summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2020-01-27 15:18:25 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2020-01-27 15:18:25 -0800
commit03aa8c8cfaec1aab6b1ea7fde656bbc893f6cff8 (patch)
tree717587b719a5314bcfb5bf4801a022924556515f
parent16d06120d73acde39ac0d5e3c539b4f407e9e588 (diff)
parent9ea37e24d4a95dd934a0600d65caa25e409705bb (diff)
downloadlinux-03aa8c8cfaec1aab6b1ea7fde656bbc893f6cff8.tar.bz2
Merge branch 'for-5.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
Pull cgroup updates from Tejun Heo: - cgroup2 interface for hugetlb controller. I think this was the last remaining bit which was missing from cgroup2 - fixes for race and a spurious warning in threaded cgroup handling - other minor changes * 'for-5.6' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: iocost: Fix iocost_monitor.py due to helper type mismatch cgroup: Prevent double killing of css when enabling threaded cgroup cgroup: fix function name in comment mm: hugetlb controller for cgroups v2
-rw-r--r--Documentation/admin-guide/cgroup-v2.rst29
-rw-r--r--include/linux/hugetlb.h3
-rw-r--r--kernel/cgroup/cgroup.c11
-rw-r--r--kernel/cgroup/rstat.c2
-rw-r--r--mm/hugetlb_cgroup.c198
-rw-r--r--tools/cgroup/iocost_monitor.py4
6 files changed, 227 insertions, 20 deletions
diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index 0636bcb60b5a..3f801461f0f3 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -61,6 +61,8 @@ v1 is available under Documentation/admin-guide/cgroup-v1/.
5-6. Device
5-7. RDMA
5-7-1. RDMA Interface Files
+ 5-8. HugeTLB
+ 5.8-1. HugeTLB Interface Files
5-8. Misc
5-8-1. perf_event
5-N. Non-normative information
@@ -2056,6 +2058,33 @@ RDMA Interface Files
mlx4_0 hca_handle=1 hca_object=20
ocrdma1 hca_handle=1 hca_object=23
+HugeTLB
+-------
+
+The HugeTLB controller allows to limit the HugeTLB usage per control group and
+enforces the controller limit during page fault.
+
+HugeTLB Interface Files
+~~~~~~~~~~~~~~~~~~~~~~~
+
+ hugetlb.<hugepagesize>.current
+ Show current usage for "hugepagesize" hugetlb. It exists for all
+ the cgroup except root.
+
+ hugetlb.<hugepagesize>.max
+ Set/show the hard limit of "hugepagesize" hugetlb usage.
+ The default value is "max". It exists for all the cgroup except root.
+
+ hugetlb.<hugepagesize>.events
+ A read-only flat-keyed file which exists on non-root cgroups.
+
+ max
+ The number of allocation failure due to HugeTLB limit
+
+ hugetlb.<hugepagesize>.events.local
+ Similar to hugetlb.<hugepagesize>.events but the fields in the file
+ are local to the cgroup i.e. not hierarchical. The file modified event
+ generated on this file reflects only the local events.
Misc
----
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 31d4920994b9..1e897e4168ac 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -432,7 +432,8 @@ struct hstate {
unsigned int surplus_huge_pages_node[MAX_NUMNODES];
#ifdef CONFIG_CGROUP_HUGETLB
/* cgroup control files */
- struct cftype cgroup_files[5];
+ struct cftype cgroup_files_dfl[5];
+ struct cftype cgroup_files_legacy[5];
#endif
char name[HSTATE_NAME_LEN];
};
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 735af8f15f95..1e12e6928bca 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -3055,8 +3055,6 @@ static int cgroup_apply_control_enable(struct cgroup *cgrp)
for_each_subsys(ss, ssid) {
struct cgroup_subsys_state *css = cgroup_css(dsct, ss);
- WARN_ON_ONCE(css && percpu_ref_is_dying(&css->refcnt));
-
if (!(cgroup_ss_mask(dsct) & (1 << ss->id)))
continue;
@@ -3066,6 +3064,8 @@ static int cgroup_apply_control_enable(struct cgroup *cgrp)
return PTR_ERR(css);
}
+ WARN_ON_ONCE(percpu_ref_is_dying(&css->refcnt));
+
if (css_visible(css)) {
ret = css_populate_dir(css);
if (ret)
@@ -3101,11 +3101,11 @@ static void cgroup_apply_control_disable(struct cgroup *cgrp)
for_each_subsys(ss, ssid) {
struct cgroup_subsys_state *css = cgroup_css(dsct, ss);
- WARN_ON_ONCE(css && percpu_ref_is_dying(&css->refcnt));
-
if (!css)
continue;
+ WARN_ON_ONCE(percpu_ref_is_dying(&css->refcnt));
+
if (css->parent &&
!(cgroup_ss_mask(dsct) & (1 << ss->id))) {
kill_css(css);
@@ -3392,7 +3392,8 @@ static ssize_t cgroup_type_write(struct kernfs_open_file *of, char *buf,
if (strcmp(strstrip(buf), "threaded"))
return -EINVAL;
- cgrp = cgroup_kn_lock_live(of->kn, false);
+ /* drain dying csses before we re-apply (threaded) subtree control */
+ cgrp = cgroup_kn_lock_live(of->kn, true);
if (!cgrp)
return -ENOENT;
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index b48b22d4deb6..6f87352f8219 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -33,7 +33,7 @@ void cgroup_rstat_updated(struct cgroup *cgrp, int cpu)
return;
/*
- * Paired with the one in cgroup_rstat_cpu_pop_upated(). Either we
+ * Paired with the one in cgroup_rstat_cpu_pop_updated(). Either we
* see NULL updated_next or they see our updated stat.
*/
smp_mb();
diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index 2ac38bdc18a1..e434b05416c6 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -3,6 +3,10 @@
* Copyright IBM Corporation, 2012
* Author Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
*
+ * Cgroup v2
+ * Copyright (C) 2019 Red Hat, Inc.
+ * Author: Giuseppe Scrivano <gscrivan@redhat.com>
+ *
* This program is free software; you can redistribute it and/or modify it
* under the terms of version 2.1 of the GNU Lesser General Public License
* as published by the Free Software Foundation.
@@ -19,18 +23,36 @@
#include <linux/hugetlb.h>
#include <linux/hugetlb_cgroup.h>
+enum hugetlb_memory_event {
+ HUGETLB_MAX,
+ HUGETLB_NR_MEMORY_EVENTS,
+};
+
struct hugetlb_cgroup {
struct cgroup_subsys_state css;
+
/*
* the counter to account for hugepages from hugetlb.
*/
struct page_counter hugepage[HUGE_MAX_HSTATE];
+
+ atomic_long_t events[HUGE_MAX_HSTATE][HUGETLB_NR_MEMORY_EVENTS];
+ atomic_long_t events_local[HUGE_MAX_HSTATE][HUGETLB_NR_MEMORY_EVENTS];
+
+ /* Handle for "hugetlb.events" */
+ struct cgroup_file events_file[HUGE_MAX_HSTATE];
+
+ /* Handle for "hugetlb.events.local" */
+ struct cgroup_file events_local_file[HUGE_MAX_HSTATE];
};
#define MEMFILE_PRIVATE(x, val) (((x) << 16) | (val))
#define MEMFILE_IDX(val) (((val) >> 16) & 0xffff)
#define MEMFILE_ATTR(val) ((val) & 0xffff)
+#define hugetlb_cgroup_from_counter(counter, idx) \
+ container_of(counter, struct hugetlb_cgroup, hugepage[idx])
+
static struct hugetlb_cgroup *root_h_cgroup __read_mostly;
static inline
@@ -178,6 +200,19 @@ static void hugetlb_cgroup_css_offline(struct cgroup_subsys_state *css)
} while (hugetlb_cgroup_have_usage(h_cg));
}
+static inline void hugetlb_event(struct hugetlb_cgroup *hugetlb, int idx,
+ enum hugetlb_memory_event event)
+{
+ atomic_long_inc(&hugetlb->events_local[idx][event]);
+ cgroup_file_notify(&hugetlb->events_local_file[idx]);
+
+ do {
+ atomic_long_inc(&hugetlb->events[idx][event]);
+ cgroup_file_notify(&hugetlb->events_file[idx]);
+ } while ((hugetlb = parent_hugetlb_cgroup(hugetlb)) &&
+ !hugetlb_cgroup_is_root(hugetlb));
+}
+
int hugetlb_cgroup_charge_cgroup(int idx, unsigned long nr_pages,
struct hugetlb_cgroup **ptr)
{
@@ -202,8 +237,12 @@ again:
}
rcu_read_unlock();
- if (!page_counter_try_charge(&h_cg->hugepage[idx], nr_pages, &counter))
+ if (!page_counter_try_charge(&h_cg->hugepage[idx], nr_pages,
+ &counter)) {
ret = -ENOMEM;
+ hugetlb_event(hugetlb_cgroup_from_counter(counter, idx), idx,
+ HUGETLB_MAX);
+ }
css_put(&h_cg->css);
done:
*ptr = h_cg;
@@ -283,10 +322,45 @@ static u64 hugetlb_cgroup_read_u64(struct cgroup_subsys_state *css,
}
}
+static int hugetlb_cgroup_read_u64_max(struct seq_file *seq, void *v)
+{
+ int idx;
+ u64 val;
+ struct cftype *cft = seq_cft(seq);
+ unsigned long limit;
+ struct page_counter *counter;
+ struct hugetlb_cgroup *h_cg = hugetlb_cgroup_from_css(seq_css(seq));
+
+ idx = MEMFILE_IDX(cft->private);
+ counter = &h_cg->hugepage[idx];
+
+ limit = round_down(PAGE_COUNTER_MAX,
+ 1 << huge_page_order(&hstates[idx]));
+
+ switch (MEMFILE_ATTR(cft->private)) {
+ case RES_USAGE:
+ val = (u64)page_counter_read(counter);
+ seq_printf(seq, "%llu\n", val * PAGE_SIZE);
+ break;
+ case RES_LIMIT:
+ val = (u64)counter->max;
+ if (val == limit)
+ seq_puts(seq, "max\n");
+ else
+ seq_printf(seq, "%llu\n", val * PAGE_SIZE);
+ break;
+ default:
+ BUG();
+ }
+
+ return 0;
+}
+
static DEFINE_MUTEX(hugetlb_limit_mutex);
static ssize_t hugetlb_cgroup_write(struct kernfs_open_file *of,
- char *buf, size_t nbytes, loff_t off)
+ char *buf, size_t nbytes, loff_t off,
+ const char *max)
{
int ret, idx;
unsigned long nr_pages;
@@ -296,7 +370,7 @@ static ssize_t hugetlb_cgroup_write(struct kernfs_open_file *of,
return -EINVAL;
buf = strstrip(buf);
- ret = page_counter_memparse(buf, "-1", &nr_pages);
+ ret = page_counter_memparse(buf, max, &nr_pages);
if (ret)
return ret;
@@ -316,6 +390,18 @@ static ssize_t hugetlb_cgroup_write(struct kernfs_open_file *of,
return ret ?: nbytes;
}
+static ssize_t hugetlb_cgroup_write_legacy(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ return hugetlb_cgroup_write(of, buf, nbytes, off, "-1");
+}
+
+static ssize_t hugetlb_cgroup_write_dfl(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ return hugetlb_cgroup_write(of, buf, nbytes, off, "max");
+}
+
static ssize_t hugetlb_cgroup_reset(struct kernfs_open_file *of,
char *buf, size_t nbytes, loff_t off)
{
@@ -350,7 +436,36 @@ static char *mem_fmt(char *buf, int size, unsigned long hsize)
return buf;
}
-static void __init __hugetlb_cgroup_file_init(int idx)
+static int __hugetlb_events_show(struct seq_file *seq, bool local)
+{
+ int idx;
+ long max;
+ struct cftype *cft = seq_cft(seq);
+ struct hugetlb_cgroup *h_cg = hugetlb_cgroup_from_css(seq_css(seq));
+
+ idx = MEMFILE_IDX(cft->private);
+
+ if (local)
+ max = atomic_long_read(&h_cg->events_local[idx][HUGETLB_MAX]);
+ else
+ max = atomic_long_read(&h_cg->events[idx][HUGETLB_MAX]);
+
+ seq_printf(seq, "max %lu\n", max);
+
+ return 0;
+}
+
+static int hugetlb_events_show(struct seq_file *seq, void *v)
+{
+ return __hugetlb_events_show(seq, false);
+}
+
+static int hugetlb_events_local_show(struct seq_file *seq, void *v)
+{
+ return __hugetlb_events_show(seq, true);
+}
+
+static void __init __hugetlb_cgroup_file_dfl_init(int idx)
{
char buf[32];
struct cftype *cft;
@@ -360,38 +475,93 @@ static void __init __hugetlb_cgroup_file_init(int idx)
mem_fmt(buf, 32, huge_page_size(h));
/* Add the limit file */
- cft = &h->cgroup_files[0];
+ cft = &h->cgroup_files_dfl[0];
+ snprintf(cft->name, MAX_CFTYPE_NAME, "%s.max", buf);
+ cft->private = MEMFILE_PRIVATE(idx, RES_LIMIT);
+ cft->seq_show = hugetlb_cgroup_read_u64_max;
+ cft->write = hugetlb_cgroup_write_dfl;
+ cft->flags = CFTYPE_NOT_ON_ROOT;
+
+ /* Add the current usage file */
+ cft = &h->cgroup_files_dfl[1];
+ snprintf(cft->name, MAX_CFTYPE_NAME, "%s.current", buf);
+ cft->private = MEMFILE_PRIVATE(idx, RES_USAGE);
+ cft->seq_show = hugetlb_cgroup_read_u64_max;
+ cft->flags = CFTYPE_NOT_ON_ROOT;
+
+ /* Add the events file */
+ cft = &h->cgroup_files_dfl[2];
+ snprintf(cft->name, MAX_CFTYPE_NAME, "%s.events", buf);
+ cft->private = MEMFILE_PRIVATE(idx, 0);
+ cft->seq_show = hugetlb_events_show;
+ cft->file_offset = offsetof(struct hugetlb_cgroup, events_file[idx]),
+ cft->flags = CFTYPE_NOT_ON_ROOT;
+
+ /* Add the events.local file */
+ cft = &h->cgroup_files_dfl[3];
+ snprintf(cft->name, MAX_CFTYPE_NAME, "%s.events.local", buf);
+ cft->private = MEMFILE_PRIVATE(idx, 0);
+ cft->seq_show = hugetlb_events_local_show;
+ cft->file_offset = offsetof(struct hugetlb_cgroup,
+ events_local_file[idx]),
+ cft->flags = CFTYPE_NOT_ON_ROOT;
+
+ /* NULL terminate the last cft */
+ cft = &h->cgroup_files_dfl[4];
+ memset(cft, 0, sizeof(*cft));
+
+ WARN_ON(cgroup_add_dfl_cftypes(&hugetlb_cgrp_subsys,
+ h->cgroup_files_dfl));
+}
+
+static void __init __hugetlb_cgroup_file_legacy_init(int idx)
+{
+ char buf[32];
+ struct cftype *cft;
+ struct hstate *h = &hstates[idx];
+
+ /* format the size */
+ mem_fmt(buf, 32, huge_page_size(h));
+
+ /* Add the limit file */
+ cft = &h->cgroup_files_legacy[0];
snprintf(cft->name, MAX_CFTYPE_NAME, "%s.limit_in_bytes", buf);
cft->private = MEMFILE_PRIVATE(idx, RES_LIMIT);
cft->read_u64 = hugetlb_cgroup_read_u64;
- cft->write = hugetlb_cgroup_write;
+ cft->write = hugetlb_cgroup_write_legacy;
/* Add the usage file */
- cft = &h->cgroup_files[1];
+ cft = &h->cgroup_files_legacy[1];
snprintf(cft->name, MAX_CFTYPE_NAME, "%s.usage_in_bytes", buf);
cft->private = MEMFILE_PRIVATE(idx, RES_USAGE);
cft->read_u64 = hugetlb_cgroup_read_u64;
/* Add the MAX usage file */
- cft = &h->cgroup_files[2];
+ cft = &h->cgroup_files_legacy[2];
snprintf(cft->name, MAX_CFTYPE_NAME, "%s.max_usage_in_bytes", buf);
cft->private = MEMFILE_PRIVATE(idx, RES_MAX_USAGE);
cft->write = hugetlb_cgroup_reset;
cft->read_u64 = hugetlb_cgroup_read_u64;
/* Add the failcntfile */
- cft = &h->cgroup_files[3];
+ cft = &h->cgroup_files_legacy[3];
snprintf(cft->name, MAX_CFTYPE_NAME, "%s.failcnt", buf);
cft->private = MEMFILE_PRIVATE(idx, RES_FAILCNT);
cft->write = hugetlb_cgroup_reset;
cft->read_u64 = hugetlb_cgroup_read_u64;
/* NULL terminate the last cft */
- cft = &h->cgroup_files[4];
+ cft = &h->cgroup_files_legacy[4];
memset(cft, 0, sizeof(*cft));
WARN_ON(cgroup_add_legacy_cftypes(&hugetlb_cgrp_subsys,
- h->cgroup_files));
+ h->cgroup_files_legacy));
+}
+
+static void __init __hugetlb_cgroup_file_init(int idx)
+{
+ __hugetlb_cgroup_file_dfl_init(idx);
+ __hugetlb_cgroup_file_legacy_init(idx);
}
void __init hugetlb_cgroup_file_init(void)
@@ -433,8 +603,14 @@ void hugetlb_cgroup_migrate(struct page *oldhpage, struct page *newhpage)
return;
}
+static struct cftype hugetlb_files[] = {
+ {} /* terminate */
+};
+
struct cgroup_subsys hugetlb_cgrp_subsys = {
.css_alloc = hugetlb_cgroup_css_alloc,
.css_offline = hugetlb_cgroup_css_offline,
.css_free = hugetlb_cgroup_css_free,
+ .dfl_cftypes = hugetlb_files,
+ .legacy_cftypes = hugetlb_files,
};
diff --git a/tools/cgroup/iocost_monitor.py b/tools/cgroup/iocost_monitor.py
index f79b23582a1d..7427a5ee761b 100644
--- a/tools/cgroup/iocost_monitor.py
+++ b/tools/cgroup/iocost_monitor.py
@@ -72,7 +72,7 @@ class BlkgIterator:
name = BlkgIterator.blkcg_name(blkcg)
path = parent_path + '/' + name if parent_path else name
blkg = drgn.Object(prog, 'struct blkcg_gq',
- address=radix_tree_lookup(blkcg.blkg_tree, q_id))
+ address=radix_tree_lookup(blkcg.blkg_tree.address_of_(), q_id))
if not blkg.address_:
return
@@ -228,7 +228,7 @@ q_id = None
root_iocg = None
ioc = None
-for i, ptr in radix_tree_for_each(blkcg_root.blkg_tree):
+for i, ptr in radix_tree_for_each(blkcg_root.blkg_tree.address_of_()):
blkg = drgn.Object(prog, 'struct blkcg_gq', address=ptr)
try:
if devname == blkg.q.kobj.parent.name.string_().decode('utf-8'):