| Message ID | 20210120061235.148637-3-ying.huang@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | numa balancing: Migrate on fault among multiple bound nodes | expand |
Hi Huang Ying, On 1/20/21 7:12 AM, Huang Ying wrote: > Signed-off-by: "Huang, Ying" <ying.huang@intel.com> > Cc: "Alejandro Colomar" <alx.mailinglists@gmail.com> Sorry, for the confusion. I have a different email for reading lists. I use alx.manpages@ for everything, and alx.mailinglists@ just for reading lists, but sometimes, when I answer emails not sent to me, I forget to change the reply address, and you see that address (which I intended to be readonly). Please, use alx.manpages@gmail.com, or your mail might get lost between many list emails ;) > --- > man2/set_mempolicy.2 | 22 ++++++++++++++++++++++ > 1 file changed, 22 insertions(+) > > diff --git a/man2/set_mempolicy.2 b/man2/set_mempolicy.2 > index 68011eecb..fa64a1820 100644 > --- a/man2/set_mempolicy.2 > +++ b/man2/set_mempolicy.2 > @@ -113,6 +113,22 @@ A nonempty > .I nodemask > specifies node IDs that are relative to the set of > node IDs allowed by the process's current cpuset. > +.TP > +.BR MPOL_F_NUMA_BALANCING " (since Linux 5.12)" > +When > +.I mode > +is > +.BR MPOL_BIND , > +enable the kernel NUMA balancing for the task if it is supported by > +the kernel. > +If the flag isn't supported by the kernel, or is used with > +.I mode > +other than > +.BR MPOL_BIND , > +return \-1 and > +.I errno > +is set to > +.BR EINVAL . The wording here is a bit weird: [return // is set]. It would be better as [return // set] or [returns // sets] or [is returned // is set]. The same page, has: [ RETURN VALUE On success, set_mempolicy() returns 0; on error, -1 is re‐ turned and errno is set to indicate the error. ] so I'd use the latter for consistency. > .PP > .I nodemask > points to a bit mask of node IDs that contains up to > @@ -293,6 +309,12 @@ argument specified both > .B MPOL_F_STATIC_NODES > and > .BR MPOL_F_RELATIVE_NODES . > +Or, the > +.B MPOL_F_NUMA_BALANCING > +isn't supported by the kernel, or is used with > +.I mode > +other than > +.BR MPOL_BIND . > .TP > .B ENOMEM > Insufficient kernel memory was available. > Other than that, it's good for me. Thanks, Alex Just a reminder for myself (please ignore it): - Break EINVAL into multiple paragraphs. - (Maybe) reorder lists to be in alphabetical order.
"Alejandro Colomar (man-pages)" <alx.manpages@gmail.com> writes: > Hi Huang Ying, > > On 1/20/21 7:12 AM, Huang Ying wrote: >> Signed-off-by: "Huang, Ying" <ying.huang@intel.com> >> Cc: "Alejandro Colomar" <alx.mailinglists@gmail.com> > > Sorry, for the confusion. > I have a different email for reading lists. > I use alx.manpages@ for everything, > and alx.mailinglists@ just for reading lists, but sometimes, > when I answer emails not sent to me, > I forget to change the reply address, > and you see that address (which I intended to be readonly). > > Please, use alx.manpages@gmail.com, > or your mail might get lost between many list emails ;) Got it! >> --- >> man2/set_mempolicy.2 | 22 ++++++++++++++++++++++ >> 1 file changed, 22 insertions(+) >> >> diff --git a/man2/set_mempolicy.2 b/man2/set_mempolicy.2 >> index 68011eecb..fa64a1820 100644 >> --- a/man2/set_mempolicy.2 >> +++ b/man2/set_mempolicy.2 >> @@ -113,6 +113,22 @@ A nonempty >> .I nodemask >> specifies node IDs that are relative to the set of >> node IDs allowed by the process's current cpuset. >> +.TP >> +.BR MPOL_F_NUMA_BALANCING " (since Linux 5.12)" >> +When >> +.I mode >> +is >> +.BR MPOL_BIND , >> +enable the kernel NUMA balancing for the task if it is supported by >> +the kernel. >> +If the flag isn't supported by the kernel, or is used with >> +.I mode >> +other than >> +.BR MPOL_BIND , >> +return \-1 and >> +.I errno >> +is set to >> +.BR EINVAL . > > The wording here is a bit weird: > [return // is set]. It would be better as > [return // set] or [returns // sets] or [is returned // is set]. > > The same page, has: > > [ > RETURN VALUE > On success, set_mempolicy() returns 0; on error, -1 is re‐ > turned and errno is set to indicate the error. > ] > > so I'd use the latter for consistency. Sure. Will change this. Thanks for comments! Best Regards, Huang, Ying >> .PP >> .I nodemask >> points to a bit mask of node IDs that contains up to >> @@ -293,6 +309,12 @@ argument specified both >> .B MPOL_F_STATIC_NODES >> and >> .BR MPOL_F_RELATIVE_NODES . >> +Or, the >> +.B MPOL_F_NUMA_BALANCING >> +isn't supported by the kernel, or is used with >> +.I mode >> +other than >> +.BR MPOL_BIND . >> .TP >> .B ENOMEM >> Insufficient kernel memory was available. >> > > Other than that, it's good for me. > > Thanks, > > Alex > > Just a reminder for myself (please ignore it): > - Break EINVAL into multiple paragraphs. > - (Maybe) reorder lists to be in alphabetical order.
diff --git a/man2/set_mempolicy.2 b/man2/set_mempolicy.2 index 68011eecb..fa64a1820 100644 --- a/man2/set_mempolicy.2 +++ b/man2/set_mempolicy.2 @@ -113,6 +113,22 @@ A nonempty .I nodemask specifies node IDs that are relative to the set of node IDs allowed by the process's current cpuset. +.TP +.BR MPOL_F_NUMA_BALANCING " (since Linux 5.12)" +When +.I mode +is +.BR MPOL_BIND , +enable the kernel NUMA balancing for the task if it is supported by +the kernel. +If the flag isn't supported by the kernel, or is used with +.I mode +other than +.BR MPOL_BIND , +return \-1 and +.I errno +is set to +.BR EINVAL . .PP .I nodemask points to a bit mask of node IDs that contains up to @@ -293,6 +309,12 @@ argument specified both .B MPOL_F_STATIC_NODES and .BR MPOL_F_RELATIVE_NODES . +Or, the +.B MPOL_F_NUMA_BALANCING +isn't supported by the kernel, or is used with +.I mode +other than +.BR MPOL_BIND . .TP .B ENOMEM Insufficient kernel memory was available.
Signed-off-by: "Huang, Ying" <ying.huang@intel.com> Cc: "Alejandro Colomar" <alx.mailinglists@gmail.com> --- man2/set_mempolicy.2 | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)