Re: Turmoil Continues in XP land
Bill Blanton wrote:
| "PCR" <pcrrcp@netzero.net> wrote in message
| news:%231ZLVhZ1IHA.416@TK2MSFTNGP04.phx.gbl...
|> Bill Blanton wrote:
|> | "PCR" <pcrrcp@netzero.net> wrote in message
|> | news:u56nv$J1IHA.5472@TK2MSFTNGP06.phx.gbl...
|> |> Bill Blanton wrote:
|> |> | "PCR" <pcrrcp@netzero.net> wrote in message
|> |> | news:uRU9L%23$0IHA.4004@TK2MSFTNGP03.phx.gbl...
|> |> |> Bill Blanton wrote:
|> |> |> | "PCR" <pcrrcp@netzero.net> wrote in message
|> |> |> | news:O4UZkKz0IHA.4912@TK2MSFTNGP03.phx.gbl...
|> |> |> |> Bill Blanton wrote:
|> |> |> |
|> |> |> |> | The "real" computer hardware doesn't matter at all. The VM
|> |> |> |> | communicates only
|> |> |> |> | with the host OS.
|> |> |> |>
|> |> |> |> OK. So, the host OS must first be able to recognize a new
|> |> |> |> device. Then, it's a matter of whether the VPC has something
|> |> |> |> for that too already built in. OK, then.
|> |> |> |
|> |> |> | No, the VM will operate even if the host OS does not have a
|> |> |> | similar device.
|> |> |> | However, if your host OS does not have a display adapter, for
|> |> |> | example, you're not going to "see" it. Or if your host OS
|> |> |> | does not have a sound card, you're not going to "hear" the
|> |> |> | virtual output.
|> |> |>
|> |> |> I see, then, VM makes no attempt to recognize any device at
|> |> |> all. It presumes the device exists & it's up to the host OS to
|> |> |> direct input & output to them.
|> |> |
|> |> | guest OS > virtual devices > host OS > real device
|> |> |
|> |> | Something like that. It's a lot more complicated I'm sure,
|> |>
|> |> Alright. Are you sure the VM doesn't make an attempt to see what
|> |> the host OS has to offer?
|> |
|> | For the most part, yes. There are special cases such as the CPU,
|> | mouse,
|> | and KB
|>
|> OK.
|>
|> |> Doesn't it have to know how many partitions are
|> |> available at least?
|> |
|> | No, you set up and use a virtual disk. You partition and format it
|> | just as a real disk. The VM doesn't need to know what file system
|> | the host is using or how its disk structure is set up. Virtual
|> | disk I/O is translated
|> | to file I/O before it reaches the host, and vice versa
|>
|> Oh. So, the guest OS doesn't get drive letters for the pre-existing
|> partitions of the host, (except as may be set up in a kind of
|> Networking situation you say below). OK, I see. For full partitions,
|> it can only set up & use its own virtual partitions. Alright. But
|> the ones it does set up do remain past one session. Fine.
|
| Yes, If you do a "save" on the session.
|
|
|>
|> |> What if there are two printers or two of anything?
|> |
|> | I think you can use 2 or more printers, but as I said before
|> | printers aren't fully virtualized. I've never had a need to set up
|> | a printer on a VM, so don't know much about that aspect. If I need
|> | to print something
|> | I copy/paste it into the host.
|>
|> OK. It can be done, but it's easier to copy/paste. Someone who did
|> want to run a word processor in the guest OS could do so by going
|> through the extra trouble. OK, fine.
|>
|> | As far as fully virtualized devices such as the display adapter,
|> | there is only
|> | one per VM. It doesn't matter that your real machine has two.
|>
|> Very well. The virtual OS doesn't get to pick/choose.
|>
|> |>
|> |> |> Alright, then. How is a USB flash drive unavailable for
|> |> |> use by VM then? Wouldn't it just be a drive letter presented to
|> |> |> VM by the host OS, & visa versa?
|> |> |
|> |> | There is no virtual USB device in the virtual machine. You can
|> |> | use a USB device (KB, mouse, drive), but not tied directly to
|> |> | the VM.
|> |>
|> |> So, a USB flash drive can be used for input/output, but not as a
|> |> boot media? Then, a drive letter for it is presented to the guest
|> |> OS but not the to the VM BIOS?
|> |
|> | You can "share" a USB storage device, which is analogous to
|> | networking,
|> | or you can put your virtual HD on a USB device, but you cannot
|> | access the USB device at the USB level.
|>
|> OK. One could transfer downloads & other data from a real Win98
|> machine through a USB drive, then. Very good.
|
| Yes, you can "share" drives/folders (not just USB) through a virtual
| network
| between the host and guest OS. You can also network through the
| host to get the guest online.
|
|
|
|
|>
|> |> |>
|> |> |> |> |> But the virtual BIOS does access the real hardware at
|> |> |> |> |> some point -- doesn't it? --
|> |> |> |> |
|> |> |> |> | No, never. The virtual BIOS is just a set of software
|> |> |> |> | routines within the VM .
|> |> |> |> | The VM accesses Windows. Windows accesses the hardware.
|> |> |> |> |
|> |> |> |> | That's the point of it. Physical hardware doesn't matter.
|> |> |> |> | As long as the
|> |> |> |> | program running the VM is supported by the OS, the OS
|> |> |> |> | loaded within the VM will operate.
|> |> |> |>
|> |> |> |> Alright. I think I get it now. The VM BIOS communicates only
|> |> |> |> with the host OS to know what devices are available.
|> |> |> |
|> |> |> | (Excluding the CPU) It doesn't care what devices are on the
|> |> |> | host. It has its
|> |> |> | own set of virtual devices. It's its own "machine"
|> |> |>
|> |> |> Would this for instance be a single virtual printer device? Or
|> |> |> could there be a number of them depending on type/model of
|> |> |> printer?
|> |> |
|> |> | Printers are treated differently, and aren't really fully
|> |> | virtualized.
|> |>
|> |> That was just an example. I suppose the VM has a frozen set of
|> |> device types it can use.
|> |
|> | Right. It has a set of built in devices.
|> |
|> |> I'm trying to determine whether there is also a frozen
|> |> set of drivers for any particular device type, or whether there is
|> |> only one virtual driver per device type.
|> |
|> | It emulates the devices, and the OS installed within the VM does
|> | the device
|> | discovery and installs the drivers for the virtual devices. They
|> | are devices
|> | for which Windows already has drivers. If you install an OS that
|> | does not
|> | have drivers for the virtual devices, you would have to hunt them
|> | down and install them yourself Just like in real life ;-)
|> | .
|> |
|> | Here's the list for VPC2004
|> |
|> | BIOS - AMI BIOS
|> | Chipset - Intel 440BX
|> | Sound card - Creative Labs Sound Blaster 16 ISA Plug and Play
|> | Network adapter - (multi-function) DEC 21140A 10/100
|> | Video card - S3 Trio 32/64 PCI with 8 MB Video RAM
|> |
|> |
|> |
|> |
|> |> Is it one size fits all per device
|> |> type? I guess it would have to be that way, if the VM makes no
|> |> attempt at discovery-- unless the host OS is very specific in what
|> |> it presents to the VM on a per device basis.
|> |
|> | The guest OS finds and communicates with the virtual devices. The
|> | virtual devices
|> | do I/O with the host OS. The guest and host OS are oblivious of one
|> | another.
|> | As far as the host OS is concerned, it (the VM) is a Windows
|> | program requesting
|> | I/O through normal Windows program calls. Not low level functions.
|>
|> Alright. Then, I guess the guest OS is more/less stuck with generic
|> capabilities common to various devices that can be connected to the
|> host machine. The guest OS won't know of anything special in them.
|> It won't have specialized drivers. However, on the plus side, it
|> does get to use devices that otherwise might not work on a Win98
|> machine for lack of drivers.
|>
|> | Heh, we should probably wrap up this way OT thread soon. Besides
|> | that, OE is going to start falling over itself
|>
|> I agree. I think I've found out enough for now. Thanks for all of
|> this information. It's enough for me.
.
|
| OK then, and later..
| I'll bid a goodnight,, with a wiki-link ;-]
|
http://en.wikipedia.org/wiki/X86_virtualization
| It's a general overview of virtualization, and touches on hardware
| support
| for processor virtualization..(which we didn't discuss). Interesting
| stuff.
OK, then. Goodnight. If it is a free download, surely I will take it a
day after the horrible day this Win98 crumbles to dust & I must puchase
a used XP-irradiated machine. I'm not half as stubborn as Colorado!
--
Thanks or Good Luck,
There may be humor in this post, and,
Naturally, you will not sue,
Should things get worse after this,
PCR
pcrrcp@netzero.net