Date: Mon, 30 Jun 2014 23:18:55 +0000 From: "Rang, Anton" <anton.rang@isilon.com> To: Konstantin Belousov <kostikbel@gmail.com>, "performance@freebsd.org" <performance@freebsd.org> Cc: "current@freebsd.org" <current@freebsd.org> Subject: RE: PostgreSQL performance on FreeBSD Message-ID: <F21EDC44C64DB34B90AF485AC3CEDD4B353982C1@MX104CL01.corp.emc.com> In-Reply-To: <20140627125613.GT93733@kib.kiev.ua> References: <20140627125613.GT93733@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
Thanks for this.
The cpu_search problem you reference came up here at Isilon as well. Here'=
s a patch which should get clang to do the "right thing" (inlining 3 specia=
lized copies of cpu_search); I haven't checked to make sure it doesn't hurt=
gcc, though.
Anton
Index: sched_ule.c
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
--- sched_ule.c (revision 268043)
+++ sched_ule.c (working copy)
@@ -622,11 +622,11 @@
for ((cpu) =3D 0; (cpu) <=3D mp_maxid; (cpu)++) \
if (CPU_ISSET(cpu, &mask))
=20
-static __inline int cpu_search(const struct cpu_group *cg, struct cpu_sear=
ch *low,
+static __always_inline int cpu_search(const struct cpu_group *cg, struct c=
pu_search *low,
struct cpu_search *high, const int match);
-int cpu_search_lowest(const struct cpu_group *cg, struct cpu_search *low);
-int cpu_search_highest(const struct cpu_group *cg, struct cpu_search *high=
);
-int cpu_search_both(const struct cpu_group *cg, struct cpu_search *low,
+int __noinline cpu_search_lowest(const struct cpu_group *cg, struct cpu_se=
arch *low);
+int __noinline cpu_search_highest(const struct cpu_group *cg, struct cpu_s=
earch *high);
+int __noinline cpu_search_both(const struct cpu_group *cg, struct cpu_sear=
ch *low,
struct cpu_search *high);
=20
/*
@@ -640,7 +640,7 @@
* match argument. It is reduced to the minimum set for each case. It is
* also recursive to the depth of the tree.
*/
-static __inline int
+static __always_inline int
cpu_search(const struct cpu_group *cg, struct cpu_search *low,
struct cpu_search *high, const int match)
{
-----Original Message-----
From: owner-freebsd-current@freebsd.org [mailto:owner-freebsd-current@freeb=
sd.org] On Behalf Of Konstantin Belousov
Sent: Friday, June 27, 2014 7:56 AM
To: performance@freebsd.org
Cc: current@freebsd.org
Subject: PostgreSQL performance on FreeBSD
Hi,
I did some measurements and hacks to see about the performance and scalabil=
ity of PostgreSQL 9.3 on FreeBSD, sponsored by The FreeBSD Foundation.
The results are described in https://kib.kiev.ua/kib/pgsql_perf.pdf.
The uncommitted patches, referenced in the article, are available as https:=
//kib.kiev.ua/kib/pig1.patch.txt
https://kib.kiev.ua/kib/patch-2
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?F21EDC44C64DB34B90AF485AC3CEDD4B353982C1>