On Wed, Oct 27, 2021 at 4:36 PM Mike Kravetz <mike.kravetz@xxxxxxxxxx> wrote:
>
> On 10/20/21 12:09 PM, Mina Almasry wrote:
> > For hugetlb backed jobs/VMs it's critical to understand the numa
> > information for the memory backing these jobs to deliver optimal
> > performance.
> >
> > Currently this techinically can be queried from /proc/self/numa_maps, but
>
> technically
>
> > there are significant issues with that. Namely:
> > 1. Memory can be mapped on unmapped.
> > 2. numa_maps are per process and need to be aggregaged across all
>
> aggregated
>
> > proceses in the cgroup. For shared memory this is more involved as
>
> processes
>
Very sorry for the typos.
> > the userspace needs to make sure it doesn't double count shared
> > mappings.
> > 3. I believe querying numa_maps needs to hold the mmap_lock which adds
> > to the contention on this lock.
> >
> > For these reasons I propose simply adding hugetlb.*.numa_stat file,
> > which shows the numa information of the cgroup similarly to
> > memory.numa_stat.
> >
> > On cgroup-v2:
> > cat /dev/cgroup/memory/test/hugetlb.2MB.numa_stat
>
> I know that cgroup can be mounted anywhere, but I think the conventional
> locations are mentioned in the man page. Why not use them?
>
> v2 '/sys/fs/cgroup/unified/test/hugetlb.2MB.numa_stat'
>
> > total=2097152 N0=2097152 N1=0
> >
> > On cgroup-v1:
> > cat /dev/cgroup/memory/test/hugetlb.2MB.numa_stat
>
> v1 '/sys/fs/cgroup/hugetlb/test/hugetlb.2MB.numa_stat'
>
> > total=2097152 N0=2097152 N1=0
> > hierarichal_total=2097152 N0=2097152 N1=0
> >
> > This patch was tested manually by allocating hugetlb memory and querying
> > the hugetlb.*.numa_stat file of the cgroup and its parents.
>
> I have no objections to adding this functionality, and do not see any
> blocking issues in hugetlb code. However, it would be GREAT if someone
> more familiar/experienced with cgroups would comment. My cgroup
> experience is very limited.
>
I will send V2 to Shakeel as well from our team and ask him to take a
look and he has more than enough experience to review. If anyone else
reading with cgroups knowledge can Review/Ack that would be great.
It's possible I'm a bit off the mark here, but FWIW I don't think
there is much that is continuous or ambiguous here. For
memory.numa_stat it's a bit nuanced because there is anon, rss, shmem,
etc.. but for hugetlb we just have hugetlb memory and the only care
needed is hierarchical vs not in cgroups v1.
> > 
> > Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
> > Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> > Cc: Shuah Khan <shuah@xxxxxxxxxx>
> > Cc: Miaohe Lin <linmiaohe@xxxxxxxxxx>
> > Cc: Oscar Salvador <osalvador@xxxxxxx>
> > Cc: Michal Hocko <mhocko@xxxxxxxx>
> > Cc: Muchun Song <songmuchun@xxxxxxxxxxxxx>
> > Cc: David Rientjes <rientjes@xxxxxxxxxx>
> > Cc: Jue Wang <juew@xxxxxxxxxx>
> > Cc: Yang Yao <ygyao@xxxxxxxxxx>
> > Cc: Joanna Li <joannali@xxxxxxxxxx>
> > Cc: Cannon Matthews <cannonmatthews@xxxxxxxxxx>
> > Cc: linux-mm@xxxxxxxxx
> > Cc: linux-kernel@xxxxxxxxxxxxxxx
> >
> > Signed-off-by: Mina Almasry <almasrymina@xxxxxxxxxx>
> >
> > ---
> >
> > Changes in v2:
> > - Fix warning Reported-by: kernel test robot <lkp@xxxxxxxxx>
> > ---
> > .../admin-guide/cgroup-v1/hugetlb.rst | 4 +
> > Documentation/admin-guide/cgroup-v2.rst | 7 ++
> > include/linux/hugetlb.h | 4 +-
> > include/linux/hugetlb_cgroup.h | 7 ++
> > mm/hugetlb_cgroup.c | 93 +++++++++++++++++--
> > .../testing/selftests/vm/write_to_hugetlbfs.c | 9 +-
> > 6 files changed, 113 insertions(+), 11 deletions(-)
> >
> > diff --git a/Documentation/admin-guide/cgroup-v1/hugetlb.rst b/Documentation/admin-guide/cgroup-v1/hugetlb.rst
> > index 338f2c7d7a1c..0fa724d82abb 100644
> > --- a/Documentation/admin-guide/cgroup-v1/hugetlb.rst
> > +++ b/Documentation/admin-guide/cgroup-v1/hugetlb.rst
> > @@ -29,12 +29,14 @@ Brief summary of control files::
> > hugetlb.<hugepagesize>.max_usage_in_bytes # show max "hugepagesize" hugetlb usage recorded
> > hugetlb.<hugepagesize>.usage_in_bytes # show current usage for "hugepagesize" hugetlb
> > hugetlb.<hugepagesize>.failcnt # show the number of allocation failure due to HugeTLB usage limit
> > + hugetlb.<hugepagesize>.numa_stat # show the numa information of the hugetlb memory charged to this cgroup
> >
> > For a system supporting three hugepage sizes (64k, 32M and 1G), the control
> > files include::
> >
> > hugetlb.1GB.limit_in_bytes
> > hugetlb.1GB.max_usage_in_bytes
> > + hugetlb.1GB.numa_stat
> > hugetlb.1GB.usage_in_bytes
> > hugetlb.1GB.failcnt
> > hugetlb.1GB.rsvd.limit_in_bytes
> > @@ -43,6 +45,7 @@ files include::
> > hugetlb.1GB.rsvd.failcnt
> > hugetlb.64KB.limit_in_bytes
> > hugetlb.64KB.max_usage_in_bytes
> > + hugetlb.64KB.numa_stat
> > hugetlb.64KB.usage_in_bytes
> > hugetlb.64KB.failcnt
> > hugetlb.64KB.rsvd.limit_in_bytes
> > @@ -51,6 +54,7 @@ files include::
> > hugetlb.64KB.rsvd.failcnt
> > hugetlb.32MB.limit_in_bytes
> > hugetlb.32MB.max_usage_in_bytes
> > + hugetlb.32MB.numa_stat
> > hugetlb.32MB.usage_in_bytes
> > hugetlb.32MB.failcnt
> > hugetlb.32MB.rsvd.limit_in_bytes
> > diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
> > index 4d8c27eca96b..8ba0d6aadd2c 100644
> > --- a/Documentation/admin-guide/cgroup-v2.rst
> > +++ b/Documentation/admin-guide/cgroup-v2.rst
> > @@ -2252,6 +2252,13 @@ HugeTLB Interface Files
> > are local to the cgroup i.e. not hierarchical. The file modified event
> > generated on this file reflects only the local events.
> >
> > + hugetlb.<hugepagesize>.numa_stat
> > + Similar to memory.numa_stat, it shows the numa information of the
> > + memory in this cgroup:
>
> hugetlb pages of <hugepagesize> in this cgroup. Only active in
> use hugetlb pages are included. The per-node values are in bytes.
>
> > +
> > + /dev/cgroup/memory/test # cat hugetlb.2MB.numa_stat
> > + total=0 N0=0 N1=0
>
> I would not include the 'example' above as nothing like this is used
> elsewhere in the file.
>
> > +
> > Misc
> > ----
> >
> > diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
> > index 1faebe1cd0ed..0445faaa636e 100644
> > --- a/include/linux/hugetlb.h
> > +++ b/include/linux/hugetlb.h
> > @@ -613,8 +613,8 @@ struct hstate {
> > #endif
> > #ifdef CONFIG_CGROUP_HUGETLB
> > /* cgroup control files */
> > - struct cftype cgroup_files_dfl[7];
> > - struct cftype cgroup_files_legacy[9];
> > + struct cftype cgroup_files_dfl[8];
> > + struct cftype cgroup_files_legacy[10];
> > #endif
> > char name[HSTATE_NAME_LEN];
> > };
> > diff --git a/include/linux/hugetlb_cgroup.h b/include/linux/hugetlb_cgroup.h
> > index c137396129db..54ff6ec68ed3 100644
> > --- a/include/linux/hugetlb_cgroup.h
> > +++ b/include/linux/hugetlb_cgroup.h
> > @@ -36,6 +36,11 @@ enum hugetlb_memory_event {
> > HUGETLB_NR_MEMORY_EVENTS,
> > };
> >
> > +struct hugetlb_cgroup_per_node {
> > + /* hugetlb usage in bytes over all hstates. */
> > + unsigned long usage[HUGE_MAX_HSTATE];
> > +};
> > +
> > struct hugetlb_cgroup {
> > struct cgroup_subsys_state css;
> >
> > @@ -57,6 +62,8 @@ struct hugetlb_cgroup {
> >
> > /* Handle for "hugetlb.events.local" */
> > struct cgroup_file events_local_file[HUGE_MAX_HSTATE];
> > +
> > + struct hugetlb_cgroup_per_node *nodeinfo[];
> > };
> >
> > static inline struct hugetlb_cgroup *
> > diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> > index 5383023d0cca..4b807292f7e8 100644
> > --- a/mm/hugetlb_cgroup.c
> > +++ b/mm/hugetlb_cgroup.c
> > @@ -92,6 +92,7 @@ static void hugetlb_cgroup_init(struct hugetlb_cgroup *h_cgroup,
> > struct hugetlb_cgroup *parent_h_cgroup)
> > {
> > int idx;
> > + int node;
> >
> > for (idx = 0; idx < HUGE_MAX_HSTATE; idx++) {
> > struct page_counter *fault_parent = NULL;
> > @@ -124,6 +125,15 @@ static void hugetlb_cgroup_init(struct hugetlb_cgroup *h_cgroup,
> > limit);
> > VM_BUG_ON(ret);
> > }
> > +
> > + for_each_node(node) {
> > + /* Set node_to_alloc to -1 for offline nodes. */
> > + int node_to_alloc =
> > + node_state(node, N_NORMAL_MEMORY) ? node : -1;
> > + h_cgroup->nodeinfo[node] =
> > + kzalloc_node(sizeof(struct hugetlb_cgroup_per_node),
> > + GFP_KERNEL, node_to_alloc);
>
> Do you need to handle kzalloc_node failure here?
>
Will do.
> alloc_mem_cgroup_per_node_info provides similar functionality and has
> the following comment.
>
> * TODO: this routine can waste much memory for nodes which will
> * never be onlined. It's better to use memory hotplug callback
> * function.
>
So the memory allocated per node in total is (HUGE_MAX_HSTATE *
unsigned long * num_css_on_the system). HUGE_MAX_HSTATE is 2 on x86.
This is significant, but to address this comment I have to complicate
the hugetlb_cgroup code quite a bit so I thought I'd check with you if
you think it's still OK/worth it. slub.c implements these changes
(slab_memory_callback) and they are:
- When creating a new hugetlb_cgroup, I create per_node data for only
online nodes.
- On node online I need to loop over all existing hugetlb_cgroups and
allocate per_node data. I need rcu_read_lock() here and below.
- On node offline I need to loop over all existing hugetlb_cgroups and
free per_node data.
- If I follow the slub example, I need a lock to synchronize
onlining/offlining with references to per_node data in commit_charge()
uncharge_page() and show_numa_stats().
I don't think it's worth it TBH, but I'm happy to oblige if you think so.
> Extra credit if you do this here and in alloc_mem_cgroup_per_node_info. :)
>
> > + }
> > }
> >
> > static struct cgroup_subsys_state *
> > @@ -132,7 +142,10 @@ hugetlb_cgroup_css_alloc(struct cgroup_subsys_state *parent_css)
> > struct hugetlb_cgroup *parent_h_cgroup = hugetlb_cgroup_from_css(parent_css);
> > struct hugetlb_cgroup *h_cgroup;
> >
> > - h_cgroup = kzalloc(sizeof(*h_cgroup), GFP_KERNEL);
> > + unsigned int size =
> > + sizeof(*h_cgroup) +
> > + MAX_NUMNODES * sizeof(struct hugetlb_cgroup_per_node *);
> > + h_cgroup = kzalloc(size, GFP_KERNEL);
> > if (!h_cgroup)
> > return ERR_PTR(-ENOMEM);
> >
> > @@ -292,7 +305,9 @@ static void __hugetlb_cgroup_commit_charge(int idx, unsigned long nr_pages,
> > return;
> >
> > __set_hugetlb_cgroup(page, h_cg, rsvd);
> > - return;
> > + if (!rsvd && h_cg)
> > + h_cg->nodeinfo[page_to_nid(page)]->usage[idx] += nr_pages
> > + << PAGE_SHIFT;
> > }
> >
> > void hugetlb_cgroup_commit_charge(int idx, unsigned long nr_pages,
> > @@ -331,7 +346,9 @@ static void __hugetlb_cgroup_uncharge_page(int idx, unsigned long nr_pages,
> >
> > if (rsvd)
> > css_put(&h_cg->css);
> > -
> > + else
> > + h_cg->nodeinfo[page_to_nid(page)]->usage[idx] -= nr_pages
> > + << PAGE_SHIFT;
> > return;
> > }
> >
> > @@ -421,6 +438,56 @@ enum {
> > RES_RSVD_FAILCNT,
> > };
> >
> > +static int hugetlb_cgroup_read_numa_stat(struct seq_file *seq, void *dummy)
> > +{
> > + int nid;
> > + struct cftype *cft = seq_cft(seq);
> > + int idx = MEMFILE_IDX(cft->private);
> > + bool legacy = MEMFILE_ATTR(cft->private);
> > + struct hugetlb_cgroup *h_cg = hugetlb_cgroup_from_css(seq_css(seq));
> > + struct cgroup_subsys_state *css;
> > + unsigned long usage;
> > +
> > + if (legacy) {
> > + /* Add up usage across all nodes for the non-hierarchical total. */
> > + usage = 0;
> > + for_each_node_state(nid, N_MEMORY)
> > + usage += h_cg->nodeinfo[nid]->usage[idx];
> > + seq_printf(seq, "total=%lu", usage);
> > +
> > + /* Simply print the per-node usage for the non-hierarchical total. */
> > + for_each_node_state(nid, N_MEMORY)
> > + seq_printf(seq, " N%d=%lu", nid,
> > + h_cg->nodeinfo[nid]->usage[idx]);
> > + seq_putc(seq, '\n');
> > + }
> > +
> > + /* The hierarchical total is pretty much the value recorded by the
> > + * counter, so use that.
>
> 'pretty much' ... so use that. Here is one place in particular where
> input from someone with more cgroup experience would be appreciated.
>
> --
> Mike Kravetz
>
> > + */
> > + seq_printf(seq, "%stotal=%lu", legacy ? "hierarichal_" : "",
> > + page_counter_read(&h_cg->hugepage[idx]) * PAGE_SIZE);
> > +
> > + /* For each node, transverse the css tree to obtain the hierarichal
> > + * node usage.
> > + */
> > + for_each_node_state(nid, N_MEMORY) {
> > + usage = 0;
> > + rcu_read_lock();
> > + css_for_each_descendant_pre(css, &h_cg->css) {
> > + usage += hugetlb_cgroup_from_css(css)
> > + ->nodeinfo[nid]
> > + ->usage[idx];
> > + }
> > + rcu_read_unlock();
> > + seq_printf(seq, " N%d=%lu", nid, usage);
> > + }
> > +
> > + seq_putc(seq, '\n');
> > +
> > + return 0;
> > +}
> > +
> > static u64 hugetlb_cgroup_read_u64(struct cgroup_subsys_state *css,
> > struct cftype *cft)
> > {
> > @@ -654,8 +721,14 @@ static void __init __hugetlb_cgroup_file_dfl_init(int idx)
> > cft->seq_show = hugetlb_cgroup_read_u64_max;
> > cft->flags = CFTYPE_NOT_ON_ROOT;
> >
> > - /* Add the events file */
> > + /* Add the numa stat file */
> > cft = &h->cgroup_files_dfl[4];
> > + snprintf(cft->name, MAX_CFTYPE_NAME, "%s.numa_stat", buf);
> > + cft->seq_show = hugetlb_cgroup_read_numa_stat;
> > + cft->flags = CFTYPE_NOT_ON_ROOT;
> > +
> > + /* Add the events file */
> > + cft = &h->cgroup_files_dfl[5];
> > snprintf(cft->name, MAX_CFTYPE_NAME, "%s.events", buf);
> > cft->private = MEMFILE_PRIVATE(idx, 0);
> > cft->seq_show = hugetlb_events_show;
> > @@ -663,7 +736,7 @@ static void __init __hugetlb_cgroup_file_dfl_init(int idx)
> > cft->flags = CFTYPE_NOT_ON_ROOT;
> >
> > /* Add the events.local file */
> > - cft = &h->cgroup_files_dfl[5];
> > + cft = &h->cgroup_files_dfl[6];
> > snprintf(cft->name, MAX_CFTYPE_NAME, "%s.events.local", buf);
> > cft->private = MEMFILE_PRIVATE(idx, 0);
> > cft->seq_show = hugetlb_events_local_show;
> > @@ -672,7 +745,7 @@ static void __init __hugetlb_cgroup_file_dfl_init(int idx)
> > cft->flags = CFTYPE_NOT_ON_ROOT;
> >
> > /* NULL terminate the last cft */
> > - cft = &h->cgroup_files_dfl[6];
> > + cft = &h->cgroup_files_dfl[7];
> > memset(cft, 0, sizeof(*cft));
> >
> > WARN_ON(cgroup_add_dfl_cftypes(&hugetlb_cgrp_subsys,
> > @@ -742,8 +815,14 @@ static void __init __hugetlb_cgroup_file_legacy_init(int idx)
> > cft->write = hugetlb_cgroup_reset;
> > cft->read_u64 = hugetlb_cgroup_read_u64;
> >
> > + /* Add the numa stat file */
> > + cft = &h->cgroup_files_dfl[8];
> > + snprintf(cft->name, MAX_CFTYPE_NAME, "%s.numa_stat", buf);
> > + cft->private = MEMFILE_PRIVATE(idx, 1);
> > + cft->seq_show = hugetlb_cgroup_read_numa_stat;
> > +
> > /* NULL terminate the last cft */
> > - cft = &h->cgroup_files_legacy[8];
> > + cft = &h->cgroup_files_legacy[9];
> > memset(cft, 0, sizeof(*cft));
> >
> > WARN_ON(cgroup_add_legacy_cftypes(&hugetlb_cgrp_subsys,
> > diff --git a/tools/testing/selftests/vm/write_to_hugetlbfs.c b/tools/testing/selftests/vm/write_to_hugetlbfs.c
> > index 6a2caba19ee1..d2da6315a40c 100644
> > --- a/tools/testing/selftests/vm/write_to_hugetlbfs.c
> > +++ b/tools/testing/selftests/vm/write_to_hugetlbfs.c
> > @@ -37,8 +37,8 @@ static int shmid;
> > static void exit_usage(void)
> > {
> > printf("Usage: %s -p <path to hugetlbfs file> -s <size to map> "
> > - "[-m <0=hugetlbfs | 1=mmap(MAP_HUGETLB)>] [-l] [-r] "
> > - "[-o] [-w] [-n]\n",
> > + "[-m <0=hugetlbfs | 1=mmap(MAP_HUGETLB)>] [-l(sleep)] [-r(private)] "
> > + "[-o(populate)] [-w(rite)] [-n(o-reserve)]\n",
> > self);
> > exit(EXIT_FAILURE);
> > }
> > @@ -161,6 +161,11 @@ int main(int argc, char **argv)
> > else
> > printf("RESERVE mapping.\n");
> >
> > + if (want_sleep)
> > + printf("Sleeping\n");
> > + else
> > + printf("Not sleeping\n");
> > +
> > switch (method) {
> > case HUGETLBFS:
> > printf("Allocating using HUGETLBFS.\n");
> > --
> > 2.33.0.1079.g6e70778dc9-goog
> >
