mtrr and fresh install of Fc6 on IBM T23 labtop

[jbk]

I am getting an error message at login warning that "MTRR 1 
is not used". Reading the logs it is repeated several times. 
I have done some googling and have found old posts circa 
1999 to 2003 on the subject. I have also looked into the 
documentation in the kernel source. Before I do anything 
about it I'd like some input on evaluating the problem.

output from proc
# cat /proc/mtrr
reg00: base=0x00000000 (   0MB), size= 512MB: write-back, 
count=1
reg01: base=0xe8000000 (3712MB), size=  16MB: 
write-combining, count=1
reg02: base=0xe4000000 (3648MB), size=  64MB: 
write-combining, count=1
reg03: base=0xd0000000 (3328MB), size= 256MB: 
write-combining, count=1

output from xorg.0.log
(--) SAVAGE(0): mapping framebuffer @ 0xe8000000 with size 
0x1000000
(==) SAVAGE(0): Write-combining range (0xe8000000,0x1000000)
(==) SAVAGE(0): Removed MMIO write-combining range 
(0xe8000000,0x1000000)
(II) SAVAGE(0): Splitting WC range: base: 0xe4000000, size: 
0x5000000
(==) SAVAGE(0): Write-combining range (0xe8000000,0x1000000)
(==) SAVAGE(0): Write-combining range (0xe4000000,0x5000000)
(II) SAVAGE(0): map aperture:0xb1e0c000
(II) SAVAGE(0): 4740 kB of Videoram needed for 3D; 16384 kB 
of Videoram available
(II) SAVAGE(0): Sufficient Videoram available for 3D

output from /var/log/messages
]# less /var/log/messages | grep -i mtrr
Jan 28 06:14:20 t23 kernel: mtrr: MTRR 1 not used
Jan 28 06:14:23 t23 kernel: mtrr: 0xe8000000,0x4000000 
overlaps existing 0xe8000000,0x1000000
Jan 28 06:14:23 t23 kernel: mtrr: base(0xe4000000) is not 
aligned on a size(0x5000000) boundary
Jan 28 06:14:23 t23 kernel: mtrr: 0xe8000000,0x4000000 
overlaps existing 0xe8000000,0x1000000
Jan 28 06:14:23 t23 kernel: mtrr: base(0xe4000000) is not 
aligned on a size(0x5000000) boundary
Jan 28 06:14:23 t23 kernel: mtrr: 0xe8000000,0x4000000 
overlaps existing 0xe8000000,0x1000000
Jan 28 06:14:23 t23 kernel: mtrr: base(0xe4000000) is not 
aligned on a size(0x5000000) boundary

My assumption is that I would have to re echo to proc the 
correct addresses based on the info from xorg.log. But why 
are they wrong in the first place?

Jim Kelly-Rand

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.