Commit Graph

44 Commits

Author SHA1 Message Date
Stanislav Shwartsman
362b99eba0 fixed typo in last commit 2013-08-04 19:47:19 +00:00
Stanislav Shwartsman
3fabcb00b7 VMX: CMPXHG instructions should always write to the memory destination, even if the value unchanged - it affects VMEXIT conditions for the full apic virtualization
Fixed also CMPXHG16B instruction (last one, others were fixed earlier)
2013-08-04 19:37:04 +00:00
Stanislav Shwartsman
2dbe81db51 first infrastructure changes to support EVEX prefix and AVX-512 extensions recently published by Intel 2013-07-26 12:50:56 +00:00
Stanislav Shwartsman
2357dc5ccc Fixed number of invocations of the BX_INSTR_LIN_ACCESS instrumentation callback in cpu/access32.cc, cpu/access64.cc and cpu/paging.cc specify the BX_READ memory access type where BX_RW really applies.
SF Patch #1335 by Mateusz Jurczyk
2013-07-24 18:54:18 +00:00
Stanislav Shwartsman
4bed791ccb Added year 2013 to Copyright in all files already modified in new year 2013-01-19 20:45:03 +00:00
Stanislav Shwartsman
eda28b95f4 unfortunately this change is rquired to make SMAP and SMEP features to work.
I observed ~5% emulation slowdown ... thinking about possible mitigations

this fixes TLB issue with SMAP and SMEP features.
these features introduce a new behavior when page can be inaccessible by System (CPL=0).
Current behavior is accessBits was not supporting it but legacy (from Bochs 2.3.6) was.
The wrong behavior can be observed if user access a user page and system access the same page later.
user access is fine and pass SMEP/SMA checks and stores the translation in TLB.
the system access will hit the TLB and nobody could detect that system cannot access that page.
2013-01-16 17:28:20 +00:00
Stanislav Shwartsman
182ad65ea3 changes in avx emulation code 2012-12-09 16:42:48 +00:00
Stanislav Shwartsman
c7c431f88e bx_instr_mem_data_access became completely obsolete with new stack optimization merged into SVN.
It already had limited usability before. With stack direct access optimization the callback won't be called for stack accesses as well.
See note by Brian Slechta:

=== Cut Hete ===
While using Bochs as a reference model for simulations, the simulator needs
information about what loads/stores are taking place with each instruction.
Presumably,  that  is  what  the BX_INSTR_MEM_DATA() instrumentation macros
cover (which is the place where our simulator hooks up).

The RETnear_xxx() functions call access_linear() directly, rather than call
read_virtual_xxx()  functions. This is a problem for code making use of the
BX_INSTR_MEM_DATA()   hook  because  it  does  not  get  called  for  these
instructions.  Should  this  be  changed along with some other instructions
that exhibit this?
=== Cut Hete ===

For Bryan's usage bx_instr_lin_access and bx_instr_phy_read/bx_instr_phy_write callbacks should be used.
2012-04-11 19:01:25 +00:00
Stanislav Shwartsman
279c61dc67 updated + fixed instrumentation example for instr histogram, code cleanup in the cpu 2012-03-28 21:11:19 +00:00
Stanislav Shwartsman
b5a33e82ac fixed a lot of code duplication in debugging/instrumentation of mem access 2012-03-20 18:26:04 +00:00
Stanislav Shwartsman
2b7894de7b fixed dbg print mentioned in SF bug 3029271 2011-09-22 22:08:18 +00:00
Stanislav Shwartsman
31be835056 bugfix + rename function 2011-06-14 19:56:28 +00:00
Volker Ruppert
c78026a9a2 - deleted executable properties from source files 2011-04-03 10:29:19 +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
fe0685c7f9 fine granular SMC detection (128b granularity used)
significant reduction (>80%) of false SMC flushes
2011-01-04 16:17:20 +00:00
Stanislav Shwartsman
d60b7c0919 rename accessor for opcodeReg() in instruction 2010-12-06 21:45:56 +00:00
Stanislav Shwartsman
cffe32dd2c remove unused param from exception() call 2010-03-14 15:51:27 +00:00
Stanislav Shwartsman
7254ea36a1 copyright fixes + small optimization 2009-10-14 20:45:29 +00:00
Stanislav Shwartsman
712e5a6f18 fixed priority between #AC and #GP on canonical access 2009-06-20 09:38:51 +00:00
Stanislav Shwartsman
4470c6a1c8 make ICACHE always enabled option and deprecate it in the configure script
Trace cache still can be turned off
2009-03-13 18:48:08 +00:00
Stanislav Shwartsman
b9de22961c minimize SSE tables, minor speedup in SSE code 2009-02-26 21:57:01 +00:00
Stanislav Shwartsman
9929e6ed78 - updated FSF address 2009-01-16 18:18:59 +00:00
Stanislav Shwartsman
a2e07ff971 - Removed --enable-guest2hos-tlb configure option. The option will be
always enabled for any Bochs configuration.
2008-12-11 21:19:38 +00:00
Stanislav Shwartsman
39d4db5564 - No alignment check required for 16-byte 'unaligned' access 2008-09-18 17:37:28 +00:00
Stanislav Shwartsman
db664c4012 more optimizations after fetchdecode 2008-09-16 20:57:16 +00:00
Stanislav Shwartsman
f5ba90da55 Misaligned check small optimization 2008-09-08 15:45:57 +00:00
Stanislav Shwartsman
398f80a214 - Fixed CALL FAR between legacy and 64-bit modes 2008-08-31 06:04:14 +00:00
Stanislav Shwartsman
aea946b4a3 One more change to speedup memory access through HostPtr check 2008-08-14 22:26:15 +00:00
Stanislav Shwartsman
3e03443405 Fixed speed bug when alignment access is disabled 2008-08-12 20:24:24 +00:00
Stanislav Shwartsman
e5cb22772d Fixed compilation error 2008-08-10 19:40:47 +00:00
Stanislav Shwartsman
bbf02a8bc5 More clean rewrite of the TLB access bits 2008-08-07 22:14:38 +00:00
Stanislav Shwartsman
4808a0d581 Fixed accessBits algebra 2008-08-04 05:30:37 +00:00
Stanislav Shwartsman
6398ebb1d4 First step of access bits cleanup and optimization - no perf gain yet 2008-08-03 19:53:09 +00:00
Stanislav Shwartsman
2e8bc558d1 Speedup SSE by introducing read/write_virtual_dqword_aligned methods 2008-08-02 10:16:47 +00:00
Stanislav Shwartsman
c388f48fff - Fixed memory bug in tripple fault detection
- Implement 16-byte memory accessor for SSEx - speedup SSE code emulation by >20%
2008-07-26 14:19:06 +00:00
Stanislav Shwartsman
8674659f9b - Fixed missed canonical fault when mem access split cross canonical boundary 2008-07-13 11:19:21 +00:00
Stanislav Shwartsman
380a4eaa56 Removed unused variables 2008-07-13 08:43:54 +00:00
Stanislav Shwartsman
c1f308d80d Push error code if segment violation occurs when pushing arguments into a new stack 2008-06-25 02:28:31 +00:00
Stanislav Shwartsman
92568f7525 Faster 32-bit emulation wwith 64-bit enabled mode.
~10% speedup byu optimization of 32-bit mem access
2008-06-12 19:14:40 +00:00
Stanislav Shwartsman
c3f96973ba Added debug prints 2008-05-12 19:19:03 +00:00
Stanislav Shwartsman
c8f870c9ac Remove incorrect assertion 2008-05-10 21:30:30 +00:00
Stanislav Shwartsman
d3528cccd6 Style fixes - name convention for push to new stack methods 2008-05-10 20:35:03 +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