Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 21 Mar 2013 20:14:51 +0000 (UTC)
From:      Gleb Smirnoff <glebius@FreeBSD.org>
To:        src-committers@freebsd.org, svn-src-projects@freebsd.org
Subject:   svn commit: r248592 - projects/counters/share/man/man9
Message-ID:  <201303212014.r2LKEphU011663@svn.freebsd.org>

next in thread | raw e-mail | index | archive | help
Author: glebius
Date: Thu Mar 21 20:14:51 2013
New Revision: 248592
URL: http://svnweb.freebsd.org/changeset/base/248592

Log:
  Document UMA_ZONE_PCPU.

Modified:
  projects/counters/share/man/man9/zone.9

Modified: projects/counters/share/man/man9/zone.9
==============================================================================
--- projects/counters/share/man/man9/zone.9	Thu Mar 21 19:58:25 2013	(r248591)
+++ projects/counters/share/man/man9/zone.9	Thu Mar 21 20:14:51 2013	(r248592)
@@ -153,6 +153,19 @@ See
 .Fn uma_find_refcnt .
 .It Dv UMA_ZONE_NODUMP
 Pages belonging to the zone will not be included into mini-dumps.
+.It Dv UMA_ZONE_PCPU
+An allocation from zone would have
+.Va mp_ncpu
+shadow copies, that are privately assigned to CPUs.
+A CPU can address its private copy using base allocation address plus
+multiple of current CPU id and
+.Fn sizeof "struct pcpu" :
+.Bd -literal -offset indent
+critical_enter();
+foo_pcpu = (foo *)((char *)foo_base + sizeof(struct pcpu) * curcpu);
+/* do something with foo_pcpu */
+critical_exit();
+.Ed
 .It Dv UMA_ZONE_OFFPAGE
 By default book-keeping of items within a slab is done in the slab page itself.
 This flag explicitly tells subsystem that book-keeping structure should be



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201303212014.r2LKEphU011663>