[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Video issues with T42 running Etch and more


for thinkpads and linux there is really good site..


On Don, 01 Feb 2007, justin joseph wrote:

> Have now come up against a stone wall and need help.  The video card
> on this machine is a ATI Technologies Inc Radeon Mobility M7 LW
> [Radeon Mobility 7500].  X is working but I don't think its working
> the way it should.
> I am unable to find from google, why lspci says that this card is
> 128MB whereas it is 32MB.

this is only the memory area use by your card.
your cpu is capable to address at least 4gb of ram, and a physicall  devices
grab just some of the addresses so the cpu can work with them.

512mb are grabbed by your system memory
and every pci device takes some memory

> justin@thinkpad:~$ free -m
>             total       used       free     shared    buffers     cached
> Mem:           504        230        273          0          8         96
> -/+ buffers/cache:        125        379
> Swap:         1011          0       1011
> justin@thinkpad:~$
> I see that there is only 504MB of ram where as it must have been 512.
> where has the rest 8mb gone? I understand that this happens only when
> its shared by the VGA controller and in this case why would the VGA
> controller share it because its got 32 MB of vram.

i bet 8mb are lost because auf AGP, the agpgart driver in linux by default
grabs 8mb of memory.

> I can provide the full lspci output if required and would be nice if
> someone could send me an xorg.conf which fully utilizes this system.

as above .. take a look to thinkwiki

> on the sidelines, I were wondering why *most* of the subsystems on
> this T42 shows up as R50e on the lspci output.  It that usual or has
> Lenovo conned me into purchasing R50e machine with just T42 name with
> better lid material.

lspci just reads the PCI IDs of your devices, then uses some sort of databae
to print _you_ a human readable device name. So don't take the text displayed
to serious.

Florian Reitmeir

Reply to: