perf: Convert perf_event_context.refcount to refcount_t

atomic_t variables are currently used to implement reference
counters with the following properties:

 - counter is initialized to 1 using atomic_set()
 - a resource is freed upon counter reaching zero
 - once counter reaches zero, its further
   increments aren't allowed
 - counter schema uses basic atomic operations
   (set, inc, inc_not_zero, dec_and_test, etc.)

Such atomic variables should be converted to a newly provided
refcount_t type and API that prevents accidental counter overflows
and underflows. This is important since overflows and underflows
can lead to use-after-free situation and be exploitable.

The variable perf_event_context.refcount is used as pure reference counter.
Convert it to refcount_t and fix up the operations.

** Important note for maintainers:

Some functions from refcount_t API defined in lib/refcount.c
have different memory ordering guarantees than their atomic
counterparts. Please check Documentation/core-api/refcount-vs-atomic.rst
for more information.

Normally the differences should not matter since refcount_t provides
enough guarantees to satisfy the refcounting use cases, but in
some rare cases it might matter.
Please double check that you don't have some undocumented
memory guarantees for this variable usage.

For the perf_event_context.refcount it might make a difference
in following places:

 - get_ctx(), perf_event_ctx_lock_nested(), perf_lock_task_context()
   and __perf_event_ctx_lock_double(): increment in
   refcount_inc_not_zero() only guarantees control dependency
   on success vs. fully ordered atomic counterpart
 - put_ctx(): decrement in refcount_dec_and_test() provides
   RELEASE ordering and ACQUIRE ordering + control dependency on success
   vs. fully ordered atomic counterpart

Suggested-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Elena Reshetova <elena.reshetova@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: David Windsor <dwindsor@gmail.com>
Reviewed-by: Hans Liljestrand <ishkamiel@gmail.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: acme@kernel.org
Cc: namhyung@kernel.org
Link: https://lkml.kernel.org/r/1548678448-24458-2-git-send-email-elena.reshetova@intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index a79e59fc3b7d075b30865b3f7e15a36ced79991c..6cb5d483ab34d32281ad62c6a41ad21f36cf77b0 100644 (file)
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -54,6 +54,7 @@ struct perf_guest_info_callbacks {
 #include <linux/sysfs.h>
 #include <linux/perf_regs.h>
 #include <linux/cgroup.h>
+#include <linux/refcount.h>
 #include <asm/local.h>
 
 struct perf_callchain_entry {
@@ -737,7 +738,7 @@ struct perf_event_context {
        int                             nr_stat;
        int                             nr_freq;
        int                             rotate_disable;
-       atomic_t                        refcount;
+       refcount_t                      refcount;
        struct task_struct              *task;
 
        /*

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 5b89de7918d0679628658b38a649fb084d8a8fb7..677164d54547834d7fee6e222bca91430734e41d 100644 (file)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1172,7 +1172,7 @@ static void perf_event_ctx_deactivate(struct perf_event_context *ctx)
 
 static void get_ctx(struct perf_event_context *ctx)
 {
-       WARN_ON(!atomic_inc_not_zero(&ctx->refcount));
+       refcount_inc(&ctx->refcount);
 }
 
 static void free_ctx(struct rcu_head *head)
@@ -1186,7 +1186,7 @@ static void free_ctx(struct rcu_head *head)
 
 static void put_ctx(struct perf_event_context *ctx)
 {
-       if (atomic_dec_and_test(&ctx->refcount)) {
+       if (refcount_dec_and_test(&ctx->refcount)) {
                if (ctx->parent_ctx)
                        put_ctx(ctx->parent_ctx);
                if (ctx->task && ctx->task != TASK_TOMBSTONE)
@@ -1268,7 +1268,7 @@ perf_event_ctx_lock_nested(struct perf_event *event, int nesting)
 again:
        rcu_read_lock();
        ctx = READ_ONCE(event->ctx);
-       if (!atomic_inc_not_zero(&ctx->refcount)) {
+       if (!refcount_inc_not_zero(&ctx->refcount)) {
                rcu_read_unlock();
                goto again;
        }
@@ -1401,7 +1401,7 @@ retry:
                }
 
                if (ctx->task == TASK_TOMBSTONE ||
-                   !atomic_inc_not_zero(&ctx->refcount)) {
+                   !refcount_inc_not_zero(&ctx->refcount)) {
                        raw_spin_unlock(&ctx->lock);
                        ctx = NULL;
                } else {
@@ -4057,7 +4057,7 @@ static void __perf_event_init_context(struct perf_event_context *ctx)
        INIT_LIST_HEAD(&ctx->event_list);
        INIT_LIST_HEAD(&ctx->pinned_active);
        INIT_LIST_HEAD(&ctx->flexible_active);
-       atomic_set(&ctx->refcount, 1);
+       refcount_set(&ctx->refcount, 1);
 }
 
 static struct perf_event_context *
@@ -10613,7 +10613,7 @@ __perf_event_ctx_lock_double(struct perf_event *group_leader,
 again:
        rcu_read_lock();
        gctx = READ_ONCE(group_leader->ctx);
-       if (!atomic_inc_not_zero(&gctx->refcount)) {
+       if (!refcount_inc_not_zero(&gctx->refcount)) {
                rcu_read_unlock();
                goto again;
        }