Lee, On Thu, Aug 7, 2008 at 6:17 PM, Lee Schermerhorn <Lee.Schermerhorn@xxxxxx> wrote: > On Thu, 2008-08-07 at 15:35 +0200, Michael Kerrisk wrote: >> Lee, >> >> Lee Schermerhorn wrote: >> > Add [minimal?] discussion of mode flags to mbind(2). >> >> I can make the changes required below, if you'd just clarify/ACK. >> >> > Signed-off-by: Lee Schermerhorn <lee.schermerhorn@xxxxxx> >> > >> > man2/mbind.2 | 45 ++++++++++++++++++++++++++++++++++++++++++++- >> > 1 files changed, 44 insertions(+), 1 deletion(-) >> > >> > Index: man-pages-3.05/man2/mbind.2 >> > =================================================================== >> > --- man-pages-3.05.orig/man2/mbind.2 2008-07-29 16:52:23.000000000 -0400 >> > +++ man-pages-3.05/man2/mbind.2 2008-07-29 16:53:48.000000000 -0400 >> > @@ -120,6 +120,30 @@ require the caller to specify via the >> > argument, >> > the node or nodes to which the mode applies. >> > >> > +The >> > +.I mode >> > +argument may also include an optional >> > +.IR "mode flag ". >> > +The supported >> > +.I "mode flags" >> > +are: >> > +.TP >> > +.B MPOL_F_STATIC_NODES >> >> Kernel 2.6.26, right? > > Yes. > >> >> > +A non-empty >> > +.I nodemask >> > +specifies physical node ids. >> > +Linux does will not remap the >> > +.I nodemask >> > +when the task moves to a different cpuset context, >> >> task? --> thread/ process. > > Discussed this in response to previous comments. I'd prefer 'process' > with a note that cpusets can be per thread. I've applied this patch for 3.07, and used "process" Cheers, Michael. >> > +nor when the set of nodes allowed by the task's >> >> See previous. >> >> > +current cpuset context changes. >> > +.TP >> > +.B MPOL_F_RELATIVE_NODES >> >> Kernel 2.6.26, right? > > Yes. > >> >> > +A non-empty >> > +.I nodemask >> > +specifies node ids that are relative to the set of >> > +node ids allowed by the task's current cpuset. >> >> task? --> thread/ process. >> >> > +.PP >> > .I nodemask >> > points to a bitmask of nodes containing up to >> > .I maxnode >> > @@ -139,11 +163,24 @@ is zero, >> > the >> > .I nodemask >> > argument is ignored. >> > + >> > Where a >> > .I nodemask >> > is required, it must contain at least one node that is on-line, >> > -allowed by the process' current cpuset context, >> > +allowed by the task's current cpuset context >> > +[unless the >> > +.B MPOL_F_STATIC_NODES >> > +mode flag is specified], >> > and contains memory. >> > +If the >> > +.B MPOL_F_STATIC_NODES >> > +is set in >> > +.I mode >> > +and a required >> > +.I nodemask >> > +contains no nodes that are allowed by the task's current cpuset context, >> > +the memory policy reverts to >> > +.IR "local allocation" . >> > The >> > .B MPOL_DEFAULT >> > @@ -319,6 +356,12 @@ Or, none of the node IDs specified by >> > .I nodemask >> > are on-line and allowed by the process' current cpuset context, >> > or none of the specified nodes contain memory. >> > +Or, the >> > +.I mode >> > +argument specified both >> > +.B MPOL_F_STATIC_NODES >> > +and >> > +.BR MPOL_F_RELATIVE_NODES . >> > .TP >> > .B EIO >> > .B MPOL_MF_STRICT >> >> Cheers, >> >> Michael >> > > -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ man-pages online: http://www.kernel.org/doc/man-pages/online_pages.html Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html -- To unsubscribe from this list: send the line "unsubscribe linux-numa" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
