Accessing high PCI memory

Nathalie Barat (nbarat@kaptech.com)
Mon, 13 Sep 1999 17:55:23 +0200


Message en plusieurs parties et au format MIME.

------=_NextPart_000_002C_01BEFE11.2740A640
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I am trying to read from and write to the I/O memory, in which a PCI =
board is mapped. As I could read in this mailing list, I used the =
function vremap (I am using Linux 2.0.36) to get the virtual address =
associated with the high PCI memory address, and then, I used the =
functions readb and writeb. Unfortunately, it seems not to work :

When I try to get the data page back from pte (using pgd_offset, =
pmd_offset, pte_offset and pte_page), I obtain something like =
0xF000F000, while the address I wanted to vremap was 0xFEA00000.
Furthermore, I noticed that the result of pgd_val was always 0.
(I control the fact that the read-access or the write-access doesn't =
work by observing some signals on the board).

Where is the mistake?
Thank you for your help

Nathalie Barat

PS : please CC to me (nbarat@kaptech.com)

------=_NextPart_000_002C_01BEFE11.2740A640
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">

I am trying to read from and write to = the I/O=20 memory, in which a PCI board is mapped. As I could read in this mailing = list, I=20 used the function vremap (I am using Linux 2.0.36) to get the virtual = address=20 associated with the high PCI memory address, and then, I used the = functions=20 readb and writeb. Unfortunately, it seems not to work :
 
When I try to get the data page back = from pte=20 (using pgd_offset, pmd_offset, pte_offset and pte_page), I obtain = something like=20 0xF000F000, while the address I wanted to vremap was = 0xFEA00000.
Furthermore, I noticed that the result = of pgd_val=20 was always 0.
(I control the fact = that the=20 read-access or the write-access doesn't work by observing some signals = on the=20 board).
 
Where is the mistake?
Thank you for your help
 
Nathalie Barat
 
PS : please CC to me=20 (nbarat@kaptech.com)
------=_NextPart_000_002C_01BEFE11.2740A640-- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/