﻿ticket,summary,version,host,guest,created,modified,_changetime,_description,_reporter
5768,[feature-request] Allow VM to run faster or slower than normal time,VirtualBox 3.1.0,other,other,2009-12-16T16:04:15Z,2010-02-03T12:04:16Z,2010-02-03T12:04:16Z,"bochs and DOSbox can speed up or slow down VM by emulating CPU and/or timer slower or faster than normal. I would like to have a similar feature here.

In virtualization world, a possible way to achieve this, is to increase/decrease timer interrupt injections, so VM can either run at 50% speed or 200% speed.

In this case clock time sync in Guest Additions must be turned off.

It is (sometimes) useful for gaming.

-Technologov",Technologov
21221,[regression][7.0.2?] Windows XP guest has suddenly stopped working and causes VBox Internal Error,VirtualBox-7.0.2,Linux,Windows,2022-10-30T16:34:06Z,2022-10-31T15:15:10Z,2022-10-31T15:15:10Z,"All of a sudden my Windows XP guest which I've used for almost a decade without any modifications (I always revert to a snapshot I created back then) has stopped working.

It did work with VirtualBox 7.0.
It probably even worked with VirtualBox 7.0.2 with Linux kernel 5.19.16.
I cannot be sure. I will attach a log file.

I can even send this guest in its entirety to VirtualBox developers if necessary. There's nothing private/personal about it.",Artem S. Tashkinov
21241,Almalinux 8.6 aborts (virt-what),VirtualBox-7.0.2,Windows,other,2022-11-06T07:57:59Z,2022-11-08T14:43:45Z,2022-11-08T14:43:45Z,"After installing AlmaLinux 8.6 (RHEL 64-bit), normal start fails.
When using systemd.unit=rescue.target, I can get to a shell and when executing virt-what, the VM is aborted. Works on VirtualBox 6.1.40.",mtheiss
21353,apic timer does not trigger in time,VirtualBox 6.1.40,Windows,Linux,2022-12-25T01:27:42Z,2022-12-25T01:27:42Z,2022-12-25T01:27:42Z,"Rencently,I found a bug of APIC timer.Since the vbox set the last VCPU thread as the thread to wakeup all the timer and other VCPU threads will sleep 500ms until the Last VCPU thread wakeup them.
If the last VCPU thread sleep a long time firstly,other VCPU thread's apic timer will not able to trigger in timer.This make other VCPU thread sleep at least 500ms.

here is may patch to the funtion apicSetTimerIcr in VMM/VMMALL/APICAll.cpp to fix the bug

{{{
#include ""TMInternal.h""
static VBOXSTRICTRC apicSetTimerIcr(PPDMDEVINS pDevIns, PVMCPUCC pVCpu, int rcBusy, uint32_t uInitialCount)
{
    VMCPU_ASSERT_EMT(pVCpu);

    PAPIC      pApic      = VM_TO_APIC(pVCpu->CTX_SUFF(pVM));
    PAPICCPU   pApicCpu   = VMCPU_TO_APICCPU(pVCpu);
    PXAPICPAGE pXApicPage = VMCPU_TO_XAPICPAGE(pVCpu);

    Log2((""APIC%u: apicSetTimerIcr: uInitialCount=%#RX32\n"", pVCpu->idCpu, uInitialCount));
    STAM_COUNTER_INC(&pApicCpu->StatTimerIcrWrite);

    /* In TSC-deadline mode, timer ICR writes are ignored, see Intel spec. 10.5.4.1 ""TSC-Deadline Mode"". */
    if (   pApic->fSupportsTscDeadline
        && pXApicPage->lvt_timer.u.u2TimerMode == XAPIC_TIMER_MODE_TSC_DEADLINE)
        return VINF_SUCCESS;

    /*
     * The timer CCR may be modified by apicR3TimerCallback() in parallel,
     * so obtain the lock -before- updating it here to be consistent with the
     * timer ICR. We rely on CCR being consistent in apicGetTimerCcr().
     */
    TMTIMERHANDLE hTimer = pApicCpu->hTimer;
    VBOXSTRICTRC rc = PDMDevHlpTimerLockClock(pDevIns, hTimer, rcBusy);
    if (rc == VINF_SUCCESS)
    {
        pXApicPage->timer_icr.u32InitialCount = uInitialCount;
        pXApicPage->timer_ccr.u32CurrentCount = uInitialCount;
        if (uInitialCount)
            apicStartTimer(pVCpu, uInitialCount);
        else
            apicStopTimer(pVCpu);


        PVMCC pVM   = pVCpu->CTX_SUFF(pVM);
        PVMCPUCC pVCpuDst = VMCC_GET_CPU(pVM, pVM->tm.s.idTimerCpu);

        #ifdef IN_RING3
            VMR3NotifyCpuFFU(pVCpuDst->pUVCpu, VMNOTIFYFF_FLAGS_DONE_REM | VMNOTIFYFF_FLAGS_POKE);
        #elif defined(IN_RING0)

           if (VMMGetCpu(pVM) != pVCpuDst)
            {
                switch (VMCPU_GET_STATE(pVCpuDst))
                {
                    case VMCPUSTATE_STARTED_EXEC:
                        GVMMR0SchedPokeNoGVMNoLock(pVM, pVM->tm.s.idTimerCpu);
                        break;

                    case VMCPUSTATE_STARTED_HALTED:
                        GVMMR0SchedWakeUpNoGVMNoLock(pVM, pVM->tm.s.idTimerCpu);
                        break;

                    default:
                        break; /* nothing to do in other states. */
                }
            }
        #endif

        PDMDevHlpTimerUnlockClock(pDevIns, hTimer);
    }
    return rc;
}
}}}
",liowmark
19641,Assigning more vCPU (within the physical core limits) causes random freezes on guests,VirtualBox 6.1.8,Linux,Linux,2020-06-07T04:04:45Z,2021-05-19T05:19:07Z,2021-05-19T05:19:07Z,"'''Host Info'''
-------------------
Host: Fedora 32 (64 bit)[[BR]]

CPU: Intel® Core™ i9-9880H CPU @ 2.30GHz × 8 Cores[[BR]]

GPU: Intel UHD-630[[BR]]

HyperThreading: Disabled[[BR]]

Physical Cores: 8[[BR]]

VirtualBox Version: 6.1.8[[BR]]

Display Protocol: X11[[BR]]



'''Guest Info'''
------------------------------
Guest: Debian 10 (Kernel 4.19.0-9-amd64) (Also reproducible in Kubuntu 20.04)[[BR]]

Display Adapter: VMSVGA[[BR]]

3D Acceleration: Enabled[[BR]]

VBox Guest Additions: Installed[[BR]]


----

When I assign 4 cores to the guest, on random starts the guest's animation and scrolling lags. When I checked the Session Information it showed Guest Load about ~0-1% and VMM Load about ~25%. The next restart of the guest doesn't have this behavior. Guest is smooth and doesn't cause pauses.[[BR]]


I tried changing few settings to find the root cause of the issue and surprisingly reducing the CPU count to 1 or 2 fixes the issue. The issue was not reproducible at all.
",LienCaspari
19279,BSOD (DRIVER_OVERRAN_STACK_BUFFER ) in guest when trying to run nested virtual machine,VirtualBox 6.1.2,Windows,Windows,2020-01-29T18:33:43Z,2020-01-29T18:41:17Z,2020-01-29T18:41:17Z,"I am unable to make any 64 bit virtual machine to work in guest using nested virtualization provided by Virtualbox. I am using Vbox version 6.1.2 in both Host and Guest (guest have the vbox additions installed). As soon as I try to run any 64 bit VM in guest the guest OS crashes with BSOD (but Host is stable). I am using latest Windows 10 64 bit 1909 in both host and guest. Host is running AMD Ryzen 7 3700X CPU in ASUS x570 TUF Gaming PLUS motherboard.
I have attached logs of the Guest machine and the kernel memory dump from the guest OS that crashed.
The BSOD from guest is related to VMMR0.r0 driver with memory_corruption message. The bug check code is DRIVER_OVERRAN_STACK_BUFFER ",proboszcz
20114,Crash on guest VM,VirtualBox 6.1.14,Linux,Windows,2021-01-01T13:31:41Z,2021-01-01T13:40:31Z,2021-01-01T13:40:31Z,"I have a crash VM[[BR]]
Please solve bug ![[BR]]
I have installed microsoft net framework 4.0.[[BR]]
Guest : Windows XP[[BR]]


Thank
",Sizy458
3821,Estimate required and available disk space when discarding snapshots,VirtualBox 2.2.0,Linux,Windows,2009-04-23T12:21:10Z,2009-04-23T12:21:10Z,2009-04-23T12:21:10Z,"When discarding snapshots in the Virtual Media Manager, it would be good to estimate the required and available disk space before the operation starts and refuse it if there is not enough disk space.

Today I have deleted 40Gig .vdi file in this way. The Manager said that there is not enough disk space and removed the file. This was a key file, as the VirtualBox then has refused to starg Windows.

Happily, I have had a backup copy of that file.",Vladimir
19815,Extremely low performance and high CPU consumption on Ryzen 3600X,VirtualBox 6.1.10,Windows,Windows,2020-08-15T21:29:08Z,2020-08-15T21:29:08Z,2020-08-15T21:29:08Z,"I started having this problem after migrating my VMs to the new hardware - i.e. Ryzen 3600X and Asus B550 motherboard. Simply booting a VM takes about 6 minutes with 30-40% CPU utilization, and running a VM in idle mode consumes about 10-20% CPU.
Host and guest - Windows 10 both.
See the logs.",fandrei
19575,Failed to open a session for virtual machine,VirtualBox 6.1.6,other,Linux,2020-05-09T10:11:04Z,2020-05-09T10:11:04Z,2020-05-09T10:11:04Z,"I'm getting this error message:
""
''Failed to open a session for the virtual machine Ubuntu.

The virtual machine 'Ubuntu' has terminated unexpectedly during startup with exit code 1 (0x1).  More details may be available in 'C:\Machines\Ubuntu\Logs\VBoxHardening.log'.

Result Code: E_FAIL (0x80004005)
Component: MachineWrap
Interface: IMachine {5047460a-265d-4538-b23e-ddba5fb84976}''
""
I've tried everything. I tried downloading other versions of VB (6.1 and 5.2). I don't have an anti-virus. I also tried both graphics cards (built-in and dedicated) and i still get the same message.
I have attached the VBoxHardening.log (zipped)",asabbagh4
18602,Failure Restoring machine,VirtualBox 6.0.4,Linux,Windows,2019-04-26T15:20:30Z,2019-04-26T15:20:30Z,2019-04-26T15:20:30Z,"On a virtual machine when i restore[[BR]]
I have error ""Session avorted""[[BR]]


[[BR]]

AND
[[BR]]
i have a other error in log :
''00:00:01.865624 !!Assertion Failed!!
00:00:01.865624 Expression: !VMCPU_FF_IS_SET(pVCpu, VMCPU_FF_INTERRUPT_PIC)
00:00:01.865625 Location  : /home/vbox/tinderbox/6.0-lnx64-rel/src/VBox/VMM/VMMR3/PDM.cpp(1006) int pdmR3LoadExec(VM*, SSMHANDLE*, uint32_t, uint32_t)
00:00:01.865658 Stack     :
00:00:01.865658 00007fd31c0334e3 VBoxRT.so + 0x1f04e3''
",Sizy458
1397,Feature request: Overcommitting of memory (swap in/out),VirtualBox 1.5.6,Windows,other,2008-04-08T13:32:45Z,2016-08-22T16:35:36Z,2016-08-22T16:35:36Z,"I always get HostMemoryLow e.g. Knoppix 5.2. VMware and Microsoft Virtual PC work on this PC (IBM Thinkpad T40, 512 Mb RAM). I have enclosed the Logfile. How can I help?",IuK
2107,Feature request: Predefined absolute time for Guest,VirtualBox 2.0.0,Windows,Windows,2008-09-05T22:03:12Z,2009-02-10T23:08:26Z,2009-02-10T23:08:26Z,Please add an option to set Guest time to a fixed value at startup. This is a feature for repeatable software tests.,A20_user
19887,frequently the app(VM VirtualBox) does not respond,VirtualBox 6.1.10,Windows,Linux,2020-09-13T04:35:10Z,2020-09-16T12:59:25Z,2020-09-16T12:59:25Z,"Whenever I open the app, it hangs for sure and also when I open any virtual OS it does the same kind of thing.
I am doubtful that it is causing my system to slow down
Help me asap",jishu
19423,full crash of host machine,VirtualBox 6.1.4,Linux,Linux,2020-03-23T02:23:17Z,2020-03-23T02:23:17Z,2020-03-23T02:23:17Z,"This report is going to be very incomplete for now.  I'll update it, if I can get some instruction on what and how to collect some better detail.  I am getting this crash half a dozen times a day when working in the environment

Environment:
Host Fedora 31 workstations 5.5.10-200.fc31.x86_64
dual monitors
16GB memory, 4 cores
Guest Fedora 31 workstations 5.5.10-200.fc31.x86_64
single monitor
6GB memory, 2 cores
Large window, slightly extending past the physical monitor size
scaled mode
Running guest application: qtcreator, debugging an application

The symptoms, just before the hard (black screen) crash of the host:

Attempt some operation in qtcreator, like clicking to create a new breakpoint.  The application freezes, the VM freezes, and all other applications windows except shells freeze.  Freeze might not be quite right.  Clicking on an application window on the host (2nd monitor) that is occluding a shell window activates the shell not the application window.  So the application might already be dead, and just a ghost of the window left on the desktop.  The shell window seems active, but not much time to look around.  In a few seconds, the host goes into full black screen crash, with a blinking text cursor.  Sometimes, in a few more seconds, the host login windows appears without any additional user actions.  Sometimes, ctrl+alt+del will initiate a reboot.

So, is there anything I can enable ahead of time, to get a log that might be useful for tracking this down?  What data can I collect after host reboot to include here?

I have other fc31 guest vms that are not crashing, so there is some interaction with the running applications.  But a misbehaving application should not crash the VM, and definitely should not crash the host machine.",mMerlin
2652,Gentoo hardened (grsecurity) hosts not supported,VirtualBox 4.2.2,Linux,Linux,2008-11-25T12:59:06Z,2012-10-22T08:45:10Z,2012-10-22T08:45:10Z,"Hello,


I am running VirtualBox 2.0.6 on a Gentoo Hardened Host (2.6.23-r7).

Running normal linux systems as guest works fine, however when I try to boot a Gentoo Hardened _Guest_ system, VirtualBox reports a Guru Mediation Error and the VM freezes on kernel bootup. I tried with various hardened Guest kernels (2.6.23-r7, 2.6.24-r3 and 2.6.25-r9, all hardened-sources from Gentoo) but the error remains. Grsecurity/PAX presetting in the kernel is set to ""high"". The produced log is attached.


I will try to narrow down the search on why the kernel crashes by disabling some of the hardened options but so far I didn't have much success.

Best regards,


Christian Holler",Christian Holler
17772,Guru meditation 1155 on loading driver in guest,VirtualBox 5.2.12,Windows,Windows,2018-05-17T09:35:19Z,2018-07-12T11:45:41Z,2018-07-12T11:45:41Z,"So we trying to load driver inside Windows 7 (6.1.7601.24000) guest machine on Windows 7 host machine with VirtualBox:

{{{
sc create ltest type= kernel start= demand error= ignore binpath= ""<full path to sample>"" DisplayName= ltest
sc start ltest
}}}
and we catch ""Guru Meditation 1155 (VINF_EM_TRIPLE_FAULT)""

Actually the driver is windows kernel 6.1.7601.24000 (win7sp1_ldr.171231-1547), so it's like trying load second kernel to system. Yes, I known it's crazy, but anyway it's not ok catch guru meditation on this crazy actions.
",gim
21027,Guru Meditation on installation,VirtualBox 6.1.36,Linux,all,2022-07-21T09:50:16Z,2022-07-21T09:50:16Z,2022-07-21T09:50:16Z,"Hi,

i tried version 6.0, 6.1 from repository and the latest testing 152223 Version. With each i get a error, when i try to install Windows 10 or Kubuntu. See both logs.

Can someone give me a hint, where to search?
",murrcoss
477,Hardware breakpoints don't work in Virtualbox (software virtualization only),VirtualBox 3.1.6,other,other,2007-07-02T14:32:40Z,2020-03-04T10:11:20Z,2020-03-04T10:11:20Z,"It seems that doesn't work setting hardware breakpoints in Virtualbox (Olly Debugger).
[Hardware, on Execution]
WinXP guest",marius_mtm
20873,HDAT2 crashes under SysLinux 6.04 in VirtualBox 6.1.32 r149290 (Qt5.15.2),VirtualBox 6.1.32,Linux,other,2022-03-17T15:59:04Z,2022-10-22T11:20:35Z,2022-10-22T11:20:35Z,"Steps to reproduce:

* Create a HDD with the Syslinux 6.04 bootloader and add HDAT2 (HDAT2IMG_74.IMG from www.hdat2.com) to the menu
* Start it.

Result: VBox crashes.

NB: This crash does not happen when using the floppy image directly mounted as a floppy",Artem S. Tashkinov
20525,High VMM Load on Intel CPUs introduced in 6.1.20,VirtualBox 6.1.20,Windows,Windows,2021-08-26T18:15:06Z,2022-07-27T19:24:30Z,2022-07-27T19:24:30Z,"Beginning in version 6.1.20, and also happening in .22, and .24, there is a significant increased VMM Load reported in the Session Information dialog with hosts running Intel processors when using 2 or more vCPUs, getting progressively worse as the vCPU count is increased up to the recommended threshold shown in the VM Settings dialog.

I reverted to 6.1.18 and confirmed the issue is not present in this version, with low to minimal VMM Load.

I tested and am not seeing the issue on a Ryzen 5 3400G host at this time. I have performed the same test with multiple vCPUs with 6.1.20, and 6.1.24 and the VMM Load is low in all these cases on a Ryzen host.

All hosts tested are running Windows 10 x64. Tested and confirmed issue with Windows 10 x64 and Debian x86-64 guests on Intel 3rd gen Core i5, Core i7 5820K, Xeon E5-1620 v2, and Xeon E5-2620 v4 CPUs.",lbgaus
20526,High VMM load on VMs with more than 1 core,VirtualBox 6.1.26,Windows,all,2021-08-28T02:27:31Z,2021-08-28T02:27:31Z,2021-08-28T02:27:31Z,"The VMM load is stuck about 60% when running VMs assigned with more than 1 core although the guest is totally idle.
I'm using 6.1.x revision 146512 but the problem persist on all the recent versions.
I have enabled ioapic,apic,x2apic but no use.
Note that the VM process does not use much cpu actually, but it is a bit more lag than usual.

image
[[Image(https://i.ibb.co/L0jFyJM/image.png)]]",ebk
18193,"Increase vCore limit, 32 Cores are not enough",VirtualBox 5.2.22,other,other,2018-12-17T14:34:25Z,2018-12-19T09:54:06Z,2018-12-19T09:54:06Z,"Increase the CPU Limit from 32 processors to actually available cores.

The current 32 Cores Limit does not account for modern processors in combination with multisocket systems.",multicoreuser
21222,Issue running virtual machines with VT-X enabled,VirtualBox-7.0.2,Mac OS X,Linux,2022-10-30T17:48:26Z,2022-11-24T15:14:44Z,2022-11-24T15:14:44Z,"Hi,

I am running macOS Ventura version 13.0, existing Ubuntu 22.04 virtual machines with nested VT-X enabled (and KVM installed), are failing to start.

Result code: VBoxManage: error: Details: code NS_ERROR_FAILURE (0x80004005), component SessionMachine, interface ISession.

",ecer
20223,Linux/64bit host XMM register usage in kernel randomly crashes Linux guest,VirtualBox 6.1.16,Linux,Linux,2021-02-24T18:57:46Z,2021-02-24T18:57:46Z,2021-02-24T18:57:46Z,"'''Synopsis:''' src/VBox/VMM/Makefile.kmk 'knows' flag VBOX_WITH_KERNEL_USING_XMM, which is added for Windows/64bit builds only.
Problem is that Linux kernel DOES MAKE USE of XMM registers, too. Examples include aes-ni kernel module (AES-NI instructions implicitly use XMM0..15) or RAID, respectively).
VMM at ring 0 only implements XMM wrappers for Windows/64bit and does not have conditionals to abort build if VBOX_WITH_KERNEL_USING_XMM is set in src/VBox/VMM/Makefile.kmk for linux.amd64. This yields a kernel panic because Microsoft ABI is completely different from System V ABI (aka gcc ABI).
The 6.1.18 trunk seems to rely on a complete rewrite of src/VBox/VMM/VMMR0/HMR0A.asm but 6.1.13 / 6.1.16 (6.0 also) still have this problem.

'''Solution is:'''

a) Add VBOX_WITH_KERNEL_USING_XMM to src/VBox/VMM/Makefile.kmk for linux.amd64


b) Provide conditional implementation in src/VBox/VMM/VMMR0/HMR0A.asm for gcc/amd64 expecting function parameters in System V ABI registers rdi, rsi, rdx, rcx, r8, and r9. The two wrappers hmR0VMXStartVMWrapXMM and hmR0SVMRunWrapXMM need this.


c) calling CPUMIsGuestFPUStateActive from yasm assembler to C++ requires suffix ""wrt ..plt""

I have patches for this but optimized only for ""my"" Intel CPU. VBox users either have to switch to 6.1.18 or apply the above mentioned modifications.

'''Recommendation:'''

VBOX maintainers should think about making VBOX_WITH_KERNEL_USING_XMM the default for Linux, too (because of possible AES-NI or RAID usage, respectively).",vboxcomapijunkie
5428,Mac OS X: Deleting Virtual Hard Disks Should Move to Trash,,Mac OS X,other,2009-11-09T03:43:29Z,2009-11-09T03:43:29Z,2009-11-09T03:43:29Z,"It would be nice if, when removing a virtual hard disk from the Virtual Media Manager, deleting the hard disk would move it to the Trash under Mac OS X. This would, among other things, allow the operating system to securely erase the disks if the user has opted to do that. More information may be found in [http://developer.apple.com/mac/library/documentation/Cocoa/Conceptual/LowLevelFileMgmt/Articles/FileManagement.html Apple's File Management documentation]. If this is implemented, the dialog box should change the word ""Delete"" to ""Move to Trash"" to indicate to the user what is occurring.",Jeff Kelley
18262,Not enough video memory to restart gnome after loading from save state,VirtualBox 6.0.0,Mac OS X,Linux,2019-01-02T12:10:52Z,2019-01-22T13:22:51Z,2019-01-22T13:22:51Z,"After saving state, upon reloading a black screen is shown and VM is totally unresponsive.

This is fixed by increasing the default video memory from 16mb to something like 48mb.

Took me a long time to figure out - maybe the default should be larger.

Host: VBox 6.0 MacOS Mojave 10.14.2
Guest: Ubuntu Bionic Beaver 18.04.1",0x7766
19078,"NT4 Guest: Weirdness with ""Enable Nested VT-x/AMD-V"" checkbox in guest settings",VirtualBox 6.0.14,Windows,Windows,2019-11-11T06:26:29Z,2019-11-11T06:26:29Z,2019-11-11T06:26:29Z,"I have a good working install of Windows NT4 Server that I toy around with every few months.  I am not sure when this issue began, but I am assuming during 6.x.

From a known good snapshot, I can boot into NT4 successfully and login as the administrator account.  However, when I reboot the guest, the *next* reboot will fail with this error message:

    HAL: MPS MP Structure not found[[BR]]
    HAL: No MPS Table Found.[[BR]]
    
    HAL: This HAL.DLL requires an MPS version 1.1 system[[BR]]
    Replace HAL.DLL with the correct hal for this system[[BR]]
    The system is halting

The solution that I found is to go into the guest's settings, and under System --> Processor, enable OR disable the ""Enable Nested VT-x/AMD-V"" checkbox.  That means, if the checkbox is currently checked, uncheck it.  If it is not currently checked, check it.  Save and then exit the settings dialog.

After this, the next boot will work fine.  But!  The next time you restart the guest, you must toggle the state of that checkbox again before the guest will successfully boot.

Definitely one of the weirder bugs I've ran into.  I'll attach a screenshot of the HAL message, but I'm not sure what else the VBox developers might need to figure this one out.",kumba
20005,POPCNT cpu flag not supported in 6.1 Ubuntu 16 guest,VirtualBox 6.1.16,other,other,2020-10-27T22:37:42Z,2021-12-22T10:37:35Z,2021-12-22T10:37:35Z,Ubuntu 16 guest with Enable Nested Paging enabled.  On VirtualBox 6.0 guest Ubuntu 16.04 reports having the POPCNT CPU flag.  On 6.1 POPCNT is no longer there (causing my OpenCV app to fail).  I downgraded back to 6.0 and it runs again.  If this is in the release notes I didn't find it.,eben
16801,Running a Virtual Machine when Windows Hyper-V is enabled should NOT Crash Windows,VirtualBox 5.1.22,Windows,all,2017-05-31T03:26:52Z,2020-06-04T16:28:57Z,2020-06-04T16:28:57Z,"VirtualBox can not run on Windows 10 when Hyper-V mode is ACTIVE (ON).  However the existing behaviour is pathological and must be corrected.  

Running a VM when Hyper-V is ON causes the BSOD!



Conditions:

 1. Windows boots with Hyper-V enabled
 2. Start-up VirtualBox Manager GUI
 3. START an existing VM

Outcome:

 1. VM begins to load
 2. The dread Windows BSOD (Blue Screen of Death) appears
 3. Windows dies -- generates crash log, etc.
 4. Windows reboots
 5. NOTE: Hyper-V is still active and must be deactivated manually.


Resolution Sought:

 1. Check for Hyper-V mode is: OFF, at start-up for all VirtualBox tools; especially:
     * VirtualBox Manager GUI and 
     * VM shell  itself

 2. Issue a WARNING pop-up or error message for command line
    Reporting that Hyper-V mode is active and VirtualBox cannot
    run while there is another hypervisor active.  Or some such.

 3. Exit the VirtualMachine (shell) once the Hyper-V Active 
    status is reported.  
      * Thus Avoiding the BSOD


Optional / suggestion:

 * It would be useful for the warning to tell users how to 
   disable/enable Hyper-V as part of the warning message(s).

 * Another helpful feature is for the VirtualBox management 
   GUI to turn Hyper-V off or on from a menu.


Related / supplementary notes:

There seem to be several issues that mention Hyper-V troubles.  Some are real bugs and some look like misunderstandings.

In my opinion no software should cause a system crash-dump when it can be prevented by checking some system configuration attributes.

See also:

 * https://www.virtualbox.org/ticket/16675
 * https://www.virtualbox.org/ticket/15780


",Will
10754,Running xperf -on latency -stackwalk profile doesn't capture CPU Sampling data in Windows 8,VirtualBox 4.1.18,Windows,Windows,2012-07-13T07:00:04Z,2019-04-09T15:49:32Z,2019-04-09T15:49:32Z,"When I run xperf -on latency -stackwalk profile in Windows 8 it doesn't capture CPU Sampling data (http://msdn.microsoft.com/en-us/library/ff191014.aspx, http://blogs.msdn.com/b/pigscanfly/archive/2009/08/06/stack-walking-in-xperf.aspx )

I've asked Microsoft and they told my that VirtualBox reports CPU h/w counters but then in fact not supporting them.",Andre.Ziegler
16127,Setting up virtual LPT port to connect through to real LPT port stops DOS VM starting,VirtualBox 5.1.8,Windows,other,2016-10-29T23:19:37Z,2018-10-28T19:59:43Z,2018-10-28T19:59:43Z,"I run VirtualBox on a Windows 7 Ultimate 64 bit host.
I find that with VirtualBox 5.1.4, 5.1.6 and 5.1.8 and 5.1.9-111573 (with matching Extension Pack installed) my FreeDOS 1.1 VM will not start at all after issuing the following VBoxManage commands:
VBoxManage modifyvm ""FreeDOS 1.1"" --lptmode1 LPT1
VBoxManage modifyvm ""FreeDOS 1.1"" --lpt1 0x378 7
to create a virtual LPT port and connnect it through to my real LPT1 port (on a PCI card, at address I/O address C000)
I am reverting to VirtualBox 5.1.2 for now.
I am attaching the log files and configuration of the VM in a ZIP file.
The oldest log file ""VBox.log.3"" is one from when it last worked, the others are fails.
",Brian Gregory
15405,Severe time drift with KVM paravirt and Linux guests,VirtualBox 5.0.20,Mac OS X,Linux,2016-05-11T09:32:49Z,2019-01-05T10:42:46Z,2019-01-05T10:42:46Z,"I see very strong time drift with the kvm paravirt interface, on an OS X host - reproducible with both ""debian/jessie64"" and ""centos/7"". I let it run overnight for a more precise measurement. The guest time drifts very fast:
* 60 host seconds -> 8 guest seconds
* 8h37m host -> 0h53m38s guest

Raw measurements (host time as comment - OS X is set to the Europe/Berlin timezone):
{{{
[vagrant@localhost ~]$ date # 0:59:00
Tue May 10 18:56:18 EDT 2016
[vagrant@localhost ~]$ date # 1:00:00
Tue May 10 18:56:26 EDT 2016
[vagrant@localhost ~]$ date # 9:37:00
Tue May 10 19:50:04 EDT 2016
}}}

I found two workarounds to get a stable time:
1. set the paravirt interface to legacy, instead of kvm
2. use `kvm` as the paravirt interface, but set the guest kernel `clocksource` to `acpi_pm` (`hpet` also works, if i enable it - `tsc` too, i think).

Both the Debian Jessie and CentOS kernels default to `kvm-clock` as its `clocksource`, if KVM is available. I can reproduce this problem with VirtualBox 5.0.20 and 5.0.18 on OS X 10.11.4, both with and without the Guest Additions (I use a NTP daemon in the guest, when the Guest Additions are not installed, but the drift is too big to compensate). The attached log is with the CentOS guest.

The official Vagrant images for CentOS don't come with the Guest Additions preinstalled. If you have Vagrant and Ansible installed on the host, you can use https://github.com/lpancescu/cloud-instance-starter-kit for automatically updating the system and installing the guest additions (you'll need to issue `vagrant reload` after it's done).",lpancescu
20665,Shutting down VM crashes all other Hyper-V instances on Windows 11,VirtualBox 6.1.28,Windows,all,2021-11-12T14:37:06Z,2021-11-12T17:49:24Z,2021-11-12T17:49:24Z,"Attempting to shut down any running VM in any fashion (either from within VM or via power off) on Windows 11 with Hyper-V running will result in all other Hyper-V machines crashing.

Attaching logs from VirtualBox 6.1.28, one machine powering off and second crashing. This is not exclusive to Hyper-V machines however - any running Hyper-V VM, either native Hyper-V, WSL2 or VMWare (also using VMP) will crash(!).

I can attach vmware-vmx.exe dumps if necessary too.",eider
18778,STM32CubeIDE GDB start-up fails in VirtualBox with Ubuntu guests,VirtualBox 6.0.6,Linux,Linux,2019-07-21T19:39:47Z,2019-07-21T19:39:47Z,2019-07-21T19:39:47Z,"
I have tested all supported Ubuntu versions listed in UM2563 when installed as a VirtualBox guest
and STM32CubeIDE is installed in the guest operating system (Ubuntu 14.04, 16.04, 18.04).
Also Linux Mint Mate 19.1 and 19 are tested with similar results as VirtualBox guests.

VirtualBox versions is:

VirtualBox 6.0
VirtualBox Graphical User Interface
Version 6.0.6 r130049 (Qt5.9.5)
Copyright © 2019 Oracle Corporation and/or its affiliates. All rights reserved.

VirtualBoxGuestAdditions VBox_GAs_6.0.6 is installed.
Oracle VM VirtualBox Extension Pack 6.0.8r130620 is installed in Oracle VM VirtualBox Manager.

THE ISSUE IS:
============

Only an unmodified code generated as a STM32Project.
All have the same behaviour where installation and compilation succeed but GDB start-up fails with the message:

Could not determine GDB version using command: /opt/st/stm32cubeide_1.0.2/plugins/com.st.stm32cube.ide.mcu.externaltools.gnu-tools-for-stm32.7-2018-q2-update.linux64_1.0.0.201904181610/tools/bin/arm-none-eabi-gdb --version

This is consistent in all Ubuntu versions and same with IDE versions 1.0.1 and 1.0.2.

When running the same command in a terminal session the response is as follows:

xxxxx@ubuntu1604base:~$ /opt/st/stm32cubeide_1.0.2/plugins/com.st.stm32cube.ide.mcu.externaltools.gnu-tools-for-stm32.7-2018-q2-update.linux64_1.0.0.201904181610/tools/bin/arm-none-eabi-gdb --versionGNU gdb (GNU Tools for STM32 7-2018-q2-update.20190328-1800) 8.1.0.20180315-git
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type ""show copying""
and ""show warranty"" for details.
This GDB was configured as ""--host=x86_64-linux-gnu --target=arm-none-eabi"".
Type ""show configuration"" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type ""help"".
Type ""apropos word"" to search for commands related to ""word"".

When installed without virtualisation (in Linux Mint 19.1 Mate only) everything works as expected (no errors).

",mnu
20836,Time Machine won't back up VirtualBox VMs folder on Mac Monterey,VirtualBox 6.1.32,Mac OS X,Windows,2022-02-21T23:02:57Z,2022-02-21T23:02:57Z,2022-02-21T23:02:57Z,"I have a MacBook Pro running VirtualBox 6.1.32 with a Windows 10 VM. If I try to use Time Machine to back up the 'VirtualBox VMs' directory, Time Machine hangs after transferring a trivial amount of data. The backup drive is an external drive mounted on a networked 2021 Mac mini M1 machine. If I exclude the directory from the backup list, the backup proceeds normally. Specifically, the .vdi file must be excluded for the backup to work.

Both Macs are running MacOS 12.2.1. Backup disk drive is formatted using APFS to be 3 Volumes within one Container.",LarryW-PDX
9717,Unable to restore from state saved with memory balloon active,VirtualBox 4.1.4,Windows,Windows,2011-10-07T15:24:08Z,2018-07-18T15:19:40Z,2018-07-18T15:19:40Z,"Steps to reproduce:

 1. Create a VM, install an operating system that supports the VirtualBox Guest Additions (Windows, Linux)
 2. Install Guest Additions, Reboot guest
 3. VBoxManage controlvm [vnname] [balloon size][[BR]]
   (any balloon size grater than zero causes the issue)
 4. Save the VM state
 5. Attempt to start the VM

Error message is Failed to load unit 'pgm' (VERR_SSM_LOADED_TOO_LITTLE).",cb1
18115,VBoxSVC segfault when ending Windows guest,VirtualBox 5.2.20,Linux,Windows,2018-11-07T20:37:50Z,2020-03-08T20:03:39Z,2020-03-08T20:03:39Z,"When cleanly shutting down Windows guests (on a CentOS Host), which were started with:
 VBoxManage startvm MACHINENAME -type vrdp

I see afterwards in the Linux syslog this kind of crash:
 VBoxSVC[5805]: segfault at 0 ip           (null) sp 00007ffc9097f068 error 14 in VBoxSVC[400000+4de000]

The crash is reproducible, and has been there since a long time (all of 5.2.x VirtualBox versions).",Helge Deller
20084,Virtual Box issues with kernel 5.9.12-lp152.50-preempt?,VirtualBox 6.1.16,Linux,Solaris,2020-12-07T13:56:46Z,2021-04-02T05:22:04Z,2021-04-02T05:22:04Z,"Hi, folks,

I am encountering problems when using the kernel version described in the summary. I am running opensuse-Linux 15.2 as host and Solaris 11.4 as guest. (an image provided by Oracle) Installation of VirtualBox 6.1.16 goes fine, just booting the Solaris virtual machine doesn´t happen, it just hangs when reaching 20%. Any ideas? 

Yours,

Reinhard",Reinardo
20014,VirtualBox GUI broken after update from 5.2.44 to 6.1.14,VirtualBox 6.1.14,Linux,Linux,2020-11-02T15:50:37Z,2020-11-02T15:56:52Z,2020-11-02T15:56:52Z,"After struggling all day trying to keep 5.2.44 working with VVV, I decided to try updating to 6.1.14.  As a result, when I open up the VirtualBox Manager GUI, I get this broken display.  The contents are in the incorrect location and aren't clickable.  If I expand the window, all of the content disappears.  Luckily, I don't need to use the configuration tools there at the moment but this will be a problem for me quite soon.",artlytic_dave
19582,"VirtualBox host, memory management issue, unable to release unused memory of guest",VirtualBox 6.1.6,Linux,Linux,2020-05-14T10:42:00Z,2020-05-18T15:23:13Z,2020-05-18T15:23:13Z,"I have a workstation with 32 GB installed memory.

I create a new Debian virtual machine, and allocate 16 GB to it. When I boot the machine, it only take less than 800 MB memory. As the virtual machine running, especially for heavy tasks. The memory occupation increase. But when I end those tasks, the memory usage of guest machine's OS suddenly reduce, but the memory occupation of virtual machine on host RAM, still keep as the memory upper bound. 

I believe that it is the problem of VirtualBox, because even if I reboot (not same as power off and boot) the virtual machine, the RAM usage still keep as Maximum.

I hope VirtualBox can add ability to automatically release unused guest memory.",stl0hello
19326,VirtualBox Manager 6.1.x causes BSOD when docking into HP Universal dock G2,VirtualBox 6.1.4,Windows,other,2020-02-20T16:25:26Z,2022-11-09T21:56:24Z,2022-11-09T21:56:24Z,"Host Operating system is newly imaged Windows 10.

To reproduce:

* download and install VirtualBox 6.1.4 with all defaults
* start the VirtualBox Manager
* plug the HP Univeral G2 into the laptop

This will crash the Windows 10 host with a Blue Screen Of Death: ""PNP DETECTED FATAL ERROR"".

The problem happens only when docking.
Operations while undocked are fine.
Operations while docked are fine.
Undocking works fine.

There isn't any VM involved (therefore no VM log). This is only from having the VirtualBox Manager started.

This is specific to versions 6.1.x. I have checked with 6.1.0, 6.1.2 and 6.1.4. I have tried installing VirtualBox both docked and undocked. I have also installed as an Administrator. The BSOD happened in all cases.

The problem does not happen with 6.0.x. I am currently using 6.0.16 without any problems.

This has been verified on two different HP Zbook 15u G5 (i7-8650u - 32GB RAM) and one HP EliteBook 840 G5. I tried both with the factory BIOS and the latest available BIOS (Jan 17 2020). It has also been verified on two different universal docks G2 - again, both with the factory firmware and with the latest available firmware.

I am attaching one of the VBoxSVC log file right agter undocking and docking. The last 3 lines of the log files were produced while undocking. No additional logging was produced when docking.
",come_raczy
9806,VirtualBox Manager does not update Machine status when the VM is started in a different desktop (sysinternals),VirtualBox 4.1.4,Windows,Windows,2011-10-25T02:21:01Z,2011-10-25T02:21:01Z,2011-10-25T02:21:01Z,"I use SysInternals desktops on Windows 7 64, which gives Windows four desktops.  I nominally run VMM on desktop 1 and VMs on each of the other three desktops.  All works fine, except for the VM status does not get updated on the VMM running in desktop 1.  

This has led to corrupted VM Linux filesystems due to executing a VM management task while the VM was in a state that normally would not allow the task to be executed.  

If the VMM is run in the same desktop as the VM, then the status gets updated.

WORKAROUND: verify that the VM is shutdown before performing any management tasks in the VMM.
",george lawton
19082,"VirtualBox Manager scaled *down*, when a mix of HiDPI and non-HiDPI screens are attached",VirtualBox 6.0.14,Linux,all,2019-11-13T13:39:07Z,2020-12-06T07:48:12Z,2020-12-06T07:48:12Z,"Steps to reproduce:
1. Set up a machine with a mix of HiDPI and non-HiDPI screens
2. Start virtualbox
3. Observe failure

Expected result: The GUI of virtualbox looks normal
Also acceptable: The GUI is scaled up for the HiDPI screen

Observed behavior:
The GUI of virtualbox is scaled *down* by roughly a factor of two. Making the text hard to read even on a low-resolution screen. The virtual machines themselves are not impacted.

My setup:
VirtualBox Version is 6.0.14 r133895 (Qt5.6.1) on Debian 10 Buster. The device is a Lenovo T490 notebook, with an internal 14"" 1920x1080 (HiDPI) screen. The behavior is observed when attaching a low-res 1920*1200 external display, but not when attaching a HiDPI 3840*2160 display. The behavior occurs both with the Debian-Package, and with the distribution-agnostic installation. I am not aware of any custom configuration on my system that might impact this.

Additional notes:
I have posted about this on the forums here: [https://forums.virtualbox.org/viewtopic.php?f=7&t=95297]. There are descriptions of similar behavior (may or may not be the same issue) here: [https://forums.virtualbox.org/viewtopic.php?f=7&t=92965]; and here: [https://forums.virtualbox.org/viewtopic.php?f=7&t=88116]. All of these are recent. I first observed this behavior after installing an update using apt, however I do not know which version I had before that, sorry.

Temporary workaround:
Virtualbox can be forcibly scaled up by setting QT_SCALE_FACTOR=2, however this cannot be set globally, as it will break all other Qt-based applications. Also when this bug is not triggered (b/c only low-res or HiDPI screens are attached) the extra scaling will make everything unusably big.

Is there any log files that I should post to help reproduce this?",Nuvanda
19334,VirtualBox version 6.1.2.,VirtualBox 6.1.2,Windows,all,2020-02-21T21:50:16Z,2020-02-21T21:50:16Z,2020-02-21T21:50:16Z,"Hello Oracle,

I have 2 complaints that I consider bugs in the new version of Virtual box.

but first I like to thank you so much for VirtualBox, I use it since 10 years and cannot live with out it, and this even I am specialized on Microsoft products since 1982!

Complaint/Problem-1: on the front page of ""Oracle VM VirtualBox Manager"" the virtual machine was showing the path including the drive where the VirtualBox guest is located, however with the new version it was taken away, which means that I have to press settings and then press storage and then press the *.vdi file to see where the file is located, and that is more than ammoying and delaying me, because4 I use over 30 Virtual Machines and cannot remember where is what and always need to know where it is located. Can you please place the full path back in the front page of the manager?

Complaint/Problem2: also on the front page of ""Oracle VM VirtualBox Manager"", they scrollbar is very annoying because I cannot see it and have a tough time to make it visible enable I can scroll up and down to see where is the Virtual Machine that I want to use. Can you please make the scroll bar always visible and therewith make my life easier as well as the lives of hundreds of thousands of others users?

I thank you for your understanding and I really hope you make these changes.

Should you require more information then please do not hesitate to contact me at M@Najar.eu.

Thank you and I wish you a very pleasant and most successful year 2020.

With my sincere regards,

Michael Najar
M@Najar.eu
0031-6-8442 6363
Erasmusplein 80
2532 RE The Hague",Najar
20276,Virtualbox6.1.18 don't work with Windows 10 (Failed to load R0 module),VirtualBox 6.1.18,Windows,Linux,2021-03-26T15:04:45Z,2021-03-26T15:14:09Z,2021-03-26T15:14:09Z,"Hello,
When I try to open my Ubunto VM the Virtual machine pop up the following error:

Failed to open a session for the virtual machine UbuntuES.

Failed to load R0 module C:\Program Files\Oracle\VirtualBox/VMMR0.r0: SUP_IOCTL_LDR_OPEN failed (VERR_LDR_GENERAL_FAILURE).

Failed to load VMMR0.r0 (VERR_LDR_GENERAL_FAILURE).

Result Code: E_FAIL (0x80004005)
Component: ConsoleWrap
Interface: IConsole {872da645-4a9b-1727-bee2-5585105b9eed}


I have already tried to reinstall the VM software with no Success;
I've tried to Disable Windows Hyper-V;

Please advise.",Fpedrosa
20566,VirtualBoxVM.exe: unknown software exception (0x80000003) on AMD FX-8300,VirtualBox 6.1.26,Windows,all,2021-09-17T02:25:41Z,2021-09-22T05:57:33Z,2021-09-22T05:57:33Z,"My computer CPU is AMD FX-8300.

When I open a VM and press ""Start"" button, the VirtualBoxVM.exe will stop after 2 seconds, and Windows pops up an ""Application Error"" dialog, which wrote:

""The exception unknown software exception (0x80000003) occurred in the application at location 0xac502432.""

VirtualBox is totally unusable on AMD FX-8300.
But same computer, different CPU: AMD Phenom II X4 945, VirtualBox works fine.",VB6
20980,"VM launch failed when ""Virtual Machine Platform"" (WSL2/WSA) is enabled",VirtualBox 6.1.34,Windows,Windows,2022-06-12T15:33:50Z,2022-06-12T15:33:50Z,2022-06-12T15:33:50Z,"In VirtualBox version before 6.1.28, there is a problem that once a feature of Windows named ""Virtual Machine Platform"" is enabled, VMs will get stuck when launching without showing anyting. And then in version 6.1.30, this problem has been soved. There is an entry in the changelog: ""Fixed 6.1.28 regression preventing VMs starting when using Hyper-V mode on Windows 10"". But now in Version 6.1.32 and 6.1.34, this problem appears again. In other words, any Windows features relies on ""Virtual Machine Platform"", such as WSL2 or WSA, can't work together with VirtualBox version 6.1.32 or later.",SolAbc
21348,VM with Linux crashes worse than DOS or Windows 95. Your dog can crash it,VirtualBox 6.1.40,Windows,other,2022-12-19T16:05:33Z,2022-12-21T12:56:29Z,2022-12-21T12:56:29Z,"Just write elementary program which creates large files, say, 1GB in size or one large file larger than 32 GB. And when your program running created 32GB of data in files VM part with Linux 100% crashes. 

Or if you have already large files aaa1, aaa2... and you combine them into one large file bbb which will be larger than 32GB this will also crash VB with its Linux like it is a dumb DOS 2.1. Here is syntax how to do that.   

$ ssh aaa* > bbb

I can try to attach this few lines Fortran program which is creating multiple files 1GB but if this will fail (because i  do not see any links or buttons how attach the file) you also can take it from the Forum (Windows Host part) where i was just published it.

P.S.
I failed to attach the file. OK here is the text. Just compile it with gFortran or any other Fortran compiler and run telling it to create more than 32 files (each will be 1 GB size file). It will crash at file #31",DanV
17961,VMM missing rflags.TF handling when advance rip that abused by virtual machine detection => fixed in SVN,VirtualBox 5.2.18,all,all,2018-08-30T00:43:52Z,2023-01-18T14:08:35Z,2023-01-18T14:08:35Z,"Found a virtual-box detection vector that abused by VMProtect 3.1+ and Safengine, confirmed by using custom hypervisor to monitor CPUID instruction.

see [https://github.com/tandasat/HyperPlatform/issues/26] for more discussion about this.

{{{#!cpp
IEM_STATIC void iemRegAddToRipAndClearRF(PVMCPU pVCpu, uint8_t cbInstr)
{
    pVCpu->cpum.GstCtx.eflags.Bits.u1RF = 0;

    AssertCompile(IEMMODE_16BIT == 0 && IEMMODE_32BIT == 1 && IEMMODE_64BIT == 2);
#if ARCH_BITS >= 64
    static uint64_t const s_aRipMasks[] = { UINT64_C(0xffffffff), UINT64_C(0xffffffff), UINT64_MAX };
    Assert(pVCpu->cpum.GstCtx.rip <= s_aRipMasks[(unsigned)pVCpu->iem.s.enmCpuMode]);
    pVCpu->cpum.GstCtx.rip = (pVCpu->cpum.GstCtx.rip + cbInstr) & s_aRipMasks[(unsigned)pVCpu->iem.s.enmCpuMode];
#else
    if (pVCpu->iem.s.enmCpuMode == IEMMODE_64BIT)
        pVCpu->cpum.GstCtx.rip += cbInstr;
    else
        pVCpu->cpum.GstCtx.eip += cbInstr;
#endif
}
}}}

Here are the test code that you can use to repro ""cpuid in real"" in real machine but ""cpuid run in vmm"" inside any virtualbox guest. (though the code is for Windows)

{{{#!cpp
int filter(EXCEPTION_POINTERS * pException, BOOL *pFound)
{
    auto code = *(BYTE*)pException->ContextRecord->Eip;

    printf(""cpuid exception eip = %p code = %02x\n"", pException->ContextRecord->Eip, code);

    //code != nop, in vmm
    if (code != 0x90)
        *pFound = TRUE;

    return EXCEPTION_EXECUTE_HANDLER;
}

BOOL cpuid_test()
{
    BOOL bFound = FALSE;
    __try
    {
        __asm
        {
            mov eax, 0
            pushfd;
            or DWORD ptr[esp], 0x100;
            popfd;//set TF=1
            cpuid;
            nop;//normal TF
            int 3//in vmm...
        }
    }
    __except (filter(GetExceptionInformation(), &bFound))
    {
        ;
    }
    return bFound;
}
}}}

[[Image(https://user-images.githubusercontent.com/12287588/44799149-d4273580-abe5-11e8-8b06-a2f1230360cc.png)]]
[[Image(https://user-images.githubusercontent.com/12287588/44799388-629bb700-abe6-11e8-8667-fff3f2b1cc8b.png)]]

what cpu do in real os:
execute CPUID->set EIP at nop->raise a single step exception(#DB) with error code 0x80000004->set TF=0->the exception would be handled by SEH->you see EIP at nop

what cpu do in vmm:
execute cpuid->cpuid is handled by vm-exit handler->you advance the EIP by one instrution in the handler (so EIP would be at nop)->execute nop->set EIP at int3->raise a single step exception->set TF=0->the exception would be handled by SEH->you see EIP at int3

the solution is to inject #DB when Guest's TF is set, and adjusting RIP normally, also writing the instruction length of CPUID in the vm entry instruction len which is 2 in the case of cpuid.

NB: You need to do this generically though.",hzqst
18800,vmR3HaltGlobal1Halt hogs host CPU,VirtualBox 6.0.10,all,Windows,2019-07-31T18:53:48Z,2019-07-31T19:11:19Z,2019-07-31T19:11:19Z,"Running a single CPU VM Windows 10 1803/1809/1903 64 bit or 32 bit (default VirtualBox VM settings for Windows 10 32 or 64 bit guest), the VM host process hogs the CPU, while the guest idles.
If I start a 2nd VM (no OS, no boot medium, single CPU), it's enough to lower the CPU usage of the Windows 10 VM. On Windows, the CPU usage is lower if I start the VM while the PC is playing a video

To reproduce, just start the Windows 10 VM (no other VMs running or suspended), booting the Windows 10 1803/1809/1903 installation ISO, until it has completely loaded the setup and you are shown the first window. To check the CPU load in the guest, press Shift+F10 and type in Command Prompt: wmic cpu get loadpercentage[[br]]
Booting Windows 10 1709 installation ISO is a very different experience. The host CPU is idling while the guest is idling[[br]]

I am able to build VirtualBox OSE on Windows and I have observed that it spends a most of time in vmR3HaltGlobal1Halt. I can also record an ETL log with the official VirtualBox release build, if needed.[[br]]
I can reproduce this issue on Windows 10 1803/1903 64 bit host OS and Debian 64 bit host OS.
I first mentioned about this issue on #12851[[br]]
There's also a forum thread, where some of the users reported similar issues [[https://forums.virtualbox.org/viewtopic.php?f=6&t=87991|link]][[br]]
",Mihai Hanor
20552,VMSVGA extremely laggy keyboard input for Linux guests,VirtualBox 6.1.26,Linux,Linux,2021-09-16T11:20:30Z,2021-09-24T06:53:29Z,2021-09-24T06:53:29Z,"Pretty much all Linux guests with this video adapter have an extremely laggy and random keyboard input: while you're typing in any graphical terminal sometimes what you're typing instantly shows on the screen, sometimes there's a random delay for up to three seconds.

It's annoying and maddening.

Host OS: Fedora 33, Linux 5.14.3[[BR]]
Environment: XFCE with '''no''' compositing[[BR]]
CPU: Ryzen 7 5800X[[BR]]
Guest OS: any modern Linux OS without 3D acceleration

Changing the amount of video memory from default 16MB to 128MB makes the issue less pronounced but it's still there.

VBoxSVGA is more or less free from this issue. Keyboard input is not perfect (there's quite a considerable input lag) but at least it's not as bad as with VMSVGA.",Artem S. Tashkinov
19841,WarpDrive doesn't work on VirtualBox 6.x anymore,VirtualBox 6.1.10,Linux,Windows,2020-08-28T20:57:13Z,2020-08-28T20:57:13Z,2020-08-28T20:57:13Z,"So in VirtualBox 5.x warp drive was working fine. When you make:

{{{
VBoxManage setextradata VM-name ""VBoxInternal/TM/WarpDrivePercentage"" 200
}}}

the time inside VM was going faster. But in VirtualBox 6.x it  doesn't work at all...

This happens because of this change:

{{{
    rc = CFGMR3QueryU32(pCfgHandle, ""TimerMillies"", &u32Millies);
    if (rc == VERR_CFGM_VALUE_NOT_FOUND)
        u32Millies = VM_IS_HM_ENABLED(pVM) ? 1000 : 10;
}}}

In 5.x default value of TimerMillies was 10 ms always (not 1 sec). But it is not the root of the problem.
The root lies near semaphore timeout calculation which trigger return from HLT. When WarpDrive is enabled this semaphore timeout calculated in wrong way.",gim
21408,Windows: High host cpu usage with 7.0.4,VirtualBox-7.0.4,Windows,Windows,2023-01-13T11:25:35Z,2023-01-13T11:25:35Z,2023-01-13T11:25:35Z,"Even though the guest OS is idle the VBoxHeadless process uses up a whole core for running (see attached screenshot).

The VM was running for a whole day and it's also the same if the start the VM normal or headless - it is constantly consuming the whole CPU core.

Both guest and host use the latest Windows 10 version (22H2 19045.2486).

I have also attached the configuration files of my VM, VBox logs and the sys-info of my host system.",IndyJones
9559,Winsat fails with TSC Frequency,VirtualBox 4.1.2,Windows,Windows,2011-09-05T12:26:08Z,2021-07-09T09:34:33Z,2021-07-09T09:34:33Z,"I can not get WinSat to run (Windows System Assessment Tool)

Pretty much every test receives this result:

>winsat formal
> Running the Formal Assessment
Error: Cannot Measure Time Stamp Counter (TSC) Frequency

>winsat cpu
Windows System Assessment Tool
Error: Cannot Measure Time Stamp Counter (TSC) Frequency

The Host machine is a Dell 6510 laptop with i7 q720 win7 x64.
The Guest is win7 x32.
Both are at SP1 and have current windows updates.


I've had no luck running winsat on any version 4 it did run on version 3 except for graphic.

I believe this same issue also leads to long guest OS pauses (for 30 - 90 second) where no activity in the guest happens and then it resumes to normal operations.

I have current guest additions installed,




",steve richards
19702,在开启了  Hyper-V  的  Windows 10 上运行  VirtualBox  虚拟机，会发生崩溃,VirtualBox 6.1.10,Windows,Linux,2020-07-04T08:56:06Z,2020-07-04T12:39:53Z,2020-07-04T12:39:53Z,"我在  VirtualBox  上运行  Arch  Linux  操作系统时，虚拟机出现错误，导致我无法安装  Arch  Linux  操作系统。  

具体现象：虚拟机能够成功启动，但是尝试运行位于  Live  CD  的操作系统时，会发生崩溃

镜像源：  https://cdimage.debian.org/debian-cd/current/amd64/iso-dvd/debian-10.4.0-amd64-DVD-3.iso

",FANminiworld
