login
Header Space

 
 

Re: MTRR initialization

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: Yinghai Lu <yhlu.kernel@...>
Cc: Howard Chu <hyc@...>, Andi Kleen <ak@...>, linux-kernel <linux-kernel@...>
Date: Sunday, September 16, 2007 - 2:40 pm

"Yinghai Lu" <yhlu.kernel@gmail.com> writes:


There are a couple issues with changing the MTRR configuration.
- You may not have perfect information on the cpu, the AMD revF is a good
  example.
- Code in SMM mode may actually depend on the current mtrr configuration.
- The BIOS's need to fixed to setup MTRRs properly.

So the sanest approach appears to be.
- In linux only use ram that is mapped by a write-back mtrr.
  This preserves performance and is always safe.
- If you need write-combining set it up in the page tables with PAT.

There is some difficulty there but software can always do those things
safely.

Eric
-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
MTRR initialization, Howard Chu, (Fri Sep 14, 12:33 pm)
Re: MTRR initialization, Siddha, Suresh B, (Fri Sep 21, 8:27 pm)
Re: MTRR initialization, Howard Chu, (Fri Sep 21, 8:46 pm)
Re: MTRR initialization, Jesse Barnes, (Wed Sep 19, 5:52 pm)
Re: MTRR initialization, Andi Kleen, (Thu Sep 20, 2:50 am)
Re: MTRR initialization, Jesse Barnes, (Thu Sep 20, 11:03 am)
Re: MTRR initialization, Jesse Barnes, (Thu Sep 20, 1:08 pm)
Re: MTRR initialization, Avi Kivity, (Thu Sep 20, 8:05 am)
Re: MTRR initialization, Andi Kleen, (Thu Sep 20, 8:10 am)
Re: MTRR initialization, Avi Kivity, (Thu Sep 20, 8:46 am)
Re: MTRR initialization, Yinghai Lu, (Fri Sep 14, 1:12 pm)
Re: MTRR initialization, Howard Chu, (Sun Sep 16, 12:08 pm)
Re: MTRR initialization, Yinghai Lu, (Sun Sep 16, 1:53 pm)
Re: MTRR initialization, Eric W. Biederman, (Sun Sep 16, 2:40 pm)
Re: MTRR initialization, Howard Chu, (Tue Sep 18, 1:53 pm)
Re: MTRR initialization, Eric W. Biederman, (Tue Sep 18, 2:18 pm)
speck-geostationary