Hi Svetly, On Mon, Mar 18, 2024 at 09:32:08AM -0700, Svetly Todorov wrote: > --- > Adding documentation for the new MPOL_WEIGHTED_INTERLEAVE mode in the same > manpages that mention MPOL_INTERLEAVE; namely, mbind(2), set_mempolicy(2), > and get_mempolicy(2). > > Descriptions were based on the changes introduced in this patch: > https://lore.kernel.org/all/20240202170238.90004-4-gregory.price@xxxxxxxxxxxx/ > > Which was upstreamed to 6.9 here: > https://lore.kernel.org/linux-mm/20240313200532.34e4cff216acd3db8def4637@xxxxxxxxxxxxxxxxxxxx/ > > To: alx@xxxxxxxxxx > Cc: linux-man@xxxxxxxxxxxxxxx > Cc: gregory.price@xxxxxxxxxxxx > Cc: ying.huang@xxxxxxxxx > Reviewed-by: "Huang, Ying" <ying.huang@xxxxxxxxx> > Signed-off-by: Svetly Todorov <svetly.todorov@xxxxxxxxxxxx> > > Changes in v5: > - document kernel version where WEIGHTED_INTERLEAVE was added > - Link to v4: https://lore.kernel.org/r/20240316-weighted_interleave-v4-1-6ab516f12ced@xxxxxxxxxxxx > > Changes in v4: > - BR -> B in get_mempolicy > - use roman for ellipses; format using \| > - Link to v3: https://lore.kernel.org/r/20240315-weighted_interleave-v3-1-416a1ab01524@xxxxxxxxxxxx > > Changes in v3: > - italicize paths > - Link to v2: https://lore.kernel.org/r/20240315-weighted_interleave-v2-1-b742a48750b0@xxxxxxxxxxxx > > Changes in v2: > - make flag documentation implementation-agnostic > - Link to v1: https://lore.kernel.org/r/20240314-weighted_interleave-v1-1-ce85d64db0d4@xxxxxxxxxxxx > --- > man2/get_mempolicy.2 | 12 ++++++++++-- > man2/mbind.2 | 18 ++++++++++++++++++ > man2/set_mempolicy.2 | 18 ++++++++++++++++++ > 3 files changed, 46 insertions(+), 2 deletions(-) > > diff --git a/man2/get_mempolicy.2 b/man2/get_mempolicy.2 > index 5248f04ba..00ce287c9 100644 > --- a/man2/get_mempolicy.2 > +++ b/man2/get_mempolicy.2 > @@ -137,7 +137,11 @@ specifies > but not > .BR MPOL_F_ADDR , > and the thread's current policy is > -.BR MPOL_INTERLEAVE , > +.B MPOL_INTERLEAVE > +or > +.B MPOL_WEIGHTED_INTERLEAVE > +(available since Linux 6.9), For consistency, I try to use the same phrase everywhere: (since Linux x.y) But, in simple mentions, we usually don't specify when a flag was added. That's already documented in the proper documentation of the flag. So here, I'd just remove that (and the comment below). > +.\" commit fa3bea4e1f8202d787709b7e3654eb0a99aed758 > then > .BR get_mempolicy () > will return in the location pointed to by a non-NULL > @@ -206,7 +210,11 @@ specified > but not > .B MPOL_F_ADDR > and the current thread policy is not > -.BR MPOL_INTERLEAVE . > +.B MPOL_INTERLEAVE > +or I prefer "nor". > +.B MPOL_WEIGHTED_INTERLEAVE > +(added in Linux 6.9). Same here. > +.\" commit fa3bea4e1f8202d787709b7e3654eb0a99aed758 > Or, > .I flags > specified > diff --git a/man2/mbind.2 b/man2/mbind.2 > index b0e961f9c..6f2449946 100644 > --- a/man2/mbind.2 > +++ b/man2/mbind.2 > @@ -105,6 +105,7 @@ argument must specify one of > .BR MPOL_DEFAULT , > .BR MPOL_BIND , > .BR MPOL_INTERLEAVE , > +.BR MPOL_WEIGHTED_INTERLEAVE , > .BR MPOL_PREFERRED , > or > .B MPOL_LOCAL > @@ -243,6 +244,23 @@ at least 1\ MB or bigger with a fairly uniform access pattern. > Accesses to a single page of the area will still be limited to > the memory bandwidth of a single node. > .TP > +.BR MPOL_WEIGHTED_INTERLEAVE " (since Linux 6.9)" > +.\" commit fa3bea4e1f8202d787709b7e3654eb0a99aed758 Here is where we usually document the kernel versions. So here (and in set_mempolicy.2) is good, and enough, I think. Have a lovely night! Alex > +This mode interleaves page allocations across the nodes specified in > +.I nodemask > +according to the weights in > +.IR /sys/kernel/mm/mempolicy/weighted_interleave . > +For example, if bits 0, 2, and 5 are raised in > +.IR nodemask , > +and the contents of > +.IR /sys/kernel/mm/mempolicy/weighted_interleave/node0 , > +.IR /sys/ .\|.\|. /node2 , > +and > +.IR /sys/ .\|.\|. /node5 > +are 4, 7, and 9, respectively, > +then pages in this region will be allocated on nodes 0, 2, and 5 > +in a 4:7:9 ratio. > +.TP > .B MPOL_PREFERRED > This mode sets the preferred node for allocation. > The kernel will try to allocate pages from this > diff --git a/man2/set_mempolicy.2 b/man2/set_mempolicy.2 > index fc3ad9df8..3c08b6bf0 100644 > --- a/man2/set_mempolicy.2 > +++ b/man2/set_mempolicy.2 > @@ -63,6 +63,7 @@ argument must specify one of > .BR MPOL_DEFAULT , > .BR MPOL_BIND , > .BR MPOL_INTERLEAVE , > +.BR MPOL_WEIGHTED_INTERLEAVE , > .BR MPOL_PREFERRED , > or > .B MPOL_LOCAL > @@ -199,6 +200,23 @@ the memory bandwidth of a single node. > .\" To be effective the memory area should be fairly large, > .\" at least 1 MB or bigger. > .TP > +.BR MPOL_WEIGHTED_INTERLEAVE " (since Linux 6.9)" > +.\" commit fa3bea4e1f8202d787709b7e3654eb0a99aed758 > +This mode interleaves page allocations across the nodes specified in > +.I nodemask > +according to the weights in > +.IR /sys/kernel/mm/mempolicy/weighted_interleave . > +For example, if bits 0, 2, and 5 are raised in > +.IR nodemask , > +and the contents of > +.IR /sys/kernel/mm/mempolicy/weighted_interleave/node0 , > +.IR /sys/ .\|.\|. /node2 , > +and > +.IR /sys/ .\|.\|. /node5 > +are 4, 7, and 9, respectively, > +then pages in this region will be allocated on nodes 0, 2, and 5 > +in a 4:7:9 ratio. > +.TP > .B MPOL_PREFERRED > This mode sets the preferred node for allocation. > The kernel will try to allocate pages from this node first > > --- > base-commit: a4ea5f76215dcf1d8d017eb65ab12919ccf09f3f > change-id: 20240313-weighted_interleave-e8141ed754f9 > > Best regards, > -- > Svetly Todorov <svetly.todorov@xxxxxxxxxxxx> > > -- <https://www.alejandro-colomar.es/> Looking for a remote C programming job at the moment.
Attachment:
signature.asc
Description: PGP signature
