mirrors/Bochs
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
12,633 Commits 11 Branches 185 Tags 48 MiB
25ad0b804a
Commit Graph

118 Commits

Author SHA1 Message Date
Stanislav Shwartsman
e6822c81a1 fixed behavior of MMX PSRAW/PSRAD instructions when shift count is zero - still has to invalidate x87 tags for dest register 2020-12-15 20:05:54 +00:00
Stanislav Shwartsman
773f1b7e42 cleanup return value of all instruction handlers 2018-02-16 07:57:32 +00:00
Stanislav Shwartsman
b468316250 re-style old resolve macros after resolve function inlining 2015-05-16 21:06:59 +00:00
Stanislav Shwartsman
9f18573740 Rename BX_CPU_CALL_METHODR to BX_CPU_RESOLVE_ADDR and introduce special cases BX_CPU_RESOLVE_ADDR_64 (for 64-bit mode only) and BX_CPU_RESOLVE_ADDR_32 (for 32-bit mode only) 2015-05-11 19:23:09 +00:00
Stanislav Shwartsman
1e1c893041 introduce new 64bit packed register type and implement pat/mtrr and mmx registers through it 2015-02-23 21:17:33 +00:00
Stanislav Shwartsman
5e6955c5e7 Major rewrite of memory access methods to avoid massive code duplication and enable inlining of memory access methods 2015-01-25 20:55:10 +00:00
Stanislav Shwartsman
d8d4d2f0c1 Implemented VPSRLVW/VPSRAVW/VPSLLVW AVX512BW instructions 
The only missing AVX512BW/AVX512DQ opcodes are now:

"512.66.0F3A.W1 0F VPALIGNR"
"NDS.66.0F3A.W0 42 VDBPSADBW"

"NDS.512.66.0F3A.W0 50 VRANGEPS
 NDS.512.66.0F3A.W1 50 VRANGEPD"
"NDS.512.66.0F3A.W0 51 VRANGESS
 NDS.512.66.0F3A.W1 51 VRANGESD"

"NDS.512.66.0F3A.W0 56 VREDUCEPS
 NDS.512.66.0F3A.W1 56 VREDUCEPD"
"NDS.512.66.0F3A.W0 57 VREDUCESS
 NDS.512.66.0F3A.W1 57 VREDUCESD"
 2014-07-25 21:15:48 +00:00
Stanislav Shwartsman
79456eb7e1 Implemented VPCMP* AVX512 instructions 
Implemented VMOVNTPS/PD/DQ AVX512 instructions
Implemented VMOVNTDQA AVX512 instruction
Bugfixes for AVX-512
 2013-12-02 18:05:18 +00:00
Stanislav Shwartsman
940c2a1c8e fixes for disasm 2013-10-15 17:19:18 +00:00
Stanislav Shwartsman
ff79cbd596 Infrstructure change to support disasm of BxInstruction_c directly (without calling disasm) 
The end goal will be also merging of disasm and cpu decoder to one module and remove the disasm.

Two bug fixes on the way:
TBM: fixed 64-bit TBM instructions with memory access (did 32-bit load instead of 64-bit)
BMI2: fixed operands order for PEXT/PDEP instructions
AVX2: fixed gather instruction decoding bug from decoder alias commit
 2013-09-24 05:21:00 +00:00
Stanislav Shwartsman
df90b80352 set of small cpu fixes 2012-08-09 13:11:25 +00:00
Stanislav Shwartsman
cc694377b9 Standartization of Bochs instruction handlers. 
Bochs instruction emulation handlers won't refer to direct fields of instructions like MODRM.NNN or MODRM.RM anymore.
Use generic source/destination indications like SRC1, SRC2 and DST.
All handlers are modified to support new notation. In addition fetchDecode module was modified to assign sources to instructions properly.

Immediate benefits:
- Removal of several duplicated handlers (FMA3 duplicated with FMA4 is a trivial example)
- Simpler to understand fetch-decode code

Future benefits:
- Integration of disassembler into Bochs CPU module, ability to disasm bx_instruction_c instance (planned)

Huge patch. Almost all source files wre modified.
 2012-08-05 13:52:40 +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
87953711b1 cleanup in mmx code 2011-06-26 19:31:42 +00:00
Stanislav Shwartsman
2f582db722 compile less stuff for cpu-level=5 2011-06-26 19:15:30 +00:00
Stanislav Shwartsman
7d80a6ebe0 Adding Id and Rev property to all files 2011-02-24 21:54:04 +00:00
Stanislav Shwartsman
5915d92775 very small optimizations + indent 2011-01-25 20:59:26 +00:00
Stanislav Shwartsman
5917eb29ab sse + mmx optimizations 2011-01-16 21:01:28 +00:00
Stanislav Shwartsman
8c5c078b13 optimize sse and mmx code 2011-01-16 20:42:28 +00:00
Stanislav Shwartsman
c5aca5ac21 move function to inline 2011-01-08 19:50:22 +00:00
Stanislav Shwartsman
f705cbbc63 rename functions 2010-12-25 19:34:43 +00:00
Stanislav Shwartsman
1bd512e98d split more SSE ops, optimizations in MMX code 2010-12-25 17:04:36 +00:00
Stanislav Shwartsman
471d33fc1d fix BE issue 2010-09-26 20:20:27 +00:00
Stanislav Shwartsman
d36e6d78e7 typo fix 2010-03-02 06:49:41 +00:00
Stanislav Shwartsman
01cfbdccbc Move MMX to be runtime option 2010-03-01 18:53:53 +00:00
Stanislav Shwartsman
927c3594d6 enable compilation with CPU_LEVEL <= 6 
converted SEP to runtime option as well
 2010-02-26 11:44:50 +00:00
Stanislav Shwartsman
033a20b3b2 allow to configure CPU features at runtime - implemened on example of SSE/AES/MOVBE/POPCNT 2010-02-25 22:04:31 +00:00
Stanislav Shwartsman
70dc124b3a 1st step of moving CPU options to runtime 2010-02-24 19:27:51 +00:00
Stanislav Shwartsman
5f89b554aa split few more opcodes 2010-02-10 17:21:15 +00:00
Stanislav Shwartsman
edaf19f0a1 Split MOVQ_PqQq opcode 2009-12-14 11:55:42 +00:00
Stanislav Shwartsman
67e4f97e73 make maskmov fault order like in real HWQ 2009-11-13 09:55:22 +00:00
Stanislav Shwartsman
78e4b3d616 split SSE move instructions 2009-10-24 11:17:51 +00:00
Stanislav Shwartsman
7254ea36a1 copyright fixes + small optimization 2009-10-14 20:45:29 +00:00
Stanislav Shwartsman
f1eb1d00fd do not produce fpu2mmx transition if mem write faults 2009-02-13 10:15:16 +00:00
Stanislav Shwartsman
9929e6ed78 - updated FSF address 2009-01-16 18:18:59 +00:00
Stanislav Shwartsman
ab716f62aa inline prepareMMX method 2008-10-08 11:14:35 +00:00
Stanislav Shwartsman
489447ae57 Fixed FPU2MMX state transition - should be done only fater all memory faults already checked 2008-10-08 10:51:38 +00:00
Stanislav Shwartsman
5dd02b26e3 Make even more efficient RmAddr calculation - good optimizing compiler could make more efficient code than it was before 2008-08-08 09:22:49 +00:00
Stanislav Shwartsman
709d74728d Call #UD exception directly instead of UndefinedOpcode function - for future use 2008-07-13 15:35:10 +00:00
Stanislav Shwartsman
ec1ff39a5f Splitted memory access methods for 32 and 64-bit code. 
The 64-bit code got >10% speedup, the 32-bit code also got about 2% because laddr cacluation optimization
 2008-05-10 18:10:53 +00:00
Stanislav Shwartsman
b3167d1a8f Docs for MASKMOVQ were also not correct :( 2008-04-16 05:45:45 +00:00
Stanislav Shwartsman
4f3f8608f7 Fixed MASKMOVDQU instruction decoding 2008-04-16 05:41:43 +00:00
Stanislav Shwartsman
420f30816d inline integer saturation code - speedup for MMX/SSE integer 2008-04-06 13:56:22 +00:00
Stanislav Shwartsman
167c7075fb Use fastcall gcc attribute for all cpu execution functions - this pure "compiler helper" optimization brings additional 2% speedup to Bochs code 2008-03-22 21:29:41 +00:00
Stanislav Shwartsman
457152334e step2 in XSAVE implementation 2008-02-13 16:45:21 +00:00
Stanislav Shwartsman
a2897933a3 white space cleanup 2008-02-02 21:46:54 +00:00
Stanislav Shwartsman
37fbb82baa Cleanups. Move bxInstruction_c definition to separate file instr.h 2008-01-29 17:13:10 +00:00
Stanislav Shwartsman
d9984bb3a1 Eliminate BxResolve call from the heart of cpu loop and move into instructions that really require this calculation. Yes, it blows the code of EVERY CPU method but it has >15% speedup ! 2008-01-10 19:37:56 +00:00
Stanislav Shwartsman
79fc57dec8 Fixed more VCPP2008 warnings 2007-12-26 23:07:44 +00:00
Stanislav Shwartsman
838fb2a048 Fixing V2008 warnings - they found a bug in sse_pfp.cc ! 2007-12-23 17:21:28 +00:00