22 files changed, 374 insertions, 60 deletions
diff --git a/lib/Kconfig b/lib/Kconfig
index 5e7165e6a346..655b0e43f260 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -680,6 +680,11 @@ config STACK_HASH_ORDER
 	 Select the hash size as a power of 2 for the stackdepot hash table.
 	 Choose a lower value to reduce the memory impact.
 
+config REF_TRACKER
+	bool
+	depends on STACKTRACE_SUPPORT
+	select STACKDEPOT
+
 config SBITMAP
 	bool
 
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 9ef7ce18b4f5..c77fe36bb3d8 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -316,6 +316,7 @@ config DEBUG_INFO_BTF
 	bool "Generate BTF typeinfo"
 	depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED
 	depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST
+	depends on BPF_SYSCALL
 	help
 	  Generate deduplicated BTF type information from DWARF debug info.
 	  Turning this on expects presence of pahole tool, which will convert
@@ -346,8 +347,9 @@ config FRAME_WARN
 	int "Warn for stack frames larger than"
 	range 0 8192
 	default 2048 if GCC_PLUGIN_LATENT_ENTROPY
-	default 1536 if (!64BIT && (PARISC || XTENSA))
-	default 1024 if (!64BIT && !PARISC)
+	default 2048 if PARISC
+	default 1536 if (!64BIT && XTENSA)
+	default 1024 if !64BIT
 	default 2048 if 64BIT
 	help
 	  Tell gcc to warn at build time for stack frames larger than this.
@@ -597,6 +599,11 @@ config DEBUG_MISC
 	  Say Y here if you need to enable miscellaneous debug code that should
 	  be under a more specific debug option but isn't.
 
+menu "Networking Debugging"
+
+source "net/Kconfig.debug"
+
+endmenu # "Networking Debugging"
 
 menu "Memory Debugging"
 
@@ -2105,6 +2112,16 @@ config BACKTRACE_SELF_TEST
 
 	  Say N if you are unsure.
 
+config TEST_REF_TRACKER
+	tristate "Self test for reference tracker"
+	depends on DEBUG_KERNEL && STACKTRACE_SUPPORT
+	select REF_TRACKER
+	help
+	  This option provides a kernel module performing tests
+	  using reference tracker infrastructure.
+
+	  Say N if you are unsure.
+
 config RBTREE_TEST
 	tristate "Red-Black tree test"
 	depends on DEBUG_KERNEL
diff --git a/lib/Kconfig.kcsan b/lib/Kconfig.kcsan
index e0a93ffdef30..63b70b8c5551 100644
--- a/lib/Kconfig.kcsan
+++ b/lib/Kconfig.kcsan
@@ -191,6 +191,26 @@ config KCSAN_STRICT
 	  closely aligns with the rules defined by the Linux-kernel memory
 	  consistency model (LKMM).
 
+config KCSAN_WEAK_MEMORY
+	bool "Enable weak memory modeling to detect missing memory barriers"
+	default y
+	depends on KCSAN_STRICT
+	# We can either let objtool nop __tsan_func_{entry,exit}() and builtin
+	# atomics instrumentation in .noinstr.text, or use a compiler that can
+	# implement __no_kcsan to really remove all instrumentation.
+	depends on STACK_VALIDATION || CC_IS_GCC || CLANG_VERSION >= 140000
+	help
+	  Enable support for modeling a subset of weak memory, which allows
+	  detecting a subset of data races due to missing memory barriers.
+
+	  Depends on KCSAN_STRICT, because the options strenghtening certain
+	  plain accesses by default (depending on !KCSAN_STRICT) reduce the
+	  ability to detect any data races invoving reordered accesses, in
+	  particular reordered writes.
+
+	  Weak memory modeling relies on additional instrumentation and may
+	  affect performance.
+
 config KCSAN_REPORT_VALUE_CHANGE_ONLY
 	bool "Only report races where watcher observed a data value change"
 	default y
diff --git a/lib/Makefile b/lib/Makefile
index 364c23f15578..b213a7bbf3fd 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -101,7 +101,7 @@ obj-$(CONFIG_TEST_LOCKUP) += test_lockup.o
 obj-$(CONFIG_TEST_HMM) += test_hmm.o
 obj-$(CONFIG_TEST_FREE_PAGES) += test_free_pages.o
 obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
-
+obj-$(CONFIG_TEST_REF_TRACKER) += test_ref_tracker.o
 #
 # CFLAGS for compiling floating point code inside the kernel. x86/Makefile turns
 # off the generation of FPU/SSE* instructions for kernel proper but FPU_FLAGS
@@ -270,6 +270,8 @@ obj-$(CONFIG_STACKDEPOT) += stackdepot.o
 KASAN_SANITIZE_stackdepot.o := n
 KCOV_INSTRUMENT_stackdepot.o := n
 
+obj-$(CONFIG_REF_TRACKER) += ref_tracker.o
+
 libfdt_files = fdt.o fdt_ro.o fdt_wip.o fdt_rw.o fdt_sw.o fdt_strerror.o \
 	       fdt_empty_tree.o fdt_addresses.o
 $(foreach file, $(libfdt_files), \
diff --git a/lib/crypto/Kconfig b/lib/crypto/Kconfig
index 545ccbddf6a1..8620f38e117c 100644
--- a/lib/crypto/Kconfig
+++ b/lib/crypto/Kconfig
@@ -1,7 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
 
-comment "Crypto library routines"
-
 config CRYPTO_LIB_AES
 	tristate
 
@@ -9,14 +7,14 @@ config CRYPTO_LIB_ARC4
 	tristate
 
 config CRYPTO_ARCH_HAVE_LIB_BLAKE2S
-	tristate
+	bool
 	help
 	  Declares whether the architecture provides an arch-specific
 	  accelerated implementation of the Blake2s library interface,
 	  either builtin or as a module.
 
 config CRYPTO_LIB_BLAKE2S_GENERIC
-	tristate
+	def_bool !CRYPTO_ARCH_HAVE_LIB_BLAKE2S
 	help
 	  This symbol can be depended upon by arch implementations of the
 	  Blake2s library interface that require the generic code as a
@@ -24,15 +22,6 @@ config CRYPTO_LIB_BLAKE2S_GENERIC
 	  implementation is enabled, this implementation serves the users
 	  of CRYPTO_LIB_BLAKE2S.
 
-config CRYPTO_LIB_BLAKE2S
-	tristate "BLAKE2s hash function library"
-	depends on CRYPTO_ARCH_HAVE_LIB_BLAKE2S || !CRYPTO_ARCH_HAVE_LIB_BLAKE2S
-	select CRYPTO_LIB_BLAKE2S_GENERIC if CRYPTO_ARCH_HAVE_LIB_BLAKE2S=n
-	help
-	  Enable the Blake2s library interface. This interface may be fulfilled
-	  by either the generic implementation or an arch-specific one, if one
-	  is available and enabled.
-
 config CRYPTO_ARCH_HAVE_LIB_CHACHA
 	tristate
 	help
@@ -51,7 +40,7 @@ config CRYPTO_LIB_CHACHA_GENERIC
 	  of CRYPTO_LIB_CHACHA.
 
 config CRYPTO_LIB_CHACHA
-	tristate "ChaCha library interface"
+	tristate
 	depends on CRYPTO_ARCH_HAVE_LIB_CHACHA || !CRYPTO_ARCH_HAVE_LIB_CHACHA
 	select CRYPTO_LIB_CHACHA_GENERIC if CRYPTO_ARCH_HAVE_LIB_CHACHA=n
 	help
@@ -76,7 +65,7 @@ config CRYPTO_LIB_CURVE25519_GENERIC
 	  of CRYPTO_LIB_CURVE25519.
 
 config CRYPTO_LIB_CURVE25519
-	tristate "Curve25519 scalar multiplication library"
+	tristate
 	depends on CRYPTO_ARCH_HAVE_LIB_CURVE25519 || !CRYPTO_ARCH_HAVE_LIB_CURVE25519
 	select CRYPTO_LIB_CURVE25519_GENERIC if CRYPTO_ARCH_HAVE_LIB_CURVE25519=n
 	help
@@ -111,7 +100,7 @@ config CRYPTO_LIB_POLY1305_GENERIC
 	  of CRYPTO_LIB_POLY1305.
 
 config CRYPTO_LIB_POLY1305
-	tristate "Poly1305 library interface"
+	tristate
 	depends on CRYPTO_ARCH_HAVE_LIB_POLY1305 || !CRYPTO_ARCH_HAVE_LIB_POLY1305
 	select CRYPTO_LIB_POLY1305_GENERIC if CRYPTO_ARCH_HAVE_LIB_POLY1305=n
 	help
@@ -120,7 +109,7 @@ config CRYPTO_LIB_POLY1305
 	  is available and enabled.
 
 config CRYPTO_LIB_CHACHA20POLY1305
-	tristate "ChaCha20-Poly1305 AEAD support (8-byte nonce library version)"
+	tristate
 	depends on CRYPTO_ARCH_HAVE_LIB_CHACHA || !CRYPTO_ARCH_HAVE_LIB_CHACHA
 	depends on CRYPTO_ARCH_HAVE_LIB_POLY1305 || !CRYPTO_ARCH_HAVE_LIB_POLY1305
 	select CRYPTO_LIB_CHACHA
diff --git a/lib/crypto/Makefile b/lib/crypto/Makefile
index 73205ed269ba..ed43a41f2dcc 100644
--- a/lib/crypto/Makefile
+++ b/lib/crypto/Makefile
@@ -10,11 +10,10 @@ libaes-y					:= aes.o
 obj-$(CONFIG_CRYPTO_LIB_ARC4)			+= libarc4.o
 libarc4-y					:= arc4.o
 
-obj-$(CONFIG_CRYPTO_LIB_BLAKE2S_GENERIC)	+= libblake2s-generic.o
-libblake2s-generic-y				+= blake2s-generic.o
-
-obj-$(CONFIG_CRYPTO_LIB_BLAKE2S)		+= libblake2s.o
-libblake2s-y					+= blake2s.o
+# blake2s is used by the /dev/random driver which is always builtin
+obj-y						+= libblake2s.o
+libblake2s-y					:= blake2s.o
+libblake2s-$(CONFIG_CRYPTO_LIB_BLAKE2S_GENERIC)	+= blake2s-generic.o
 
 obj-$(CONFIG_CRYPTO_LIB_CHACHA20POLY1305)	+= libchacha20poly1305.o
 libchacha20poly1305-y				+= chacha20poly1305.o
diff --git a/lib/crypto/blake2s-generic.c b/lib/crypto/blake2s-generic.c
index 04ff8df24513..75ccb3e633e6 100644
--- a/lib/crypto/blake2s-generic.c
+++ b/lib/crypto/blake2s-generic.c
@@ -37,7 +37,11 @@ static inline void blake2s_increment_counter(struct blake2s_state *state,
 	state->t[1] += (state->t[0] < inc);
 }
 
-void blake2s_compress_generic(struct blake2s_state *state,const u8 *block,
+void blake2s_compress(struct blake2s_state *state, const u8 *block,
+		      size_t nblocks, const u32 inc)
+		      __weak __alias(blake2s_compress_generic);
+
+void blake2s_compress_generic(struct blake2s_state *state, const u8 *block,
 			      size_t nblocks, const u32 inc)
 {
 	u32 m[16];
diff --git a/lib/crypto/blake2s.c b/lib/crypto/blake2s.c
index 4055aa593ec4..93f2ae051370 100644
--- a/lib/crypto/blake2s.c
+++ b/lib/crypto/blake2s.c
@@ -16,12 +16,6 @@
 #include <linux/init.h>
 #include <linux/bug.h>
 
-#if IS_ENABLED(CONFIG_CRYPTO_ARCH_HAVE_LIB_BLAKE2S)
-#  define blake2s_compress blake2s_compress_arch
-#else
-#  define blake2s_compress blake2s_compress_generic
-#endif
-
 void blake2s_update(struct blake2s_state *state, const u8 *in, size_t inlen)
 {
 	__blake2s_update(state, in, inlen, blake2s_compress);
diff --git a/lib/kunit/test.c b/lib/kunit/test.c
index 3bd741e50a2d..c7ed4aabec04 100644
--- a/lib/kunit/test.c
+++ b/lib/kunit/test.c
@@ -504,25 +504,28 @@ int kunit_run_tests(struct kunit_suite *suite)
 		struct kunit_result_stats param_stats = { 0 };
 		test_case->status = KUNIT_SKIPPED;
 
-		if (test_case->generate_params) {
+		if (!test_case->generate_params) {
+			/* Non-parameterised test. */
+			kunit_run_case_catch_errors(suite, test_case, &test);
+			kunit_update_stats(&param_stats, test.status);
+		} else {
 			/* Get initial param. */
 			param_desc[0] = '\0';
 			test.param_value = test_case->generate_params(NULL, param_desc);
-		}
+			kunit_log(KERN_INFO, &test, KUNIT_SUBTEST_INDENT KUNIT_SUBTEST_INDENT
+				  "# Subtest: %s", test_case->name);
 
-		do {
-			kunit_run_case_catch_errors(suite, test_case, &test);
+			while (test.param_value) {
+				kunit_run_case_catch_errors(suite, test_case, &test);
 
-			if (test_case->generate_params) {
 				if (param_desc[0] == '\0') {
 					snprintf(param_desc, sizeof(param_desc),
 						 "param-%d", test.param_index);
 				}
 
 				kunit_log(KERN_INFO, &test,
-					  KUNIT_SUBTEST_INDENT
-					  "# %s: %s %d - %s",
-					  test_case->name,
+					  KUNIT_SUBTEST_INDENT KUNIT_SUBTEST_INDENT
+					  "%s %d - %s",
 					  kunit_status_to_ok_not_ok(test.status),
 					  test.param_index + 1, param_desc);
 
@@ -530,11 +533,11 @@ int kunit_run_tests(struct kunit_suite *suite)
 				param_desc[0] = '\0';
 				test.param_value = test_case->generate_params(test.param_value, param_desc);
 				test.param_index++;
-			}
 
-			kunit_update_stats(&param_stats, test.status);
+				kunit_update_stats(&param_stats, test.status);
+			}
+		}
 
-		} while (test.param_value);
 
 		kunit_print_test_stats(&test, param_stats);
 
diff --git a/lib/mpi/mpi-mod.c b/lib/mpi/mpi-mod.c
index 47bc59edd4ff..54fcc01564d9 100644
--- a/lib/mpi/mpi-mod.c
+++ b/lib/mpi/mpi-mod.c
@@ -40,6 +40,8 @@ mpi_barrett_t mpi_barrett_init(MPI m, int copy)
 
 	mpi_normalize(m);
 	ctx = kcalloc(1, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return NULL;
 
 	if (copy) {
 		ctx->m = mpi_copy(m);
diff --git a/lib/nmi_backtrace.c b/lib/nmi_backtrace.c
index f9e89001b52e..199ab201d501 100644
--- a/lib/nmi_backtrace.c
+++ b/lib/nmi_backtrace.c
@@ -75,6 +75,12 @@ void nmi_trigger_cpumask_backtrace(const cpumask_t *mask,
 		touch_softlockup_watchdog();
 	}
 
+	/*
+	 * Force flush any remote buffers that might be stuck in IRQ context
+	 * and therefore could not run their irq_work.
+	 */
+	printk_trigger_flush();
+
 	clear_bit_unlock(0, &backtrace_flag);
 	put_cpu();
 }
diff --git a/lib/objagg.c b/lib/objagg.c
index 5e1676ccdadd..1e248629ed64 100644
--- a/lib/objagg.c
+++ b/lib/objagg.c
@@ -781,7 +781,6 @@ static struct objagg_tmp_graph *objagg_tmp_graph_create(struct objagg *objagg)
 	struct objagg_tmp_node *node;
 	struct objagg_tmp_node *pnode;
 	struct objagg_obj *objagg_obj;
-	size_t alloc_size;
 	int i, j;
 
 	graph = kzalloc(sizeof(*graph), GFP_KERNEL);
@@ -793,9 +792,7 @@ static struct objagg_tmp_graph *objagg_tmp_graph_create(struct objagg *objagg)
 		goto err_nodes_alloc;
 	graph->nodes_count = nodes_count;
 
-	alloc_size = BITS_TO_LONGS(nodes_count * nodes_count) *
-		     sizeof(unsigned long);
-	graph->edges = kzalloc(alloc_size, GFP_KERNEL);
+	graph->edges = bitmap_zalloc(nodes_count * nodes_count, GFP_KERNEL);
 	if (!graph->edges)
 		goto err_edges_alloc;
 
@@ -833,7 +830,7 @@ err_nodes_alloc:
 
 static void objagg_tmp_graph_destroy(struct objagg_tmp_graph *graph)
 {
-	kfree(graph->edges);
+	bitmap_free(graph->edges);
 	kfree(graph->nodes);
 	kfree(graph);
 }
diff --git a/lib/ref_tracker.c b/lib/ref_tracker.c
new file mode 100644
index 000000000000..0ae2e66dcf0f
--- /dev/null
+++ b/lib/ref_tracker.c
@@ -0,0 +1,140 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+#include <linux/export.h>
+#include <linux/ref_tracker.h>
+#include <linux/slab.h>
+#include <linux/stacktrace.h>
+#include <linux/stackdepot.h>
+
+#define REF_TRACKER_STACK_ENTRIES 16
+
+struct ref_tracker {
+	struct list_head	head;   /* anchor into dir->list or dir->quarantine */
+	bool			dead;
+	depot_stack_handle_t	alloc_stack_handle;
+	depot_stack_handle_t	free_stack_handle;
+};
+
+void ref_tracker_dir_exit(struct ref_tracker_dir *dir)
+{
+	struct ref_tracker *tracker, *n;
+	unsigned long flags;
+	bool leak = false;
+
+	spin_lock_irqsave(&dir->lock, flags);
+	list_for_each_entry_safe(tracker, n, &dir->quarantine, head) {
+		list_del(&tracker->head);
+		kfree(tracker);
+		dir->quarantine_avail++;
+	}
+	list_for_each_entry_safe(tracker, n, &dir->list, head) {
+		pr_err("leaked reference.\n");
+		if (tracker->alloc_stack_handle)
+			stack_depot_print(tracker->alloc_stack_handle);
+		leak = true;
+		list_del(&tracker->head);
+		kfree(tracker);
+	}
+	spin_unlock_irqrestore(&dir->lock, flags);
+	WARN_ON_ONCE(leak);
+	WARN_ON_ONCE(refcount_read(&dir->untracked) != 1);
+}
+EXPORT_SYMBOL(ref_tracker_dir_exit);
+
+void ref_tracker_dir_print(struct ref_tracker_dir *dir,
+			   unsigned int display_limit)
+{
+	struct ref_tracker *tracker;
+	unsigned long flags;
+	unsigned int i = 0;
+
+	spin_lock_irqsave(&dir->lock, flags);
+	list_for_each_entry(tracker, &dir->list, head) {
+		if (i < display_limit) {
+			pr_err("leaked reference.\n");
+			if (tracker->alloc_stack_handle)
+				stack_depot_print(tracker->alloc_stack_handle);
+			i++;
+		} else {
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&dir->lock, flags);
+}
+EXPORT_SYMBOL(ref_tracker_dir_print);
+
+int ref_tracker_alloc(struct ref_tracker_dir *dir,
+		      struct ref_tracker **trackerp,
+		      gfp_t gfp)
+{
+	unsigned long entries[REF_TRACKER_STACK_ENTRIES];
+	struct ref_tracker *tracker;
+	unsigned int nr_entries;
+	unsigned long flags;
+
+	*trackerp = tracker = kzalloc(sizeof(*tracker), gfp | __GFP_NOFAIL);
+	if (unlikely(!tracker)) {
+		pr_err_once("memory allocation failure, unreliable refcount tracker.\n");
+		refcount_inc(&dir->untracked);
+		return -ENOMEM;
+	}
+	nr_entries = stack_trace_save(entries, ARRAY_SIZE(entries), 1);
+	nr_entries = filter_irq_stacks(entries, nr_entries);
+	tracker->alloc_stack_handle = stack_depot_save(entries, nr_entries, gfp);
+
+	spin_lock_irqsave(&dir->lock, flags);
+	list_add(&tracker->head, &dir->list);
+	spin_unlock_irqrestore(&dir->lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ref_tracker_alloc);
+
+int ref_tracker_free(struct ref_tracker_dir *dir,
+		     struct ref_tracker **trackerp)
+{
+	unsigned long entries[REF_TRACKER_STACK_ENTRIES];
+	struct ref_tracker *tracker = *trackerp;
+	depot_stack_handle_t stack_handle;
+	unsigned int nr_entries;
+	unsigned long flags;
+
+	if (!tracker) {
+		refcount_dec(&dir->untracked);
+		return -EEXIST;
+	}
+	nr_entries = stack_trace_save(entries, ARRAY_SIZE(entries), 1);
+	nr_entries = filter_irq_stacks(entries, nr_entries);
+	stack_handle = stack_depot_save(entries, nr_entries, GFP_ATOMIC);
+
+	spin_lock_irqsave(&dir->lock, flags);
+	if (tracker->dead) {
+		pr_err("reference already released.\n");
+		if (tracker->alloc_stack_handle) {
+			pr_err("allocated in:\n");
+			stack_depot_print(tracker->alloc_stack_handle);
+		}
+		if (tracker->free_stack_handle) {
+			pr_err("freed in:\n");
+			stack_depot_print(tracker->free_stack_handle);
+		}
+		spin_unlock_irqrestore(&dir->lock, flags);
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
+	tracker->dead = true;
+
+	tracker->free_stack_handle = stack_handle;
+
+	list_move_tail(&tracker->head, &dir->quarantine);
+	if (!dir->quarantine_avail) {
+		tracker = list_first_entry(&dir->quarantine, struct ref_tracker, head);
+		list_del(&tracker->head);
+	} else {
+		dir->quarantine_avail--;
+		tracker = NULL;
+	}
+	spin_unlock_irqrestore(&dir->lock, flags);
+
+	kfree(tracker);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ref_tracker_free);
diff --git a/lib/siphash.c b/lib/siphash.c
index a90112ee72a1..72b9068ab57b 100644
--- a/lib/siphash.c
+++ b/lib/siphash.c
@@ -49,6 +49,7 @@
 	SIPROUND; \
 	return (v0 ^ v1) ^ (v2 ^ v3);
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));
@@ -80,8 +81,8 @@ u64 __siphash_aligned(const void *data, size_t len, const siphash_key_t *key)
 	POSTAMBLE
 }
 EXPORT_SYMBOL(__siphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));
@@ -113,7 +114,6 @@ u64 __siphash_unaligned(const void *data, size_t len, const siphash_key_t *key)
 	POSTAMBLE
 }
 EXPORT_SYMBOL(__siphash_unaligned);
-#endif
 
 /**
  * siphash_1u64 - compute 64-bit siphash PRF value of a u64
@@ -250,6 +250,7 @@ EXPORT_SYMBOL(siphash_3u32);
 	HSIPROUND; \
 	return (v0 ^ v1) ^ (v2 ^ v3);
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u64));
@@ -280,8 +281,8 @@ u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_unaligned(const void *data, size_t len,
 			 const hsiphash_key_t *key)
 {
@@ -313,7 +314,6 @@ u32 __hsiphash_unaligned(const void *data, size_t len,
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_unaligned);
-#endif
 
 /**
  * hsiphash_1u32 - compute 64-bit hsiphash PRF value of a u32
@@ -418,6 +418,7 @@ EXPORT_SYMBOL(hsiphash_4u32);
 	HSIPROUND; \
 	return v1 ^ v3;
 
+#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 {
 	const u8 *end = data + len - (len % sizeof(u32));
@@ -438,8 +439,8 @@ u32 __hsiphash_aligned(const void *data, size_t len, const hsiphash_key_t *key)
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_aligned);
+#endif
 
-#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
 u32 __hsiphash_unaligned(const void *data, size_t len,
 			 const hsiphash_key_t *key)
 {
@@ -461,7 +462,6 @@ u32 __hsiphash_unaligned(const void *data, size_t len,
 	HPOSTAMBLE
 }
 EXPORT_SYMBOL(__hsiphash_unaligned);
-#endif
 
 /**
  * hsiphash_1u32 - compute 32-bit hsiphash PRF value of a u32
diff --git a/lib/test_bpf.c b/lib/test_bpf.c
index adae39567264..0c5cb2d6436a 100644
--- a/lib/test_bpf.c
+++ b/lib/test_bpf.c
@@ -14683,7 +14683,7 @@ static struct tail_call_test tail_call_tests[] = {
 			BPF_EXIT_INSN(),
 		},
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
-		.result = (MAX_TAIL_CALL_CNT + 1 + 1) * MAX_TESTRUNS,
+		.result = (MAX_TAIL_CALL_CNT + 1) * MAX_TESTRUNS,
 	},
 	{
 		"Tail call count preserved across function calls",
@@ -14705,7 +14705,7 @@ static struct tail_call_test tail_call_tests[] = {
 		},
 		.stack_depth = 8,
 		.flags = FLAG_NEED_STATE | FLAG_RESULT_IN_STATE,
-		.result = (MAX_TAIL_CALL_CNT + 1 + 1) * MAX_TESTRUNS,
+		.result = (MAX_TAIL_CALL_CNT + 1) * MAX_TESTRUNS,
 	},
 	{
 		"Tail call error path, NULL target",
diff --git a/lib/test_kasan.c b/lib/test_kasan.c
index 67ed689a0b1b..0643573f8686 100644
--- a/lib/test_kasan.c
+++ b/lib/test_kasan.c
@@ -869,6 +869,7 @@ static void kasan_memchr(struct kunit *test)
 	ptr = kmalloc(size, GFP_KERNEL | __GFP_ZERO);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
+	OPTIMIZER_HIDE_VAR(size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		kasan_ptr_result = memchr(ptr, '1', size + 1));
 
@@ -894,6 +895,7 @@ static void kasan_memcmp(struct kunit *test)
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 	memset(arr, 0, sizeof(arr));
 
+	OPTIMIZER_HIDE_VAR(size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		kasan_int_result = memcmp(ptr, arr, size+1));
 	kfree(ptr);
diff --git a/lib/test_ref_tracker.c b/lib/test_ref_tracker.c
new file mode 100644
index 000000000000..19d7dec70cc6
--- /dev/null
+++ b/lib/test_ref_tracker.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Referrence tracker self test.
+ *
+ * Copyright (c) 2021 Eric Dumazet <edumazet@google.com>
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/ref_tracker.h>
+#include <linux/slab.h>
+#include <linux/timer.h>
+
+static struct ref_tracker_dir ref_dir;
+static struct ref_tracker *tracker[20];
+
+#define TRT_ALLOC(X) static noinline void 				\
+	alloctest_ref_tracker_alloc##X(struct ref_tracker_dir *dir, 	\
+				    struct ref_tracker **trackerp)	\
+	{								\
+		ref_tracker_alloc(dir, trackerp, GFP_KERNEL);		\
+	}
+
+TRT_ALLOC(1)
+TRT_ALLOC(2)
+TRT_ALLOC(3)
+TRT_ALLOC(4)
+TRT_ALLOC(5)
+TRT_ALLOC(6)
+TRT_ALLOC(7)
+TRT_ALLOC(8)
+TRT_ALLOC(9)
+TRT_ALLOC(10)
+TRT_ALLOC(11)
+TRT_ALLOC(12)
+TRT_ALLOC(13)
+TRT_ALLOC(14)
+TRT_ALLOC(15)
+TRT_ALLOC(16)
+TRT_ALLOC(17)
+TRT_ALLOC(18)
+TRT_ALLOC(19)
+
+#undef TRT_ALLOC
+
+static noinline void
+alloctest_ref_tracker_free(struct ref_tracker_dir *dir,
+			   struct ref_tracker **trackerp)
+{
+	ref_tracker_free(dir, trackerp);
+}
+
+
+static struct timer_list test_ref_tracker_timer;
+static atomic_t test_ref_timer_done = ATOMIC_INIT(0);
+
+static void test_ref_tracker_timer_func(struct timer_list *t)
+{
+	ref_tracker_alloc(&ref_dir, &tracker[0], GFP_ATOMIC);
+	atomic_set(&test_ref_timer_done, 1);
+}
+
+static int __init test_ref_tracker_init(void)
+{
+	int i;
+
+	ref_tracker_dir_init(&ref_dir, 100);
+
+	timer_setup(&test_ref_tracker_timer, test_ref_tracker_timer_func, 0);
+	mod_timer(&test_ref_tracker_timer, jiffies + 1);
+
+	alloctest_ref_tracker_alloc1(&ref_dir, &tracker[1]);
+	alloctest_ref_tracker_alloc2(&ref_dir, &tracker[2]);
+	alloctest_ref_tracker_alloc3(&ref_dir, &tracker[3]);
+	alloctest_ref_tracker_alloc4(&ref_dir, &tracker[4]);
+	alloctest_ref_tracker_alloc5(&ref_dir, &tracker[5]);
+	alloctest_ref_tracker_alloc6(&ref_dir, &tracker[6]);
+	alloctest_ref_tracker_alloc7(&ref_dir, &tracker[7]);
+	alloctest_ref_tracker_alloc8(&ref_dir, &tracker[8]);
+	alloctest_ref_tracker_alloc9(&ref_dir, &tracker[9]);
+	alloctest_ref_tracker_alloc10(&ref_dir, &tracker[10]);
+	alloctest_ref_tracker_alloc11(&ref_dir, &tracker[11]);
+	alloctest_ref_tracker_alloc12(&ref_dir, &tracker[12]);
+	alloctest_ref_tracker_alloc13(&ref_dir, &tracker[13]);
+	alloctest_ref_tracker_alloc14(&ref_dir, &tracker[14]);
+	alloctest_ref_tracker_alloc15(&ref_dir, &tracker[15]);
+	alloctest_ref_tracker_alloc16(&ref_dir, &tracker[16]);
+	alloctest_ref_tracker_alloc17(&ref_dir, &tracker[17]);
+	alloctest_ref_tracker_alloc18(&ref_dir, &tracker[18]);
+	alloctest_ref_tracker_alloc19(&ref_dir, &tracker[19]);
+
+	/* free all trackers but first 0 and 1. */
+	for (i = 2; i < ARRAY_SIZE(tracker); i++)
+		alloctest_ref_tracker_free(&ref_dir, &tracker[i]);
+
+	/* Attempt to free an already freed tracker. */
+	alloctest_ref_tracker_free(&ref_dir, &tracker[2]);
+
+	while (!atomic_read(&test_ref_timer_done))
+		msleep(1);
+
+	/* This should warn about tracker[0] & tracker[1] being not freed. */
+	ref_tracker_dir_exit(&ref_dir);
+
+	return 0;
+}
+
+static void __exit test_ref_tracker_exit(void)
+{
+}
+
+module_init(test_ref_tracker_init);
+module_exit(test_ref_tracker_exit);
+
+MODULE_LICENSE("GPL v2");
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index 58d5e567f836..53d6081f9e8b 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -3564,7 +3564,7 @@ int vsscanf(const char *buf, const char *fmt, va_list args)
 				++fmt;
 
 			for ( ; *fmt && *fmt != ']'; ++fmt, ++len)
-				set_bit((u8)*fmt, set);
+				__set_bit((u8)*fmt, set);
 
 			/* no ']' or no character set found */
 			if (!*fmt || !len)
@@ -3574,7 +3574,7 @@ int vsscanf(const char *buf, const char *fmt, va_list args)
 			if (negate) {
 				bitmap_complement(set, set, 256);
 				/* exclude null '\0' byte */
-				clear_bit(0, set);
+				__clear_bit(0, set);
 			}
 
 			/* match must be non-empty */
diff --git a/lib/zstd/Makefile b/lib/zstd/Makefile
index 65218ec5b8f2..fc45339fc3a3 100644
--- a/lib/zstd/Makefile
+++ b/lib/zstd/Makefile
@@ -11,8 +11,6 @@
 obj-$(CONFIG_ZSTD_COMPRESS) += zstd_compress.o
 obj-$(CONFIG_ZSTD_DECOMPRESS) += zstd_decompress.o
 
-ccflags-y += -O3
-
 zstd_compress-y := \
 		zstd_compress_module.o \
 		common/debug.o \
diff --git a/lib/zstd/common/compiler.h b/lib/zstd/common/compiler.h
index a1a051e4bce6..f5a9c70a228a 100644
--- a/lib/zstd/common/compiler.h
+++ b/lib/zstd/common/compiler.h
@@ -16,6 +16,7 @@
 *********************************************************/
 /* force inlining */
 
+#if !defined(ZSTD_NO_INLINE)
 #if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L   /* C99 */
 #  define INLINE_KEYWORD inline
 #else
@@ -24,6 +25,12 @@
 
 #define FORCE_INLINE_ATTR __attribute__((always_inline))
 
+#else
+
+#define INLINE_KEYWORD
+#define FORCE_INLINE_ATTR
+
+#endif
 
 /*
   On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC).
diff --git a/lib/zstd/compress/zstd_compress_superblock.c b/lib/zstd/compress/zstd_compress_superblock.c
index ee03e0aedb03..b0610b255653 100644
--- a/lib/zstd/compress/zstd_compress_superblock.c
+++ b/lib/zstd/compress/zstd_compress_superblock.c
@@ -411,6 +411,8 @@ static size_t ZSTD_seqDecompressedSize(seqStore_t const* seqStore, const seqDef*
     const seqDef* sp = sstart;
     size_t matchLengthSum = 0;
     size_t litLengthSum = 0;
+    /* Only used by assert(), suppress unused variable warnings in production. */
+    (void)litLengthSum;
     while (send-sp > 0) {
         ZSTD_sequenceLength const seqLen = ZSTD_getSequenceLength(seqStore, sp);
         litLengthSum += seqLen.litLength;
diff --git a/lib/zstd/compress/zstd_opt.c b/lib/zstd/compress/zstd_opt.c
index 04337050fe9a..dfc55e3e8119 100644
--- a/lib/zstd/compress/zstd_opt.c
+++ b/lib/zstd/compress/zstd_opt.c
@@ -8,6 +8,18 @@
  * You may select, at your option, one of the above-listed licenses.
  */
 
+/*
+ * Disable inlining for the optimal parser for the kernel build.
+ * It is unlikely to be used in the kernel, and where it is used
+ * latency shouldn't matter because it is very slow to begin with.
+ * We prefer a ~180KB binary size win over faster optimal parsing.
+ *
+ * TODO(https://github.com/facebook/zstd/issues/2862):
+ * Improve the code size of the optimal parser in general, so we
+ * don't need this hack for the kernel build.
+ */
+#define ZSTD_NO_INLINE 1
+
 #include "zstd_compress_internal.h"
 #include "hist.h"
 #include "zstd_opt.h"