|
|
@36196
|
14 years |
vboxsync |
VMM: don't die hard during PDPTR preloading (VT-x), try guru meditate …
|
|
|
@35346
|
14 years |
vboxsync |
VMM reorg: Moving the public include files from include/VBox to …
|
|
|
@35023
|
14 years |
vboxsync |
Introduce sane upper and lower limits for the preemption timer
|
|
|
@35005
|
14 years |
vboxsync |
Fixed massive performance regression for nested paging with 32-bit SMP …
|
|
|
@34958
|
14 years |
vboxsync |
Backed out 68777: previous code was safe
|
|
|
@34957
|
14 years |
vboxsync |
Recommit 68775
|
|
|
@34956
|
14 years |
vboxsync |
Compile problem; postpone
|
|
|
@34955
|
14 years |
vboxsync |
Can't load a minimal state if we were rescheduled to a new cpu (!!)
|
|
|
@33595
|
14 years |
vboxsync |
src/*: more spelling fixes (logging), thanks Timeless!
|
|
|
@32847
|
14 years |
vboxsync |
Clear VMCPU_FF_TO_R3 on the way out
|
|
|
@32525
|
14 years |
vboxsync |
Updated comments
|
|
|
@32512
|
14 years |
vboxsync |
Enabled minimal guest state loading (VT-x; again). Watch for regressions.
|
|
|
@32506
|
14 years |
vboxsync |
Extra stats
|
|
|
@32458
|
14 years |
vboxsync |
VMM: Return to ring-3 when DMA work is pending.
|
|
|
@32419
|
14 years |
vboxsync |
TM,HWVMXR0: Use the preemption timer to make sure we stop executing …
|
|
|
@31786
|
14 years |
vboxsync |
PGM: AMD-V + NP optimization (IOM sometimes makes use of uErr).
|
|
|
@31646
|
14 years |
vboxsync |
HWVMXR0.cpp: Made the PGM/MMIO hack work with EPT.
|
|
|
@31636
|
14 years |
vboxsync |
EM: Addressed VBOXSTRICTRC todo from r64673. Decided to only use …
|
|
|
@31569
|
14 years |
vboxsync |
HWACCM: burn fix / VBOXSTRICTRC cleanup - a bit risky…
|
|
|
@31565
|
14 years |
vboxsync |
PGM,HWACCM: MMIO optimizations - coding in progress.
|
|
|
@31402
|
14 years |
vboxsync |
PGM: Replaced the hazzardous raw-mode context dynamic mapping code …
|
|
|
@31394
|
14 years |
vboxsync |
Put back (disabled) r64154
|
|
|
@31197
|
14 years |
vboxsync |
Temporarily backed out r64154 as everyone is complaining about debug …
|
|
|
@31169
|
14 years |
vboxsync |
Only sync a minimal guest state when nothing has changed.
|
|
|
@31099
|
14 years |
vboxsync |
VMM: RT_UNLIKELY misusage (disabled code)
|
|
|
@31053
|
14 years |
vboxsync |
log fix
|
|
|
@30915
|
14 years |
vboxsync |
logging fix
|
|
|
@30768
|
14 years |
vboxsync |
VT-x: fix 32-bit EFI guests
|
|
|
@30590
|
14 years |
vboxsync |
HWVMXR0.cpp,HWSVMR0.cpp: Realigned and simplified the profiling.
|
|
|
@30390
|
14 years |
vboxsync |
VT-x: when the guest's FPU state is active, then we no longer care …
|
|
|
@30388
|
14 years |
vboxsync |
Statistic correction
|
|
|
@30375
|
14 years |
vboxsync |
Experimental code (disabled)
|
|
|
@30181
|
15 years |
vboxsync |
Allow 64-bit SMP guests on 32-bit hosts
|
|
|
@30180
|
15 years |
vboxsync |
Make 32->64 switcher SMP aware.
|
|
|
@30160
|
15 years |
vboxsync |
VMMGetStackRC change
|
|
|
@30105
|
15 years |
vboxsync |
More stats
|
|
|
@30091
|
15 years |
vboxsync |
If we're somehow not in VMX root mode, then we shouldn't dare leaving it.
|
|
|
@30089
|
15 years |
vboxsync |
Don't execute vmread after vmxon fails. (-> #UD)
|
|
|
@29737
|
15 years |
vboxsync |
Fixed dropping back to the recompiler too often when running …
|
|
|
@29250
|
15 years |
vboxsync |
iprt/asm*.h: split out asm-math.h, don't include asm-*.h from asm.h, …
|
|
|
@28800
|
15 years |
vboxsync |
Automated rebranding to Oracle copyright/license strings via filemuncher
|
|
|
@27976
|
15 years |
vboxsync |
*: scm cleans up whitespace and adds a new line at the end of …
|
|
|
@27445
|
15 years |
vboxsync |
Logging update
|
|
|
@27444
|
15 years |
vboxsync |
Resync DR6 if the debug state is active.
|
|
|
@27442
|
15 years |
vboxsync |
Paranoia
|
|
|
@27440
|
15 years |
vboxsync |
More logging
|
|
|
@27438
|
15 years |
vboxsync |
We must intercept X86_XCPT_DB for VT-x as DR6 is modified by X86_EFL_TF.
|
|
|
@27231
|
15 years |
vboxsync |
Implemented mwait extension for breaking on external interrupt when …
|
|
|
@26296
|
15 years |
vboxsync |
more warnings.
|
|
|
@26180
|
15 years |
vboxsync |
*: The rest of the %V* format specifiers are history.
|
|
|
@26152
|
15 years |
vboxsync |
VMM: pdm.h and @copydoc cleanups.
|
|
|
@26146
|
15 years |
vboxsync |
Make sure we fall back to the recompiler if the VMM device heap is …
|
|
|
@26066
|
15 years |
vboxsync |
Guest SMP: force all VCPUs to go back to ring 3 when a pgm pool flush …
|
|
|
@26062
|
15 years |
vboxsync |
Handle pending pgm sync operations in the main execution loop (vt-x/amd-v)
|
|
|
@26025
|
15 years |
vboxsync |
build fix.
|
|
|
@26018
|
15 years |
vboxsync |
Missing pending pgm sync handling in VT-x and AMD-V page fault …
|
|
|
@26014
|
15 years |
vboxsync |
HWVMXR0.cpp: -Wshadow.
|
|
|
@25837
|
15 years |
vboxsync |
VMM: More micro optimizations around CPUM getters and predicates.
|
|
|
@25553
|
15 years |
vboxsync |
Inverted check (unrestricted guest commit regression)
|
|
|
@25397
|
15 years |
vboxsync |
Backed out 56022; not the real problem
|
|
|
@25396
|
15 years |
vboxsync |
Must set uCurrentASID or else invvpid might fail
|
|
|
@25392
|
15 years |
vboxsync |
Unrestricted guest execution update
|
|
|
@25389
|
15 years |
vboxsync |
More unrestricted guest execution changes
|
|
|
@25386
|
15 years |
vboxsync |
Initial commit for unrestricted guest execution support (VT-x)
|
|
|
@24829
|
15 years |
vboxsync |
Removed excessive logging
|
|
|
@24809
|
15 years |
vboxsync |
HWVMXR0.cpp: fixed logging.
|
|
|
@24731
|
15 years |
vboxsync |
Logging changes
|
|
|
@24284
|
15 years |
vboxsync |
Compile fix
|
|
|
@24283
|
15 years |
vboxsync |
Cleaned up
|
|
|
@24251
|
15 years |
vboxsync |
Another tpr patch fix
|
|
|
@24249
|
15 years |
vboxsync |
TPR patch fixes
|
|
|
@24248
|
15 years |
vboxsync |
VT-x: Must save the host's LSTAR msr in case the guest changed it (TPR …
|
|
|
@24243
|
15 years |
vboxsync |
TPR patching for VT-x without VMX_VMCS_CTRL_PROC_EXEC2_VIRT_APIC. Untested.
|
|
|
@24216
|
15 years |
vboxsync |
VMM: Backed out r54155 & r54154 due to XP breakage.
|
|
|
@24207
|
15 years |
vboxsync |
Check if we need to intercept invlpg; we catch all page table updates, …
|
|
|
@24029
|
15 years |
vboxsync |
Compile fix
|
|
|
@24027
|
15 years |
vboxsync |
Fail if VT-x or AMD-V are already enabled (Windows 7)
|
|
|
@23983
|
15 years |
vboxsync |
Must flush all contexts when restarting with asid 1
|
|
|
@23977
|
15 years |
vboxsync |
Cleaned up
|
|
|
@23976
|
15 years |
vboxsync |
Flush page instead of context (code path not used)
|
|
|
@23944
|
15 years |
vboxsync |
Format specifier fix
|
|
|
@23943
|
15 years |
vboxsync |
More verbose assertion
|
|
|
@23927
|
15 years |
vboxsync |
netflt/win: alternative loopback handling mechanism (disabled for now)
|
|
|
@23641
|
15 years |
vboxsync |
Wrong else
|
|
|
@23517
|
15 years |
vboxsync |
Noisy logging
|
|
|
@23366
|
15 years |
vboxsync |
Wait for the target VCPU to finish its world switch.
|
|
|
@23220
|
15 years |
vboxsync |
Properly emulate task switches caused by IDT gates (VT-x & AMD-V)
|
|
|
@23058
|
15 years |
vboxsync |
Noisy logging
|
|
|
@23034
|
15 years |
vboxsync |
Always clear IF in VMXR0Execute64BitsHandler
|
|
|
@23033
|
15 years |
vboxsync |
Reversed ifdef corrected (not harmful though)
|
|
|
@22890
|
15 years |
vboxsync |
VM::cCPUs -> VM::cCpus so it matches all the other cCpus and aCpus members.
|
|
|
@22493
|
15 years |
vboxsync |
VMM,DevPCI,VBox/types.h: Added a VBOXSTRICTRC type for indicating …
|
|
|
@22429
|
15 years |
vboxsync |
VMM,x86.h/mac: Be more careful with X86DESCHC.
|
|
|
@22257
|
15 years |
vboxsync |
Guestimate too high; was still seeing decreasing rdtsc values
|
|
|
@22243
|
15 years |
vboxsync |
Apply the same tsc underflow check to AMD-V.
|
|
|
@22242
|
15 years |
vboxsync |
Made TSC underflow checking more generic.
|
|
|
@22241
|
15 years |
vboxsync |
Attempt to get rid of decreasing rdtsc return values.
|
|
|
@22079
|
15 years |
vboxsync |
Back to manual msr save and restore.
|
|
|
@22063
|
15 years |
vboxsync |
Removed unnecessary check
|
|
|
@22060
|
15 years |
vboxsync |
More cleanup
|
|
|