* [PATCH RESEND v3 1/3] fs/resctrl: Prevent deadlock and use-after-free in info file handlers
2026-05-26 22:38 [PATCH RESEND v3 0/3] x86,fs/resctrl: Fix long-standing issues Reinette Chatre
@ 2026-05-26 22:38 ` Reinette Chatre
2026-05-26 22:38 ` [PATCH RESEND v3 2/3] x86/resctrl: Ensure domain fully initialized before placed on RCU list Reinette Chatre
2026-05-26 22:38 ` [PATCH RESEND v3 3/3] fs/resctrl: Fix UAF from worker threads when domains are removed Reinette Chatre
2 siblings, 0 replies; 5+ messages in thread
From: Reinette Chatre @ 2026-05-26 22:38 UTC (permalink / raw)
To: reinette.chatre; +Cc: linux-kernel, patches
resctrl provides files under the info/ directory to expose global
configuration and capabilities to userspace. These files are instantiated
statically during filesystem mount and expose data associated with internal
schema structures via kernfs private pointers.
A potential deadlock exists between userspace readers of these info files
and the unmount filesystem teardown process. Reading an info file invokes
kernfs which acquires an active reference, after which the handler typically
attempts to acquire the rdtgroup_mutex. Concurrently, unmounting the
filesystem holds the rdtgroup_mutex and then attempts to recursively
remove the info kernfs nodes involving kernfs_drain() which blocks until
all active references are released. Another problem exists where info files
might be accessed from an outdated mount if the filesystem is unmounted and
remounted during a reader's execution, leading to a use-after-free when
reading the now-deleted private schema data.
Introduce info_kn_lock() and info_kn_unlock() helpers to coordinate locking
across all info handlers. These helpers mirror similar logic used by resource
group handlers by deliberately breaking the kernfs active protection before
attempting to acquire the rdtgroup_mutex, preventing the deadlock. To guard
against the vulnerability from rapid mount cycling, info_kn_lock() securely
walks the parent lineage of the kernfs node under an RCU section to confirm
the node belongs to the globally active root before permitting the operation
to proceed. Convert all info file handlers to use this helper and only
de-reference the schema after it determined safe to do so.
Make no attempt to output error message to last_cmd_status on failure
since failure implies there is no filesystem with which to display error
to user space.
Reported-by: Sashiko <sashiko-bot@kernel.org>
Closes: https://sashiko.dev/#/patchset/20260515193944.15114-1-tony.luck%40intel.com?part=3
Assisted-by: GitHub_Copilot:gemini-3.1-pro
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
---
Changes since V2:
- New patch
---
fs/resctrl/ctrlmondata.c | 38 ++++----
fs/resctrl/internal.h | 3 +-
fs/resctrl/monitor.c | 48 +++++-----
fs/resctrl/rdtgroup.c | 192 ++++++++++++++++++++++++++++++++-------
4 files changed, 203 insertions(+), 78 deletions(-)
diff --git a/fs/resctrl/ctrlmondata.c b/fs/resctrl/ctrlmondata.c
index 9a7dfc48cb2e..b95bf6208be2 100644
--- a/fs/resctrl/ctrlmondata.c
+++ b/fs/resctrl/ctrlmondata.c
@@ -769,10 +769,12 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg)
int resctrl_io_alloc_show(struct kernfs_open_file *of, struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
if (r->cache.io_alloc_capable) {
if (resctrl_arch_get_io_alloc_enabled(r))
seq_puts(seq, "enabled\n");
@@ -782,7 +784,7 @@ int resctrl_io_alloc_show(struct kernfs_open_file *of, struct seq_file *seq, voi
seq_puts(seq, "not supported\n");
}
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -847,7 +849,7 @@ ssize_t resctrl_io_alloc_write(struct kernfs_open_file *of, char *buf,
size_t nbytes, loff_t off)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
char const *grp_name;
u32 io_alloc_closid;
bool enable;
@@ -857,9 +859,10 @@ ssize_t resctrl_io_alloc_write(struct kernfs_open_file *of, char *buf,
if (ret)
return ret;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
rdt_last_cmd_clear();
if (!r->cache.io_alloc_capable) {
@@ -907,8 +910,7 @@ ssize_t resctrl_io_alloc_write(struct kernfs_open_file *of, char *buf,
}
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -916,14 +918,15 @@ ssize_t resctrl_io_alloc_write(struct kernfs_open_file *of, char *buf,
int resctrl_io_alloc_cbm_show(struct kernfs_open_file *of, struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
int ret = 0;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
+ r = s->res;
if (!r->cache.io_alloc_capable) {
rdt_last_cmd_printf("io_alloc is not supported on %s\n", s->name);
ret = -ENODEV;
@@ -945,8 +948,7 @@ int resctrl_io_alloc_cbm_show(struct kernfs_open_file *of, struct seq_file *seq,
show_doms(seq, s, NULL, resctrl_io_alloc_closid(r));
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret;
}
@@ -1013,7 +1015,7 @@ ssize_t resctrl_io_alloc_cbm_write(struct kernfs_open_file *of, char *buf,
size_t nbytes, loff_t off)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
u32 io_alloc_closid;
int ret = 0;
@@ -1023,10 +1025,11 @@ ssize_t resctrl_io_alloc_cbm_write(struct kernfs_open_file *of, char *buf,
buf[nbytes - 1] = '\0';
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
+ r = s->res;
if (!r->cache.io_alloc_capable) {
rdt_last_cmd_printf("io_alloc is not supported on %s\n", s->name);
ret = -ENODEV;
@@ -1051,8 +1054,7 @@ ssize_t resctrl_io_alloc_cbm_write(struct kernfs_open_file *of, char *buf,
out_clear_configs:
rdt_staged_configs_clear();
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
diff --git a/fs/resctrl/internal.h b/fs/resctrl/internal.h
index 1a9b29119f88..081406ab9714 100644
--- a/fs/resctrl/internal.h
+++ b/fs/resctrl/internal.h
@@ -335,8 +335,9 @@ __printf(1, 2)
void rdt_last_cmd_printf(const char *fmt, ...);
struct rdtgroup *rdtgroup_kn_lock_live(struct kernfs_node *kn);
-
void rdtgroup_kn_unlock(struct kernfs_node *kn);
+bool info_kn_lock(struct kernfs_node *kn);
+void info_kn_unlock(struct kernfs_node *kn);
int rdtgroup_kn_mode_restrict(struct rdtgroup *r, const char *name);
diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c
index 9fd901c78dc6..005d2babe692 100644
--- a/fs/resctrl/monitor.c
+++ b/fs/resctrl/monitor.c
@@ -1052,7 +1052,8 @@ int event_filter_show(struct kernfs_open_file *of, struct seq_file *seq, void *v
bool sep = false;
int ret = 0, i;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
r = resctrl_arch_get_resource(mevt->rid);
@@ -1073,7 +1074,7 @@ int event_filter_show(struct kernfs_open_file *of, struct seq_file *seq, void *v
seq_putc(seq, '\n');
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return ret;
}
@@ -1084,7 +1085,8 @@ int resctrl_mbm_assign_on_mkdir_show(struct kernfs_open_file *of, struct seq_fil
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
int ret = 0;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
if (!resctrl_arch_mbm_cntr_assign_enabled(r)) {
@@ -1096,7 +1098,7 @@ int resctrl_mbm_assign_on_mkdir_show(struct kernfs_open_file *of, struct seq_fil
seq_printf(s, "%u\n", r->mon.mbm_assign_on_mkdir);
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return ret;
}
@@ -1112,7 +1114,8 @@ ssize_t resctrl_mbm_assign_on_mkdir_write(struct kernfs_open_file *of, char *buf
if (ret)
return ret;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
if (!resctrl_arch_mbm_cntr_assign_enabled(r)) {
@@ -1124,7 +1127,7 @@ ssize_t resctrl_mbm_assign_on_mkdir_write(struct kernfs_open_file *of, char *buf
r->mon.mbm_assign_on_mkdir = value;
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -1411,8 +1414,8 @@ ssize_t event_filter_write(struct kernfs_open_file *of, char *buf, size_t nbytes
buf[nbytes - 1] = '\0';
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
@@ -1430,8 +1433,7 @@ ssize_t event_filter_write(struct kernfs_open_file *of, char *buf, size_t nbytes
}
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -1442,7 +1444,8 @@ int resctrl_mbm_assign_mode_show(struct kernfs_open_file *of,
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
bool enabled;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
enabled = resctrl_arch_mbm_cntr_assign_enabled(r);
if (r->mon.mbm_cntr_assignable) {
@@ -1461,7 +1464,7 @@ int resctrl_mbm_assign_mode_show(struct kernfs_open_file *of,
seq_puts(s, "[default]\n");
}
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1480,8 +1483,8 @@ ssize_t resctrl_mbm_assign_mode_write(struct kernfs_open_file *of, char *buf,
buf[nbytes - 1] = '\0';
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
@@ -1533,8 +1536,7 @@ ssize_t resctrl_mbm_assign_mode_write(struct kernfs_open_file *of, char *buf,
}
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -1546,8 +1548,8 @@ int resctrl_num_mbm_cntrs_show(struct kernfs_open_file *of,
struct rdt_l3_mon_domain *dom;
bool sep = false;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
list_for_each_entry(dom, &r->mon_domains, hdr.list) {
if (sep)
@@ -1558,8 +1560,7 @@ int resctrl_num_mbm_cntrs_show(struct kernfs_open_file *of,
}
seq_putc(s, '\n');
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1572,8 +1573,8 @@ int resctrl_available_mbm_cntrs_show(struct kernfs_open_file *of,
u32 cntrs, i;
int ret = 0;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
@@ -1599,8 +1600,7 @@ int resctrl_available_mbm_cntrs_show(struct kernfs_open_file *of,
seq_putc(s, '\n');
out_unlock:
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret;
}
diff --git a/fs/resctrl/rdtgroup.c b/fs/resctrl/rdtgroup.c
index 5dfdaa6f9d8f..8eaedf31adcd 100644
--- a/fs/resctrl/rdtgroup.c
+++ b/fs/resctrl/rdtgroup.c
@@ -971,13 +971,14 @@ static int rdt_last_cmd_status_show(struct kernfs_open_file *of,
{
int len;
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
len = seq_buf_used(&last_cmd_status);
if (len)
seq_printf(seq, "%.*s", len, last_cmd_status_buf);
else
seq_puts(seq, "ok\n");
- mutex_unlock(&rdtgroup_mutex);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -996,7 +997,11 @@ static int rdt_num_closids_show(struct kernfs_open_file *of,
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
seq_printf(seq, "%u\n", s->num_closid);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1004,9 +1009,14 @@ static int rdt_default_ctrl_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%x\n", resctrl_get_default_ctrl(r));
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1014,9 +1024,15 @@ static int rdt_min_cbm_bits_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%u\n", r->cache.min_cbm_bits);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1024,9 +1040,14 @@ static int rdt_shareable_bits_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%x\n", r->cache.shareable_bits);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1054,15 +1075,16 @@ static int rdt_bit_usage_show(struct kernfs_open_file *of,
*/
unsigned long sw_shareable = 0, hw_shareable = 0;
unsigned long exclusive = 0, pseudo_locked = 0;
- struct rdt_resource *r = s->res;
struct rdt_ctrl_domain *dom;
int i, hwb, swb, excl, psl;
+ struct rdt_resource *r;
enum rdtgrp_mode mode;
bool sep = false;
u32 ctrl_val;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
list_for_each_entry(dom, &r->ctrl_domains, hdr.list) {
if (sep)
seq_putc(seq, ';');
@@ -1138,8 +1160,7 @@ static int rdt_bit_usage_show(struct kernfs_open_file *of,
sep = true;
}
seq_putc(seq, '\n');
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1147,9 +1168,14 @@ static int rdt_min_bw_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%u\n", r->membw.min_bw);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1158,8 +1184,12 @@ static int rdt_num_rmids_show(struct kernfs_open_file *of,
{
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
seq_printf(seq, "%u\n", r->mon.num_rmid);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1169,6 +1199,8 @@ static int rdt_mon_features_show(struct kernfs_open_file *of,
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
struct mon_evt *mevt;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
for_each_mon_event(mevt) {
if (mevt->rid != r->rid || !mevt->enabled)
continue;
@@ -1178,6 +1210,8 @@ static int rdt_mon_features_show(struct kernfs_open_file *of,
seq_printf(seq, "%s_config\n", mevt->name);
}
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1185,9 +1219,14 @@ static int rdt_bw_gran_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%u\n", r->membw.bw_gran);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1195,16 +1234,24 @@ static int rdt_delay_linear_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%u\n", r->membw.delay_linear);
+ info_kn_unlock(of->kn);
+
return 0;
}
static int max_threshold_occ_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
seq_printf(seq, "%u\n", resctrl_rmid_realloc_threshold);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1213,22 +1260,28 @@ static int rdt_thread_throttle_mode_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+
+ r = s->res;
switch (r->membw.throttle_mode) {
case THREAD_THROTTLE_PER_THREAD:
seq_puts(seq, "per-thread\n");
- return 0;
+ break;
case THREAD_THROTTLE_MAX:
seq_puts(seq, "max\n");
- return 0;
+ break;
case THREAD_THROTTLE_UNDEFINED:
seq_puts(seq, "undefined\n");
- return 0;
+ break;
+ default:
+ WARN_ON_ONCE(1);
+ break;
}
- WARN_ON_ONCE(1);
-
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1242,12 +1295,20 @@ static ssize_t max_threshold_occ_write(struct kernfs_open_file *of,
if (ret)
return ret;
- if (bytes > resctrl_rmid_realloc_limit)
- return -EINVAL;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+
+ if (bytes > resctrl_rmid_realloc_limit) {
+ ret = -EINVAL;
+ goto out_unlock;
+ }
resctrl_rmid_realloc_threshold = resctrl_arch_round_mon_val(bytes);
- return nbytes;
+out_unlock:
+ info_kn_unlock(of->kn);
+
+ return ret ?: nbytes;
}
/*
@@ -1287,10 +1348,15 @@ static int rdt_has_sparse_bitmasks_show(struct kernfs_open_file *of,
struct seq_file *seq, void *v)
{
struct resctrl_schema *s = rdt_kn_parent_priv(of->kn);
- struct rdt_resource *r = s->res;
+ struct rdt_resource *r;
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+ r = s->res;
seq_printf(seq, "%u\n", r->cache.arch_has_sparse_bitmasks);
+ info_kn_unlock(of->kn);
+
return 0;
}
@@ -1646,8 +1712,8 @@ static int mbm_config_show(struct seq_file *s, struct rdt_resource *r, u32 evtid
struct rdt_l3_mon_domain *dom;
bool sep = false;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ lockdep_assert_cpus_held();
+ lockdep_assert_held(&rdtgroup_mutex);
list_for_each_entry(dom, &r->mon_domains, hdr.list) {
if (sep)
@@ -1664,8 +1730,6 @@ static int mbm_config_show(struct seq_file *s, struct rdt_resource *r, u32 evtid
}
seq_puts(s, "\n");
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
return 0;
}
@@ -1675,8 +1739,12 @@ static int mbm_total_bytes_config_show(struct kernfs_open_file *of,
{
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+
mbm_config_show(seq, r, QOS_L3_MBM_TOTAL_EVENT_ID);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1685,8 +1753,12 @@ static int mbm_local_bytes_config_show(struct kernfs_open_file *of,
{
struct rdt_resource *r = rdt_kn_parent_priv(of->kn);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
+
mbm_config_show(seq, r, QOS_L3_MBM_LOCAL_EVENT_ID);
+ info_kn_unlock(of->kn);
return 0;
}
@@ -1784,8 +1856,8 @@ static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
if (nbytes == 0 || buf[nbytes - 1] != '\n')
return -EINVAL;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
@@ -1793,8 +1865,7 @@ static ssize_t mbm_total_bytes_config_write(struct kernfs_open_file *of,
ret = mon_config_write(r, buf, QOS_L3_MBM_TOTAL_EVENT_ID);
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -1810,8 +1881,8 @@ static ssize_t mbm_local_bytes_config_write(struct kernfs_open_file *of,
if (nbytes == 0 || buf[nbytes - 1] != '\n')
return -EINVAL;
- cpus_read_lock();
- mutex_lock(&rdtgroup_mutex);
+ if (!info_kn_lock(of->kn))
+ return -ENOENT;
rdt_last_cmd_clear();
@@ -1819,8 +1890,7 @@ static ssize_t mbm_local_bytes_config_write(struct kernfs_open_file *of,
ret = mon_config_write(r, buf, QOS_L3_MBM_LOCAL_EVENT_ID);
- mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
+ info_kn_unlock(of->kn);
return ret ?: nbytes;
}
@@ -2634,6 +2704,58 @@ void rdtgroup_kn_unlock(struct kernfs_node *kn)
rdtgroup_kn_put(rdtgrp, kn);
}
+/*
+ * Accessing the kn after breaking active protection is safe since the open
+ * of resctrl file holds a kernfs base reference (different from active
+ * protection) on the kn ensuring that it remains accessible even if it was
+ * unlinked. Each kn in turn holds base reference to parent so the kn's
+ * genealogy remains in memory until all base references dropped.
+ */
+static bool is_active_resctrl_node(struct kernfs_node *kn)
+{
+ struct kernfs_node *p;
+ bool match = false;
+
+ guard(rcu)();
+ p = kn;
+ while (p) {
+ if (p == rdtgroup_default.kn) {
+ match = true;
+ break;
+ }
+ p = rcu_dereference(p->__parent);
+ }
+
+ return match;
+}
+
+bool info_kn_lock(struct kernfs_node *kn)
+{
+ kernfs_break_active_protection(kn);
+ cpus_read_lock();
+ mutex_lock(&rdtgroup_mutex);
+
+ /*
+ * Check both if resctrl is torn down (!rdtgroup_default.kn) and
+ * if the reader's kernfs_node originates from a dead mount.
+ */
+ if (!rdtgroup_default.kn || !is_active_resctrl_node(kn)) {
+ mutex_unlock(&rdtgroup_mutex);
+ cpus_read_unlock();
+ kernfs_unbreak_active_protection(kn);
+ return false;
+ }
+
+ return true;
+}
+
+void info_kn_unlock(struct kernfs_node *kn)
+{
+ mutex_unlock(&rdtgroup_mutex);
+ cpus_read_unlock();
+ kernfs_unbreak_active_protection(kn);
+}
+
static int mkdir_mondata_all(struct kernfs_node *parent_kn,
struct rdtgroup *prgrp,
struct kernfs_node **mon_data_kn);
--
2.50.1
^ permalink raw reply [flat|nested] 5+ messages in thread* [PATCH RESEND v3 3/3] fs/resctrl: Fix UAF from worker threads when domains are removed
2026-05-26 22:38 [PATCH RESEND v3 0/3] x86,fs/resctrl: Fix long-standing issues Reinette Chatre
2026-05-26 22:38 ` [PATCH RESEND v3 1/3] fs/resctrl: Prevent deadlock and use-after-free in info file handlers Reinette Chatre
2026-05-26 22:38 ` [PATCH RESEND v3 2/3] x86/resctrl: Ensure domain fully initialized before placed on RCU list Reinette Chatre
@ 2026-05-26 22:38 ` Reinette Chatre
2 siblings, 0 replies; 5+ messages in thread
From: Reinette Chatre @ 2026-05-26 22:38 UTC (permalink / raw)
To: reinette.chatre; +Cc: linux-kernel, patches
The mbm_handle_overflow() and cqm_handle_limbo() workers read event
counters and may sleep while doing so. They are scheduled via
delayed_work embedded in struct rdt_l3_mon_domain. Architecture allocates
and frees these domains from CPU hotplug callbacks under cpus_write_lock(),
and the workers acquire cpus_read_lock() to keep the domain alive across
their access.
A use-after-free can occur when a worker is blocked waiting for
cpus_read_lock() while the hotplug core holds cpus_write_lock():
the architecture frees the rdt_l3_mon_domain that contains the worker's
work_struct. When the worker unblocks, the container_of() it performs on
the embedded work pointer dereferences freed memory.
Drop cpus_read_lock() from the workers and instead drain pending and
in-flight work synchronously before the architecture can free the domain.
Since architecture offlines the domain under cpus_write_lock() after it has
been unlinked from the RCU list and a grace period has elapsed no new work
can be scheduled. The cancel only needs to wait out existing work.
Drop rdtgroup_mutex during CPU offline around cancel_delayed_work_sync()
so that a worker waiting on the mutex can complete before re-pinning the
work on a different CPU.
Fixes: 24247aeeabe9 ("x86/intel_rdt/cqm: Improve limbo list processing")
Reported-by: Sashiko <sashiko-bot@kernel.org>
Closes: https://sashiko.dev/#/patchset/20260429184858.36423-1-tony.luck%40intel.com # [1]
Co-developed-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
---
Changes since v2:
- Rewrite changelog
- v2 attempted to solve the issue by using is_percpu_thread() within the
worker to learn if CPU worker was running on is going offline. A
Sashiko (https://sashiko.dev/#/patchset/20260515193944.15114-1-tony.luck%40intel.com?part=5)
pointed out that this would not be able to handle the scenario if one
of the hotplug handlers following the resctrl offline handlers failed.
- Some other fixes attempted that failed:
- Switch to accessing domain structure in handler via RCU so that CPU
hotplug lock no longer needed. Use cancel_delayed_work_sync() with
mutex dropped to cancel worker. Running worker from RCU read-side
critical section is a problem since the worker needs to be
able to sleep (mbm_handle_overflow()->mbm_update()->
mbm_update_one_event()->resctrl_arch_mon_ctx_alloc()->
might_sleep())
- Adding a reference count to the domain structure to avoid the worker
needing to take CPU hotplug lock. This ended up being very complicated
with the architecture needing new APIs to manage the reference count
which cannot cleanly integrate into MPAM since it uses a single
architecture domain structure to contain both the control and monitoring
domain structures. Managing the references across mount, unmount,
online, offline, as well as worker self exit resulted in several
asymmetrical and complicated paths that were error prone. Locking also
proved to be complicated since architecture would need to initiate
domain free that will need to call back into resctrl that will take
rdtgroup_mutex which means that references need to be taken/released
without locking.
---
fs/resctrl/monitor.c | 52 ++++++++++++++++++++++++++++++++++---------
fs/resctrl/rdtgroup.c | 52 ++++++++++++++++++++++++++++++++++++++-----
2 files changed, 89 insertions(+), 15 deletions(-)
diff --git a/fs/resctrl/monitor.c b/fs/resctrl/monitor.c
index 005d2babe692..ae8bc9953591 100644
--- a/fs/resctrl/monitor.c
+++ b/fs/resctrl/monitor.c
@@ -623,14 +623,22 @@ void mon_event_count(void *info)
rr->err = 0;
}
-static struct rdt_ctrl_domain *get_ctrl_domain_from_cpu(int cpu,
- struct rdt_resource *r)
+/*
+ * Find the software controller's ctrl domain that contains @cpu on resource @r.
+ *
+ * Only called from the mbm_over worker via update_mba_bw() where the returned
+ * domain is kept alive by cancel_delayed_work_sync() in
+ * resctrl_offline_ctrl_domain(). This drains this worker and then waits on
+ * rdtgroup_mutex held here before the architecture can free the ctrl domain.
+ *
+ * Context: Call from RCU read-side critical section.
+ */
+static struct rdt_ctrl_domain *get_sc_ctrl_domain_from_cpu(int cpu,
+ struct rdt_resource *r)
{
struct rdt_ctrl_domain *d;
- lockdep_assert_cpus_held();
-
- list_for_each_entry(d, &r->ctrl_domains, hdr.list) {
+ list_for_each_entry_rcu(d, &r->ctrl_domains, hdr.list) {
/* Find the domain that contains this CPU */
if (cpumask_test_cpu(cpu, &d->hdr.cpu_mask))
return d;
@@ -691,7 +699,8 @@ static void update_mba_bw(struct rdtgroup *rgrp, struct rdt_l3_mon_domain *dom_m
if (WARN_ON_ONCE(!pmbm_data))
return;
- dom_mba = get_ctrl_domain_from_cpu(smp_processor_id(), r_mba);
+ guard(rcu)();
+ dom_mba = get_sc_ctrl_domain_from_cpu(smp_processor_id(), r_mba);
if (!dom_mba) {
pr_warn_once("Failure to get domain for MBA update\n");
return;
@@ -794,9 +803,19 @@ void cqm_handle_limbo(struct work_struct *work)
unsigned long delay = msecs_to_jiffies(CQM_LIMBOCHECK_INTERVAL);
struct rdt_l3_mon_domain *d;
- cpus_read_lock();
+ /*
+ * Safe to run without CPU hotplug lock. Work is guaranteed to be
+ * canceled before the domain structure is removed.
+ */
mutex_lock(&rdtgroup_mutex);
+ /*
+ * Ensure the worker is dedicated to a CPU as intended and not
+ * relocated by workqueue subsystem as part of CPU going offline.
+ */
+ if (!is_percpu_thread())
+ goto out_unlock;
+
d = container_of(work, struct rdt_l3_mon_domain, cqm_limbo.work);
__check_limbo(d, false);
@@ -808,8 +827,8 @@ void cqm_handle_limbo(struct work_struct *work)
delay);
}
+out_unlock:
mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
}
/**
@@ -841,7 +860,10 @@ void mbm_handle_overflow(struct work_struct *work)
struct list_head *head;
struct rdt_resource *r;
- cpus_read_lock();
+ /*
+ * Safe to run without CPU hotplug lock. Work is guaranteed to be
+ * canceled before the domain structure is removed.
+ */
mutex_lock(&rdtgroup_mutex);
/*
@@ -851,6 +873,17 @@ void mbm_handle_overflow(struct work_struct *work)
if (!resctrl_mounted || !resctrl_arch_mon_capable())
goto out_unlock;
+ /*
+ * Ensure the worker is dedicated to a CPU and not relocated by
+ * workqueue subsystem as part of CPU going offline since reading
+ * events depend on smp_processor_id(). After passing this check
+ * smp_processor_id() is valid for entire duration of this worker
+ * since it runs with rdtgroup_mutex held and the offline handler needs
+ * rdtgroup_mutex to offline the CPU being run on here.
+ */
+ if (!is_percpu_thread())
+ goto out_unlock;
+
r = resctrl_arch_get_resource(RDT_RESOURCE_L3);
d = container_of(work, struct rdt_l3_mon_domain, mbm_over.work);
@@ -875,7 +908,6 @@ void mbm_handle_overflow(struct work_struct *work)
out_unlock:
mutex_unlock(&rdtgroup_mutex);
- cpus_read_unlock();
}
/**
diff --git a/fs/resctrl/rdtgroup.c b/fs/resctrl/rdtgroup.c
index 8eaedf31adcd..05ff14b8dc9d 100644
--- a/fs/resctrl/rdtgroup.c
+++ b/fs/resctrl/rdtgroup.c
@@ -4426,6 +4426,29 @@ static void domain_destroy_l3_mon_state(struct rdt_l3_mon_domain *d)
void resctrl_offline_ctrl_domain(struct rdt_resource *r, struct rdt_ctrl_domain *d)
{
+ /*
+ * mbm_handle_overflow() may dereference this ctrl domain via
+ * update_mba_bw()->get_sc_ctrl_domain_from_cpu(). The architecture has
+ * unlinked the domain from the RCU list and waited a grace period, so
+ * no new worker iteration can find it; drain any worker that already
+ * holds a pointer to it before the architecture frees the domain.
+ *
+ * Software controller is enabled/disabled on mount/unmount with
+ * cpus_read_lock() held. Running here with cpus_write_lock() so
+ * there are no concurrent changes to software controller status.
+ */
+ if (r->rid == RDT_RESOURCE_MBA && is_mba_sc(r)) {
+ struct rdt_resource *l3 = resctrl_arch_get_resource(RDT_RESOURCE_L3);
+ struct rdt_l3_mon_domain *mon_d;
+
+ list_for_each_entry(mon_d, &l3->mon_domains, hdr.list) {
+ if (mon_d->hdr.id == d->hdr.id) {
+ cancel_delayed_work_sync(&mon_d->mbm_over);
+ break;
+ }
+ }
+ }
+
mutex_lock(&rdtgroup_mutex);
if (supports_mba_mbps() && r->rid == RDT_RESOURCE_MBA)
@@ -4438,6 +4461,24 @@ void resctrl_offline_mon_domain(struct rdt_resource *r, struct rdt_domain_hdr *h
{
struct rdt_l3_mon_domain *d;
+ /*
+ * Called by architecture under CPU hotplug lock as it prepares to remove
+ * the domain which is guaranteed to be accessible here.
+ * The domain has been unlinked from the RCU list and a grace period
+ * has elapsed, so no new worker can be scheduled. Drain any worker that
+ * is in flight or pending before letting architecture proceed to free
+ * the domain that has the workers' struct delayed_work embedded.
+ * Do so before taking rdtgroup_mutex since the workers also acquire it.
+ */
+ if (r->rid == RDT_RESOURCE_L3 &&
+ domain_header_is_valid(hdr, RESCTRL_MON_DOMAIN, RDT_RESOURCE_L3)) {
+ d = container_of(hdr, struct rdt_l3_mon_domain, hdr);
+ if (resctrl_is_mbm_enabled())
+ cancel_delayed_work_sync(&d->mbm_over);
+ if (resctrl_is_mon_event_enabled(QOS_L3_OCCUP_EVENT_ID))
+ cancel_delayed_work_sync(&d->cqm_limbo);
+ }
+
mutex_lock(&rdtgroup_mutex);
/*
@@ -4454,8 +4495,6 @@ void resctrl_offline_mon_domain(struct rdt_resource *r, struct rdt_domain_hdr *h
goto out_unlock;
d = container_of(hdr, struct rdt_l3_mon_domain, hdr);
- if (resctrl_is_mbm_enabled())
- cancel_delayed_work(&d->mbm_over);
if (resctrl_is_mon_event_enabled(QOS_L3_OCCUP_EVENT_ID) && has_busy_rmid(d)) {
/*
* When a package is going down, forcefully
@@ -4466,7 +4505,6 @@ void resctrl_offline_mon_domain(struct rdt_resource *r, struct rdt_domain_hdr *h
* package never comes back.
*/
__check_limbo(d, true);
- cancel_delayed_work(&d->cqm_limbo);
}
domain_destroy_l3_mon_state(d);
@@ -4647,12 +4685,16 @@ void resctrl_offline_cpu(unsigned int cpu)
d = get_mon_domain_from_cpu(cpu, l3);
if (d) {
if (resctrl_is_mbm_enabled() && cpu == d->mbm_work_cpu) {
- cancel_delayed_work(&d->mbm_over);
+ mutex_unlock(&rdtgroup_mutex);
+ cancel_delayed_work_sync(&d->mbm_over);
+ mutex_lock(&rdtgroup_mutex);
mbm_setup_overflow_handler(d, 0, cpu);
}
if (resctrl_is_mon_event_enabled(QOS_L3_OCCUP_EVENT_ID) &&
cpu == d->cqm_work_cpu && has_busy_rmid(d)) {
- cancel_delayed_work(&d->cqm_limbo);
+ mutex_unlock(&rdtgroup_mutex);
+ cancel_delayed_work_sync(&d->cqm_limbo);
+ mutex_lock(&rdtgroup_mutex);
cqm_setup_limbo_handler(d, 0, cpu);
}
}
--
2.50.1
^ permalink raw reply [flat|nested] 5+ messages in thread