Re: [PATCHES] update i386 spinlock for hyperthreading

Tom Lane wrote:

>Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
>
>
>>Manfred Spraul wrote:
>>
>>
>>>I remember there were patches that tried other algorithms instead of the
>>>simple LRU for the buffer manager. Has anyone tried to change the
>>>locking of the buffer manager?
>>>
>>>
>
>
>
>>CVS HEAD already has an Adaptive Replacement Cache (ARC) for buffer
>>replacement.
>>
>>
>
>That's irrelevant to the problem, though. Unless the ARC code uses data
>structures that are more amenable to localized locking than the old
>global buffer freelist. (Jan?)
>
> regards, tom lane
>
>

Not that I know of. The new strategy uses one shared hash table like the
old, and one buffer pool as well. It grabs the same old Bufmgr lock
during the lookup+replacement decision process, gives it up during
eventual IO, grabs it again when done with the IO. As a matter of fact,
the strategy itself does no locking at all. Like the old LRU code it
simply assumes that the buffer manager holds the lock during calls.

Jan

--

#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== JanWieck(at)Yahoo(dot)com #