From: | Manfred Spraul <manfred(at)colorfullife(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | PostgreSQL-patches <pgsql-patches(at)postgresql(dot)org> |
Subject: | Re: update i386 spinlock for hyperthreading |
Date: | 2003-12-27 21:36:36 |
Message-ID: | 3FEDFB64.8090104@colorfullife.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general pgsql-hackers pgsql-patches |
Tom Lane wrote:
>Anyway, I've committed your patch with some changes.
>
>
Thanks.
>BTW, I noticed a lot of concern in the Intel app notes about reserving
>64 or even 128 bytes for each spinlock to avoid cache line conflicts.
>That seems excessive to me (we use a lot of spinlocks for buffers), but
>perhaps it is worth looking into.
>
This recommendation usually ignored in the Linux kernel. A few very hot
spinlocks have an exclusive cacheline, but most don't.
>>Is there an easy way find out which LWLock is contended?
>>
>>
>
>Not from oprofile output, as far as I can think. I've suspected for
>some time that the BufMgrLock is a major bottleneck, but have no proof.
>
>
I'll try to write a patch that dumps the LWLock usage and ask mark to
run it.
--
Manfred
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Treat | 2003-12-27 21:57:54 | Re: Is my MySQL Gaining ? |
Previous Message | Joe Conway | 2003-12-27 21:18:18 | Re: [GENERAL] Strange permission problem regarding pg_settings |
From | Date | Subject | |
---|---|---|---|
Next Message | A E | 2003-12-27 21:57:29 | Re: Error with returning SETOF Record |
Previous Message | Tom Lane | 2003-12-27 21:19:17 | Re: Error with returning SETOF Record |
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2003-12-27 23:14:43 | Re: fork/exec patch: pre-CreateProcess finalization |
Previous Message | Tom Lane | 2003-12-27 21:11:55 | Re: update i386 spinlock for hyperthreading |