From owner-cvs-all Fri Aug 13 12:59:26 1999 Delivered-To: cvs-all@freebsd.org Received: from gate.keisu.t.u-tokyo.ac.jp (ns06.t.u-tokyo.ac.jp [133.11.68.1]) by hub.freebsd.org (Postfix) with SMTP id 4846214E0B for ; Fri, 13 Aug 1999 12:59:16 -0700 (PDT) (envelope-from simokawa@sat.t.u-tokyo.ac.jp) Received: (qmail 89353 invoked from network); 13 Aug 1999 19:57:48 -0000 Received: from sylph.sat.t.u-tokyo.ac.jp (10.6.1.20) by ns06.t.u-tokyo.ac.jp with SMTP; 13 Aug 1999 19:57:48 -0000 Received: from ett.sat.t.u-tokyo.ac.jp (ett.sat.t.u-tokyo.ac.jp [10.6.1.30]) by sylph.sat.t.u-tokyo.ac.jp (Postfix) with ESMTP id 109B02DAA9; Sat, 14 Aug 1999 04:57:46 +0900 (JST) Received: from ett.sat.t.u-tokyo.ac.jp by ett.sat.t.u-tokyo.ac.jp (8.9.3/sat-V0.6) id EAA36442; Sat, 14 Aug 1999 04:57:46 +0900 (JST) Date: Sat, 14 Aug 1999 04:57:45 +0900 Message-ID: <14260.30905.562825.38243A@ett.sat.t.u-tokyo.ac.jp> From: Hidetoshi Shimokawa To: alc@cs.rice.edu Cc: rgrimes@gndrsh.aac.dev.com, cvs-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/vm vm_page.h In-Reply-To: In your message of "Fri, 13 Aug 1999 14:01:28 -0500" <19990813140128.C27982@cs.rice.edu> References: <199908122116.OAA56955@freefall.freebsd.org> <199908131458.HAA01346@gndrsh.aac.dev.com> <19990813140128.C27982@cs.rice.edu> User-Agent: Wanderlust/1.0.0 (Kokomo) SEMI/1.13.3 (Komaiko) FLIM/1.12.5 (Hirahata) MULE XEmacs/21.2 (beta13) (Demeter) (i386-unknown-freebsd3.1) X-Face: OE([KxWyJI0r[R~S/>7ia}SJ)i%a,$-9%7{*yihQk|]gl}2p#"oXmX/fT}Bn7:#j7i14gu$ jgR\S*&C3R/pJX wrote: > > On Fri, Aug 13, 1999 at 07:58:43AM -0700, Rodney W. Grimes wrote: > > > alc 1999/08/12 14:16:55 PDT > > > > > > Modified files: > > > sys/vm vm_page.h > > > Log: > > > Make the default page coloring parameters match a (non-Xeon) Pentium II/III. > > > > > > This setting is also acceptable for Celerons and Pentium Pros > > > with less than 1MB L2 caches. > > > > > > Note: PQ_L2_SIZE is a misnomer. The correct number of colors is > > > a function of the cache's degree of associativity as well as its size. > > > > Yea... has anyone looked at creating the right color scheme for the > > AMD K6-III 4-way L2 cache? If access to the correct hardware is an issue > > I should be able to help with that. > > > > In theory, a Celeron needs only 8 colors, a Pentium Pro with 256KB cache > needs only 16 colors, and a K6-III is complicated. :-) > > The answer for a K6-III depends on whether the coloring wants > or needs to take into account the direct-map L3 cache. If it doesn't, > then 16 colors should suffice. If it does, then 128 (512KB L3) or > 256 (1MB L3) are necessary. > > Do you know if the inclusion property holds between the on-chip L2 > and the motherboard-level L3? > > As a rule, using a few too many colors is okay, but using too few > colors is no good. As for 21164 Alpha, it has 96KB 3-way L2 cache. So it needs only 4 (= 96KB / 8KB / 3) colors. My Alpha box has 4MB direct-map L3 cache and it needs 4MB / 8KB = 512 colors. I have the following entries in my vm_page.h JFYI, Index: vm_page.h =================================================================== RCS file: /pub/FreeBSD-CVS/src/sys/vm/vm_page.h,v retrieving revision 1.65 diff -u -r1.65 vm_page.h --- vm_page.h 1999/08/12 21:16:53 1.65 +++ vm_page.h 1999/08/13 19:56:28 @@ -150,6 +150,12 @@ /* Each of PQ_FREE, and PQ_CACHE have PQ_HASH_SIZE entries */ /* Define one of the following */ +#if defined(PQ_JUMBOCACHE) +#define PQ_PRIME1 31 +#define PQ_PRIME2 23 +#define PQ_L2_SIZE 512 /* A number of colors opt for 4M cache / 8K page */ +#endif + #if defined(PQ_HUGECACHE) #define PQ_PRIME1 31 /* Prime number somewhat less than PQ_HASH_SIZE */ #define PQ_PRIME2 23 /* Prime number somewhat less than PQ_HASH_SIZE */ -- /\ Hidetoshi Shimokawa \/ simokawa@sat.t.u-tokyo.ac.jp PGP public key: finger -l simokawa@sat.t.u-tokyo.ac.jp To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message