Re: Stack at high memory locations

Rene Herman <rene.herman@xxxxxxxxxxxx> · Tue, 06 May 2008 17:46:39 +0200

On 06-05-08 05:15, Peter Teoh wrote:

Greg: added to the CC as one of the LDD3 authors. perhaps it's interesting 

to note for a possible followup edition.

Please be aware that /proc/iomem lists physical addresses, while his
request was for availability of a virtual address.

I was aware of the difference, that is why before I posted, I referred
to LDD3, and found this mapping between /proc/*/maps (which is virtual
address) and iomem.   And since Greg in the book speaked as if the two
address are the same (they are identical in the first place - could it
be due to identity mapping for the low mem?   I suspect so.....) and
that's where I deduced that it is virtual memory.

Here (the whole paragraph under "The mmap Device Operation"):

http://people.freedesktop.org/~nagappan/LDD3/LDD3.html

Ok...logically it should physical address....being IOMEM...and being
process independent....

Ah, I see. That "definitive" example is actually a fairly poor chosen one 

due this exact potential for confusion indeed. If I remember right the fact 

that (in the book) you see that identify for the first megabyte is only due 

to the X server setting up a VM86 environment ("a DOS box", if that 

terminology still means anything to you) in which to call the VGA BIOS. The 

VGA BIOS expects to find a real-mode environment; specifically, expects to 

be called in an environment where it itself lives at 0xc0000 (expressed in 

real-mode terms as 0xc000:0x0) and the VGA videomemory at 0xa0000 (0xa000:0x0).

These days it appears X is either no longer using VM86 or using segmentation 

register tricks (google for "unreal mode" if you're interested) to get 

around that issue:

root@7ixe4:~# cat /proc/$(pidof X)/maps | grep /dev/mem
b4a24000-b5224000 rw-s ef000000 08:05 905647     /dev/mem
b5224000-b5264000 rw-s 000c0000 08:05 905647     /dev/mem
b5b14000-b5b24000 rw-s 000a0000 08:05 905647     /dev/mem
b5b24000-b7b24000 rw-s e4000000 08:05 905647     /dev/mem
b7b24000-b7b28000 rw-s efefc000 08:05 905647     /dev/mem

Not much into X, so I don't know. In any case, in the book it's a very 

non-generic example indeed. Badly picked.

Rene.

--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ