Commit Graph

335 Commits

Author SHA1 Message Date
Stanislav Shwartsman
3ca29cbdf3 stack direct access optimization - 5% emu speedup to all 32-bit guests, for 64-bit guests speedup is less because they have less stack accesses 2012-03-25 11:54:32 +00:00
Stanislav Shwartsman
d4541f1a88 removed dedicated handler for MOVNTI - can be replaced with existing handlers 2012-02-27 15:50:43 +00:00
Stanislav Shwartsman
9bebe91826 eliminate duplicated cpu methods by adding extra param to opcodes with no modrm 2012-02-03 10:24:59 +00:00
Stanislav Shwartsman
e7ed8aca5c move inhibit interrrupts functionality to icount interface 2011-12-21 06:17:45 +00:00
Stanislav Shwartsman
f09bdf353a RDMSR can also read TSC so make it end-of-trace as well (same as RDTSC) 2011-11-24 16:03:51 +00:00
Stanislav Shwartsman
5d9bbae71c bugfix: cant use ib2 it is overlap with disp32 2011-10-19 21:28:36 +00:00
Stanislav Shwartsman
5cc04b9955 Implemented AMDs Buldozer XOP and TBM extensions.
XOP: few instructions are still missing, coming soon

  BX_PANIC(("VPERMILPS_VpsHpsWpsVIbR: not implemented yet"));
  BX_PANIC(("VPERMILPD_VpdHpdWpdVIbR: not implemented yet"));
  BX_PANIC(("VPMADCSSWD_VdqHdqWdqVIbR: not implemented yet"));
  BX_PANIC(("VPMADCSWD_VdqHdqWdqVIbR: not implemented yet"));
  BX_PANIC(("VFRCZPS_VpsWpsR: not implemented yet"));
  BX_PANIC(("VFRCZPD_VpdWpdR: not implemented yet"));
  BX_PANIC(("VFRCZSS_VssWssR: not implemented yet"));
  BX_PANIC(("VFRCZSD_VsdWsdR: not implemented yet"));
2011-10-19 20:54:04 +00:00
Stanislav Shwartsman
e282b5e88d Move DAZ handling into softfloat library (for float16, float32 and float64 only) and cleanup SSE and AVX code
Fix decoding of opcodes with VEX.W=1 in 32-bit mode (AVX2, FMA)
2011-10-01 15:40:36 +00:00
Stanislav Shwartsman
62d0c8abf7 - Now you could disable x86-64 from .bochsrc so now it is possible to emulate
32-bit CPU using Bochs binary compiled with x86-64 support.

The commit also fixes some init.cc issues with initialization of SYSCALL/SYSRET MSR in AMD hosts and also includes code reorg.
2011-09-25 17:36:20 +00:00
Stanislav Shwartsman
c6d07ae1b5 store modrm() for x87 in Ib() byte because x87 have no Ib() 2011-09-20 06:02:27 +00:00
Stanislav Shwartsman
50207eeb90 - Added support for AMD SSE4A emulation, the instructions can be enabled
using .bochsrc CPUID option.
2011-09-18 16:18:22 +00:00
Stanislav Shwartsman
e2f0880f1c support more than 32-bit cpu features vector 2011-09-14 20:22:24 +00:00
Stanislav Shwartsman
939aee87c9 handle special case - BSF/BSR vs TZCNT/LZCNT 2011-09-06 19:18:21 +00:00
Stanislav Shwartsman
e000b61cfd make RDTSC 'end of trace' instruction - guarantee that any pair ot RDTSC/RDTSCP on same CPU do not belong to same trace (and therefore will always update time and etc) - required for Win7 Thin 2011-09-06 14:13:39 +00:00
Stanislav Shwartsman
c0f5919787 small optimization 2011-09-03 15:36:40 +00:00
Stanislav Shwartsman
8099fd9efd implemented alternative access to CR8: AMD feature. Lock CR0 -> CR8 2011-09-03 15:22:56 +00:00
Stanislav Shwartsman
1f5e036695 lzcnt/tzcnt bmi instructions implemented 2011-08-31 20:43:47 +00:00
Stanislav Shwartsman
fb9da23f9b syscall/sysret are not supported outside long64 mode in Intel CPUs 2011-08-30 21:32:40 +00:00
Stanislav Shwartsman
0f73ff39df bug fix 2011-08-30 19:16:08 +00:00
Stanislav Shwartsman
c30275016e avx2 added broadcast from register 2011-08-29 21:00:25 +00:00
Stanislav Shwartsman
6bdfbeeffa fixed for gather VSIB calculation 2011-08-28 20:14:53 +00:00
Stanislav Shwartsman
239c5a449d added 'locked' information to bxInstruction_c for instrumentation and other future use 2011-08-27 20:09:18 +00:00
Stanislav Shwartsman
44241a1e56 - Added support for AVX and AVX2 instructions emulation, to enable configure
with --enable-avx option. When compiled in, AVX still has to be enabled
    using .bochsrc CPUID option. AVX2 FMA instructions still not implemented.

  - Added support for Bit Manipulation Instructions (BMI) emulation. The BMI
    instructions support can be enabled using .bochsrc CPUID option.
2011-08-27 13:47:16 +00:00
Stanislav Shwartsman
e48765a511 VMX fixed, cleanups 2011-07-29 20:22:35 +00:00
Stanislav Shwartsman
002c86660a reword all the CPU code in preparation for future CPU speedup implementation.
Bochs emulation can be another 10-15% faster using technique described in paper
"Fast Microcode Interpretation with Transactional Commit/Abort"
http://amas-bt.cs.virginia.edu/2011proceedings/amasbt2011-p3.pdf
2011-07-06 20:01:18 +00:00
Stanislav Shwartsman
08ba847ce4 fix bug inserted with prev commit + cleanup 2011-06-28 16:04:40 +00:00
Stanislav Shwartsman
68f96846fe no need to define 3b opcodes on < 80x686 2011-06-26 19:39:48 +00:00
Stanislav Shwartsman
2f582db722 compile less stuff for cpu-level=5 2011-06-26 19:15:30 +00:00
Stanislav Shwartsman
beafa7c88b improved x86 hw code bp handling 2011-06-24 13:38:34 +00:00
Stanislav Shwartsman
29e3f6e762 remove trace cache disabled mode from the code. next step going to be - introducing new optimization features based on trace cache 2011-06-01 20:34:04 +00:00
Stanislav Shwartsman
a02ddb36d2 undo a change from 2 weeks ago that cause correctness failure 2011-05-06 08:03:45 +00:00
Stanislav Shwartsman
c44f82f4ac small cleanup 2011-04-25 20:26:22 +00:00
Stanislav Shwartsman
024a1ace38 move X2APIC to be .bochsrc option, rework of the cpuid code 2011-04-21 13:27:42 +00:00
Stanislav Shwartsman
69b829a935 small fixes 2011-04-12 06:05:31 +00:00
Stanislav Shwartsman
0de9a5f75d compilation fix 2011-04-08 16:20:26 +00:00
Stanislav Shwartsman
4de76b0571 introduced victim cache for a trace cache structure.
Allows to significantly  cut trace cache miss latenct and find data in victim cahe instead of redoding it 
8 entries VC in parallel with direct map 64K entries
2011-03-25 23:06:34 +00:00
Stanislav Shwartsman
7664c55b08 first fixups after AVX
(and for AVX)
2011-03-20 18:27:31 +00:00
Stanislav Shwartsman
7ced718040 implemented AVX instructions support
many changes - some cleanup will follow
please report ay found bugs !
2011-03-19 20:09:34 +00:00
Stanislav Shwartsman
7d80a6ebe0 Adding Id and Rev property to all files 2011-02-24 21:54:04 +00:00
Stanislav Shwartsman
2d3f3668c7 Fixed IRET 64-bit mode bug
Support for 32 float copare methods for AVX
ckeanups in fetchdecode
2011-02-13 06:10:11 +00:00
Stanislav Shwartsman
a31103e7d8 optimize fetchdecode tables - part2 2011-01-21 16:07:51 +00:00
Stanislav Shwartsman
fbc9b8b190 phase1 of opcode tables optimization 2011-01-20 16:24:42 +00:00
Stanislav Shwartsman
8c5c078b13 optimize sse and mmx code 2011-01-16 20:42:28 +00:00
Stanislav Shwartsman
a80b44b6db split more sse ops 2011-01-09 20:18:02 +00:00
Stanislav Shwartsman
a1bc92a46b split more SSE opcodes 2011-01-08 11:20:29 +00:00
Stanislav Shwartsman
205351f44e Split R/M all SSE fetchdecode tables
- next step optimize tables
2011-01-08 09:53:52 +00:00
Stanislav Shwartsman
f9f868247a split more SSE ops 2010-12-30 20:35:10 +00:00
Stanislav Shwartsman
25b1e2e58d split more SSE ops 2010-12-26 20:41:47 +00:00
Stanislav Shwartsman
c005444d5b split more SSE opcodes 2010-12-25 07:59:15 +00:00
Stanislav Shwartsman
fee7a91d86 Fixed compilation with cpu-level=3 2010-12-24 16:34:55 +00:00