Re: race on multi-processor solaris
From: Casper H.S. Dik (Casper.Dik_at_Sun.COM)
Date: 12/02/03
- Next message: SenderX: "Re: race on multi-processor solaris"
- Previous message: Michael Laajanen: "Re: NC03-Q4 announcements webcast"
- In reply to: SenderX: "Re: race on multi-processor solaris"
- Next in thread: SenderX: "Re: race on multi-processor solaris"
- Reply: SenderX: "Re: race on multi-processor solaris"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 02 Dec 2003 22:59:10 GMT
"SenderX" <xxx@xxx.xxx> writes:
>To clarify...
>The lock-based test WILL NOT USE ANY lock-free optimizations.
>The test is to compare lock-based tech with lock-free tech, so lock-based
>has to be true to itself.
I'm hoping you're using the "system provided locking
primitives"? Not something you wrote yourself?
I'd prefer it if it was:
- C
- "CAS" code called and specified as "C" functions
(which obviously won't work) and asm "proper"
implementations in a separate file
Casper
-- Expressed in this posting are my opinions. They are in no way related to opinions held by my employer, Sun Microsystems. Statements on Sun products included here are not gospel and may be fiction rather than truth.
- Next message: SenderX: "Re: race on multi-processor solaris"
- Previous message: Michael Laajanen: "Re: NC03-Q4 announcements webcast"
- In reply to: SenderX: "Re: race on multi-processor solaris"
- Next in thread: SenderX: "Re: race on multi-processor solaris"
- Reply: SenderX: "Re: race on multi-processor solaris"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|