On Sun, 2012-09-09 at 23:54 -0400, zhenzhong.duan wrote:Hi,On 2012-09-08 02:40, H. Peter Anvin wrote:Without this patch, mtrr_cleanup could not optimize. The original MTRROn 09/07/2012 10:44 AM, Peter Hurley wrote:For a large memory system, mtrr_cleanup offten fail in most case. Even
diff --git a/arch/x86/kernel/cpu/mtrr/cleanup.cI really don't like it as it introduces yet another user of max_pfn,
b/arch/x86/kernel/cpu/mtrr/cleanup.c
which should be going away. Furthermore, the better question is what
remaining needs there are for MTRR cleanup; historically the reason
was that it prevented the display from being mapped WC via MTRR due to
the MTRR conflict resolution rules favoring UC.
if it succeed, it often occupy all of MTRR entrys.
How was display mapped as WC in above case?
setup from BIOS remained, which left the display as UC (and a lot of log
spew).
From manual, it's readonly, writing it will lead to #GP.
Why did bios give a lot of space then real mem, for hotplug?I assume the reason was for hotplug.
An interesting side note: more recent revisions of this BIOS (rev. A11)
report one less variable MTRR (so, IA32_MTRRCAP is writable?)
Oh, I see, WC in page level take precedence. Is the fix already in upstream?
Well, since PAT entries marked WC override all MTRR settings, whateverHowever, the right way to fix that is to use the PAT interfaces, whichDo you mean disable MTRR totally here?
doesn't have this drawback -- then MTRR cleanup becomes entirely
superfluous and the problem goes away.
the BIOS set the variable MTRRs to becomes irrelevant, so not disabled
but rather ignored.