Can someone clarify its behavior?
First couple of times when I try to allocate from that lookaside list
I’m getting a “miss” condition. I.e. instead of allocating from the list
the system allocates from the NP pool. After a while (like 4-5 misses) I
see that allocations start coming from the list. I could speculate why
this is happening (like system collects some stats to decide if it
“worth” to “cut” NP for the entire list) but I believe that somebody in
this list know for sure 
I didn’t see anything similar to that for
Paged lookaside lists though. The system that I have is W2K SP3
TIA,
Vladimir
Hard to say exactly without source access, so perhaps others could comment.
I do know the system tracks lookaside list usage and adjusts the sizes on a
frequent basis. Also, the Depth parameter for the
ExInitializeNPagedLookasideList call would allow you to specify a minimum
list size (this parameter was not specified as reserved in the original NT4
DDK documentation). Specifying a depth would prevent the misses at start.
However, because the system does monitor the lists frequently MS has now
made the depth parameter reserved. I suppose they don’t want drivers
specifying memory that might not get used.
–
Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm
“Vladimir Chtchetkine” wrote in message
news:xxxxx@ntdev…
Can someone clarify its behavior?
First couple of times when I try to allocate from that lookaside list
I’m getting a “miss” condition. I.e. instead of allocating from the list
the system allocates from the NP pool. After a while (like 4-5 misses) I
see that allocations start coming from the list. I could speculate why
this is happening (like system collects some stats to decide if it
“worth” to “cut” NP for the entire list) but I believe that somebody in
this list know for sure I didn’t see anything similar to that for
Paged lookaside lists though. The system that I have is W2K SP3
TIA,
Vladimir
Thanks, Bill!
I do remember that Depth param was “valid” (as an opposite to
“reserved”) for NT4 but when I experimented with my W2K specifying this
param didn’t do any difference.
-----Original Message-----
From: Bill McKenzie [mailto:xxxxx@compuware.com]
Sent: Thursday, November 06, 2003 4:50 PM
To: Windows System Software Devs Interest List
Subject: [ntdev] Re: NPLookaside list
Hard to say exactly without source access, so perhaps others could
comment.
I do know the system tracks lookaside list usage and adjusts the sizes
on a
frequent basis. Also, the Depth parameter for the
ExInitializeNPagedLookasideList call would allow you to specify a
minimum
list size (this parameter was not specified as reserved in the original
NT4
DDK documentation). Specifying a depth would prevent the misses at
start.
However, because the system does monitor the lists frequently MS has now
made the depth parameter reserved. I suppose they don’t want drivers
specifying memory that might not get used.
–
Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm
“Vladimir Chtchetkine” wrote in
message
news:xxxxx@ntdev…
Can someone clarify its behavior?
First couple of times when I try to allocate from that lookaside list
I’m getting a “miss” condition. I.e. instead of allocating from the list
the system allocates from the NP pool. After a while (like 4-5 misses) I
see that allocations start coming from the list. I could speculate why
this is happening (like system collects some stats to decide if it
“worth” to “cut” NP for the entire list) but I believe that somebody in
this list know for sure I didn’t see anything similar to that for
Paged lookaside lists though. The system that I have is W2K SP3
TIA,
Vladimir
—
Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256
You are currently subscribed to ntdev as:
xxxxx@borland.com
To unsubscribe send a blank email to xxxxx@lists.osr.com