summaryrefslogtreecommitdiff
path: root/include/linux/refcount.h
diff options
context:
space:
mode:
authorWill Deacon <will@kernel.org>2019-11-21 14:58:57 +0300
committerIngo Molnar <mingo@kernel.org>2019-11-25 11:15:10 +0300
commitdcb786493f3e48da3272b710028d42ec608cfda1 (patch)
tree9058d4cfc8c638f8e1799f0c24b48d18fc7f0262 /include/linux/refcount.h
parent77e9971c79c29542ab7dd4140f9343bf2ff36158 (diff)
downloadlinux-dcb786493f3e48da3272b710028d42ec608cfda1.tar.xz
locking/refcount: Improve performance of generic REFCOUNT_FULL code
Rewrite the generic REFCOUNT_FULL implementation so that the saturation point is moved to INT_MIN / 2. This allows us to defer the sanity checks until after the atomic operation, which removes many uses of cmpxchg() in favour of atomic_fetch_{add,sub}(). Some crude perf results obtained from lkdtm show substantially less overhead, despite the checking: $ perf stat -r 3 -B -- echo {ATOMIC,REFCOUNT}_TIMING >/sys/kernel/debug/provoke-crash/DIRECT # arm64 ATOMIC_TIMING: 46.50451 +- 0.00134 seconds time elapsed ( +- 0.00% ) REFCOUNT_TIMING (REFCOUNT_FULL, mainline): 77.57522 +- 0.00982 seconds time elapsed ( +- 0.01% ) REFCOUNT_TIMING (REFCOUNT_FULL, this series): 48.7181 +- 0.0256 seconds time elapsed ( +- 0.05% ) # x86 ATOMIC_TIMING: 31.6225 +- 0.0776 seconds time elapsed ( +- 0.25% ) REFCOUNT_TIMING (!REFCOUNT_FULL, mainline/x86 asm): 31.6689 +- 0.0901 seconds time elapsed ( +- 0.28% ) REFCOUNT_TIMING (REFCOUNT_FULL, mainline): 53.203 +- 0.138 seconds time elapsed ( +- 0.26% ) REFCOUNT_TIMING (REFCOUNT_FULL, this series): 31.7408 +- 0.0486 seconds time elapsed ( +- 0.15% ) Signed-off-by: Will Deacon <will@kernel.org> Reviewed-by: Ard Biesheuvel <ardb@kernel.org> Reviewed-by: Kees Cook <keescook@chromium.org> Tested-by: Hanjun Guo <guohanjun@huawei.com> Tested-by: Jan Glauber <jglauber@marvell.com> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org> Cc: Elena Reshetova <elena.reshetova@intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20191121115902.2551-6-will@kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'include/linux/refcount.h')
-rw-r--r--include/linux/refcount.h131
1 files changed, 75 insertions, 56 deletions
diff --git a/include/linux/refcount.h b/include/linux/refcount.h
index e719b5b1220e..e3b218d669ce 100644
--- a/include/linux/refcount.h
+++ b/include/linux/refcount.h
@@ -47,8 +47,8 @@ static inline unsigned int refcount_read(const refcount_t *r)
#ifdef CONFIG_REFCOUNT_FULL
#include <linux/bug.h>
-#define REFCOUNT_MAX (UINT_MAX - 1)
-#define REFCOUNT_SATURATED UINT_MAX
+#define REFCOUNT_MAX INT_MAX
+#define REFCOUNT_SATURATED (INT_MIN / 2)
/*
* Variant of atomic_t specialized for reference counts.
@@ -56,9 +56,47 @@ static inline unsigned int refcount_read(const refcount_t *r)
* The interface matches the atomic_t interface (to aid in porting) but only
* provides the few functions one should use for reference counting.
*
- * It differs in that the counter saturates at REFCOUNT_SATURATED and will not
- * move once there. This avoids wrapping the counter and causing 'spurious'
- * use-after-free issues.
+ * Saturation semantics
+ * ====================
+ *
+ * refcount_t differs from atomic_t in that the counter saturates at
+ * REFCOUNT_SATURATED and will not move once there. This avoids wrapping the
+ * counter and causing 'spurious' use-after-free issues. In order to avoid the
+ * cost associated with introducing cmpxchg() loops into all of the saturating
+ * operations, we temporarily allow the counter to take on an unchecked value
+ * and then explicitly set it to REFCOUNT_SATURATED on detecting that underflow
+ * or overflow has occurred. Although this is racy when multiple threads
+ * access the refcount concurrently, by placing REFCOUNT_SATURATED roughly
+ * equidistant from 0 and INT_MAX we minimise the scope for error:
+ *
+ * INT_MAX REFCOUNT_SATURATED UINT_MAX
+ * 0 (0x7fff_ffff) (0xc000_0000) (0xffff_ffff)
+ * +--------------------------------+----------------+----------------+
+ * <---------- bad value! ---------->
+ *
+ * (in a signed view of the world, the "bad value" range corresponds to
+ * a negative counter value).
+ *
+ * As an example, consider a refcount_inc() operation that causes the counter
+ * to overflow:
+ *
+ * int old = atomic_fetch_add_relaxed(r);
+ * // old is INT_MAX, refcount now INT_MIN (0x8000_0000)
+ * if (old < 0)
+ * atomic_set(r, REFCOUNT_SATURATED);
+ *
+ * If another thread also performs a refcount_inc() operation between the two
+ * atomic operations, then the count will continue to edge closer to 0. If it
+ * reaches a value of 1 before /any/ of the threads reset it to the saturated
+ * value, then a concurrent refcount_dec_and_test() may erroneously free the
+ * underlying object. Given the precise timing details involved with the
+ * round-robin scheduling of each thread manipulating the refcount and the need
+ * to hit the race multiple times in succession, there doesn't appear to be a
+ * practical avenue of attack even if using refcount_add() operations with
+ * larger increments.
+ *
+ * Memory ordering
+ * ===============
*
* Memory ordering rules are slightly relaxed wrt regular atomic_t functions
* and provide only what is strictly required for refcounts.
@@ -109,25 +147,19 @@ static inline unsigned int refcount_read(const refcount_t *r)
*/
static inline __must_check bool refcount_add_not_zero(int i, refcount_t *r)
{
- unsigned int new, val = atomic_read(&r->refs);
+ int old = refcount_read(r);
do {
- if (!val)
- return false;
-
- if (unlikely(val == REFCOUNT_SATURATED))
- return true;
-
- new = val + i;
- if (new < val)
- new = REFCOUNT_SATURATED;
+ if (!old)
+ break;
+ } while (!atomic_try_cmpxchg_relaxed(&r->refs, &old, old + i));
- } while (!atomic_try_cmpxchg_relaxed(&r->refs, &val, new));
-
- WARN_ONCE(new == REFCOUNT_SATURATED,
- "refcount_t: saturated; leaking memory.\n");
+ if (unlikely(old < 0 || old + i < 0)) {
+ refcount_set(r, REFCOUNT_SATURATED);
+ WARN_ONCE(1, "refcount_t: saturated; leaking memory.\n");
+ }
- return true;
+ return old;
}
/**
@@ -148,7 +180,13 @@ static inline __must_check bool refcount_add_not_zero(int i, refcount_t *r)
*/
static inline void refcount_add(int i, refcount_t *r)
{
- WARN_ONCE(!refcount_add_not_zero(i, r), "refcount_t: addition on 0; use-after-free.\n");
+ int old = atomic_fetch_add_relaxed(i, &r->refs);
+
+ WARN_ONCE(!old, "refcount_t: addition on 0; use-after-free.\n");
+ if (unlikely(old <= 0 || old + i <= 0)) {
+ refcount_set(r, REFCOUNT_SATURATED);
+ WARN_ONCE(old, "refcount_t: saturated; leaking memory.\n");
+ }
}
/**
@@ -166,23 +204,7 @@ static inline void refcount_add(int i, refcount_t *r)
*/
static inline __must_check bool refcount_inc_not_zero(refcount_t *r)
{
- unsigned int new, val = atomic_read(&r->refs);
-
- do {
- new = val + 1;
-
- if (!val)
- return false;
-
- if (unlikely(!new))
- return true;
-
- } while (!atomic_try_cmpxchg_relaxed(&r->refs, &val, new));
-
- WARN_ONCE(new == REFCOUNT_SATURATED,
- "refcount_t: saturated; leaking memory.\n");
-
- return true;
+ return refcount_add_not_zero(1, r);
}
/**
@@ -199,7 +221,7 @@ static inline __must_check bool refcount_inc_not_zero(refcount_t *r)
*/
static inline void refcount_inc(refcount_t *r)
{
- WARN_ONCE(!refcount_inc_not_zero(r), "refcount_t: increment on 0; use-after-free.\n");
+ refcount_add(1, r);
}
/**
@@ -224,26 +246,19 @@ static inline void refcount_inc(refcount_t *r)
*/
static inline __must_check bool refcount_sub_and_test(int i, refcount_t *r)
{
- unsigned int new, val = atomic_read(&r->refs);
-
- do {
- if (unlikely(val == REFCOUNT_SATURATED))
- return false;
+ int old = atomic_fetch_sub_release(i, &r->refs);
- new = val - i;
- if (new > val) {
- WARN_ONCE(new > val, "refcount_t: underflow; use-after-free.\n");
- return false;
- }
-
- } while (!atomic_try_cmpxchg_release(&r->refs, &val, new));
-
- if (!new) {
+ if (old == i) {
smp_acquire__after_ctrl_dep();
return true;
}
- return false;
+ if (unlikely(old < 0 || old - i < 0)) {
+ refcount_set(r, REFCOUNT_SATURATED);
+ WARN_ONCE(1, "refcount_t: underflow; use-after-free.\n");
+ }
+
+ return false;
}
/**
@@ -276,9 +291,13 @@ static inline __must_check bool refcount_dec_and_test(refcount_t *r)
*/
static inline void refcount_dec(refcount_t *r)
{
- WARN_ONCE(refcount_dec_and_test(r), "refcount_t: decrement hit 0; leaking memory.\n");
-}
+ int old = atomic_fetch_sub_release(1, &r->refs);
+ if (unlikely(old <= 1)) {
+ refcount_set(r, REFCOUNT_SATURATED);
+ WARN_ONCE(1, "refcount_t: decrement hit 0; leaking memory.\n");
+ }
+}
#else /* CONFIG_REFCOUNT_FULL */
#define REFCOUNT_MAX INT_MAX