* [PATCH kcsan 0/19] KCSAN updates for v5.10
@ 2020-08-31 18:17 Paul E. McKenney
2020-08-31 18:17 ` [PATCH kcsan 01/19] kcsan: Add support for atomic builtins paulmck
` (18 more replies)
0 siblings, 19 replies; 23+ messages in thread
From: Paul E. McKenney @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng
Hello!
This series provides KCSAN updates:
1. Add support for atomic builtins.
2. Add atomic builtin TSAN instrumentation to uaccess whitelist.
3. Add atomic builtin test case.
4. Support compounded read-write instrumentation.
5. objtool, kcsan: Add __tsan_read_write to uaccess whitelist.
6. Skew delay to be longer for certain access types.
7. Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks.
8. Test support for compound instrumentation.
9. instrumented.h: Introduce read-write instrumentation hooks.
10. asm-generic/bitops: Use instrument_read_write() where appropriate.
11. locking/atomics: Use read-write instrumentation for atomic RMWs.
12. Simplify debugfs counter to name mapping.
13. Simplify constant string handling.
14. Remove debugfs test command.
15. Show message if enabled early.
16. Use pr_fmt for consistency.
17. Optimize debugfs stats counters.
18. bitops, kcsan: Partially revert instrumentation for non-atomic bitops.
19. Use tracing-safe version of prandom.
Thanx, Paul
------------------------------------------------------------------------
include/asm-generic/atomic-instrumented.h | 330 +++++++++----------
include/asm-generic/bitops/instrumented-atomic.h | 6
include/asm-generic/bitops/instrumented-lock.h | 2
include/asm-generic/bitops/instrumented-non-atomic.h | 36 +-
include/linux/instrumented.h | 30 +
include/linux/kcsan-checks.h | 45 +-
kernel/kcsan/core.c | 238 +++++++++++--
kernel/kcsan/debugfs.c | 136 +------
kernel/kcsan/kcsan-test.c | 128 ++++++-
kernel/kcsan/kcsan.h | 12
kernel/kcsan/report.c | 10
kernel/kcsan/selftest.c | 8
lib/Kconfig.kcsan | 5
scripts/Makefile.kcsan | 2
scripts/atomic/gen-atomic-instrumented.sh | 21 -
tools/objtool/check.c | 55 +++
16 files changed, 697 insertions(+), 367 deletions(-)
^ permalink raw reply [flat|nested] 23+ messages in thread
* [PATCH kcsan 01/19] kcsan: Add support for atomic builtins
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 02/19] objtool: Add atomic builtin TSAN instrumentation to uaccess whitelist paulmck
` (17 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Some architectures (currently e.g. s390 partially) implement atomics
using the compiler's atomic builtins (__atomic_*, __sync_*). To support
enabling KCSAN on such architectures in future, or support experimental
use of these builtins, implement support for them.
We should also avoid breaking KCSAN kernels due to use (accidental or
otherwise) of atomic builtins in drivers, as has happened in the past:
https://lkml.kernel.org/r/5231d2c0-41d9-6721-e15f-a7eedf3ce69e@infradead.org
The instrumentation is subtly different from regular reads/writes: TSAN
instrumentation replaces the use of atomic builtins with a call into the
runtime, and the runtime's job is to also execute the desired atomic
operation. We rely on the __atomic_* compiler builtins, available with
all KCSAN-supported compilers, to implement each TSAN atomic
instrumentation function.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 110 ++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 110 insertions(+)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 9147ff6..682d9fd 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -879,3 +879,113 @@ void __tsan_init(void)
{
}
EXPORT_SYMBOL(__tsan_init);
+
+/*
+ * Instrumentation for atomic builtins (__atomic_*, __sync_*).
+ *
+ * Normal kernel code _should not_ be using them directly, but some
+ * architectures may implement some or all atomics using the compilers'
+ * builtins.
+ *
+ * Note: If an architecture decides to fully implement atomics using the
+ * builtins, because they are implicitly instrumented by KCSAN (and KASAN,
+ * etc.), implementing the ARCH_ATOMIC interface (to get instrumentation via
+ * atomic-instrumented) is no longer necessary.
+ *
+ * TSAN instrumentation replaces atomic accesses with calls to any of the below
+ * functions, whose job is to also execute the operation itself.
+ */
+
+#define DEFINE_TSAN_ATOMIC_LOAD_STORE(bits) \
+ u##bits __tsan_atomic##bits##_load(const u##bits *ptr, int memorder); \
+ u##bits __tsan_atomic##bits##_load(const u##bits *ptr, int memorder) \
+ { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_ATOMIC); \
+ return __atomic_load_n(ptr, memorder); \
+ } \
+ EXPORT_SYMBOL(__tsan_atomic##bits##_load); \
+ void __tsan_atomic##bits##_store(u##bits *ptr, u##bits v, int memorder); \
+ void __tsan_atomic##bits##_store(u##bits *ptr, u##bits v, int memorder) \
+ { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ __atomic_store_n(ptr, v, memorder); \
+ } \
+ EXPORT_SYMBOL(__tsan_atomic##bits##_store)
+
+#define DEFINE_TSAN_ATOMIC_RMW(op, bits, suffix) \
+ u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder); \
+ u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder) \
+ { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ return __atomic_##op##suffix(ptr, v, memorder); \
+ } \
+ EXPORT_SYMBOL(__tsan_atomic##bits##_##op)
+
+/*
+ * Note: CAS operations are always classified as write, even in case they
+ * fail. We cannot perform check_access() after a write, as it might lead to
+ * false positives, in cases such as:
+ *
+ * T0: __atomic_compare_exchange_n(&p->flag, &old, 1, ...)
+ *
+ * T1: if (__atomic_load_n(&p->flag, ...)) {
+ * modify *p;
+ * p->flag = 0;
+ * }
+ *
+ * The only downside is that, if there are 3 threads, with one CAS that
+ * succeeds, another CAS that fails, and an unmarked racing operation, we may
+ * point at the wrong CAS as the source of the race. However, if we assume that
+ * all CAS can succeed in some other execution, the data race is still valid.
+ */
+#define DEFINE_TSAN_ATOMIC_CMPXCHG(bits, strength, weak) \
+ int __tsan_atomic##bits##_compare_exchange_##strength(u##bits *ptr, u##bits *exp, \
+ u##bits val, int mo, int fail_mo); \
+ int __tsan_atomic##bits##_compare_exchange_##strength(u##bits *ptr, u##bits *exp, \
+ u##bits val, int mo, int fail_mo) \
+ { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ return __atomic_compare_exchange_n(ptr, exp, val, weak, mo, fail_mo); \
+ } \
+ EXPORT_SYMBOL(__tsan_atomic##bits##_compare_exchange_##strength)
+
+#define DEFINE_TSAN_ATOMIC_CMPXCHG_VAL(bits) \
+ u##bits __tsan_atomic##bits##_compare_exchange_val(u##bits *ptr, u##bits exp, u##bits val, \
+ int mo, int fail_mo); \
+ u##bits __tsan_atomic##bits##_compare_exchange_val(u##bits *ptr, u##bits exp, u##bits val, \
+ int mo, int fail_mo) \
+ { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ __atomic_compare_exchange_n(ptr, &exp, val, 0, mo, fail_mo); \
+ return exp; \
+ } \
+ EXPORT_SYMBOL(__tsan_atomic##bits##_compare_exchange_val)
+
+#define DEFINE_TSAN_ATOMIC_OPS(bits) \
+ DEFINE_TSAN_ATOMIC_LOAD_STORE(bits); \
+ DEFINE_TSAN_ATOMIC_RMW(exchange, bits, _n); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_add, bits, ); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_sub, bits, ); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_and, bits, ); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_or, bits, ); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_xor, bits, ); \
+ DEFINE_TSAN_ATOMIC_RMW(fetch_nand, bits, ); \
+ DEFINE_TSAN_ATOMIC_CMPXCHG(bits, strong, 0); \
+ DEFINE_TSAN_ATOMIC_CMPXCHG(bits, weak, 1); \
+ DEFINE_TSAN_ATOMIC_CMPXCHG_VAL(bits)
+
+DEFINE_TSAN_ATOMIC_OPS(8);
+DEFINE_TSAN_ATOMIC_OPS(16);
+DEFINE_TSAN_ATOMIC_OPS(32);
+DEFINE_TSAN_ATOMIC_OPS(64);
+
+void __tsan_atomic_thread_fence(int memorder);
+void __tsan_atomic_thread_fence(int memorder)
+{
+ __atomic_thread_fence(memorder);
+}
+EXPORT_SYMBOL(__tsan_atomic_thread_fence);
+
+void __tsan_atomic_signal_fence(int memorder);
+void __tsan_atomic_signal_fence(int memorder) { }
+EXPORT_SYMBOL(__tsan_atomic_signal_fence);
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 02/19] objtool: Add atomic builtin TSAN instrumentation to uaccess whitelist
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
2020-08-31 18:17 ` [PATCH kcsan 01/19] kcsan: Add support for atomic builtins paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 03/19] kcsan: Add atomic builtin test case paulmck
` (16 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng,
Paul E . McKenney, Josh Poimboeuf, Peter Zijlstra
From: Marco Elver <elver@google.com>
Adds the new TSAN functions that may be emitted for atomic builtins to
objtool's uaccess whitelist.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
---
tools/objtool/check.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 50 insertions(+)
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index e034a8f..7546a9d 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -528,6 +528,56 @@ static const char *uaccess_safe_builtin[] = {
"__tsan_write4",
"__tsan_write8",
"__tsan_write16",
+ "__tsan_atomic8_load",
+ "__tsan_atomic16_load",
+ "__tsan_atomic32_load",
+ "__tsan_atomic64_load",
+ "__tsan_atomic8_store",
+ "__tsan_atomic16_store",
+ "__tsan_atomic32_store",
+ "__tsan_atomic64_store",
+ "__tsan_atomic8_exchange",
+ "__tsan_atomic16_exchange",
+ "__tsan_atomic32_exchange",
+ "__tsan_atomic64_exchange",
+ "__tsan_atomic8_fetch_add",
+ "__tsan_atomic16_fetch_add",
+ "__tsan_atomic32_fetch_add",
+ "__tsan_atomic64_fetch_add",
+ "__tsan_atomic8_fetch_sub",
+ "__tsan_atomic16_fetch_sub",
+ "__tsan_atomic32_fetch_sub",
+ "__tsan_atomic64_fetch_sub",
+ "__tsan_atomic8_fetch_and",
+ "__tsan_atomic16_fetch_and",
+ "__tsan_atomic32_fetch_and",
+ "__tsan_atomic64_fetch_and",
+ "__tsan_atomic8_fetch_or",
+ "__tsan_atomic16_fetch_or",
+ "__tsan_atomic32_fetch_or",
+ "__tsan_atomic64_fetch_or",
+ "__tsan_atomic8_fetch_xor",
+ "__tsan_atomic16_fetch_xor",
+ "__tsan_atomic32_fetch_xor",
+ "__tsan_atomic64_fetch_xor",
+ "__tsan_atomic8_fetch_nand",
+ "__tsan_atomic16_fetch_nand",
+ "__tsan_atomic32_fetch_nand",
+ "__tsan_atomic64_fetch_nand",
+ "__tsan_atomic8_compare_exchange_strong",
+ "__tsan_atomic16_compare_exchange_strong",
+ "__tsan_atomic32_compare_exchange_strong",
+ "__tsan_atomic64_compare_exchange_strong",
+ "__tsan_atomic8_compare_exchange_weak",
+ "__tsan_atomic16_compare_exchange_weak",
+ "__tsan_atomic32_compare_exchange_weak",
+ "__tsan_atomic64_compare_exchange_weak",
+ "__tsan_atomic8_compare_exchange_val",
+ "__tsan_atomic16_compare_exchange_val",
+ "__tsan_atomic32_compare_exchange_val",
+ "__tsan_atomic64_compare_exchange_val",
+ "__tsan_atomic_thread_fence",
+ "__tsan_atomic_signal_fence",
/* KCOV */
"write_comp_data",
"check_kcov_mode",
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 03/19] kcsan: Add atomic builtin test case
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
2020-08-31 18:17 ` [PATCH kcsan 01/19] kcsan: Add support for atomic builtins paulmck
2020-08-31 18:17 ` [PATCH kcsan 02/19] objtool: Add atomic builtin TSAN instrumentation to uaccess whitelist paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 04/19] kcsan: Support compounded read-write instrumentation paulmck
` (15 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Adds test case to kcsan-test module, to test atomic builtin
instrumentation works.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/kcsan-test.c | 63 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 63 insertions(+)
diff --git a/kernel/kcsan/kcsan-test.c b/kernel/kcsan/kcsan-test.c
index fed6fcb..721180c 100644
--- a/kernel/kcsan/kcsan-test.c
+++ b/kernel/kcsan/kcsan-test.c
@@ -390,6 +390,15 @@ static noinline void test_kernel_seqlock_writer(void)
write_sequnlock_irqrestore(&test_seqlock, flags);
}
+static noinline void test_kernel_atomic_builtins(void)
+{
+ /*
+ * Generate concurrent accesses, expecting no reports, ensuring KCSAN
+ * treats builtin atomics as actually atomic.
+ */
+ __atomic_load_n(&test_var, __ATOMIC_RELAXED);
+}
+
/* ===== Test cases ===== */
/* Simple test with normal data race. */
@@ -853,6 +862,59 @@ static void test_seqlock_noreport(struct kunit *test)
}
/*
+ * Test atomic builtins work and required instrumentation functions exist. We
+ * also test that KCSAN understands they're atomic by racing with them via
+ * test_kernel_atomic_builtins(), and expect no reports.
+ *
+ * The atomic builtins _SHOULD NOT_ be used in normal kernel code!
+ */
+static void test_atomic_builtins(struct kunit *test)
+{
+ bool match_never = false;
+
+ begin_test_checks(test_kernel_atomic_builtins, test_kernel_atomic_builtins);
+ do {
+ long tmp;
+
+ kcsan_enable_current();
+
+ __atomic_store_n(&test_var, 42L, __ATOMIC_RELAXED);
+ KUNIT_EXPECT_EQ(test, 42L, __atomic_load_n(&test_var, __ATOMIC_RELAXED));
+
+ KUNIT_EXPECT_EQ(test, 42L, __atomic_exchange_n(&test_var, 20, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 20L, test_var);
+
+ tmp = 20L;
+ KUNIT_EXPECT_TRUE(test, __atomic_compare_exchange_n(&test_var, &tmp, 30L,
+ 0, __ATOMIC_RELAXED,
+ __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, tmp, 20L);
+ KUNIT_EXPECT_EQ(test, test_var, 30L);
+ KUNIT_EXPECT_FALSE(test, __atomic_compare_exchange_n(&test_var, &tmp, 40L,
+ 1, __ATOMIC_RELAXED,
+ __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, tmp, 30L);
+ KUNIT_EXPECT_EQ(test, test_var, 30L);
+
+ KUNIT_EXPECT_EQ(test, 30L, __atomic_fetch_add(&test_var, 1, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 31L, __atomic_fetch_sub(&test_var, 1, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 30L, __atomic_fetch_and(&test_var, 0xf, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 14L, __atomic_fetch_xor(&test_var, 0xf, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 1L, __atomic_fetch_or(&test_var, 0xf0, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, 241L, __atomic_fetch_nand(&test_var, 0xf, __ATOMIC_RELAXED));
+ KUNIT_EXPECT_EQ(test, -2L, test_var);
+
+ __atomic_thread_fence(__ATOMIC_SEQ_CST);
+ __atomic_signal_fence(__ATOMIC_SEQ_CST);
+
+ kcsan_disable_current();
+
+ match_never = report_available();
+ } while (!end_test_checks(match_never));
+ KUNIT_EXPECT_FALSE(test, match_never);
+}
+
+/*
* Each test case is run with different numbers of threads. Until KUnit supports
* passing arguments for each test case, we encode #threads in the test case
* name (read by get_num_threads()). [The '-' was chosen as a stylistic
@@ -891,6 +953,7 @@ static struct kunit_case kcsan_test_cases[] = {
KCSAN_KUNIT_CASE(test_assert_exclusive_access_scoped),
KCSAN_KUNIT_CASE(test_jiffies_noreport),
KCSAN_KUNIT_CASE(test_seqlock_noreport),
+ KCSAN_KUNIT_CASE(test_atomic_builtins),
{},
};
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 04/19] kcsan: Support compounded read-write instrumentation
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (2 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 03/19] kcsan: Add atomic builtin test case paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 05/19] objtool, kcsan: Add __tsan_read_write to uaccess whitelist paulmck
` (14 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Add support for compounded read-write instrumentation if supported by
the compiler. Adds the necessary instrumentation functions, and a new
type which is used to generate a more descriptive report.
Furthermore, such compounded memory access instrumentation is excluded
from the "assume aligned writes up to word size are atomic" rule,
because we cannot assume that the compiler emits code that is atomic for
compound ops.
LLVM/Clang added support for the feature in:
https://github.com/llvm/llvm-project/commit/785d41a261d136b64ab6c15c5d35f2adc5ad53e3
The new instrumentation is emitted for sets of memory accesses in the
same basic block to the same address with at least one read appearing
before a write. These typically result from compound operations such as
++, --, +=, -=, |=, &=, etc. but also equivalent forms such as "var =
var + 1". Where the compiler determines that it is equivalent to emit a
call to a single __tsan_read_write instead of separate __tsan_read and
__tsan_write, we can then benefit from improved performance and better
reporting for such access patterns.
The new reports now show that the ops are both reads and writes, for
example:
read-write to 0xffffffff90548a38 of 8 bytes by task 143 on cpu 3:
test_kernel_rmw_array+0x45/0xa0
access_thread+0x71/0xb0
kthread+0x21e/0x240
ret_from_fork+0x22/0x30
read-write to 0xffffffff90548a38 of 8 bytes by task 144 on cpu 2:
test_kernel_rmw_array+0x45/0xa0
access_thread+0x71/0xb0
kthread+0x21e/0x240
ret_from_fork+0x22/0x30
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
include/linux/kcsan-checks.h | 45 +++++++++++++++++++++++++++++---------------
kernel/kcsan/core.c | 23 +++++++++++++++++-----
kernel/kcsan/report.c | 4 ++++
scripts/Makefile.kcsan | 2 +-
4 files changed, 53 insertions(+), 21 deletions(-)
diff --git a/include/linux/kcsan-checks.h b/include/linux/kcsan-checks.h
index c5f6c1d..cf14840 100644
--- a/include/linux/kcsan-checks.h
+++ b/include/linux/kcsan-checks.h
@@ -7,19 +7,13 @@
#include <linux/compiler_attributes.h>
#include <linux/types.h>
-/*
- * ACCESS TYPE MODIFIERS
- *
- * <none>: normal read access;
- * WRITE : write access;
- * ATOMIC: access is atomic;
- * ASSERT: access is not a regular access, but an assertion;
- * SCOPED: access is a scoped access;
- */
-#define KCSAN_ACCESS_WRITE 0x1
-#define KCSAN_ACCESS_ATOMIC 0x2
-#define KCSAN_ACCESS_ASSERT 0x4
-#define KCSAN_ACCESS_SCOPED 0x8
+/* Access types -- if KCSAN_ACCESS_WRITE is not set, the access is a read. */
+#define KCSAN_ACCESS_WRITE (1 << 0) /* Access is a write. */
+#define KCSAN_ACCESS_COMPOUND (1 << 1) /* Compounded read-write instrumentation. */
+#define KCSAN_ACCESS_ATOMIC (1 << 2) /* Access is atomic. */
+/* The following are special, and never due to compiler instrumentation. */
+#define KCSAN_ACCESS_ASSERT (1 << 3) /* Access is an assertion. */
+#define KCSAN_ACCESS_SCOPED (1 << 4) /* Access is a scoped access. */
/*
* __kcsan_*: Always calls into the runtime when KCSAN is enabled. This may be used
@@ -205,6 +199,15 @@ static inline void __kcsan_disable_current(void) { }
__kcsan_check_access(ptr, size, KCSAN_ACCESS_WRITE)
/**
+ * __kcsan_check_read_write - check regular read-write access for races
+ *
+ * @ptr: address of access
+ * @size: size of access
+ */
+#define __kcsan_check_read_write(ptr, size) \
+ __kcsan_check_access(ptr, size, KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE)
+
+/**
* kcsan_check_read - check regular read access for races
*
* @ptr: address of access
@@ -221,18 +224,30 @@ static inline void __kcsan_disable_current(void) { }
#define kcsan_check_write(ptr, size) \
kcsan_check_access(ptr, size, KCSAN_ACCESS_WRITE)
+/**
+ * kcsan_check_read_write - check regular read-write access for races
+ *
+ * @ptr: address of access
+ * @size: size of access
+ */
+#define kcsan_check_read_write(ptr, size) \
+ kcsan_check_access(ptr, size, KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE)
+
/*
* Check for atomic accesses: if atomic accesses are not ignored, this simply
* aliases to kcsan_check_access(), otherwise becomes a no-op.
*/
#ifdef CONFIG_KCSAN_IGNORE_ATOMICS
-#define kcsan_check_atomic_read(...) do { } while (0)
-#define kcsan_check_atomic_write(...) do { } while (0)
+#define kcsan_check_atomic_read(...) do { } while (0)
+#define kcsan_check_atomic_write(...) do { } while (0)
+#define kcsan_check_atomic_read_write(...) do { } while (0)
#else
#define kcsan_check_atomic_read(ptr, size) \
kcsan_check_access(ptr, size, KCSAN_ACCESS_ATOMIC)
#define kcsan_check_atomic_write(ptr, size) \
kcsan_check_access(ptr, size, KCSAN_ACCESS_ATOMIC | KCSAN_ACCESS_WRITE)
+#define kcsan_check_atomic_read_write(ptr, size) \
+ kcsan_check_access(ptr, size, KCSAN_ACCESS_ATOMIC | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_COMPOUND)
#endif
/**
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 682d9fd..4c8b40b 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -223,7 +223,7 @@ is_atomic(const volatile void *ptr, size_t size, int type, struct kcsan_ctx *ctx
if (IS_ENABLED(CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC) &&
(type & KCSAN_ACCESS_WRITE) && size <= sizeof(long) &&
- IS_ALIGNED((unsigned long)ptr, size))
+ !(type & KCSAN_ACCESS_COMPOUND) && IS_ALIGNED((unsigned long)ptr, size))
return true; /* Assume aligned writes up to word size are atomic. */
if (ctx->atomic_next > 0) {
@@ -793,7 +793,17 @@ EXPORT_SYMBOL(__kcsan_check_access);
EXPORT_SYMBOL(__tsan_write##size); \
void __tsan_unaligned_write##size(void *ptr) \
__alias(__tsan_write##size); \
- EXPORT_SYMBOL(__tsan_unaligned_write##size)
+ EXPORT_SYMBOL(__tsan_unaligned_write##size); \
+ void __tsan_read_write##size(void *ptr); \
+ void __tsan_read_write##size(void *ptr) \
+ { \
+ check_access(ptr, size, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE); \
+ } \
+ EXPORT_SYMBOL(__tsan_read_write##size); \
+ void __tsan_unaligned_read_write##size(void *ptr) \
+ __alias(__tsan_read_write##size); \
+ EXPORT_SYMBOL(__tsan_unaligned_read_write##size)
DEFINE_TSAN_READ_WRITE(1);
DEFINE_TSAN_READ_WRITE(2);
@@ -916,7 +926,8 @@ EXPORT_SYMBOL(__tsan_init);
u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder); \
u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
return __atomic_##op##suffix(ptr, v, memorder); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_##op)
@@ -944,7 +955,8 @@ EXPORT_SYMBOL(__tsan_init);
int __tsan_atomic##bits##_compare_exchange_##strength(u##bits *ptr, u##bits *exp, \
u##bits val, int mo, int fail_mo) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
return __atomic_compare_exchange_n(ptr, exp, val, weak, mo, fail_mo); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_compare_exchange_##strength)
@@ -955,7 +967,8 @@ EXPORT_SYMBOL(__tsan_init);
u##bits __tsan_atomic##bits##_compare_exchange_val(u##bits *ptr, u##bits exp, u##bits val, \
int mo, int fail_mo) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
__atomic_compare_exchange_n(ptr, &exp, val, 0, mo, fail_mo); \
return exp; \
} \
diff --git a/kernel/kcsan/report.c b/kernel/kcsan/report.c
index 9d07e17..3e83a69 100644
--- a/kernel/kcsan/report.c
+++ b/kernel/kcsan/report.c
@@ -228,6 +228,10 @@ static const char *get_access_type(int type)
return "write";
case KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC:
return "write (marked)";
+ case KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE:
+ return "read-write";
+ case KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC:
+ return "read-write (marked)";
case KCSAN_ACCESS_SCOPED:
return "read (scoped)";
case KCSAN_ACCESS_SCOPED | KCSAN_ACCESS_ATOMIC:
diff --git a/scripts/Makefile.kcsan b/scripts/Makefile.kcsan
index c50f27b..c37f951 100644
--- a/scripts/Makefile.kcsan
+++ b/scripts/Makefile.kcsan
@@ -11,5 +11,5 @@ endif
# of some options does not break KCSAN nor causes false positive reports.
CFLAGS_KCSAN := -fsanitize=thread \
$(call cc-option,$(call cc-param,tsan-instrument-func-entry-exit=0) -fno-optimize-sibling-calls) \
- $(call cc-option,$(call cc-param,tsan-instrument-read-before-write=1)) \
+ $(call cc-option,$(call cc-param,tsan-compound-read-before-write=1),$(call cc-option,$(call cc-param,tsan-instrument-read-before-write=1))) \
$(call cc-param,tsan-distinguish-volatile=1)
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 05/19] objtool, kcsan: Add __tsan_read_write to uaccess whitelist
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (3 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 04/19] kcsan: Support compounded read-write instrumentation paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 06/19] kcsan: Skew delay to be longer for certain access types paulmck
` (13 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng,
Josh Poimboeuf, Paul E . McKenney
From: Marco Elver <elver@google.com>
Adds the new __tsan_read_write compound instrumentation to objtool's
uaccess whitelist.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
tools/objtool/check.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 7546a9d..5eee156 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -528,6 +528,11 @@ static const char *uaccess_safe_builtin[] = {
"__tsan_write4",
"__tsan_write8",
"__tsan_write16",
+ "__tsan_read_write1",
+ "__tsan_read_write2",
+ "__tsan_read_write4",
+ "__tsan_read_write8",
+ "__tsan_read_write16",
"__tsan_atomic8_load",
"__tsan_atomic16_load",
"__tsan_atomic32_load",
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 06/19] kcsan: Skew delay to be longer for certain access types
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (4 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 05/19] objtool, kcsan: Add __tsan_read_write to uaccess whitelist paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 07/19] kcsan: Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks paulmck
` (12 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
For compound instrumentation and assert accesses, skew the watchpoint
delay to be longer if randomized. This is useful to improve race
detection for such accesses.
For compound accesses we should increase the delay as we've aggregated
both read and write instrumentation. By giving up 1 call into the
runtime, we're less likely to set up a watchpoint and thus less likely
to detect a race. We can balance this by increasing the watchpoint
delay.
For assert accesses, we know these are of increased interest, and we
wish to increase our chances of detecting races for such checks.
Note that, kcsan_udelay_{task,interrupt} define the upper bound delays.
When randomized, delays are uniformly distributed between [0, delay].
Skewing the delay does not break this promise as long as the defined
upper bounds are still adhered to. The current skew results in delays
uniformly distributed between [delay/2, delay].
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 4c8b40b..95a364e 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -283,11 +283,15 @@ static __always_inline bool kcsan_is_enabled(void)
return READ_ONCE(kcsan_enabled) && get_ctx()->disable_count == 0;
}
-static inline unsigned int get_delay(void)
+static inline unsigned int get_delay(int type)
{
unsigned int delay = in_task() ? kcsan_udelay_task : kcsan_udelay_interrupt;
+ /* For certain access types, skew the random delay to be longer. */
+ unsigned int skew_delay_order =
+ (type & (KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_ASSERT)) ? 1 : 0;
+
return delay - (IS_ENABLED(CONFIG_KCSAN_DELAY_RANDOMIZE) ?
- prandom_u32_max(delay) :
+ prandom_u32_max(delay >> skew_delay_order) :
0);
}
@@ -470,7 +474,7 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
* Delay this thread, to increase probability of observing a racy
* conflicting access.
*/
- udelay(get_delay());
+ udelay(get_delay(type));
/*
* Re-read value, and check if it is as expected; if not, we infer a
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 07/19] kcsan: Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (5 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 06/19] kcsan: Skew delay to be longer for certain access types paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 08/19] kcsan: Test support for compound instrumentation paulmck
` (11 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks for the builtin atomics
instrumentation.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 30 ++++++++++++++++++++++--------
1 file changed, 22 insertions(+), 8 deletions(-)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 95a364e..99e5044 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -914,14 +914,19 @@ EXPORT_SYMBOL(__tsan_init);
u##bits __tsan_atomic##bits##_load(const u##bits *ptr, int memorder); \
u##bits __tsan_atomic##bits##_load(const u##bits *ptr, int memorder) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_ATOMIC); \
+ if (!IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS)) { \
+ check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_ATOMIC); \
+ } \
return __atomic_load_n(ptr, memorder); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_load); \
void __tsan_atomic##bits##_store(u##bits *ptr, u##bits v, int memorder); \
void __tsan_atomic##bits##_store(u##bits *ptr, u##bits v, int memorder) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ if (!IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS)) { \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ } \
__atomic_store_n(ptr, v, memorder); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_store)
@@ -930,8 +935,11 @@ EXPORT_SYMBOL(__tsan_init);
u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder); \
u##bits __tsan_atomic##bits##_##op(u##bits *ptr, u##bits v, int memorder) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, \
- KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ if (!IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS)) { \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | \
+ KCSAN_ACCESS_ATOMIC); \
+ } \
return __atomic_##op##suffix(ptr, v, memorder); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_##op)
@@ -959,8 +967,11 @@ EXPORT_SYMBOL(__tsan_init);
int __tsan_atomic##bits##_compare_exchange_##strength(u##bits *ptr, u##bits *exp, \
u##bits val, int mo, int fail_mo) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, \
- KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ if (!IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS)) { \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | \
+ KCSAN_ACCESS_ATOMIC); \
+ } \
return __atomic_compare_exchange_n(ptr, exp, val, weak, mo, fail_mo); \
} \
EXPORT_SYMBOL(__tsan_atomic##bits##_compare_exchange_##strength)
@@ -971,8 +982,11 @@ EXPORT_SYMBOL(__tsan_init);
u##bits __tsan_atomic##bits##_compare_exchange_val(u##bits *ptr, u##bits exp, u##bits val, \
int mo, int fail_mo) \
{ \
- check_access(ptr, bits / BITS_PER_BYTE, \
- KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC); \
+ if (!IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS)) { \
+ check_access(ptr, bits / BITS_PER_BYTE, \
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | \
+ KCSAN_ACCESS_ATOMIC); \
+ } \
__atomic_compare_exchange_n(ptr, &exp, val, 0, mo, fail_mo); \
return exp; \
} \
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 08/19] kcsan: Test support for compound instrumentation
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (6 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 07/19] kcsan: Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 09/19] instrumented.h: Introduce read-write instrumentation hooks paulmck
` (10 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Changes kcsan-test module to support checking reports that include
compound instrumentation. Since we should not fail the test if this
support is unavailable, we have to add a config variable that the test
can use to decide what to check for.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/kcsan-test.c | 65 +++++++++++++++++++++++++++++++++++++----------
lib/Kconfig.kcsan | 5 ++++
2 files changed, 56 insertions(+), 14 deletions(-)
diff --git a/kernel/kcsan/kcsan-test.c b/kernel/kcsan/kcsan-test.c
index 721180c..ebe7fd2 100644
--- a/kernel/kcsan/kcsan-test.c
+++ b/kernel/kcsan/kcsan-test.c
@@ -27,6 +27,12 @@
#include <linux/types.h>
#include <trace/events/printk.h>
+#ifdef CONFIG_CC_HAS_TSAN_COMPOUND_READ_BEFORE_WRITE
+#define __KCSAN_ACCESS_RW(alt) (KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE)
+#else
+#define __KCSAN_ACCESS_RW(alt) (alt)
+#endif
+
/* Points to current test-case memory access "kernels". */
static void (*access_kernels[2])(void);
@@ -186,20 +192,21 @@ static bool report_matches(const struct expect_report *r)
/* Access 1 & 2 */
for (i = 0; i < 2; ++i) {
+ const int ty = r->access[i].type;
const char *const access_type =
- (r->access[i].type & KCSAN_ACCESS_ASSERT) ?
- ((r->access[i].type & KCSAN_ACCESS_WRITE) ?
- "assert no accesses" :
- "assert no writes") :
- ((r->access[i].type & KCSAN_ACCESS_WRITE) ?
- "write" :
- "read");
+ (ty & KCSAN_ACCESS_ASSERT) ?
+ ((ty & KCSAN_ACCESS_WRITE) ?
+ "assert no accesses" :
+ "assert no writes") :
+ ((ty & KCSAN_ACCESS_WRITE) ?
+ ((ty & KCSAN_ACCESS_COMPOUND) ?
+ "read-write" :
+ "write") :
+ "read");
const char *const access_type_aux =
- (r->access[i].type & KCSAN_ACCESS_ATOMIC) ?
- " (marked)" :
- ((r->access[i].type & KCSAN_ACCESS_SCOPED) ?
- " (scoped)" :
- "");
+ (ty & KCSAN_ACCESS_ATOMIC) ?
+ " (marked)" :
+ ((ty & KCSAN_ACCESS_SCOPED) ? " (scoped)" : "");
if (i == 1) {
/* Access 2 */
@@ -277,6 +284,12 @@ static noinline void test_kernel_write_atomic(void)
WRITE_ONCE(test_var, READ_ONCE_NOCHECK(test_sink) + 1);
}
+static noinline void test_kernel_atomic_rmw(void)
+{
+ /* Use builtin, so we can set up the "bad" atomic/non-atomic scenario. */
+ __atomic_fetch_add(&test_var, 1, __ATOMIC_RELAXED);
+}
+
__no_kcsan
static noinline void test_kernel_write_uninstrumented(void) { test_var++; }
@@ -439,8 +452,8 @@ static void test_concurrent_races(struct kunit *test)
const struct expect_report expect = {
.access = {
/* NULL will match any address. */
- { test_kernel_rmw_array, NULL, 0, KCSAN_ACCESS_WRITE },
- { test_kernel_rmw_array, NULL, 0, 0 },
+ { test_kernel_rmw_array, NULL, 0, __KCSAN_ACCESS_RW(KCSAN_ACCESS_WRITE) },
+ { test_kernel_rmw_array, NULL, 0, __KCSAN_ACCESS_RW(0) },
},
};
static const struct expect_report never = {
@@ -629,6 +642,29 @@ static void test_read_plain_atomic_write(struct kunit *test)
KUNIT_EXPECT_TRUE(test, match_expect);
}
+/* Test that atomic RMWs generate correct report. */
+__no_kcsan
+static void test_read_plain_atomic_rmw(struct kunit *test)
+{
+ const struct expect_report expect = {
+ .access = {
+ { test_kernel_read, &test_var, sizeof(test_var), 0 },
+ { test_kernel_atomic_rmw, &test_var, sizeof(test_var),
+ KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ATOMIC },
+ },
+ };
+ bool match_expect = false;
+
+ if (IS_ENABLED(CONFIG_KCSAN_IGNORE_ATOMICS))
+ return;
+
+ begin_test_checks(test_kernel_read, test_kernel_atomic_rmw);
+ do {
+ match_expect = report_matches(&expect);
+ } while (!end_test_checks(match_expect));
+ KUNIT_EXPECT_TRUE(test, match_expect);
+}
+
/* Zero-sized accesses should never cause data race reports. */
__no_kcsan
static void test_zero_size_access(struct kunit *test)
@@ -942,6 +978,7 @@ static struct kunit_case kcsan_test_cases[] = {
KCSAN_KUNIT_CASE(test_write_write_struct_part),
KCSAN_KUNIT_CASE(test_read_atomic_write_atomic),
KCSAN_KUNIT_CASE(test_read_plain_atomic_write),
+ KCSAN_KUNIT_CASE(test_read_plain_atomic_rmw),
KCSAN_KUNIT_CASE(test_zero_size_access),
KCSAN_KUNIT_CASE(test_data_race),
KCSAN_KUNIT_CASE(test_assert_exclusive_writer),
diff --git a/lib/Kconfig.kcsan b/lib/Kconfig.kcsan
index 3d282d5..f271ff5 100644
--- a/lib/Kconfig.kcsan
+++ b/lib/Kconfig.kcsan
@@ -40,6 +40,11 @@ menuconfig KCSAN
if KCSAN
+# Compiler capabilities that should not fail the test if they are unavailable.
+config CC_HAS_TSAN_COMPOUND_READ_BEFORE_WRITE
+ def_bool (CC_IS_CLANG && $(cc-option,-fsanitize=thread -mllvm -tsan-compound-read-before-write=1)) || \
+ (CC_IS_GCC && $(cc-option,-fsanitize=thread --param tsan-compound-read-before-write=1))
+
config KCSAN_VERBOSE
bool "Show verbose reports with more information about system state"
depends on PROVE_LOCKING
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 09/19] instrumented.h: Introduce read-write instrumentation hooks
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (7 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 08/19] kcsan: Test support for compound instrumentation paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 10/19] asm-generic/bitops: Use instrument_read_write() where appropriate paulmck
` (9 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Introduce read-write instrumentation hooks, to more precisely denote an
operation's behaviour.
KCSAN is able to distinguish compound instrumentation, and with the new
instrumentation we then benefit from improved reporting. More
importantly, read-write compound operations should not implicitly be
treated as atomic, if they aren't actually atomic.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
include/linux/instrumented.h | 30 ++++++++++++++++++++++++++++++
1 file changed, 30 insertions(+)
diff --git a/include/linux/instrumented.h b/include/linux/instrumented.h
index 43e6ea5..42faebb 100644
--- a/include/linux/instrumented.h
+++ b/include/linux/instrumented.h
@@ -43,6 +43,21 @@ static __always_inline void instrument_write(const volatile void *v, size_t size
}
/**
+ * instrument_read_write - instrument regular read-write access
+ *
+ * Instrument a regular write access. The instrumentation should be inserted
+ * before the actual write happens.
+ *
+ * @ptr address of access
+ * @size size of access
+ */
+static __always_inline void instrument_read_write(const volatile void *v, size_t size)
+{
+ kasan_check_write(v, size);
+ kcsan_check_read_write(v, size);
+}
+
+/**
* instrument_atomic_read - instrument atomic read access
*
* Instrument an atomic read access. The instrumentation should be inserted
@@ -73,6 +88,21 @@ static __always_inline void instrument_atomic_write(const volatile void *v, size
}
/**
+ * instrument_atomic_read_write - instrument atomic read-write access
+ *
+ * Instrument an atomic read-write access. The instrumentation should be
+ * inserted before the actual write happens.
+ *
+ * @ptr address of access
+ * @size size of access
+ */
+static __always_inline void instrument_atomic_read_write(const volatile void *v, size_t size)
+{
+ kasan_check_write(v, size);
+ kcsan_check_atomic_read_write(v, size);
+}
+
+/**
* instrument_copy_to_user - instrument reads of copy_to_user
*
* Instrument reads from kernel memory, that are due to copy_to_user (and
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 10/19] asm-generic/bitops: Use instrument_read_write() where appropriate
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (8 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 09/19] instrumented.h: Introduce read-write instrumentation hooks paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 11/19] locking/atomics: Use read-write instrumentation for atomic RMWs paulmck
` (8 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Use the new instrument_read_write() where appropriate.
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
include/asm-generic/bitops/instrumented-atomic.h | 6 +++---
include/asm-generic/bitops/instrumented-lock.h | 2 +-
include/asm-generic/bitops/instrumented-non-atomic.h | 6 +++---
3 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/include/asm-generic/bitops/instrumented-atomic.h b/include/asm-generic/bitops/instrumented-atomic.h
index fb2cb33..81915dc 100644
--- a/include/asm-generic/bitops/instrumented-atomic.h
+++ b/include/asm-generic/bitops/instrumented-atomic.h
@@ -67,7 +67,7 @@ static inline void change_bit(long nr, volatile unsigned long *addr)
*/
static inline bool test_and_set_bit(long nr, volatile unsigned long *addr)
{
- instrument_atomic_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_atomic_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch_test_and_set_bit(nr, addr);
}
@@ -80,7 +80,7 @@ static inline bool test_and_set_bit(long nr, volatile unsigned long *addr)
*/
static inline bool test_and_clear_bit(long nr, volatile unsigned long *addr)
{
- instrument_atomic_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_atomic_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch_test_and_clear_bit(nr, addr);
}
@@ -93,7 +93,7 @@ static inline bool test_and_clear_bit(long nr, volatile unsigned long *addr)
*/
static inline bool test_and_change_bit(long nr, volatile unsigned long *addr)
{
- instrument_atomic_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_atomic_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch_test_and_change_bit(nr, addr);
}
diff --git a/include/asm-generic/bitops/instrumented-lock.h b/include/asm-generic/bitops/instrumented-lock.h
index b9bec46..75ef606 100644
--- a/include/asm-generic/bitops/instrumented-lock.h
+++ b/include/asm-generic/bitops/instrumented-lock.h
@@ -52,7 +52,7 @@ static inline void __clear_bit_unlock(long nr, volatile unsigned long *addr)
*/
static inline bool test_and_set_bit_lock(long nr, volatile unsigned long *addr)
{
- instrument_atomic_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_atomic_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch_test_and_set_bit_lock(nr, addr);
}
diff --git a/include/asm-generic/bitops/instrumented-non-atomic.h b/include/asm-generic/bitops/instrumented-non-atomic.h
index 20f788a..f86234c 100644
--- a/include/asm-generic/bitops/instrumented-non-atomic.h
+++ b/include/asm-generic/bitops/instrumented-non-atomic.h
@@ -68,7 +68,7 @@ static inline void __change_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
{
- instrument_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch___test_and_set_bit(nr, addr);
}
@@ -82,7 +82,7 @@ static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
{
- instrument_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch___test_and_clear_bit(nr, addr);
}
@@ -96,7 +96,7 @@ static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_change_bit(long nr, volatile unsigned long *addr)
{
- instrument_write(addr + BIT_WORD(nr), sizeof(long));
+ instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
return arch___test_and_change_bit(nr, addr);
}
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 11/19] locking/atomics: Use read-write instrumentation for atomic RMWs
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (9 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 10/19] asm-generic/bitops: Use instrument_read_write() where appropriate paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 12/19] kcsan: Simplify debugfs counter to name mapping paulmck
` (7 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Will Deacon,
Arnd Bergmann, linux-arch, Paul E . McKenney
From: Marco Elver <elver@google.com>
Use instrument_atomic_read_write() for atomic RMW ops.
Cc: Will Deacon <will@kernel.org>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: <linux-arch@vger.kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
include/asm-generic/atomic-instrumented.h | 330 +++++++++++++++---------------
scripts/atomic/gen-atomic-instrumented.sh | 21 +-
2 files changed, 180 insertions(+), 171 deletions(-)
diff --git a/include/asm-generic/atomic-instrumented.h b/include/asm-generic/atomic-instrumented.h
index 379986e..cd223b6 100644
--- a/include/asm-generic/atomic-instrumented.h
+++ b/include/asm-generic/atomic-instrumented.h
@@ -60,7 +60,7 @@ atomic_set_release(atomic_t *v, int i)
static __always_inline void
atomic_add(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_add(i, v);
}
#define atomic_add atomic_add
@@ -69,7 +69,7 @@ atomic_add(int i, atomic_t *v)
static __always_inline int
atomic_add_return(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_return(i, v);
}
#define atomic_add_return atomic_add_return
@@ -79,7 +79,7 @@ atomic_add_return(int i, atomic_t *v)
static __always_inline int
atomic_add_return_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_return_acquire(i, v);
}
#define atomic_add_return_acquire atomic_add_return_acquire
@@ -89,7 +89,7 @@ atomic_add_return_acquire(int i, atomic_t *v)
static __always_inline int
atomic_add_return_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_return_release(i, v);
}
#define atomic_add_return_release atomic_add_return_release
@@ -99,7 +99,7 @@ atomic_add_return_release(int i, atomic_t *v)
static __always_inline int
atomic_add_return_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_return_relaxed(i, v);
}
#define atomic_add_return_relaxed atomic_add_return_relaxed
@@ -109,7 +109,7 @@ atomic_add_return_relaxed(int i, atomic_t *v)
static __always_inline int
atomic_fetch_add(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_add(i, v);
}
#define atomic_fetch_add atomic_fetch_add
@@ -119,7 +119,7 @@ atomic_fetch_add(int i, atomic_t *v)
static __always_inline int
atomic_fetch_add_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_add_acquire(i, v);
}
#define atomic_fetch_add_acquire atomic_fetch_add_acquire
@@ -129,7 +129,7 @@ atomic_fetch_add_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_add_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_add_release(i, v);
}
#define atomic_fetch_add_release atomic_fetch_add_release
@@ -139,7 +139,7 @@ atomic_fetch_add_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_add_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_add_relaxed(i, v);
}
#define atomic_fetch_add_relaxed atomic_fetch_add_relaxed
@@ -148,7 +148,7 @@ atomic_fetch_add_relaxed(int i, atomic_t *v)
static __always_inline void
atomic_sub(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_sub(i, v);
}
#define atomic_sub atomic_sub
@@ -157,7 +157,7 @@ atomic_sub(int i, atomic_t *v)
static __always_inline int
atomic_sub_return(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_sub_return(i, v);
}
#define atomic_sub_return atomic_sub_return
@@ -167,7 +167,7 @@ atomic_sub_return(int i, atomic_t *v)
static __always_inline int
atomic_sub_return_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_sub_return_acquire(i, v);
}
#define atomic_sub_return_acquire atomic_sub_return_acquire
@@ -177,7 +177,7 @@ atomic_sub_return_acquire(int i, atomic_t *v)
static __always_inline int
atomic_sub_return_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_sub_return_release(i, v);
}
#define atomic_sub_return_release atomic_sub_return_release
@@ -187,7 +187,7 @@ atomic_sub_return_release(int i, atomic_t *v)
static __always_inline int
atomic_sub_return_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_sub_return_relaxed(i, v);
}
#define atomic_sub_return_relaxed atomic_sub_return_relaxed
@@ -197,7 +197,7 @@ atomic_sub_return_relaxed(int i, atomic_t *v)
static __always_inline int
atomic_fetch_sub(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_sub(i, v);
}
#define atomic_fetch_sub atomic_fetch_sub
@@ -207,7 +207,7 @@ atomic_fetch_sub(int i, atomic_t *v)
static __always_inline int
atomic_fetch_sub_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_sub_acquire(i, v);
}
#define atomic_fetch_sub_acquire atomic_fetch_sub_acquire
@@ -217,7 +217,7 @@ atomic_fetch_sub_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_sub_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_sub_release(i, v);
}
#define atomic_fetch_sub_release atomic_fetch_sub_release
@@ -227,7 +227,7 @@ atomic_fetch_sub_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_sub_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_sub_relaxed(i, v);
}
#define atomic_fetch_sub_relaxed atomic_fetch_sub_relaxed
@@ -237,7 +237,7 @@ atomic_fetch_sub_relaxed(int i, atomic_t *v)
static __always_inline void
atomic_inc(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_inc(v);
}
#define atomic_inc atomic_inc
@@ -247,7 +247,7 @@ atomic_inc(atomic_t *v)
static __always_inline int
atomic_inc_return(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_return(v);
}
#define atomic_inc_return atomic_inc_return
@@ -257,7 +257,7 @@ atomic_inc_return(atomic_t *v)
static __always_inline int
atomic_inc_return_acquire(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_return_acquire(v);
}
#define atomic_inc_return_acquire atomic_inc_return_acquire
@@ -267,7 +267,7 @@ atomic_inc_return_acquire(atomic_t *v)
static __always_inline int
atomic_inc_return_release(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_return_release(v);
}
#define atomic_inc_return_release atomic_inc_return_release
@@ -277,7 +277,7 @@ atomic_inc_return_release(atomic_t *v)
static __always_inline int
atomic_inc_return_relaxed(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_return_relaxed(v);
}
#define atomic_inc_return_relaxed atomic_inc_return_relaxed
@@ -287,7 +287,7 @@ atomic_inc_return_relaxed(atomic_t *v)
static __always_inline int
atomic_fetch_inc(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_inc(v);
}
#define atomic_fetch_inc atomic_fetch_inc
@@ -297,7 +297,7 @@ atomic_fetch_inc(atomic_t *v)
static __always_inline int
atomic_fetch_inc_acquire(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_inc_acquire(v);
}
#define atomic_fetch_inc_acquire atomic_fetch_inc_acquire
@@ -307,7 +307,7 @@ atomic_fetch_inc_acquire(atomic_t *v)
static __always_inline int
atomic_fetch_inc_release(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_inc_release(v);
}
#define atomic_fetch_inc_release atomic_fetch_inc_release
@@ -317,7 +317,7 @@ atomic_fetch_inc_release(atomic_t *v)
static __always_inline int
atomic_fetch_inc_relaxed(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_inc_relaxed(v);
}
#define atomic_fetch_inc_relaxed atomic_fetch_inc_relaxed
@@ -327,7 +327,7 @@ atomic_fetch_inc_relaxed(atomic_t *v)
static __always_inline void
atomic_dec(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_dec(v);
}
#define atomic_dec atomic_dec
@@ -337,7 +337,7 @@ atomic_dec(atomic_t *v)
static __always_inline int
atomic_dec_return(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_return(v);
}
#define atomic_dec_return atomic_dec_return
@@ -347,7 +347,7 @@ atomic_dec_return(atomic_t *v)
static __always_inline int
atomic_dec_return_acquire(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_return_acquire(v);
}
#define atomic_dec_return_acquire atomic_dec_return_acquire
@@ -357,7 +357,7 @@ atomic_dec_return_acquire(atomic_t *v)
static __always_inline int
atomic_dec_return_release(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_return_release(v);
}
#define atomic_dec_return_release atomic_dec_return_release
@@ -367,7 +367,7 @@ atomic_dec_return_release(atomic_t *v)
static __always_inline int
atomic_dec_return_relaxed(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_return_relaxed(v);
}
#define atomic_dec_return_relaxed atomic_dec_return_relaxed
@@ -377,7 +377,7 @@ atomic_dec_return_relaxed(atomic_t *v)
static __always_inline int
atomic_fetch_dec(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_dec(v);
}
#define atomic_fetch_dec atomic_fetch_dec
@@ -387,7 +387,7 @@ atomic_fetch_dec(atomic_t *v)
static __always_inline int
atomic_fetch_dec_acquire(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_dec_acquire(v);
}
#define atomic_fetch_dec_acquire atomic_fetch_dec_acquire
@@ -397,7 +397,7 @@ atomic_fetch_dec_acquire(atomic_t *v)
static __always_inline int
atomic_fetch_dec_release(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_dec_release(v);
}
#define atomic_fetch_dec_release atomic_fetch_dec_release
@@ -407,7 +407,7 @@ atomic_fetch_dec_release(atomic_t *v)
static __always_inline int
atomic_fetch_dec_relaxed(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_dec_relaxed(v);
}
#define atomic_fetch_dec_relaxed atomic_fetch_dec_relaxed
@@ -416,7 +416,7 @@ atomic_fetch_dec_relaxed(atomic_t *v)
static __always_inline void
atomic_and(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_and(i, v);
}
#define atomic_and atomic_and
@@ -425,7 +425,7 @@ atomic_and(int i, atomic_t *v)
static __always_inline int
atomic_fetch_and(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_and(i, v);
}
#define atomic_fetch_and atomic_fetch_and
@@ -435,7 +435,7 @@ atomic_fetch_and(int i, atomic_t *v)
static __always_inline int
atomic_fetch_and_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_and_acquire(i, v);
}
#define atomic_fetch_and_acquire atomic_fetch_and_acquire
@@ -445,7 +445,7 @@ atomic_fetch_and_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_and_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_and_release(i, v);
}
#define atomic_fetch_and_release atomic_fetch_and_release
@@ -455,7 +455,7 @@ atomic_fetch_and_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_and_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_and_relaxed(i, v);
}
#define atomic_fetch_and_relaxed atomic_fetch_and_relaxed
@@ -465,7 +465,7 @@ atomic_fetch_and_relaxed(int i, atomic_t *v)
static __always_inline void
atomic_andnot(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_andnot(i, v);
}
#define atomic_andnot atomic_andnot
@@ -475,7 +475,7 @@ atomic_andnot(int i, atomic_t *v)
static __always_inline int
atomic_fetch_andnot(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_andnot(i, v);
}
#define atomic_fetch_andnot atomic_fetch_andnot
@@ -485,7 +485,7 @@ atomic_fetch_andnot(int i, atomic_t *v)
static __always_inline int
atomic_fetch_andnot_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_andnot_acquire(i, v);
}
#define atomic_fetch_andnot_acquire atomic_fetch_andnot_acquire
@@ -495,7 +495,7 @@ atomic_fetch_andnot_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_andnot_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_andnot_release(i, v);
}
#define atomic_fetch_andnot_release atomic_fetch_andnot_release
@@ -505,7 +505,7 @@ atomic_fetch_andnot_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_andnot_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_andnot_relaxed(i, v);
}
#define atomic_fetch_andnot_relaxed atomic_fetch_andnot_relaxed
@@ -514,7 +514,7 @@ atomic_fetch_andnot_relaxed(int i, atomic_t *v)
static __always_inline void
atomic_or(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_or(i, v);
}
#define atomic_or atomic_or
@@ -523,7 +523,7 @@ atomic_or(int i, atomic_t *v)
static __always_inline int
atomic_fetch_or(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_or(i, v);
}
#define atomic_fetch_or atomic_fetch_or
@@ -533,7 +533,7 @@ atomic_fetch_or(int i, atomic_t *v)
static __always_inline int
atomic_fetch_or_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_or_acquire(i, v);
}
#define atomic_fetch_or_acquire atomic_fetch_or_acquire
@@ -543,7 +543,7 @@ atomic_fetch_or_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_or_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_or_release(i, v);
}
#define atomic_fetch_or_release atomic_fetch_or_release
@@ -553,7 +553,7 @@ atomic_fetch_or_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_or_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_or_relaxed(i, v);
}
#define atomic_fetch_or_relaxed atomic_fetch_or_relaxed
@@ -562,7 +562,7 @@ atomic_fetch_or_relaxed(int i, atomic_t *v)
static __always_inline void
atomic_xor(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic_xor(i, v);
}
#define atomic_xor atomic_xor
@@ -571,7 +571,7 @@ atomic_xor(int i, atomic_t *v)
static __always_inline int
atomic_fetch_xor(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_xor(i, v);
}
#define atomic_fetch_xor atomic_fetch_xor
@@ -581,7 +581,7 @@ atomic_fetch_xor(int i, atomic_t *v)
static __always_inline int
atomic_fetch_xor_acquire(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_xor_acquire(i, v);
}
#define atomic_fetch_xor_acquire atomic_fetch_xor_acquire
@@ -591,7 +591,7 @@ atomic_fetch_xor_acquire(int i, atomic_t *v)
static __always_inline int
atomic_fetch_xor_release(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_xor_release(i, v);
}
#define atomic_fetch_xor_release atomic_fetch_xor_release
@@ -601,7 +601,7 @@ atomic_fetch_xor_release(int i, atomic_t *v)
static __always_inline int
atomic_fetch_xor_relaxed(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_xor_relaxed(i, v);
}
#define atomic_fetch_xor_relaxed atomic_fetch_xor_relaxed
@@ -611,7 +611,7 @@ atomic_fetch_xor_relaxed(int i, atomic_t *v)
static __always_inline int
atomic_xchg(atomic_t *v, int i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_xchg(v, i);
}
#define atomic_xchg atomic_xchg
@@ -621,7 +621,7 @@ atomic_xchg(atomic_t *v, int i)
static __always_inline int
atomic_xchg_acquire(atomic_t *v, int i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_xchg_acquire(v, i);
}
#define atomic_xchg_acquire atomic_xchg_acquire
@@ -631,7 +631,7 @@ atomic_xchg_acquire(atomic_t *v, int i)
static __always_inline int
atomic_xchg_release(atomic_t *v, int i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_xchg_release(v, i);
}
#define atomic_xchg_release atomic_xchg_release
@@ -641,7 +641,7 @@ atomic_xchg_release(atomic_t *v, int i)
static __always_inline int
atomic_xchg_relaxed(atomic_t *v, int i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_xchg_relaxed(v, i);
}
#define atomic_xchg_relaxed atomic_xchg_relaxed
@@ -651,7 +651,7 @@ atomic_xchg_relaxed(atomic_t *v, int i)
static __always_inline int
atomic_cmpxchg(atomic_t *v, int old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_cmpxchg(v, old, new);
}
#define atomic_cmpxchg atomic_cmpxchg
@@ -661,7 +661,7 @@ atomic_cmpxchg(atomic_t *v, int old, int new)
static __always_inline int
atomic_cmpxchg_acquire(atomic_t *v, int old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_cmpxchg_acquire(v, old, new);
}
#define atomic_cmpxchg_acquire atomic_cmpxchg_acquire
@@ -671,7 +671,7 @@ atomic_cmpxchg_acquire(atomic_t *v, int old, int new)
static __always_inline int
atomic_cmpxchg_release(atomic_t *v, int old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_cmpxchg_release(v, old, new);
}
#define atomic_cmpxchg_release atomic_cmpxchg_release
@@ -681,7 +681,7 @@ atomic_cmpxchg_release(atomic_t *v, int old, int new)
static __always_inline int
atomic_cmpxchg_relaxed(atomic_t *v, int old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_cmpxchg_relaxed(v, old, new);
}
#define atomic_cmpxchg_relaxed atomic_cmpxchg_relaxed
@@ -691,8 +691,8 @@ atomic_cmpxchg_relaxed(atomic_t *v, int old, int new)
static __always_inline bool
atomic_try_cmpxchg(atomic_t *v, int *old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic_try_cmpxchg(v, old, new);
}
#define atomic_try_cmpxchg atomic_try_cmpxchg
@@ -702,8 +702,8 @@ atomic_try_cmpxchg(atomic_t *v, int *old, int new)
static __always_inline bool
atomic_try_cmpxchg_acquire(atomic_t *v, int *old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic_try_cmpxchg_acquire(v, old, new);
}
#define atomic_try_cmpxchg_acquire atomic_try_cmpxchg_acquire
@@ -713,8 +713,8 @@ atomic_try_cmpxchg_acquire(atomic_t *v, int *old, int new)
static __always_inline bool
atomic_try_cmpxchg_release(atomic_t *v, int *old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic_try_cmpxchg_release(v, old, new);
}
#define atomic_try_cmpxchg_release atomic_try_cmpxchg_release
@@ -724,8 +724,8 @@ atomic_try_cmpxchg_release(atomic_t *v, int *old, int new)
static __always_inline bool
atomic_try_cmpxchg_relaxed(atomic_t *v, int *old, int new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic_try_cmpxchg_relaxed(v, old, new);
}
#define atomic_try_cmpxchg_relaxed atomic_try_cmpxchg_relaxed
@@ -735,7 +735,7 @@ atomic_try_cmpxchg_relaxed(atomic_t *v, int *old, int new)
static __always_inline bool
atomic_sub_and_test(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_sub_and_test(i, v);
}
#define atomic_sub_and_test atomic_sub_and_test
@@ -745,7 +745,7 @@ atomic_sub_and_test(int i, atomic_t *v)
static __always_inline bool
atomic_dec_and_test(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_and_test(v);
}
#define atomic_dec_and_test atomic_dec_and_test
@@ -755,7 +755,7 @@ atomic_dec_and_test(atomic_t *v)
static __always_inline bool
atomic_inc_and_test(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_and_test(v);
}
#define atomic_inc_and_test atomic_inc_and_test
@@ -765,7 +765,7 @@ atomic_inc_and_test(atomic_t *v)
static __always_inline bool
atomic_add_negative(int i, atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_negative(i, v);
}
#define atomic_add_negative atomic_add_negative
@@ -775,7 +775,7 @@ atomic_add_negative(int i, atomic_t *v)
static __always_inline int
atomic_fetch_add_unless(atomic_t *v, int a, int u)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_fetch_add_unless(v, a, u);
}
#define atomic_fetch_add_unless atomic_fetch_add_unless
@@ -785,7 +785,7 @@ atomic_fetch_add_unless(atomic_t *v, int a, int u)
static __always_inline bool
atomic_add_unless(atomic_t *v, int a, int u)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_add_unless(v, a, u);
}
#define atomic_add_unless atomic_add_unless
@@ -795,7 +795,7 @@ atomic_add_unless(atomic_t *v, int a, int u)
static __always_inline bool
atomic_inc_not_zero(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_not_zero(v);
}
#define atomic_inc_not_zero atomic_inc_not_zero
@@ -805,7 +805,7 @@ atomic_inc_not_zero(atomic_t *v)
static __always_inline bool
atomic_inc_unless_negative(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_inc_unless_negative(v);
}
#define atomic_inc_unless_negative atomic_inc_unless_negative
@@ -815,7 +815,7 @@ atomic_inc_unless_negative(atomic_t *v)
static __always_inline bool
atomic_dec_unless_positive(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_unless_positive(v);
}
#define atomic_dec_unless_positive atomic_dec_unless_positive
@@ -825,7 +825,7 @@ atomic_dec_unless_positive(atomic_t *v)
static __always_inline int
atomic_dec_if_positive(atomic_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic_dec_if_positive(v);
}
#define atomic_dec_if_positive atomic_dec_if_positive
@@ -870,7 +870,7 @@ atomic64_set_release(atomic64_t *v, s64 i)
static __always_inline void
atomic64_add(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_add(i, v);
}
#define atomic64_add atomic64_add
@@ -879,7 +879,7 @@ atomic64_add(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_add_return(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_return(i, v);
}
#define atomic64_add_return atomic64_add_return
@@ -889,7 +889,7 @@ atomic64_add_return(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_add_return_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_return_acquire(i, v);
}
#define atomic64_add_return_acquire atomic64_add_return_acquire
@@ -899,7 +899,7 @@ atomic64_add_return_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_add_return_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_return_release(i, v);
}
#define atomic64_add_return_release atomic64_add_return_release
@@ -909,7 +909,7 @@ atomic64_add_return_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_add_return_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_return_relaxed(i, v);
}
#define atomic64_add_return_relaxed atomic64_add_return_relaxed
@@ -919,7 +919,7 @@ atomic64_add_return_relaxed(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_add(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_add(i, v);
}
#define atomic64_fetch_add atomic64_fetch_add
@@ -929,7 +929,7 @@ atomic64_fetch_add(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_add_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_add_acquire(i, v);
}
#define atomic64_fetch_add_acquire atomic64_fetch_add_acquire
@@ -939,7 +939,7 @@ atomic64_fetch_add_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_add_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_add_release(i, v);
}
#define atomic64_fetch_add_release atomic64_fetch_add_release
@@ -949,7 +949,7 @@ atomic64_fetch_add_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_add_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_add_relaxed(i, v);
}
#define atomic64_fetch_add_relaxed atomic64_fetch_add_relaxed
@@ -958,7 +958,7 @@ atomic64_fetch_add_relaxed(s64 i, atomic64_t *v)
static __always_inline void
atomic64_sub(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_sub(i, v);
}
#define atomic64_sub atomic64_sub
@@ -967,7 +967,7 @@ atomic64_sub(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_sub_return(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_sub_return(i, v);
}
#define atomic64_sub_return atomic64_sub_return
@@ -977,7 +977,7 @@ atomic64_sub_return(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_sub_return_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_sub_return_acquire(i, v);
}
#define atomic64_sub_return_acquire atomic64_sub_return_acquire
@@ -987,7 +987,7 @@ atomic64_sub_return_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_sub_return_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_sub_return_release(i, v);
}
#define atomic64_sub_return_release atomic64_sub_return_release
@@ -997,7 +997,7 @@ atomic64_sub_return_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_sub_return_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_sub_return_relaxed(i, v);
}
#define atomic64_sub_return_relaxed atomic64_sub_return_relaxed
@@ -1007,7 +1007,7 @@ atomic64_sub_return_relaxed(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_sub(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_sub(i, v);
}
#define atomic64_fetch_sub atomic64_fetch_sub
@@ -1017,7 +1017,7 @@ atomic64_fetch_sub(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_sub_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_sub_acquire(i, v);
}
#define atomic64_fetch_sub_acquire atomic64_fetch_sub_acquire
@@ -1027,7 +1027,7 @@ atomic64_fetch_sub_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_sub_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_sub_release(i, v);
}
#define atomic64_fetch_sub_release atomic64_fetch_sub_release
@@ -1037,7 +1037,7 @@ atomic64_fetch_sub_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_sub_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_sub_relaxed(i, v);
}
#define atomic64_fetch_sub_relaxed atomic64_fetch_sub_relaxed
@@ -1047,7 +1047,7 @@ atomic64_fetch_sub_relaxed(s64 i, atomic64_t *v)
static __always_inline void
atomic64_inc(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_inc(v);
}
#define atomic64_inc atomic64_inc
@@ -1057,7 +1057,7 @@ atomic64_inc(atomic64_t *v)
static __always_inline s64
atomic64_inc_return(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_return(v);
}
#define atomic64_inc_return atomic64_inc_return
@@ -1067,7 +1067,7 @@ atomic64_inc_return(atomic64_t *v)
static __always_inline s64
atomic64_inc_return_acquire(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_return_acquire(v);
}
#define atomic64_inc_return_acquire atomic64_inc_return_acquire
@@ -1077,7 +1077,7 @@ atomic64_inc_return_acquire(atomic64_t *v)
static __always_inline s64
atomic64_inc_return_release(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_return_release(v);
}
#define atomic64_inc_return_release atomic64_inc_return_release
@@ -1087,7 +1087,7 @@ atomic64_inc_return_release(atomic64_t *v)
static __always_inline s64
atomic64_inc_return_relaxed(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_return_relaxed(v);
}
#define atomic64_inc_return_relaxed atomic64_inc_return_relaxed
@@ -1097,7 +1097,7 @@ atomic64_inc_return_relaxed(atomic64_t *v)
static __always_inline s64
atomic64_fetch_inc(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_inc(v);
}
#define atomic64_fetch_inc atomic64_fetch_inc
@@ -1107,7 +1107,7 @@ atomic64_fetch_inc(atomic64_t *v)
static __always_inline s64
atomic64_fetch_inc_acquire(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_inc_acquire(v);
}
#define atomic64_fetch_inc_acquire atomic64_fetch_inc_acquire
@@ -1117,7 +1117,7 @@ atomic64_fetch_inc_acquire(atomic64_t *v)
static __always_inline s64
atomic64_fetch_inc_release(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_inc_release(v);
}
#define atomic64_fetch_inc_release atomic64_fetch_inc_release
@@ -1127,7 +1127,7 @@ atomic64_fetch_inc_release(atomic64_t *v)
static __always_inline s64
atomic64_fetch_inc_relaxed(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_inc_relaxed(v);
}
#define atomic64_fetch_inc_relaxed atomic64_fetch_inc_relaxed
@@ -1137,7 +1137,7 @@ atomic64_fetch_inc_relaxed(atomic64_t *v)
static __always_inline void
atomic64_dec(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_dec(v);
}
#define atomic64_dec atomic64_dec
@@ -1147,7 +1147,7 @@ atomic64_dec(atomic64_t *v)
static __always_inline s64
atomic64_dec_return(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_return(v);
}
#define atomic64_dec_return atomic64_dec_return
@@ -1157,7 +1157,7 @@ atomic64_dec_return(atomic64_t *v)
static __always_inline s64
atomic64_dec_return_acquire(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_return_acquire(v);
}
#define atomic64_dec_return_acquire atomic64_dec_return_acquire
@@ -1167,7 +1167,7 @@ atomic64_dec_return_acquire(atomic64_t *v)
static __always_inline s64
atomic64_dec_return_release(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_return_release(v);
}
#define atomic64_dec_return_release atomic64_dec_return_release
@@ -1177,7 +1177,7 @@ atomic64_dec_return_release(atomic64_t *v)
static __always_inline s64
atomic64_dec_return_relaxed(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_return_relaxed(v);
}
#define atomic64_dec_return_relaxed atomic64_dec_return_relaxed
@@ -1187,7 +1187,7 @@ atomic64_dec_return_relaxed(atomic64_t *v)
static __always_inline s64
atomic64_fetch_dec(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_dec(v);
}
#define atomic64_fetch_dec atomic64_fetch_dec
@@ -1197,7 +1197,7 @@ atomic64_fetch_dec(atomic64_t *v)
static __always_inline s64
atomic64_fetch_dec_acquire(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_dec_acquire(v);
}
#define atomic64_fetch_dec_acquire atomic64_fetch_dec_acquire
@@ -1207,7 +1207,7 @@ atomic64_fetch_dec_acquire(atomic64_t *v)
static __always_inline s64
atomic64_fetch_dec_release(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_dec_release(v);
}
#define atomic64_fetch_dec_release atomic64_fetch_dec_release
@@ -1217,7 +1217,7 @@ atomic64_fetch_dec_release(atomic64_t *v)
static __always_inline s64
atomic64_fetch_dec_relaxed(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_dec_relaxed(v);
}
#define atomic64_fetch_dec_relaxed atomic64_fetch_dec_relaxed
@@ -1226,7 +1226,7 @@ atomic64_fetch_dec_relaxed(atomic64_t *v)
static __always_inline void
atomic64_and(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_and(i, v);
}
#define atomic64_and atomic64_and
@@ -1235,7 +1235,7 @@ atomic64_and(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_and(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_and(i, v);
}
#define atomic64_fetch_and atomic64_fetch_and
@@ -1245,7 +1245,7 @@ atomic64_fetch_and(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_and_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_and_acquire(i, v);
}
#define atomic64_fetch_and_acquire atomic64_fetch_and_acquire
@@ -1255,7 +1255,7 @@ atomic64_fetch_and_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_and_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_and_release(i, v);
}
#define atomic64_fetch_and_release atomic64_fetch_and_release
@@ -1265,7 +1265,7 @@ atomic64_fetch_and_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_and_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_and_relaxed(i, v);
}
#define atomic64_fetch_and_relaxed atomic64_fetch_and_relaxed
@@ -1275,7 +1275,7 @@ atomic64_fetch_and_relaxed(s64 i, atomic64_t *v)
static __always_inline void
atomic64_andnot(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_andnot(i, v);
}
#define atomic64_andnot atomic64_andnot
@@ -1285,7 +1285,7 @@ atomic64_andnot(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_andnot(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_andnot(i, v);
}
#define atomic64_fetch_andnot atomic64_fetch_andnot
@@ -1295,7 +1295,7 @@ atomic64_fetch_andnot(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_andnot_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_andnot_acquire(i, v);
}
#define atomic64_fetch_andnot_acquire atomic64_fetch_andnot_acquire
@@ -1305,7 +1305,7 @@ atomic64_fetch_andnot_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_andnot_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_andnot_release(i, v);
}
#define atomic64_fetch_andnot_release atomic64_fetch_andnot_release
@@ -1315,7 +1315,7 @@ atomic64_fetch_andnot_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_andnot_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_andnot_relaxed(i, v);
}
#define atomic64_fetch_andnot_relaxed atomic64_fetch_andnot_relaxed
@@ -1324,7 +1324,7 @@ atomic64_fetch_andnot_relaxed(s64 i, atomic64_t *v)
static __always_inline void
atomic64_or(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_or(i, v);
}
#define atomic64_or atomic64_or
@@ -1333,7 +1333,7 @@ atomic64_or(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_or(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_or(i, v);
}
#define atomic64_fetch_or atomic64_fetch_or
@@ -1343,7 +1343,7 @@ atomic64_fetch_or(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_or_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_or_acquire(i, v);
}
#define atomic64_fetch_or_acquire atomic64_fetch_or_acquire
@@ -1353,7 +1353,7 @@ atomic64_fetch_or_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_or_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_or_release(i, v);
}
#define atomic64_fetch_or_release atomic64_fetch_or_release
@@ -1363,7 +1363,7 @@ atomic64_fetch_or_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_or_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_or_relaxed(i, v);
}
#define atomic64_fetch_or_relaxed atomic64_fetch_or_relaxed
@@ -1372,7 +1372,7 @@ atomic64_fetch_or_relaxed(s64 i, atomic64_t *v)
static __always_inline void
atomic64_xor(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
arch_atomic64_xor(i, v);
}
#define atomic64_xor atomic64_xor
@@ -1381,7 +1381,7 @@ atomic64_xor(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_xor(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_xor(i, v);
}
#define atomic64_fetch_xor atomic64_fetch_xor
@@ -1391,7 +1391,7 @@ atomic64_fetch_xor(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_xor_acquire(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_xor_acquire(i, v);
}
#define atomic64_fetch_xor_acquire atomic64_fetch_xor_acquire
@@ -1401,7 +1401,7 @@ atomic64_fetch_xor_acquire(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_xor_release(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_xor_release(i, v);
}
#define atomic64_fetch_xor_release atomic64_fetch_xor_release
@@ -1411,7 +1411,7 @@ atomic64_fetch_xor_release(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_xor_relaxed(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_xor_relaxed(i, v);
}
#define atomic64_fetch_xor_relaxed atomic64_fetch_xor_relaxed
@@ -1421,7 +1421,7 @@ atomic64_fetch_xor_relaxed(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_xchg(atomic64_t *v, s64 i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_xchg(v, i);
}
#define atomic64_xchg atomic64_xchg
@@ -1431,7 +1431,7 @@ atomic64_xchg(atomic64_t *v, s64 i)
static __always_inline s64
atomic64_xchg_acquire(atomic64_t *v, s64 i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_xchg_acquire(v, i);
}
#define atomic64_xchg_acquire atomic64_xchg_acquire
@@ -1441,7 +1441,7 @@ atomic64_xchg_acquire(atomic64_t *v, s64 i)
static __always_inline s64
atomic64_xchg_release(atomic64_t *v, s64 i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_xchg_release(v, i);
}
#define atomic64_xchg_release atomic64_xchg_release
@@ -1451,7 +1451,7 @@ atomic64_xchg_release(atomic64_t *v, s64 i)
static __always_inline s64
atomic64_xchg_relaxed(atomic64_t *v, s64 i)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_xchg_relaxed(v, i);
}
#define atomic64_xchg_relaxed atomic64_xchg_relaxed
@@ -1461,7 +1461,7 @@ atomic64_xchg_relaxed(atomic64_t *v, s64 i)
static __always_inline s64
atomic64_cmpxchg(atomic64_t *v, s64 old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_cmpxchg(v, old, new);
}
#define atomic64_cmpxchg atomic64_cmpxchg
@@ -1471,7 +1471,7 @@ atomic64_cmpxchg(atomic64_t *v, s64 old, s64 new)
static __always_inline s64
atomic64_cmpxchg_acquire(atomic64_t *v, s64 old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_cmpxchg_acquire(v, old, new);
}
#define atomic64_cmpxchg_acquire atomic64_cmpxchg_acquire
@@ -1481,7 +1481,7 @@ atomic64_cmpxchg_acquire(atomic64_t *v, s64 old, s64 new)
static __always_inline s64
atomic64_cmpxchg_release(atomic64_t *v, s64 old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_cmpxchg_release(v, old, new);
}
#define atomic64_cmpxchg_release atomic64_cmpxchg_release
@@ -1491,7 +1491,7 @@ atomic64_cmpxchg_release(atomic64_t *v, s64 old, s64 new)
static __always_inline s64
atomic64_cmpxchg_relaxed(atomic64_t *v, s64 old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_cmpxchg_relaxed(v, old, new);
}
#define atomic64_cmpxchg_relaxed atomic64_cmpxchg_relaxed
@@ -1501,8 +1501,8 @@ atomic64_cmpxchg_relaxed(atomic64_t *v, s64 old, s64 new)
static __always_inline bool
atomic64_try_cmpxchg(atomic64_t *v, s64 *old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic64_try_cmpxchg(v, old, new);
}
#define atomic64_try_cmpxchg atomic64_try_cmpxchg
@@ -1512,8 +1512,8 @@ atomic64_try_cmpxchg(atomic64_t *v, s64 *old, s64 new)
static __always_inline bool
atomic64_try_cmpxchg_acquire(atomic64_t *v, s64 *old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic64_try_cmpxchg_acquire(v, old, new);
}
#define atomic64_try_cmpxchg_acquire atomic64_try_cmpxchg_acquire
@@ -1523,8 +1523,8 @@ atomic64_try_cmpxchg_acquire(atomic64_t *v, s64 *old, s64 new)
static __always_inline bool
atomic64_try_cmpxchg_release(atomic64_t *v, s64 *old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic64_try_cmpxchg_release(v, old, new);
}
#define atomic64_try_cmpxchg_release atomic64_try_cmpxchg_release
@@ -1534,8 +1534,8 @@ atomic64_try_cmpxchg_release(atomic64_t *v, s64 *old, s64 new)
static __always_inline bool
atomic64_try_cmpxchg_relaxed(atomic64_t *v, s64 *old, s64 new)
{
- instrument_atomic_write(v, sizeof(*v));
- instrument_atomic_write(old, sizeof(*old));
+ instrument_atomic_read_write(v, sizeof(*v));
+ instrument_atomic_read_write(old, sizeof(*old));
return arch_atomic64_try_cmpxchg_relaxed(v, old, new);
}
#define atomic64_try_cmpxchg_relaxed atomic64_try_cmpxchg_relaxed
@@ -1545,7 +1545,7 @@ atomic64_try_cmpxchg_relaxed(atomic64_t *v, s64 *old, s64 new)
static __always_inline bool
atomic64_sub_and_test(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_sub_and_test(i, v);
}
#define atomic64_sub_and_test atomic64_sub_and_test
@@ -1555,7 +1555,7 @@ atomic64_sub_and_test(s64 i, atomic64_t *v)
static __always_inline bool
atomic64_dec_and_test(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_and_test(v);
}
#define atomic64_dec_and_test atomic64_dec_and_test
@@ -1565,7 +1565,7 @@ atomic64_dec_and_test(atomic64_t *v)
static __always_inline bool
atomic64_inc_and_test(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_and_test(v);
}
#define atomic64_inc_and_test atomic64_inc_and_test
@@ -1575,7 +1575,7 @@ atomic64_inc_and_test(atomic64_t *v)
static __always_inline bool
atomic64_add_negative(s64 i, atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_negative(i, v);
}
#define atomic64_add_negative atomic64_add_negative
@@ -1585,7 +1585,7 @@ atomic64_add_negative(s64 i, atomic64_t *v)
static __always_inline s64
atomic64_fetch_add_unless(atomic64_t *v, s64 a, s64 u)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_fetch_add_unless(v, a, u);
}
#define atomic64_fetch_add_unless atomic64_fetch_add_unless
@@ -1595,7 +1595,7 @@ atomic64_fetch_add_unless(atomic64_t *v, s64 a, s64 u)
static __always_inline bool
atomic64_add_unless(atomic64_t *v, s64 a, s64 u)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_add_unless(v, a, u);
}
#define atomic64_add_unless atomic64_add_unless
@@ -1605,7 +1605,7 @@ atomic64_add_unless(atomic64_t *v, s64 a, s64 u)
static __always_inline bool
atomic64_inc_not_zero(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_not_zero(v);
}
#define atomic64_inc_not_zero atomic64_inc_not_zero
@@ -1615,7 +1615,7 @@ atomic64_inc_not_zero(atomic64_t *v)
static __always_inline bool
atomic64_inc_unless_negative(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_inc_unless_negative(v);
}
#define atomic64_inc_unless_negative atomic64_inc_unless_negative
@@ -1625,7 +1625,7 @@ atomic64_inc_unless_negative(atomic64_t *v)
static __always_inline bool
atomic64_dec_unless_positive(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_unless_positive(v);
}
#define atomic64_dec_unless_positive atomic64_dec_unless_positive
@@ -1635,7 +1635,7 @@ atomic64_dec_unless_positive(atomic64_t *v)
static __always_inline s64
atomic64_dec_if_positive(atomic64_t *v)
{
- instrument_atomic_write(v, sizeof(*v));
+ instrument_atomic_read_write(v, sizeof(*v));
return arch_atomic64_dec_if_positive(v);
}
#define atomic64_dec_if_positive atomic64_dec_if_positive
@@ -1786,4 +1786,4 @@ atomic64_dec_if_positive(atomic64_t *v)
})
#endif /* _ASM_GENERIC_ATOMIC_INSTRUMENTED_H */
-// 89bf97f3a7509b740845e51ddf31055b48a81f40
+// 9d5e6a315fb1335d02f0ccd3655a91c3dafcc63e
diff --git a/scripts/atomic/gen-atomic-instrumented.sh b/scripts/atomic/gen-atomic-instrumented.sh
index 6afadf7..2b7fec7 100755
--- a/scripts/atomic/gen-atomic-instrumented.sh
+++ b/scripts/atomic/gen-atomic-instrumented.sh
@@ -5,9 +5,10 @@ ATOMICDIR=$(dirname $0)
. ${ATOMICDIR}/atomic-tbl.sh
-#gen_param_check(arg)
+#gen_param_check(meta, arg)
gen_param_check()
{
+ local meta="$1"; shift
local arg="$1"; shift
local type="${arg%%:*}"
local name="$(gen_param_name "${arg}")"
@@ -17,17 +18,25 @@ gen_param_check()
i) return;;
esac
- # We don't write to constant parameters
- [ ${type#c} != ${type} ] && rw="read"
+ if [ ${type#c} != ${type} ]; then
+ # We don't write to constant parameters.
+ rw="read"
+ elif [ "${meta}" != "s" ]; then
+ # An atomic RMW: if this parameter is not a constant, and this atomic is
+ # not just a 's'tore, this parameter is both read from and written to.
+ rw="read_write"
+ fi
printf "\tinstrument_atomic_${rw}(${name}, sizeof(*${name}));\n"
}
-#gen_param_check(arg...)
+#gen_params_checks(meta, arg...)
gen_params_checks()
{
+ local meta="$1"; shift
+
while [ "$#" -gt 0 ]; do
- gen_param_check "$1"
+ gen_param_check "$meta" "$1"
shift;
done
}
@@ -77,7 +86,7 @@ gen_proto_order_variant()
local ret="$(gen_ret_type "${meta}" "${int}")"
local params="$(gen_params "${int}" "${atomic}" "$@")"
- local checks="$(gen_params_checks "$@")"
+ local checks="$(gen_params_checks "${meta}" "$@")"
local args="$(gen_args "$@")"
local retstmt="$(gen_ret_stmt "${meta}")"
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 12/19] kcsan: Simplify debugfs counter to name mapping
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (10 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 11/19] locking/atomics: Use read-write instrumentation for atomic RMWs paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:17 ` [PATCH kcsan 13/19] kcsan: Simplify constant string handling paulmck
` (6 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Simplify counter ID to name mapping by using an array with designated
inits. This way, we can turn a run-time BUG() into a compile-time static
assertion failure if a counter name is missing.
No functional change intended.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/debugfs.c | 33 +++++++++++++--------------------
1 file changed, 13 insertions(+), 20 deletions(-)
diff --git a/kernel/kcsan/debugfs.c b/kernel/kcsan/debugfs.c
index 023e49c..3a9566a 100644
--- a/kernel/kcsan/debugfs.c
+++ b/kernel/kcsan/debugfs.c
@@ -19,6 +19,18 @@
* Statistics counters.
*/
static atomic_long_t counters[KCSAN_COUNTER_COUNT];
+static const char *const counter_names[] = {
+ [KCSAN_COUNTER_USED_WATCHPOINTS] = "used_watchpoints",
+ [KCSAN_COUNTER_SETUP_WATCHPOINTS] = "setup_watchpoints",
+ [KCSAN_COUNTER_DATA_RACES] = "data_races",
+ [KCSAN_COUNTER_ASSERT_FAILURES] = "assert_failures",
+ [KCSAN_COUNTER_NO_CAPACITY] = "no_capacity",
+ [KCSAN_COUNTER_REPORT_RACES] = "report_races",
+ [KCSAN_COUNTER_RACES_UNKNOWN_ORIGIN] = "races_unknown_origin",
+ [KCSAN_COUNTER_UNENCODABLE_ACCESSES] = "unencodable_accesses",
+ [KCSAN_COUNTER_ENCODING_FALSE_POSITIVES] = "encoding_false_positives",
+};
+static_assert(ARRAY_SIZE(counter_names) == KCSAN_COUNTER_COUNT);
/*
* Addresses for filtering functions from reporting. This list can be used as a
@@ -39,24 +51,6 @@ static struct {
};
static DEFINE_SPINLOCK(report_filterlist_lock);
-static const char *counter_to_name(enum kcsan_counter_id id)
-{
- switch (id) {
- case KCSAN_COUNTER_USED_WATCHPOINTS: return "used_watchpoints";
- case KCSAN_COUNTER_SETUP_WATCHPOINTS: return "setup_watchpoints";
- case KCSAN_COUNTER_DATA_RACES: return "data_races";
- case KCSAN_COUNTER_ASSERT_FAILURES: return "assert_failures";
- case KCSAN_COUNTER_NO_CAPACITY: return "no_capacity";
- case KCSAN_COUNTER_REPORT_RACES: return "report_races";
- case KCSAN_COUNTER_RACES_UNKNOWN_ORIGIN: return "races_unknown_origin";
- case KCSAN_COUNTER_UNENCODABLE_ACCESSES: return "unencodable_accesses";
- case KCSAN_COUNTER_ENCODING_FALSE_POSITIVES: return "encoding_false_positives";
- case KCSAN_COUNTER_COUNT:
- BUG();
- }
- return NULL;
-}
-
void kcsan_counter_inc(enum kcsan_counter_id id)
{
atomic_long_inc(&counters[id]);
@@ -271,8 +265,7 @@ static int show_info(struct seq_file *file, void *v)
/* show stats */
seq_printf(file, "enabled: %i\n", READ_ONCE(kcsan_enabled));
for (i = 0; i < KCSAN_COUNTER_COUNT; ++i)
- seq_printf(file, "%s: %ld\n", counter_to_name(i),
- atomic_long_read(&counters[i]));
+ seq_printf(file, "%s: %ld\n", counter_names[i], atomic_long_read(&counters[i]));
/* show filter functions, and filter type */
spin_lock_irqsave(&report_filterlist_lock, flags);
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 13/19] kcsan: Simplify constant string handling
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (11 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 12/19] kcsan: Simplify debugfs counter to name mapping paulmck
@ 2020-08-31 18:17 ` paulmck
2020-08-31 18:18 ` [PATCH kcsan 14/19] kcsan: Remove debugfs test command paulmck
` (5 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:17 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Simplify checking prefixes and length calculation of constant strings.
For the former, the kernel provides str_has_prefix(), and the latter we
should just use strlen("..") because GCC and Clang have optimizations
that optimize these into constants.
No functional change intended.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/debugfs.c | 8 ++++----
kernel/kcsan/report.c | 4 ++--
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/kernel/kcsan/debugfs.c b/kernel/kcsan/debugfs.c
index 3a9566a..116bdd8 100644
--- a/kernel/kcsan/debugfs.c
+++ b/kernel/kcsan/debugfs.c
@@ -300,16 +300,16 @@ debugfs_write(struct file *file, const char __user *buf, size_t count, loff_t *o
WRITE_ONCE(kcsan_enabled, true);
} else if (!strcmp(arg, "off")) {
WRITE_ONCE(kcsan_enabled, false);
- } else if (!strncmp(arg, "microbench=", sizeof("microbench=") - 1)) {
+ } else if (str_has_prefix(arg, "microbench=")) {
unsigned long iters;
- if (kstrtoul(&arg[sizeof("microbench=") - 1], 0, &iters))
+ if (kstrtoul(&arg[strlen("microbench=")], 0, &iters))
return -EINVAL;
microbenchmark(iters);
- } else if (!strncmp(arg, "test=", sizeof("test=") - 1)) {
+ } else if (str_has_prefix(arg, "test=")) {
unsigned long iters;
- if (kstrtoul(&arg[sizeof("test=") - 1], 0, &iters))
+ if (kstrtoul(&arg[strlen("test=")], 0, &iters))
return -EINVAL;
test_thread(iters);
} else if (!strcmp(arg, "whitelist")) {
diff --git a/kernel/kcsan/report.c b/kernel/kcsan/report.c
index 3e83a69..bf1d594 100644
--- a/kernel/kcsan/report.c
+++ b/kernel/kcsan/report.c
@@ -279,8 +279,8 @@ static int get_stack_skipnr(const unsigned long stack_entries[], int num_entries
cur = strnstr(buf, "kcsan_", len);
if (cur) {
- cur += sizeof("kcsan_") - 1;
- if (strncmp(cur, "test", sizeof("test") - 1))
+ cur += strlen("kcsan_");
+ if (!str_has_prefix(cur, "test"))
continue; /* KCSAN runtime function. */
/* KCSAN related test. */
}
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 14/19] kcsan: Remove debugfs test command
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (12 preceding siblings ...)
2020-08-31 18:17 ` [PATCH kcsan 13/19] kcsan: Simplify constant string handling paulmck
@ 2020-08-31 18:18 ` paulmck
2020-08-31 18:18 ` [PATCH kcsan 15/19] kcsan: Show message if enabled early paulmck
` (4 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Remove the debugfs test command, as it is no longer needed now that we
have the KUnit+Torture based kcsan-test module. This is to avoid
confusion around how KCSAN should be tested, as only the kcsan-test
module is maintained.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/debugfs.c | 66 --------------------------------------------------
1 file changed, 66 deletions(-)
diff --git a/kernel/kcsan/debugfs.c b/kernel/kcsan/debugfs.c
index 116bdd8..de1da1b01 100644
--- a/kernel/kcsan/debugfs.c
+++ b/kernel/kcsan/debugfs.c
@@ -98,66 +98,6 @@ static noinline void microbenchmark(unsigned long iters)
current->kcsan_ctx = ctx_save;
}
-/*
- * Simple test to create conflicting accesses. Write 'test=<iters>' to KCSAN's
- * debugfs file from multiple tasks to generate real conflicts and show reports.
- */
-static long test_dummy;
-static long test_flags;
-static long test_scoped;
-static noinline void test_thread(unsigned long iters)
-{
- const long CHANGE_BITS = 0xff00ff00ff00ff00L;
- const struct kcsan_ctx ctx_save = current->kcsan_ctx;
- cycles_t cycles;
-
- /* We may have been called from an atomic region; reset context. */
- memset(¤t->kcsan_ctx, 0, sizeof(current->kcsan_ctx));
-
- pr_info("KCSAN: %s begin | iters: %lu\n", __func__, iters);
- pr_info("test_dummy@%px, test_flags@%px, test_scoped@%px,\n",
- &test_dummy, &test_flags, &test_scoped);
-
- cycles = get_cycles();
- while (iters--) {
- /* These all should generate reports. */
- __kcsan_check_read(&test_dummy, sizeof(test_dummy));
- ASSERT_EXCLUSIVE_WRITER(test_dummy);
- ASSERT_EXCLUSIVE_ACCESS(test_dummy);
-
- ASSERT_EXCLUSIVE_BITS(test_flags, ~CHANGE_BITS); /* no report */
- __kcsan_check_read(&test_flags, sizeof(test_flags)); /* no report */
-
- ASSERT_EXCLUSIVE_BITS(test_flags, CHANGE_BITS); /* report */
- __kcsan_check_read(&test_flags, sizeof(test_flags)); /* no report */
-
- /* not actually instrumented */
- WRITE_ONCE(test_dummy, iters); /* to observe value-change */
- __kcsan_check_write(&test_dummy, sizeof(test_dummy));
-
- test_flags ^= CHANGE_BITS; /* generate value-change */
- __kcsan_check_write(&test_flags, sizeof(test_flags));
-
- BUG_ON(current->kcsan_ctx.scoped_accesses.prev);
- {
- /* Should generate reports anywhere in this block. */
- ASSERT_EXCLUSIVE_WRITER_SCOPED(test_scoped);
- ASSERT_EXCLUSIVE_ACCESS_SCOPED(test_scoped);
- BUG_ON(!current->kcsan_ctx.scoped_accesses.prev);
- /* Unrelated accesses. */
- __kcsan_check_access(&cycles, sizeof(cycles), 0);
- __kcsan_check_access(&cycles, sizeof(cycles), KCSAN_ACCESS_ATOMIC);
- }
- BUG_ON(current->kcsan_ctx.scoped_accesses.prev);
- }
- cycles = get_cycles() - cycles;
-
- pr_info("KCSAN: %s end | cycles: %llu\n", __func__, cycles);
-
- /* restore context */
- current->kcsan_ctx = ctx_save;
-}
-
static int cmp_filterlist_addrs(const void *rhs, const void *lhs)
{
const unsigned long a = *(const unsigned long *)rhs;
@@ -306,12 +246,6 @@ debugfs_write(struct file *file, const char __user *buf, size_t count, loff_t *o
if (kstrtoul(&arg[strlen("microbench=")], 0, &iters))
return -EINVAL;
microbenchmark(iters);
- } else if (str_has_prefix(arg, "test=")) {
- unsigned long iters;
-
- if (kstrtoul(&arg[strlen("test=")], 0, &iters))
- return -EINVAL;
- test_thread(iters);
} else if (!strcmp(arg, "whitelist")) {
set_report_filterlist_whitelist(true);
} else if (!strcmp(arg, "blacklist")) {
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 15/19] kcsan: Show message if enabled early
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (13 preceding siblings ...)
2020-08-31 18:18 ` [PATCH kcsan 14/19] kcsan: Remove debugfs test command paulmck
@ 2020-08-31 18:18 ` paulmck
2020-08-31 18:18 ` [PATCH kcsan 16/19] kcsan: Use pr_fmt for consistency paulmck
` (3 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Show a message in the kernel log if KCSAN was enabled early.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 99e5044..b176400 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -1,5 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
+#define pr_fmt(fmt) "kcsan: " fmt
+
#include <linux/atomic.h>
#include <linux/bug.h>
#include <linux/delay.h>
@@ -463,7 +465,7 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
if (IS_ENABLED(CONFIG_KCSAN_DEBUG)) {
kcsan_disable_current();
- pr_err("KCSAN: watching %s, size: %zu, addr: %px [slot: %d, encoded: %lx]\n",
+ pr_err("watching %s, size: %zu, addr: %px [slot: %d, encoded: %lx]\n",
is_write ? "write" : "read", size, ptr,
watchpoint_slot((unsigned long)ptr),
encode_watchpoint((unsigned long)ptr, size, is_write));
@@ -623,8 +625,10 @@ void __init kcsan_init(void)
* We are in the init task, and no other tasks should be running;
* WRITE_ONCE without memory barrier is sufficient.
*/
- if (kcsan_early_enable)
+ if (kcsan_early_enable) {
+ pr_info("enabled early\n");
WRITE_ONCE(kcsan_enabled, true);
+ }
}
/* === Exported interface =================================================== */
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 16/19] kcsan: Use pr_fmt for consistency
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (14 preceding siblings ...)
2020-08-31 18:18 ` [PATCH kcsan 15/19] kcsan: Show message if enabled early paulmck
@ 2020-08-31 18:18 ` paulmck
2020-08-31 18:18 ` [PATCH kcsan 17/19] kcsan: Optimize debugfs stats counters paulmck
` (2 subsequent siblings)
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Use the same pr_fmt throughout for consistency. [ The only exception is
report.c, where the format must be kept precisely as-is. ]
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/debugfs.c | 8 +++++---
kernel/kcsan/selftest.c | 8 +++++---
2 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/kernel/kcsan/debugfs.c b/kernel/kcsan/debugfs.c
index de1da1b01..6c4914f 100644
--- a/kernel/kcsan/debugfs.c
+++ b/kernel/kcsan/debugfs.c
@@ -1,5 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
+#define pr_fmt(fmt) "kcsan: " fmt
+
#include <linux/atomic.h>
#include <linux/bsearch.h>
#include <linux/bug.h>
@@ -80,7 +82,7 @@ static noinline void microbenchmark(unsigned long iters)
*/
WRITE_ONCE(kcsan_enabled, false);
- pr_info("KCSAN: %s begin | iters: %lu\n", __func__, iters);
+ pr_info("%s begin | iters: %lu\n", __func__, iters);
cycles = get_cycles();
while (iters--) {
@@ -91,7 +93,7 @@ static noinline void microbenchmark(unsigned long iters)
}
cycles = get_cycles() - cycles;
- pr_info("KCSAN: %s end | cycles: %llu\n", __func__, cycles);
+ pr_info("%s end | cycles: %llu\n", __func__, cycles);
WRITE_ONCE(kcsan_enabled, was_enabled);
/* restore context */
@@ -154,7 +156,7 @@ static ssize_t insert_report_filterlist(const char *func)
ssize_t ret = 0;
if (!addr) {
- pr_err("KCSAN: could not find function: '%s'\n", func);
+ pr_err("could not find function: '%s'\n", func);
return -ENOENT;
}
diff --git a/kernel/kcsan/selftest.c b/kernel/kcsan/selftest.c
index d26a052..d98bc20 100644
--- a/kernel/kcsan/selftest.c
+++ b/kernel/kcsan/selftest.c
@@ -1,5 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
+#define pr_fmt(fmt) "kcsan: " fmt
+
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/printk.h>
@@ -116,16 +118,16 @@ static int __init kcsan_selftest(void)
if (do_test()) \
++passed; \
else \
- pr_err("KCSAN selftest: " #do_test " failed"); \
+ pr_err("selftest: " #do_test " failed"); \
} while (0)
RUN_TEST(test_requires);
RUN_TEST(test_encode_decode);
RUN_TEST(test_matching_access);
- pr_info("KCSAN selftest: %d/%d tests passed\n", passed, total);
+ pr_info("selftest: %d/%d tests passed\n", passed, total);
if (passed != total)
- panic("KCSAN selftests failed");
+ panic("selftests failed");
return 0;
}
postcore_initcall(kcsan_selftest);
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 17/19] kcsan: Optimize debugfs stats counters
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (15 preceding siblings ...)
2020-08-31 18:18 ` [PATCH kcsan 16/19] kcsan: Use pr_fmt for consistency paulmck
@ 2020-08-31 18:18 ` paulmck
2020-08-31 18:18 ` [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops paulmck
2020-08-31 18:18 ` [PATCH kcsan 19/19] kcsan: Use tracing-safe version of prandom paulmck
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
Remove kcsan_counter_inc/dec() functions, as they perform no other
logic, and are no longer needed.
This avoids several calls in kcsan_setup_watchpoint() and
kcsan_found_watchpoint(), as well as lets the compiler warn us about
potential out-of-bounds accesses as the array's size is known at all
usage sites at compile-time.
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 22 +++++++++++-----------
kernel/kcsan/debugfs.c | 21 +++++----------------
kernel/kcsan/kcsan.h | 12 ++++++------
kernel/kcsan/report.c | 2 +-
4 files changed, 23 insertions(+), 34 deletions(-)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index b176400..8a1ff605 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -367,13 +367,13 @@ static noinline void kcsan_found_watchpoint(const volatile void *ptr,
* already removed the watchpoint, or another thread consumed
* the watchpoint before this thread.
*/
- kcsan_counter_inc(KCSAN_COUNTER_REPORT_RACES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_REPORT_RACES]);
}
if ((type & KCSAN_ACCESS_ASSERT) != 0)
- kcsan_counter_inc(KCSAN_COUNTER_ASSERT_FAILURES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_ASSERT_FAILURES]);
else
- kcsan_counter_inc(KCSAN_COUNTER_DATA_RACES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_DATA_RACES]);
user_access_restore(flags);
}
@@ -414,7 +414,7 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
goto out;
if (!check_encodable((unsigned long)ptr, size)) {
- kcsan_counter_inc(KCSAN_COUNTER_UNENCODABLE_ACCESSES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_UNENCODABLE_ACCESSES]);
goto out;
}
@@ -434,12 +434,12 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
* with which should_watch() returns true should be tweaked so
* that this case happens very rarely.
*/
- kcsan_counter_inc(KCSAN_COUNTER_NO_CAPACITY);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_NO_CAPACITY]);
goto out_unlock;
}
- kcsan_counter_inc(KCSAN_COUNTER_SETUP_WATCHPOINTS);
- kcsan_counter_inc(KCSAN_COUNTER_USED_WATCHPOINTS);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_SETUP_WATCHPOINTS]);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_USED_WATCHPOINTS]);
/*
* Read the current value, to later check and infer a race if the data
@@ -541,16 +541,16 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
* increment this counter.
*/
if (is_assert && value_change == KCSAN_VALUE_CHANGE_TRUE)
- kcsan_counter_inc(KCSAN_COUNTER_ASSERT_FAILURES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_ASSERT_FAILURES]);
kcsan_report(ptr, size, type, value_change, KCSAN_REPORT_RACE_SIGNAL,
watchpoint - watchpoints);
} else if (value_change == KCSAN_VALUE_CHANGE_TRUE) {
/* Inferring a race, since the value should not have changed. */
- kcsan_counter_inc(KCSAN_COUNTER_RACES_UNKNOWN_ORIGIN);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_RACES_UNKNOWN_ORIGIN]);
if (is_assert)
- kcsan_counter_inc(KCSAN_COUNTER_ASSERT_FAILURES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_ASSERT_FAILURES]);
if (IS_ENABLED(CONFIG_KCSAN_REPORT_RACE_UNKNOWN_ORIGIN) || is_assert)
kcsan_report(ptr, size, type, KCSAN_VALUE_CHANGE_TRUE,
@@ -563,7 +563,7 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
* reused after this point.
*/
remove_watchpoint(watchpoint);
- kcsan_counter_dec(KCSAN_COUNTER_USED_WATCHPOINTS);
+ atomic_long_dec(&kcsan_counters[KCSAN_COUNTER_USED_WATCHPOINTS]);
out_unlock:
if (!kcsan_interrupt_watcher)
local_irq_restore(irq_flags);
diff --git a/kernel/kcsan/debugfs.c b/kernel/kcsan/debugfs.c
index 6c4914f..3c8093a 100644
--- a/kernel/kcsan/debugfs.c
+++ b/kernel/kcsan/debugfs.c
@@ -17,10 +17,7 @@
#include "kcsan.h"
-/*
- * Statistics counters.
- */
-static atomic_long_t counters[KCSAN_COUNTER_COUNT];
+atomic_long_t kcsan_counters[KCSAN_COUNTER_COUNT];
static const char *const counter_names[] = {
[KCSAN_COUNTER_USED_WATCHPOINTS] = "used_watchpoints",
[KCSAN_COUNTER_SETUP_WATCHPOINTS] = "setup_watchpoints",
@@ -53,16 +50,6 @@ static struct {
};
static DEFINE_SPINLOCK(report_filterlist_lock);
-void kcsan_counter_inc(enum kcsan_counter_id id)
-{
- atomic_long_inc(&counters[id]);
-}
-
-void kcsan_counter_dec(enum kcsan_counter_id id)
-{
- atomic_long_dec(&counters[id]);
-}
-
/*
* The microbenchmark allows benchmarking KCSAN core runtime only. To run
* multiple threads, pipe 'microbench=<iters>' from multiple tasks into the
@@ -206,8 +193,10 @@ static int show_info(struct seq_file *file, void *v)
/* show stats */
seq_printf(file, "enabled: %i\n", READ_ONCE(kcsan_enabled));
- for (i = 0; i < KCSAN_COUNTER_COUNT; ++i)
- seq_printf(file, "%s: %ld\n", counter_names[i], atomic_long_read(&counters[i]));
+ for (i = 0; i < KCSAN_COUNTER_COUNT; ++i) {
+ seq_printf(file, "%s: %ld\n", counter_names[i],
+ atomic_long_read(&kcsan_counters[i]));
+ }
/* show filter functions, and filter type */
spin_lock_irqsave(&report_filterlist_lock, flags);
diff --git a/kernel/kcsan/kcsan.h b/kernel/kcsan/kcsan.h
index 2948001..8d4bf34 100644
--- a/kernel/kcsan/kcsan.h
+++ b/kernel/kcsan/kcsan.h
@@ -8,6 +8,7 @@
#ifndef _KERNEL_KCSAN_KCSAN_H
#define _KERNEL_KCSAN_KCSAN_H
+#include <linux/atomic.h>
#include <linux/kcsan.h>
#include <linux/sched.h>
@@ -34,6 +35,10 @@ void kcsan_restore_irqtrace(struct task_struct *task);
*/
void kcsan_debugfs_init(void);
+/*
+ * Statistics counters displayed via debugfs; should only be modified in
+ * slow-paths.
+ */
enum kcsan_counter_id {
/*
* Number of watchpoints currently in use.
@@ -86,12 +91,7 @@ enum kcsan_counter_id {
KCSAN_COUNTER_COUNT, /* number of counters */
};
-
-/*
- * Increment/decrement counter with given id; avoid calling these in fast-path.
- */
-extern void kcsan_counter_inc(enum kcsan_counter_id id);
-extern void kcsan_counter_dec(enum kcsan_counter_id id);
+extern atomic_long_t kcsan_counters[KCSAN_COUNTER_COUNT];
/*
* Returns true if data races in the function symbol that maps to func_addr
diff --git a/kernel/kcsan/report.c b/kernel/kcsan/report.c
index bf1d594..d3bf87e 100644
--- a/kernel/kcsan/report.c
+++ b/kernel/kcsan/report.c
@@ -559,7 +559,7 @@ static bool prepare_report_consumer(unsigned long *flags,
* If the actual accesses to not match, this was a false
* positive due to watchpoint encoding.
*/
- kcsan_counter_inc(KCSAN_COUNTER_ENCODING_FALSE_POSITIVES);
+ atomic_long_inc(&kcsan_counters[KCSAN_COUNTER_ENCODING_FALSE_POSITIVES]);
goto discard;
}
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (16 preceding siblings ...)
2020-08-31 18:18 ` [PATCH kcsan 17/19] kcsan: Optimize debugfs stats counters paulmck
@ 2020-08-31 18:18 ` paulmck
2020-09-02 3:30 ` Boqun Feng
2020-08-31 18:18 ` [PATCH kcsan 19/19] kcsan: Use tracing-safe version of prandom paulmck
18 siblings, 1 reply; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng,
Paul E . McKenney, Will Deacon, Arnd Bergmann, Daniel Axtens,
Michael Ellerman, linux-arch
From: Marco Elver <elver@google.com>
Previous to the change to distinguish read-write accesses, when
CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC=y is set, KCSAN would consider
the non-atomic bitops as atomic. We want to partially revert to this
behaviour, but with one important distinction: report racing
modifications, since lost bits due to non-atomicity are certainly
possible.
Given the operations here only modify a single bit, assuming
non-atomicity of the writer is sufficient may be reasonable for certain
usage (and follows the permissible nature of the "assume plain writes
atomic" rule). In other words:
1. We want non-atomic read-modify-write races to be reported;
this is accomplished by kcsan_check_read(), where any
concurrent write (atomic or not) will generate a report.
2. We do not want to report races with marked readers, but -do-
want to report races with unmarked readers; this is
accomplished by the instrument_write() ("assume atomic
write" with Kconfig option set).
With the above rules, when KCSAN_ASSUME_PLAIN_WRITES_ATOMIC is selected,
it is hoped that KCSAN's reporting behaviour is better aligned with
current expected permissible usage for non-atomic bitops.
Note that, a side-effect of not telling KCSAN that the accesses are
read-writes, is that this information is not displayed in the access
summary in the report. It is, however, visible in inline-expanded stack
traces. For now, it does not make sense to introduce yet another special
case to KCSAN's runtime, only to cater to the case here.
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Paul E. McKenney <paulmck@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Daniel Axtens <dja@axtens.net>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
.../asm-generic/bitops/instrumented-non-atomic.h | 30 +++++++++++++++++++---
1 file changed, 27 insertions(+), 3 deletions(-)
diff --git a/include/asm-generic/bitops/instrumented-non-atomic.h b/include/asm-generic/bitops/instrumented-non-atomic.h
index f86234c..37363d5 100644
--- a/include/asm-generic/bitops/instrumented-non-atomic.h
+++ b/include/asm-generic/bitops/instrumented-non-atomic.h
@@ -58,6 +58,30 @@ static inline void __change_bit(long nr, volatile unsigned long *addr)
arch___change_bit(nr, addr);
}
+static inline void __instrument_read_write_bitop(long nr, volatile unsigned long *addr)
+{
+ if (IS_ENABLED(CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC)) {
+ /*
+ * We treat non-atomic read-write bitops a little more special.
+ * Given the operations here only modify a single bit, assuming
+ * non-atomicity of the writer is sufficient may be reasonable
+ * for certain usage (and follows the permissible nature of the
+ * assume-plain-writes-atomic rule):
+ * 1. report read-modify-write races -> check read;
+ * 2. do not report races with marked readers, but do report
+ * races with unmarked readers -> check "atomic" write.
+ */
+ kcsan_check_read(addr + BIT_WORD(nr), sizeof(long));
+ /*
+ * Use generic write instrumentation, in case other sanitizers
+ * or tools are enabled alongside KCSAN.
+ */
+ instrument_write(addr + BIT_WORD(nr), sizeof(long));
+ } else {
+ instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
+ }
+}
+
/**
* __test_and_set_bit - Set a bit and return its old value
* @nr: Bit to set
@@ -68,7 +92,7 @@ static inline void __change_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
{
- instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
+ __instrument_read_write_bitop(nr, addr);
return arch___test_and_set_bit(nr, addr);
}
@@ -82,7 +106,7 @@ static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
{
- instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
+ __instrument_read_write_bitop(nr, addr);
return arch___test_and_clear_bit(nr, addr);
}
@@ -96,7 +120,7 @@ static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
*/
static inline bool __test_and_change_bit(long nr, volatile unsigned long *addr)
{
- instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
+ __instrument_read_write_bitop(nr, addr);
return arch___test_and_change_bit(nr, addr);
}
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* [PATCH kcsan 19/19] kcsan: Use tracing-safe version of prandom
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
` (17 preceding siblings ...)
2020-08-31 18:18 ` [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops paulmck
@ 2020-08-31 18:18 ` paulmck
18 siblings, 0 replies; 23+ messages in thread
From: paulmck @ 2020-08-31 18:18 UTC (permalink / raw)
To: linux-kernel, kasan-dev, kernel-team, mingo
Cc: elver, andreyknvl, glider, dvyukov, cai, boqun.feng, Paul E . McKenney
From: Marco Elver <elver@google.com>
In the core runtime, we must minimize any calls to external library
functions to avoid any kind of recursion. This can happen even though
instrumentation is disabled for called functions, but tracing is
enabled.
Most recently, prandom_u32() added a tracepoint, which can cause
problems for KCSAN even if the rcuidle variant is used. For example:
kcsan -> prandom_u32() -> trace_prandom_u32_rcuidle ->
srcu_read_lock_notrace -> __srcu_read_lock -> kcsan ...
While we could disable KCSAN in kcsan_setup_watchpoint(), this does not
solve other unexpected behaviour we may get due recursing into functions
that may not be tolerant to such recursion:
__srcu_read_lock -> kcsan -> ... -> __srcu_read_lock
Therefore, switch to using prandom_u32_state(), which is uninstrumented,
and does not have a tracepoint.
Link: https://lkml.kernel.org/r/20200821063043.1949509-1-elver@google.com
Link: https://lkml.kernel.org/r/20200820172046.GA177701@elver.google.com
Signed-off-by: Marco Elver <elver@google.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
---
kernel/kcsan/core.c | 35 +++++++++++++++++++++++++++++------
1 file changed, 29 insertions(+), 6 deletions(-)
diff --git a/kernel/kcsan/core.c b/kernel/kcsan/core.c
index 8a1ff605..3994a21 100644
--- a/kernel/kcsan/core.c
+++ b/kernel/kcsan/core.c
@@ -100,6 +100,9 @@ static atomic_long_t watchpoints[CONFIG_KCSAN_NUM_WATCHPOINTS + NUM_SLOTS-1];
*/
static DEFINE_PER_CPU(long, kcsan_skip);
+/* For kcsan_prandom_u32_max(). */
+static DEFINE_PER_CPU(struct rnd_state, kcsan_rand_state);
+
static __always_inline atomic_long_t *find_watchpoint(unsigned long addr,
size_t size,
bool expect_write,
@@ -271,11 +274,28 @@ should_watch(const volatile void *ptr, size_t size, int type, struct kcsan_ctx *
return true;
}
+/*
+ * Returns a pseudo-random number in interval [0, ep_ro). See prandom_u32_max()
+ * for more details.
+ *
+ * The open-coded version here is using only safe primitives for all contexts
+ * where we can have KCSAN instrumentation. In particular, we cannot use
+ * prandom_u32() directly, as its tracepoint could cause recursion.
+ */
+static u32 kcsan_prandom_u32_max(u32 ep_ro)
+{
+ struct rnd_state *state = &get_cpu_var(kcsan_rand_state);
+ const u32 res = prandom_u32_state(state);
+
+ put_cpu_var(kcsan_rand_state);
+ return (u32)(((u64) res * ep_ro) >> 32);
+}
+
static inline void reset_kcsan_skip(void)
{
long skip_count = kcsan_skip_watch -
(IS_ENABLED(CONFIG_KCSAN_SKIP_WATCH_RANDOMIZE) ?
- prandom_u32_max(kcsan_skip_watch) :
+ kcsan_prandom_u32_max(kcsan_skip_watch) :
0);
this_cpu_write(kcsan_skip, skip_count);
}
@@ -285,16 +305,18 @@ static __always_inline bool kcsan_is_enabled(void)
return READ_ONCE(kcsan_enabled) && get_ctx()->disable_count == 0;
}
-static inline unsigned int get_delay(int type)
+/* Introduce delay depending on context and configuration. */
+static void delay_access(int type)
{
unsigned int delay = in_task() ? kcsan_udelay_task : kcsan_udelay_interrupt;
/* For certain access types, skew the random delay to be longer. */
unsigned int skew_delay_order =
(type & (KCSAN_ACCESS_COMPOUND | KCSAN_ACCESS_ASSERT)) ? 1 : 0;
- return delay - (IS_ENABLED(CONFIG_KCSAN_DELAY_RANDOMIZE) ?
- prandom_u32_max(delay >> skew_delay_order) :
- 0);
+ delay -= IS_ENABLED(CONFIG_KCSAN_DELAY_RANDOMIZE) ?
+ kcsan_prandom_u32_max(delay >> skew_delay_order) :
+ 0;
+ udelay(delay);
}
void kcsan_save_irqtrace(struct task_struct *task)
@@ -476,7 +498,7 @@ kcsan_setup_watchpoint(const volatile void *ptr, size_t size, int type)
* Delay this thread, to increase probability of observing a racy
* conflicting access.
*/
- udelay(get_delay(type));
+ delay_access(type);
/*
* Re-read value, and check if it is as expected; if not, we infer a
@@ -620,6 +642,7 @@ void __init kcsan_init(void)
BUG_ON(!in_task());
kcsan_debugfs_init();
+ prandom_seed_full_state(&kcsan_rand_state);
/*
* We are in the init task, and no other tasks should be running;
--
2.9.5
^ permalink raw reply related [flat|nested] 23+ messages in thread
* Re: [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops
2020-08-31 18:18 ` [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops paulmck
@ 2020-09-02 3:30 ` Boqun Feng
2020-09-02 6:13 ` Marco Elver
0 siblings, 1 reply; 23+ messages in thread
From: Boqun Feng @ 2020-09-02 3:30 UTC (permalink / raw)
To: paulmck
Cc: linux-kernel, kasan-dev, kernel-team, mingo, elver, andreyknvl,
glider, dvyukov, cai, Will Deacon, Arnd Bergmann, Daniel Axtens,
Michael Ellerman, linux-arch
Hi Paul and Marco,
The whole update patchset looks good to me, just one question out of
curiosity fo this one, please see below:
On Mon, Aug 31, 2020 at 11:18:04AM -0700, paulmck@kernel.org wrote:
> From: Marco Elver <elver@google.com>
>
> Previous to the change to distinguish read-write accesses, when
> CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC=y is set, KCSAN would consider
> the non-atomic bitops as atomic. We want to partially revert to this
> behaviour, but with one important distinction: report racing
> modifications, since lost bits due to non-atomicity are certainly
> possible.
>
> Given the operations here only modify a single bit, assuming
> non-atomicity of the writer is sufficient may be reasonable for certain
> usage (and follows the permissible nature of the "assume plain writes
> atomic" rule). In other words:
>
> 1. We want non-atomic read-modify-write races to be reported;
> this is accomplished by kcsan_check_read(), where any
> concurrent write (atomic or not) will generate a report.
>
> 2. We do not want to report races with marked readers, but -do-
> want to report races with unmarked readers; this is
> accomplished by the instrument_write() ("assume atomic
> write" with Kconfig option set).
>
Is there any code in kernel using the above assumption (i.e.
non-atomicity of the writer is sufficient)? IOW, have you observed
anything bad (e.g. an anoying false positive) after applying the
read_write changes but without this patch?
Regards,
Boqun
> With the above rules, when KCSAN_ASSUME_PLAIN_WRITES_ATOMIC is selected,
> it is hoped that KCSAN's reporting behaviour is better aligned with
> current expected permissible usage for non-atomic bitops.
>
> Note that, a side-effect of not telling KCSAN that the accesses are
> read-writes, is that this information is not displayed in the access
> summary in the report. It is, however, visible in inline-expanded stack
> traces. For now, it does not make sense to introduce yet another special
> case to KCSAN's runtime, only to cater to the case here.
>
> Cc: Dmitry Vyukov <dvyukov@google.com>
> Cc: Paul E. McKenney <paulmck@kernel.org>
> Cc: Will Deacon <will@kernel.org>
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Daniel Axtens <dja@axtens.net>
> Cc: Michael Ellerman <mpe@ellerman.id.au>
> Cc: <linux-arch@vger.kernel.org>
> Signed-off-by: Marco Elver <elver@google.com>
> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
> ---
> .../asm-generic/bitops/instrumented-non-atomic.h | 30 +++++++++++++++++++---
> 1 file changed, 27 insertions(+), 3 deletions(-)
>
> diff --git a/include/asm-generic/bitops/instrumented-non-atomic.h b/include/asm-generic/bitops/instrumented-non-atomic.h
> index f86234c..37363d5 100644
> --- a/include/asm-generic/bitops/instrumented-non-atomic.h
> +++ b/include/asm-generic/bitops/instrumented-non-atomic.h
> @@ -58,6 +58,30 @@ static inline void __change_bit(long nr, volatile unsigned long *addr)
> arch___change_bit(nr, addr);
> }
>
> +static inline void __instrument_read_write_bitop(long nr, volatile unsigned long *addr)
> +{
> + if (IS_ENABLED(CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC)) {
> + /*
> + * We treat non-atomic read-write bitops a little more special.
> + * Given the operations here only modify a single bit, assuming
> + * non-atomicity of the writer is sufficient may be reasonable
> + * for certain usage (and follows the permissible nature of the
> + * assume-plain-writes-atomic rule):
> + * 1. report read-modify-write races -> check read;
> + * 2. do not report races with marked readers, but do report
> + * races with unmarked readers -> check "atomic" write.
> + */
> + kcsan_check_read(addr + BIT_WORD(nr), sizeof(long));
> + /*
> + * Use generic write instrumentation, in case other sanitizers
> + * or tools are enabled alongside KCSAN.
> + */
> + instrument_write(addr + BIT_WORD(nr), sizeof(long));
> + } else {
> + instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
> + }
> +}
> +
> /**
> * __test_and_set_bit - Set a bit and return its old value
> * @nr: Bit to set
> @@ -68,7 +92,7 @@ static inline void __change_bit(long nr, volatile unsigned long *addr)
> */
> static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
> {
> - instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
> + __instrument_read_write_bitop(nr, addr);
> return arch___test_and_set_bit(nr, addr);
> }
>
> @@ -82,7 +106,7 @@ static inline bool __test_and_set_bit(long nr, volatile unsigned long *addr)
> */
> static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
> {
> - instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
> + __instrument_read_write_bitop(nr, addr);
> return arch___test_and_clear_bit(nr, addr);
> }
>
> @@ -96,7 +120,7 @@ static inline bool __test_and_clear_bit(long nr, volatile unsigned long *addr)
> */
> static inline bool __test_and_change_bit(long nr, volatile unsigned long *addr)
> {
> - instrument_read_write(addr + BIT_WORD(nr), sizeof(long));
> + __instrument_read_write_bitop(nr, addr);
> return arch___test_and_change_bit(nr, addr);
> }
>
> --
> 2.9.5
>
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops
2020-09-02 3:30 ` Boqun Feng
@ 2020-09-02 6:13 ` Marco Elver
2020-09-04 1:24 ` Boqun Feng
0 siblings, 1 reply; 23+ messages in thread
From: Marco Elver @ 2020-09-02 6:13 UTC (permalink / raw)
To: Boqun Feng
Cc: paulmck, linux-kernel, kasan-dev, kernel-team, mingo, andreyknvl,
glider, dvyukov, cai, Will Deacon, Arnd Bergmann, Daniel Axtens,
Michael Ellerman, linux-arch
On Wed, Sep 02, 2020 at 11:30AM +0800, Boqun Feng wrote:
> Hi Paul and Marco,
>
> The whole update patchset looks good to me, just one question out of
> curiosity fo this one, please see below:
>
> On Mon, Aug 31, 2020 at 11:18:04AM -0700, paulmck@kernel.org wrote:
> > From: Marco Elver <elver@google.com>
> >
> > Previous to the change to distinguish read-write accesses, when
> > CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC=y is set, KCSAN would consider
> > the non-atomic bitops as atomic. We want to partially revert to this
> > behaviour, but with one important distinction: report racing
> > modifications, since lost bits due to non-atomicity are certainly
> > possible.
> >
> > Given the operations here only modify a single bit, assuming
> > non-atomicity of the writer is sufficient may be reasonable for certain
> > usage (and follows the permissible nature of the "assume plain writes
> > atomic" rule). In other words:
> >
> > 1. We want non-atomic read-modify-write races to be reported;
> > this is accomplished by kcsan_check_read(), where any
> > concurrent write (atomic or not) will generate a report.
> >
> > 2. We do not want to report races with marked readers, but -do-
> > want to report races with unmarked readers; this is
> > accomplished by the instrument_write() ("assume atomic
> > write" with Kconfig option set).
> >
>
> Is there any code in kernel using the above assumption (i.e.
> non-atomicity of the writer is sufficient)? IOW, have you observed
> anything bad (e.g. an anoying false positive) after applying the
> read_write changes but without this patch?
We were looking for an answer to:
https://lkml.kernel.org/r/20200810124516.GM17456@casper.infradead.org
Initially we thought using atomic bitops might be required, but after a
longer offline discussion realized that simply marking the reader in
this case, but retaining the non-atomic bitop is probably all that's
needed.
The version of KCSAN that found the above was still using KCSAN from
Linux 5.8, but we realized with the changed read-write instrumentation
to bitops in this series, we'd regress and still report the race even if
the reader was marked. To avoid this with the default KCSAN config, we
determined that we need the patch here.
The bitops are indeed a bit more special, because for both the atomic
and non-atomic bitops we *can* reason about the generated code (since we
control it, although not sure about the asm-generic ones), and that
makes reasoning about accesses racing with non-atomic bitops more
feasible. At least that's our rationale for deciding that reverting
non-atomic bitops treatment to it's more relaxed version is ok.
Thanks,
-- Marco
^ permalink raw reply [flat|nested] 23+ messages in thread
* Re: [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops
2020-09-02 6:13 ` Marco Elver
@ 2020-09-04 1:24 ` Boqun Feng
0 siblings, 0 replies; 23+ messages in thread
From: Boqun Feng @ 2020-09-04 1:24 UTC (permalink / raw)
To: Marco Elver
Cc: paulmck, linux-kernel, kasan-dev, kernel-team, mingo, andreyknvl,
glider, dvyukov, cai, Will Deacon, Arnd Bergmann, Daniel Axtens,
Michael Ellerman, linux-arch
On Wed, Sep 02, 2020 at 08:13:15AM +0200, Marco Elver wrote:
> On Wed, Sep 02, 2020 at 11:30AM +0800, Boqun Feng wrote:
> > Hi Paul and Marco,
> >
> > The whole update patchset looks good to me, just one question out of
> > curiosity fo this one, please see below:
> >
> > On Mon, Aug 31, 2020 at 11:18:04AM -0700, paulmck@kernel.org wrote:
> > > From: Marco Elver <elver@google.com>
> > >
> > > Previous to the change to distinguish read-write accesses, when
> > > CONFIG_KCSAN_ASSUME_PLAIN_WRITES_ATOMIC=y is set, KCSAN would consider
> > > the non-atomic bitops as atomic. We want to partially revert to this
> > > behaviour, but with one important distinction: report racing
> > > modifications, since lost bits due to non-atomicity are certainly
> > > possible.
> > >
> > > Given the operations here only modify a single bit, assuming
> > > non-atomicity of the writer is sufficient may be reasonable for certain
> > > usage (and follows the permissible nature of the "assume plain writes
> > > atomic" rule). In other words:
> > >
> > > 1. We want non-atomic read-modify-write races to be reported;
> > > this is accomplished by kcsan_check_read(), where any
> > > concurrent write (atomic or not) will generate a report.
> > >
> > > 2. We do not want to report races with marked readers, but -do-
> > > want to report races with unmarked readers; this is
> > > accomplished by the instrument_write() ("assume atomic
> > > write" with Kconfig option set).
> > >
> >
> > Is there any code in kernel using the above assumption (i.e.
> > non-atomicity of the writer is sufficient)? IOW, have you observed
> > anything bad (e.g. an anoying false positive) after applying the
> > read_write changes but without this patch?
>
> We were looking for an answer to:
>
> https://lkml.kernel.org/r/20200810124516.GM17456@casper.infradead.org
>
> Initially we thought using atomic bitops might be required, but after a
> longer offline discussion realized that simply marking the reader in
> this case, but retaining the non-atomic bitop is probably all that's
> needed.
>
> The version of KCSAN that found the above was still using KCSAN from
> Linux 5.8, but we realized with the changed read-write instrumentation
> to bitops in this series, we'd regress and still report the race even if
> the reader was marked. To avoid this with the default KCSAN config, we
> determined that we need the patch here.
>
Thanks for the background! Now I see the point of having this patch ;-)
FWIW, feel free to add for the whole series:
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Regards,
Boqun
> The bitops are indeed a bit more special, because for both the atomic
> and non-atomic bitops we *can* reason about the generated code (since we
> control it, although not sure about the asm-generic ones), and that
> makes reasoning about accesses racing with non-atomic bitops more
> feasible. At least that's our rationale for deciding that reverting
> non-atomic bitops treatment to it's more relaxed version is ok.
>
> Thanks,
> -- Marco
^ permalink raw reply [flat|nested] 23+ messages in thread
end of thread, other threads:[~2020-09-04 1:24 UTC | newest]
Thread overview: 23+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-31 18:17 [PATCH kcsan 0/19] KCSAN updates for v5.10 Paul E. McKenney
2020-08-31 18:17 ` [PATCH kcsan 01/19] kcsan: Add support for atomic builtins paulmck
2020-08-31 18:17 ` [PATCH kcsan 02/19] objtool: Add atomic builtin TSAN instrumentation to uaccess whitelist paulmck
2020-08-31 18:17 ` [PATCH kcsan 03/19] kcsan: Add atomic builtin test case paulmck
2020-08-31 18:17 ` [PATCH kcsan 04/19] kcsan: Support compounded read-write instrumentation paulmck
2020-08-31 18:17 ` [PATCH kcsan 05/19] objtool, kcsan: Add __tsan_read_write to uaccess whitelist paulmck
2020-08-31 18:17 ` [PATCH kcsan 06/19] kcsan: Skew delay to be longer for certain access types paulmck
2020-08-31 18:17 ` [PATCH kcsan 07/19] kcsan: Add missing CONFIG_KCSAN_IGNORE_ATOMICS checks paulmck
2020-08-31 18:17 ` [PATCH kcsan 08/19] kcsan: Test support for compound instrumentation paulmck
2020-08-31 18:17 ` [PATCH kcsan 09/19] instrumented.h: Introduce read-write instrumentation hooks paulmck
2020-08-31 18:17 ` [PATCH kcsan 10/19] asm-generic/bitops: Use instrument_read_write() where appropriate paulmck
2020-08-31 18:17 ` [PATCH kcsan 11/19] locking/atomics: Use read-write instrumentation for atomic RMWs paulmck
2020-08-31 18:17 ` [PATCH kcsan 12/19] kcsan: Simplify debugfs counter to name mapping paulmck
2020-08-31 18:17 ` [PATCH kcsan 13/19] kcsan: Simplify constant string handling paulmck
2020-08-31 18:18 ` [PATCH kcsan 14/19] kcsan: Remove debugfs test command paulmck
2020-08-31 18:18 ` [PATCH kcsan 15/19] kcsan: Show message if enabled early paulmck
2020-08-31 18:18 ` [PATCH kcsan 16/19] kcsan: Use pr_fmt for consistency paulmck
2020-08-31 18:18 ` [PATCH kcsan 17/19] kcsan: Optimize debugfs stats counters paulmck
2020-08-31 18:18 ` [PATCH kcsan 18/19] bitops, kcsan: Partially revert instrumentation for non-atomic bitops paulmck
2020-09-02 3:30 ` Boqun Feng
2020-09-02 6:13 ` Marco Elver
2020-09-04 1:24 ` Boqun Feng
2020-08-31 18:18 ` [PATCH kcsan 19/19] kcsan: Use tracing-safe version of prandom paulmck
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).