VMware Communities
7CLOWN7
Contributor
Contributor

Multiple PCI Express Standard Root Ports Conflict

I've tried VMWare Workstation and created a Windows 7 RC VM. When booting up this VM it takes a good 15 minutes or longer. Once it is up there are about 32 PCI Express Standard Root Ports in Device Manager and yet the host PC does not even have PCI-E slots just normal PCI slots and a AGP port. It takes about 5 minutes to disable each one so to disable them all would take about an hour or so. With this many devices there are several conflicts because of lack of resources and some drivers that should load or work will not such as network etc.

Has anyone else seen this issue or something similar to it???

0 Kudos
11 Replies
RDPetruska
Leadership
Leadership

Why are you trying to disable the virtual hardware in the guest? Have you installed the VMware Tools yet? The Tools should contain drivers for all the virtual hardware except audio and networking... those should be on the Win7 installation media already.

0 Kudos
7CLOWN7
Contributor
Contributor

Yes everything is installed and the drivers are installed from Tools. The problem is the PCI Express Standard Root Port gets installed 32-36 times which then conflicts. Did you not look at the picture? Yes the tools are installed and the VM is configured and set up properly. I've installed serveral VM's on this same machine including a Windows 2008 and others before without this issue. This is a Windows 7 VM that is doing this. As I stated, this machines Host does not even have PCI-E period. The reason for trying to disable them is because once you disable all of them then the system boots properly and will then function properly. The problem is, all of these loading and as mentioned, to disable even just one of them takes 5 minutes and the boot with them all enabled takes 15-20 minutes. If you un-install them that also takes 5 minutes for each one but upon reboot the system would re-install them. So disabling is the best way but I'd like to clear them all permanently or even have a fresh install that wouldn't load the driver period.

0 Kudos
continuum
Immortal
Immortal

As I stated, this machines Host does not even have PCI-E period.

And this really does not matter period.

Your virtual hardware and your real hardware are completely different.

Only the CPU is more or less the same in both environments.

Suggestion:

disable the virtual PCI-bridge in your virtual hardware and see if that makes a difference.

So post your vmx-file

___________________________________

description of vmx-parameters:

VMware-liveCD:


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

0 Kudos
7CLOWN7
Contributor
Contributor

I disabled the PCI-Bridge and it took over 15 minutes for it to do so, and it killed my sound and NIC and made the system unresponsive. The 36 PCI-E items still remained and the conflicts still present. I was thinking about removing the entries within the .vmx file for the pci-e devices which shows 4 of them and rebooting to see what happens but haven't gotten to that yet.

UPDATE:

With you wanting to see the vmx file and the file I forgot had the entries in, pointed me actually to the right direction. After editing the file and doing as I mentioned up above - I REMOVED the lines pciBridge4.present = "TRUE"

pciBridge4.virtualDev = "pcieRootPort" and did this for all 4 sections showing in the vmx file. Rebooted machine and BINGO the machine booted VERY quickly and now the Device Manager looks normal and is no longer expanded and no longer has any conflicts what so ever.

Thanks for making me remember about the vmx file it totally slipped my mind to even check in there and even do any testing with it. So if anyone else sees this look into the vmx file with a text editor and remove those items pointing to pcieRootPort and all is cured.

Am I the ONLY person to see this issue with Windows 7 as a VM or any OS for that matter??? Either way it's fixed now, thanks for the posts guys.

0 Kudos
continuum
Immortal
Immortal

In short

if you have more PCI-E rootports in devicemanager than you wanted - disable them - its your VM

on:

pciBridge4.present = "TRUE"
pciBridge4.virtualDev = "pcieRootPort"

off:

pciBridge4.present = "false"
pciBridge4.virtualDev = "pcieRootPort"

___________________________________

description of vmx-parameters:

VMware-liveCD:


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

0 Kudos
7CLOWN7
Contributor
Contributor

Well to add insult to injury. I also have a Windows 2008 Server VM and it did actually have the same issue so I fixed it the same way, I just removed those lines instead of using the False change.

But with the Windows 2008 Server it also has Generic Bus over 100 times in device manager but there are NO conflicts and there is nothing in the VMX file to fix all of those.

So my next question is why are the VMs creating multiple devices over and over? I should have given the title to the thread Multiple Devices period at this point.

As I mentioned earlier is no one else seeing multiple devices over and over and I mean anywhere from 30-100 times?

0 Kudos
continuum
Immortal
Immortal

You must have a strange CPU - here those PCI-E root ports are detected in one run - no matter if its Vista, 2k8 or win7

by the way - setting devicaname.present = "false" is better than removing the line.

You may not notice if for example

guestOS = "longhorn"

also sets the silent default

pciBridge7.present = "TRUE"

pciBridge7.virtualDev = "pcieRootPort"

not all devicename.present = "TRUE" lines will be printed in vmx

___________________________________

description of vmx-parameters:

VMware-liveCD:


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

0 Kudos
7CLOWN7
Contributor
Contributor

CPU is a AMD Athlon XP 2700+ @ 2.17GHZ on an A7N8X-Deluxe Nforce 2 board. I know that anything before Nforce 4 is NOT supported in anything over XP but as someone mentioned before it should not make a difference in Virtual side since it uses only CPU from Host... But if it does use other factors of real hardware then that could be the issue...

0 Kudos
continuum
Immortal
Immortal

yep - I guess that is strange enough

you can also try to set

virtualHW.version = "6"

then you will get a much cleaner devicemanager view - only works with newly created VMs

___________________________________

description of vmx-parameters:

VMware-liveCD:


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

0 Kudos
7CLOWN7
Contributor
Contributor

I actually added a bit more above, didn't expect you to post so quick. I currenlty have both of those at virtualHW.version="7". What do the hardware levels mean exactly?

0 Kudos
continuum
Immortal
Immortal

What do the hardware levels mean exactly?

thats a long story ...

the higher the number the more modern is the main-board of the virtual machine.

version 4 for example had 6 PCI-slots (VGA + 5 free slots)

version 4 uses USB 1.1

version 6 uses USB 2

version 7 has PCI-extensions boards so that it can have up to 60 PCI-slots

If I remember right # 6 still used 6 PCI-slots only - so this may help you in this case

want some more on vmx-settings ? - read my site > vmx

___________________________________

description of vmx-parameters:

VMware-liveCD:


________________________________________________
Do you need support with a VMFS recovery problem ? - send a message via skype "sanbarrow"
I do not support Workstation 16 at this time ...

0 Kudos