From 1ba3cbf3ec3b21d866436fb46b4bb7bdc38608f9 Mon Sep 17 00:00:00 2001 From: Peng Zhang Date: Mon, 3 Apr 2023 20:27:38 +0800 Subject: mm: kfence: improve the performance of __kfence_alloc() and __kfence_free() In __kfence_alloc() and __kfence_free(), we will set and check canary. Assuming that the size of the object is close to 0, nearly 4k memory accesses are required because setting and checking canary is executed byte by byte. canary is now defined like this: KFENCE_CANARY_PATTERN(addr) ((u8)0xaa ^ (u8)((unsigned long)(addr) & 0x7)) Observe that canary is only related to the lower three bits of the address, so every 8 bytes of canary are the same. We can access 8-byte canary each time instead of byte-by-byte, thereby optimizing nearly 4k memory accesses to 4k/8 times. Use the bcc tool funclatency to measure the latency of __kfence_alloc() and __kfence_free(), the numbers (deleted the distribution of latency) is posted below. Though different object sizes will have an impact on the measurement, we ignore it for now and assume the average object size is roughly equal. Before patching: __kfence_alloc: avg = 5055 nsecs, total: 5515252 nsecs, count: 1091 __kfence_free: avg = 5319 nsecs, total: 9735130 nsecs, count: 1830 After patching: __kfence_alloc: avg = 3597 nsecs, total: 6428491 nsecs, count: 1787 __kfence_free: avg = 3046 nsecs, total: 3415390 nsecs, count: 1121 The numbers indicate that there is ~30% - ~40% performance improvement. Link: https://lkml.kernel.org/r/20230403122738.6006-1-zhangpeng.00@bytedance.com Signed-off-by: Peng Zhang Reviewed-by: Marco Elver Cc: Alexander Potapenko Cc: Dmitry Vyukov Signed-off-by: Andrew Morton --- mm/kfence/core.c | 70 ++++++++++++++++++++++++++++++++++++++---------------- mm/kfence/kfence.h | 10 +++++++- mm/kfence/report.c | 2 +- 3 files changed, 59 insertions(+), 23 deletions(-) (limited to 'mm/kfence') diff --git a/mm/kfence/core.c b/mm/kfence/core.c index 1065e0568d05..d085ac7c8ae0 100644 --- a/mm/kfence/core.c +++ b/mm/kfence/core.c @@ -297,20 +297,13 @@ metadata_update_state(struct kfence_metadata *meta, enum kfence_object_state nex WRITE_ONCE(meta->state, next); } -/* Write canary byte to @addr. */ -static inline bool set_canary_byte(u8 *addr) -{ - *addr = KFENCE_CANARY_PATTERN(addr); - return true; -} - /* Check canary byte at @addr. */ static inline bool check_canary_byte(u8 *addr) { struct kfence_metadata *meta; unsigned long flags; - if (likely(*addr == KFENCE_CANARY_PATTERN(addr))) + if (likely(*addr == KFENCE_CANARY_PATTERN_U8(addr))) return true; atomic_long_inc(&counters[KFENCE_COUNTER_BUGS]); @@ -323,15 +316,31 @@ static inline bool check_canary_byte(u8 *addr) return false; } -/* __always_inline this to ensure we won't do an indirect call to fn. */ -static __always_inline void for_each_canary(const struct kfence_metadata *meta, bool (*fn)(u8 *)) +static inline void set_canary(const struct kfence_metadata *meta) { const unsigned long pageaddr = ALIGN_DOWN(meta->addr, PAGE_SIZE); - unsigned long addr; + unsigned long addr = pageaddr; + + /* + * The canary may be written to part of the object memory, but it does + * not affect it. The user should initialize the object before using it. + */ + for (; addr < meta->addr; addr += sizeof(u64)) + *((u64 *)addr) = KFENCE_CANARY_PATTERN_U64; + + addr = ALIGN_DOWN(meta->addr + meta->size, sizeof(u64)); + for (; addr - pageaddr < PAGE_SIZE; addr += sizeof(u64)) + *((u64 *)addr) = KFENCE_CANARY_PATTERN_U64; +} + +static inline void check_canary(const struct kfence_metadata *meta) +{ + const unsigned long pageaddr = ALIGN_DOWN(meta->addr, PAGE_SIZE); + unsigned long addr = pageaddr; /* - * We'll iterate over each canary byte per-side until fn() returns - * false. However, we'll still iterate over the canary bytes to the + * We'll iterate over each canary byte per-side until a corrupted byte + * is found. However, we'll still iterate over the canary bytes to the * right of the object even if there was an error in the canary bytes to * the left of the object. Specifically, if check_canary_byte() * generates an error, showing both sides might give more clues as to @@ -339,16 +348,35 @@ static __always_inline void for_each_canary(const struct kfence_metadata *meta, */ /* Apply to left of object. */ - for (addr = pageaddr; addr < meta->addr; addr++) { - if (!fn((u8 *)addr)) + for (; meta->addr - addr >= sizeof(u64); addr += sizeof(u64)) { + if (unlikely(*((u64 *)addr) != KFENCE_CANARY_PATTERN_U64)) break; } - /* Apply to right of object. */ - for (addr = meta->addr + meta->size; addr < pageaddr + PAGE_SIZE; addr++) { - if (!fn((u8 *)addr)) + /* + * If the canary is corrupted in a certain 64 bytes, or the canary + * memory cannot be completely covered by multiple consecutive 64 bytes, + * it needs to be checked one by one. + */ + for (; addr < meta->addr; addr++) { + if (unlikely(!check_canary_byte((u8 *)addr))) break; } + + /* Apply to right of object. */ + for (addr = meta->addr + meta->size; addr % sizeof(u64) != 0; addr++) { + if (unlikely(!check_canary_byte((u8 *)addr))) + return; + } + for (; addr - pageaddr < PAGE_SIZE; addr += sizeof(u64)) { + if (unlikely(*((u64 *)addr) != KFENCE_CANARY_PATTERN_U64)) { + + for (; addr - pageaddr < PAGE_SIZE; addr++) { + if (!check_canary_byte((u8 *)addr)) + return; + } + } + } } static void *kfence_guarded_alloc(struct kmem_cache *cache, size_t size, gfp_t gfp, @@ -434,7 +462,7 @@ static void *kfence_guarded_alloc(struct kmem_cache *cache, size_t size, gfp_t g #endif /* Memory initialization. */ - for_each_canary(meta, set_canary_byte); + set_canary(meta); /* * We check slab_want_init_on_alloc() ourselves, rather than letting @@ -495,7 +523,7 @@ static void kfence_guarded_free(void *addr, struct kfence_metadata *meta, bool z alloc_covered_add(meta->alloc_stack_hash, -1); /* Check canary bytes for memory corruption. */ - for_each_canary(meta, check_canary_byte); + check_canary(meta); /* * Clear memory if init-on-free is set. While we protect the page, the @@ -751,7 +779,7 @@ static void kfence_check_all_canary(void) struct kfence_metadata *meta = &kfence_metadata[i]; if (meta->state == KFENCE_OBJECT_ALLOCATED) - for_each_canary(meta, check_canary_byte); + check_canary(meta); } } diff --git a/mm/kfence/kfence.h b/mm/kfence/kfence.h index 600f2e2431d6..2aafc46a4aaf 100644 --- a/mm/kfence/kfence.h +++ b/mm/kfence/kfence.h @@ -21,7 +21,15 @@ * lower 3 bits of the address, to detect memory corruptions with higher * probability, where similar constants are used. */ -#define KFENCE_CANARY_PATTERN(addr) ((u8)0xaa ^ (u8)((unsigned long)(addr) & 0x7)) +#define KFENCE_CANARY_PATTERN_U8(addr) ((u8)0xaa ^ (u8)((unsigned long)(addr) & 0x7)) + +/* + * Define a continuous 8-byte canary starting from a multiple of 8. The canary + * of each byte is only related to the lowest three bits of its address, so the + * canary of every 8 bytes is the same. 64-bit memory can be filled and checked + * at a time instead of byte by byte to improve performance. + */ +#define KFENCE_CANARY_PATTERN_U64 ((u64)0xaaaaaaaaaaaaaaaa ^ (u64)(0x0706050403020100)) /* Maximum stack depth for reports. */ #define KFENCE_STACK_DEPTH 64 diff --git a/mm/kfence/report.c b/mm/kfence/report.c index 60205f1257ef..197430a5be4a 100644 --- a/mm/kfence/report.c +++ b/mm/kfence/report.c @@ -168,7 +168,7 @@ static void print_diff_canary(unsigned long address, size_t bytes_to_show, pr_cont("["); for (cur = (const u8 *)address; cur < end; cur++) { - if (*cur == KFENCE_CANARY_PATTERN(cur)) + if (*cur == KFENCE_CANARY_PATTERN_U8(cur)) pr_cont(" ."); else if (no_hash_pointers) pr_cont(" 0x%02x", *cur); -- cgit v1.2.3 From 1f6ab566cb3be9e8292e34b89e8be83d75aa232e Mon Sep 17 00:00:00 2001 From: Pavankumar Kondeti Date: Thu, 13 Apr 2023 15:38:59 +0530 Subject: printk: export console trace point for kcsan/kasan/kfence/kmsan The console tracepoint is used by kcsan/kasan/kfence/kmsan test modules. Since this tracepoint is not exported, these modules iterate over all available tracepoints to find the console trace point. Export the trace point so that it can be directly used. Link: https://lkml.kernel.org/r/20230413100859.1492323-1-quic_pkondeti@quicinc.com Signed-off-by: Pavankumar Kondeti Cc: Alexander Potapenko Cc: Andrey Konovalov Cc: Andrey Ryabinin Cc: Dmitry Vyukov Cc: John Ogness Cc: Marco Elver Cc: Petr Mladek Cc: Sergey Senozhatsky Cc: Steven Rostedt Cc: Vincenzo Frascino Signed-off-by: Andrew Morton --- kernel/kcsan/kcsan_test.c | 20 ++++++-------------- kernel/printk/printk.c | 2 ++ mm/kasan/kasan_test.c | 22 ++-------------------- mm/kfence/kfence_test.c | 22 ++-------------------- mm/kmsan/kmsan_test.c | 22 ++-------------------- 5 files changed, 14 insertions(+), 74 deletions(-) (limited to 'mm/kfence') diff --git a/kernel/kcsan/kcsan_test.c b/kernel/kcsan/kcsan_test.c index a60c561724be..0ddbdab5903d 100644 --- a/kernel/kcsan/kcsan_test.c +++ b/kernel/kcsan/kcsan_test.c @@ -1572,34 +1572,26 @@ static void test_exit(struct kunit *test) } __no_kcsan -static void register_tracepoints(struct tracepoint *tp, void *ignore) +static void register_tracepoints(void) { - check_trace_callback_type_console(probe_console); - if (!strcmp(tp->name, "console")) - WARN_ON(tracepoint_probe_register(tp, probe_console, NULL)); + register_trace_console(probe_console, NULL); } __no_kcsan -static void unregister_tracepoints(struct tracepoint *tp, void *ignore) +static void unregister_tracepoints(void) { - if (!strcmp(tp->name, "console")) - tracepoint_probe_unregister(tp, probe_console, NULL); + unregister_trace_console(probe_console, NULL); } static int kcsan_suite_init(struct kunit_suite *suite) { - /* - * Because we want to be able to build the test as a module, we need to - * iterate through all known tracepoints, since the static registration - * won't work here. - */ - for_each_kernel_tracepoint(register_tracepoints, NULL); + register_tracepoints(); return 0; } static void kcsan_suite_exit(struct kunit_suite *suite) { - for_each_kernel_tracepoint(unregister_tracepoints, NULL); + unregister_tracepoints(); tracepoint_synchronize_unregister(); } diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index fd0c9f913940..1c71bb170e85 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -71,6 +71,8 @@ EXPORT_SYMBOL_GPL(console_printk); atomic_t ignore_console_lock_warning __read_mostly = ATOMIC_INIT(0); EXPORT_SYMBOL(ignore_console_lock_warning); +EXPORT_TRACEPOINT_SYMBOL_GPL(console); + /* * Low level drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it. diff --git a/mm/kasan/kasan_test.c b/mm/kasan/kasan_test.c index a375776f9896..b61cc6a42541 100644 --- a/mm/kasan/kasan_test.c +++ b/mm/kasan/kasan_test.c @@ -56,19 +56,6 @@ static void probe_console(void *ignore, const char *buf, size_t len) WRITE_ONCE(test_status.async_fault, true); } -static void register_tracepoints(struct tracepoint *tp, void *ignore) -{ - check_trace_callback_type_console(probe_console); - if (!strcmp(tp->name, "console")) - WARN_ON(tracepoint_probe_register(tp, probe_console, NULL)); -} - -static void unregister_tracepoints(struct tracepoint *tp, void *ignore) -{ - if (!strcmp(tp->name, "console")) - tracepoint_probe_unregister(tp, probe_console, NULL); -} - static int kasan_suite_init(struct kunit_suite *suite) { if (!kasan_enabled()) { @@ -86,12 +73,7 @@ static int kasan_suite_init(struct kunit_suite *suite) */ multishot = kasan_save_enable_multi_shot(); - /* - * Because we want to be able to build the test as a module, we need to - * iterate through all known tracepoints, since the static registration - * won't work here. - */ - for_each_kernel_tracepoint(register_tracepoints, NULL); + register_trace_console(probe_console, NULL); return 0; } @@ -99,7 +81,7 @@ static void kasan_suite_exit(struct kunit_suite *suite) { kasan_kunit_test_suite_end(); kasan_restore_multi_shot(multishot); - for_each_kernel_tracepoint(unregister_tracepoints, NULL); + unregister_trace_console(probe_console, NULL); tracepoint_synchronize_unregister(); } diff --git a/mm/kfence/kfence_test.c b/mm/kfence/kfence_test.c index b5d66a69200d..6aee19a79236 100644 --- a/mm/kfence/kfence_test.c +++ b/mm/kfence/kfence_test.c @@ -825,33 +825,15 @@ static void test_exit(struct kunit *test) test_cache_destroy(); } -static void register_tracepoints(struct tracepoint *tp, void *ignore) -{ - check_trace_callback_type_console(probe_console); - if (!strcmp(tp->name, "console")) - WARN_ON(tracepoint_probe_register(tp, probe_console, NULL)); -} - -static void unregister_tracepoints(struct tracepoint *tp, void *ignore) -{ - if (!strcmp(tp->name, "console")) - tracepoint_probe_unregister(tp, probe_console, NULL); -} - static int kfence_suite_init(struct kunit_suite *suite) { - /* - * Because we want to be able to build the test as a module, we need to - * iterate through all known tracepoints, since the static registration - * won't work here. - */ - for_each_kernel_tracepoint(register_tracepoints, NULL); + register_trace_console(probe_console, NULL); return 0; } static void kfence_suite_exit(struct kunit_suite *suite) { - for_each_kernel_tracepoint(unregister_tracepoints, NULL); + unregister_trace_console(probe_console, NULL); tracepoint_synchronize_unregister(); } diff --git a/mm/kmsan/kmsan_test.c b/mm/kmsan/kmsan_test.c index d9eb141c27aa..312989aa2865 100644 --- a/mm/kmsan/kmsan_test.c +++ b/mm/kmsan/kmsan_test.c @@ -626,33 +626,15 @@ static void test_exit(struct kunit *test) { } -static void register_tracepoints(struct tracepoint *tp, void *ignore) -{ - check_trace_callback_type_console(probe_console); - if (!strcmp(tp->name, "console")) - WARN_ON(tracepoint_probe_register(tp, probe_console, NULL)); -} - -static void unregister_tracepoints(struct tracepoint *tp, void *ignore) -{ - if (!strcmp(tp->name, "console")) - tracepoint_probe_unregister(tp, probe_console, NULL); -} - static int kmsan_suite_init(struct kunit_suite *suite) { - /* - * Because we want to be able to build the test as a module, we need to - * iterate through all known tracepoints, since the static registration - * won't work here. - */ - for_each_kernel_tracepoint(register_tracepoints, NULL); + register_trace_console(probe_console, NULL); return 0; } static void kmsan_suite_exit(struct kunit_suite *suite) { - for_each_kernel_tracepoint(unregister_tracepoints, NULL); + unregister_trace_console(probe_console, NULL); tracepoint_synchronize_unregister(); } -- cgit v1.2.3