache.
The constraint of reserving some bits for the page number
lowers the achievable quality of the hashing function.

However, user-mode does not have this requirement. Thus,
with this change we use for user-mode a hashing function that
is both faster and of better quality than the previous one.

Measurements:

Note: baseline (i.e. speedup == 1x) is QEMU v2.9.0.

- SPECint06 (test set), x86_64-linux-user. Host: Intel i7-6700K @ 4.00GHz

2.2x +-+--------------------------------------------------------------------------------------------------------------+-+
| |
| jr |
2x +jr+multhash +....................................................+++++...................................+-+
| jr+hash |$$$ |
| |$+$ |
| ### $ |
1.8x +-+......................................................................#|#.$...................................+-+
| ++#+# $ |
| |# # $ |
1.6x +-+....................................................................***.#.$....................++$$$..........+-+
| $$$ *+* # $ |$+$ |
| ++$$$ ### $ * * # $ +++|$ $ |
| ++###+$ # # $ * * # $ ### ****## $ |
1.4x +-+...................***+#.$.........***.#.$..........................*.*.#.$...........#+#$$.*++*|#.$..........+-+
| *+* # $ * * # $ * * # $ # # $ * *+# $ |
| * * # $ +++++ * * # $ * * # $ *** # $ * * # $ ###$$ |
1.2x +-+...................*.*.#.$.***##$$.*.*.#.$..........................*.*.#.$.........*.*.#.$.*..*.#.$.***+#+$..+-+
| * * # $ *+* # $ * * # $ +++ * * # $ ++###$$ * * # $ * * # $ * * # $ |
| ***##$$ * * # $ * * # $ * * # $ ***##$$ ++### * * # $ *** #+$ * * # $ * * # $ * * # $ |
| *+*+#+$ ***##$$$ * * # $ * * # $ * * # $ *+* # $ ++####$$ ***+# * * # $ * * # $ * * # $ * * # $ * * # $ |
1x +-++-*+*+#+$+*+*+#-+$+*+*-#+$+*+*+#+$+*+*+#+$+*-*+#+$+***++#+$+*+*+#$$+*+*+#+$+*+*+#+$+*+*-#+$+*+-*+#+$+*+*+#+$-++-+
| * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ |
| * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ * * # $ |
0.8x +-+--***##$$-***##$$$-***##$$-***##$$-***##$$-***##$$-***###$$-***##$$-***##$$-***##$$-***##$$-****##$$-***##$$--+-+
astar bzip2 gcc gobmk h264ref hmmlibquantum mcf omnetpperlbench sjengxalancbmk hmean
png: http://imgur.com/4UXTrEc

Here I also tried the hash function suggested by Paolo ("multhash"):

return ((uint64_t) (pc * 2654435761) >> 32) & (TB_JMP_CACHE_SIZE - 1);

As you can see it is just as good as the other new function ("hash"),
which is what I ended up going with.

- SPECint06 (train set), x86_64-linux-user. Host: Intel i7-6700K @ 4.00GHz

2.6x +-+--------------------------------------------------------------------------------------------------------------+-+
| |
| jr ### |
2.4x +jr+hash...........................................................................................#.#...........+-+
| # # |
| # # |
2.2x +-+................................................................................................#.#...........+-+
| # # |
| # # |
2x +-+................................................................................................#.#...........+-+
| **** # |
| * * # |
1.8x +-+.............................................................................................*..*.#...........+-+
| +++ * * # |
| #### #### * * # |
1.6x +-+......................................####.............................#..#.****..#..........*..*.#...........+-+
| +++ #++# **** # * * # #### * * # |
| ### # # * * # * * # # # * * # |
1.4x +-+...................****+#..........****..#..........................*..*..#.*..*..#....#..#..*..*.#...........+-+
| *++* # * * # * * # * * # *** # * * # #### |
| * * # #### * * # * * # * * # * * # * * # **** # |
1.2x +-+...................*..*.#..****++#.*..*..#..........................*..*..#.*..*..#..*.*..#..*..*.#..*..*..#..+-+
| ****### * * # * * # * * # * * # * * # * * # * * # * * # |
| * * # ***### * * # * * # * * # ****## * * # * * # * * # * * # * * # |
1x +-+--****###--***###--****##--****###-****###--***###--***###--****##--****###-****###--***###--****##--****###--+-+
astar bzip2 gcc gobmk h264ref hmmlibquantum mcf omnetpperlbench sjengxalancbmk hmean
png: http://imgur.com/ArCbHqo

- NBench, x86_64-linux-user. Host: Intel i7-6700K @ 4.00GHz

1.12x +-+-------------------------------------------------------------------------------------------------------------+-+
| |
| jr +++ |
1.1x +jr+hash...........................................................####.........................................+-+
| +++#| # |
| | #++# |
1.08x +-+................................+++................+++.+++..*****..#.........................................+-+
| | +++ | | * | * # |
| | | | | *+++* # |
1.06x +-+................................****###.............|...|...*...*..#.........................+++.............+-+
| *| * |# ****### * * # | |
| *| *++# *| * |# * * # #### |
1.04x +-+................................*++*..#............*|.*.|#..*...*..#........................#.|#.............+-+
| * * # *++*++# * * # +++#++# |
| * * # * * # * * # | # # +++#### |
1.02x +-+................................*..*..#......+++...*..*..#..*...*..#.....................****..#..*****++#...+-+
| +++ * * # +++ | * * # * * # +++ *| * # *+++* # |
| +++ | +++ +++ ++++++ * * # *****### * * # * * # | +++ ++++++ *++* # * * # |
1x +-++-+++++####++****###++++-+####+-*++*++#-+*+++*-+#++*++*++#++*+-+*++#+-+++####-+*****###++*++*++#++*+-+*++#+-++-+
| *****| # *++* |# *****| # * * # * *++# * * # * * # **** |# * * # * * # * * # |
| * | *| # * *++# * | *++# * * # * * # * * # * * # *| *++# * * # * * # * * # |
0.98x +-+...*.|.*++#..*..*..#..*+++*..#..*..*..#..*...*..#..*..*..#..*...*..#..*++*..#..*...*..#..*..*..#..*...*..#...+-+
| *+++* # * * # * * # * * # * * # * * # * * # * * # * * # * * # * * # |
| * * # * * # * * # * * # * * # * * # * * # * * # * * # * * # * * # |
0.96x +-+---*****###--****###--*****###--****###--*****###--****###--*****###--****###--*****###--****###--*****###---+-+
ASSIGNMENT BITFIELD FOURFP EMULATION HUFFMAN LU DECOMPOSITIONEURAL NNUMERIC SOSTRING SORT hmean
png: http://imgur.com/ZXFX0hJ

- NBench, arm-linux-user. Host: Intel i7-4790K @ 4.00GHz

1.3x +-+-------------------------------------------------------------------------------------------------------------+-+
| #### |
| jr # # +++ |
1.25x +jr+hash.....................#..#...........................................####................................+-+
| # # # # |
| # # # # |
1.2x +-+..........................#..#...........................................#..#................................+-+
| # # # # |
| # # # # |
1.15x +-+..........................#..#...........................................#..#................................+-+
| # # #### # # |
| # # # # # # |
1.1x +-+..........................#..#..................................#..#.....#..#................................+-+
| # # # # # # +++ |
| # # #### # # # # #### |
1.05x +-+..........................#..#...............#..#.....####......#..#.....#..#.........................#..#...+-+
| # # # # # # # # # # +++ # # |
| +++ ***** # #### ***** # # # +++# # **** # ****### # # |
1x +-++-+*****###++****+++++*+-+*++#+-****++#-+*+++*-+#+++++#++#++*****++#+-*++*++#-+*****-++++*++*++#++*****++#+-++-+
| * * # * * | * * # * * # * * # **** # * * # * * # * *### * *++# * * # |
| * * # * *### * * # * * # * * # * * # * * # * * # * * # * * # * * # |
0.95x +-+...*...*..#..*..*.|#..*...*..#..*..*..#..*...*..#..*..*..#..*...*..#..*..*..#..*...*..#..*..*..#..*...*..#...+-+
| * * # * * |# * * # * * # * * # * * # * * # * * # * * # * * # * * # |
| * * # * * |# * * # * * # * * # * * # * * # * * # * * # * * # * * # |
0.9x +-+---*****###--****###--*****###--****###--*****###--****###--*****###--****###--*****###--****###--*****###---+-+
ASSIGNMENT BITFIELD FOURFP EMULATION HUFFMAN LU DECOMPOSITIONEURAL NNUMERIC SOSTRING SORT hmean
png: http://imgur.com/FfD27ey

Backports commit 6f1653180f5701c6a8f1b35b89a80b1e3260928e from qemu
2018-03-03 14:11:29 -05:00
..
crypto crypto: Clean up includes 2018-02-19 00:47:40 -05:00
default-configs arm64eb: add support for ARM64 big endian. 2017-04-24 23:30:01 +08:00
docs docs: clarify memory region lifecycle 2018-02-12 15:11:21 -05:00
fpu softfloat: Use correct type in float64_to_uint64_round_to_zero() 2018-03-02 10:44:10 -05:00
hw i386: Remove AMD feature flag aliases from Opteron models 2018-03-01 23:49:04 -05:00
include tb-hash: improve tb_jmp_cache hash function in user mode 2018-03-03 14:11:29 -05:00
qapi qapi: Fix object input visit beyond end of list 2018-03-02 12:22:50 -05:00
qobject util/cutils: Rename qemu_strtoll(), qemu_strtoull() 2018-03-02 08:39:45 -05:00
qom qapi: Drop unused non-strict qobject input visitor 2018-03-02 12:14:52 -05:00
scripts qapi: rename QmpOutputVisitor to QObjectOutputVisitor 2018-02-27 08:05:33 -05:00
target target/i386: optimize indirect branches 2018-03-03 14:10:14 -05:00
tcg tcg/i386: implement goto_ptr 2018-03-02 21:08:38 -05:00
util bitmap: add bitmap_copy_and_clear_atomic 2018-03-02 18:59:11 -05:00
aarch64.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
aarch64eb.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
accel.c clean-up: removed duplicate #includes 2018-02-28 08:51:56 -05:00
arm.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
armeb.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
atomic_template.h tcg: Add atomic128 helpers 2018-02-27 21:43:48 -05:00
CODING_STYLE import 2015-08-21 15:04:50 +08:00
configure configure: remove Cygwin 2018-03-02 14:17:41 -05:00
COPYING import 2015-08-21 15:04:50 +08:00
COPYING.LIB import 2015-08-21 15:04:50 +08:00
cpu-exec-common.c tcg: Add EXCP_ATOMIC 2018-02-27 11:57:58 -05:00
cpu-exec.c tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
cpus.c tcg: handle EXCP_ATOMIC exception for system emulation 2018-03-02 09:56:43 -05:00
cputlb.c cputlb: handle first atomic write to the page 2018-03-02 18:59:12 -05:00
exec.c memory: tune last param of iommu_ops.translate() 2018-03-02 18:59:12 -05:00
gen_all_header.sh arm64eb: add support for ARM64 big endian. 2017-04-24 23:30:01 +08:00
glib_compat.c qapi: Improve qobject input visitor error reporting 2018-03-02 12:05:53 -05:00
HACKING import 2015-08-21 15:04:50 +08:00
header_gen.py tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
ioport.c hw: remove pio_addr_t 2018-02-24 02:43:16 -05:00
LICENSE import 2015-08-21 15:04:50 +08:00
m68k.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
Makefile Makefile: Add a FORCE target 2018-02-24 17:03:51 -05:00
Makefile.objs tcg: Add atomic helpers 2018-02-27 15:57:47 -05:00
Makefile.target Move target-* CPU file into a target/ folder 2018-03-01 22:50:58 -05:00
memory.c memory: Introduce DEVICE_HOST_ENDIAN for ram device 2018-03-02 11:24:32 -05:00
memory_ldst.inc.c exec: introduce memory_ldst.inc.c 2018-03-01 09:59:34 -05:00
memory_mapping.c include/qemu/osdep.h: Don't include qapi/error.h 2018-02-21 23:08:18 -05:00
mips.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
mips64.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
mips64el.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
mipsel.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
powerpc.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
qapi-schema.json qapi: add missing colon-ending for section name 2018-03-01 09:07:10 -05:00
qemu-timer.c timer/cpus: fix some typos and update some comments 2018-02-25 23:21:57 -05:00
rules.mak rules.mak: Don't extract libs from .mo-libs in link command 2018-02-26 02:08:03 -05:00
softmmu_template.h cputlb: Remove includes from softmmu_template.h 2018-02-27 12:40:43 -05:00
sparc.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
sparc64.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
tcg-runtime.c tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00
translate-all.c translate-all: exit cpu_restore_state early if translating 2018-03-02 12:46:16 -05:00
translate-all.h translate-all.c: Compute L1 page table properties at runtime 2018-02-26 11:46:58 -05:00
translate-common.c exec: Clean up includes 2018-02-19 00:49:55 -05:00
unicorn_common.h qom/cpu: Add MemoryRegion property 2018-02-18 21:54:50 -05:00
VERSION import 2015-08-21 15:04:50 +08:00
vl.c tcg: add options for enabling MTTCG 2018-03-02 09:25:01 -05:00
vl.h import 2015-08-21 15:04:50 +08:00
x86_64.h tcg: Introduce goto_ptr opcode and tcg_gen_lookup_and_goto_ptr 2018-03-02 21:05:18 -05:00