- 论坛徽章:
- 13
|
本帖最后由 dengbao2001 于 2012-07-27 10:56 编辑
Issue
Received the following message when restarted the server
kernel: mtrr: type mismatch for f8000000,400000 old: uncachable new: write-combining
After reboot, the console has messages
mtrr: type mismatch for e8000000,4000000 old: uncachable new: write-combining
Environment
Red Hat Enterprise Linux 4
Red Hat Enterprise Linux 5
Resolution:
These messages are informational messages only, seen in systems with CPUs supporting MTRR and are caused by MTRR overlaps.
The messages can be safely ignored if there are no associated functional difficulties noticed in the system.
These messages are not to be considered as an error, just a kernel message to give out the information that a type mismatch in expected access mode has been noticed in the MTRR, and is intended to be corrected, if possible.
This feature is generally controlled by BIOS settings. Some system firmwares have an "MTRR Mapping" feature in the BIOS setup, you may resolve these messages by changing this setting.
Update system BIOS to the latest available version.
For more information regarding mtrr, refer to /usr/share/doc/kernel-doc-*/Documentation/mtrr.txt
Root Cause
MTRR stands for Memory Type Range Register, which are a set of Processor Supplementary Capabilities control registers that provide system software with control of how accesses to memory ranges which the CPU has cached. MTRRs may be used to control processor access to memory ranges. This is most useful when you have a video card on a PCI or AGP bus.
There are different possible access modes to memory ranges, and they are uncached, write-through, write-combining, write-protect, and write-back. Write-combining allows bus write transfers to be combined into a larger transfer before bursting them over the bus to allow more efficient writes to system resources like graphics card memory. This often increases the speed of image write operations by several times, at the cost of losing the simple sequential read/write semantics of normal memory. This is normally preferred for the X server.
MTRRs are a finite limited resource provided by the CPU (for CPUs that support MTRRs). The BIOS may use some of these to set cacheability of memory regious during POST, and the OS can modify/add/remove regions after that. However, the OS is limited to using what the hardware provides/allows.
The messages are sometimes caused by a BIOS problem. The hardware vendor needs to be consulted regarding them further. |
|