Merge tag 'mm-nonmm-stable-2024-07-21-15-07' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull non-MM updates from Andrew Morton:

 - In the series "treewide: Refactor heap related implementation",
   Kuan-Wei Chiu has significantly reworked the min_heap library code
   and has taught bcachefs to use the new more generic implementation.

 - Yury Norov's series "Cleanup cpumask.h inclusion in core headers"
   reworks the cpumask and nodemask headers to make things generally
   more rational.

 - Kuan-Wei Chiu has sent along some maintenance work against our
   sorting library code in the series "lib/sort: Optimizations and
   cleanups".

 - More library maintainance work from Christophe Jaillet in the series
   "Remove usage of the deprecated ida_simple_xx() API".

 - Ryusuke Konishi continues with the nilfs2 fixes and clanups in the
   series "nilfs2: eliminate the call to inode_attach_wb()".

 - Kuan-Ying Lee has some fixes to the gdb scripts in the series "Fix
   GDB command error".

 - Plus the usual shower of singleton patches all over the place. Please
   see the relevant changelogs for details.

* tag 'mm-nonmm-stable-2024-07-21-15-07' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (98 commits)
  ia64: scrub ia64 from poison.h
  watchdog/perf: properly initialize the turbo mode timestamp and rearm counter
  tsacct: replace strncpy() with strscpy()
  lib/bch.c: use swap() to improve code
  test_bpf: convert comma to semicolon
  init/modpost: conditionally check section mismatch to __meminit*
  init: remove unused __MEMINIT* macros
  nilfs2: Constify struct kobj_type
  nilfs2: avoid undefined behavior in nilfs_cnt32_ge macro
  math: rational: add missing MODULE_DESCRIPTION() macro
  lib/zlib: add missing MODULE_DESCRIPTION() macro
  fs: ufs: add MODULE_DESCRIPTION()
  lib/rbtree.c: fix the example typo
  ocfs2: add bounds checking to ocfs2_check_dir_entry()
  fs: add kernel-doc comments to ocfs2_prepare_orphan_dir()
  coredump: simplify zap_process()
  selftests/fpu: add missing MODULE_DESCRIPTION() macro
  compiler.h: simplify data_race() macro
  build-id: require program headers to be right after ELF header
  resource: add missing MODULE_DESCRIPTION()
  ...

31 files changed, 247 insertions, 150 deletions

diff --git a/include/linux/cacheinfo.h b/include/linux/cacheinfo.h
index 3dde175f4108..108060612bb8 100644
--- a/include/linux/cacheinfo.h
+++ b/include/linux/cacheinfo.h
@@ -4,7 +4,7 @@
 
 #include <linux/bitops.h>
 #include <linux/cpuhplock.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/smp.h>
 
 struct device_node;
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 2150ca60394b..c60ba0ab1462 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -10,7 +10,6 @@
  */
 
 #include <linux/sched.h>
-#include <linux/cpumask.h>
 #include <linux/nodemask.h>
 #include <linux/rculist.h>
 #include <linux/cgroupstats.h>
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h
index 9aac31d856f3..b0df28ddd394 100644
--- a/include/linux/clockchips.h
+++ b/include/linux/clockchips.h
@@ -12,7 +12,7 @@
 #ifdef CONFIG_GENERIC_CLOCKEVENTS
 
 # include <linux/clocksource.h>
-# include <linux/cpumask.h>
+# include <linux/cpumask_types.h>
 # include <linux/ktime.h>
 # include <linux/notifier.h>
 
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 68a24a3a6979..2594553bb30b 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -208,10 +208,8 @@ void ftrace_likely_update(struct ftrace_likely_data *f, int val,
  */
 #define data_race(expr)							\
 ({									\
-	__unqual_scalar_typeof(({ expr; })) __v = ({			\
-		__kcsan_disable_current();				\
-		expr;							\
-	});								\
+	__kcsan_disable_current();					\
+	__auto_type __v = (expr);					\
 	__kcsan_enable_current();					\
 	__v;								\
 })
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index a8926d0a28cd..bdcec1732445 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -16,7 +16,6 @@
 
 #include <linux/node.h>
 #include <linux/compiler.h>
-#include <linux/cpumask.h>
 #include <linux/cpuhotplug.h>
 #include <linux/cpuhplock.h>
 #include <linux/cpu_smt.h>
diff --git a/include/linux/cpu_cooling.h b/include/linux/cpu_cooling.h
index a3bdc8a98f2c..2c774fb3c091 100644
--- a/include/linux/cpu_cooling.h
+++ b/include/linux/cpu_cooling.h
@@ -15,7 +15,6 @@
 
 #include <linux/of.h>
 #include <linux/thermal.h>
-#include <linux/cpumask.h>
 
 struct cpufreq_policy;
 
diff --git a/include/linux/cpu_rmap.h b/include/linux/cpu_rmap.h
index cae324d10965..20b5729903d7 100644
--- a/include/linux/cpu_rmap.h
+++ b/include/linux/cpu_rmap.h
@@ -7,7 +7,7 @@
  * Copyright 2011 Solarflare Communications Inc.
  */
 
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/gfp.h>
 #include <linux/slab.h>
 #include <linux/kref.h>
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 954d4adc8f81..099e8b32dd68 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -9,25 +9,13 @@
  */
 #include <linux/cleanup.h>
 #include <linux/kernel.h>
-#include <linux/threads.h>
 #include <linux/bitmap.h>
+#include <linux/cpumask_types.h>
 #include <linux/atomic.h>
 #include <linux/bug.h>
 #include <linux/gfp_types.h>
 #include <linux/numa.h>
 
-/* Don't assign or return these: may not be this big! */
-typedef struct cpumask { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;
-
-/**
- * cpumask_bits - get the bits in a cpumask
- * @maskp: the struct cpumask *
- *
- * You should only assume nr_cpu_ids bits of this mask are valid.  This is
- * a macro so it's const-correct.
- */
-#define cpumask_bits(maskp) ((maskp)->bits)
-
 /**
  * cpumask_pr_args - printf args to output a cpumask
  * @maskp: cpumask to be printed
@@ -925,48 +913,7 @@ static inline unsigned int cpumask_size(void)
 	return bitmap_size(large_cpumask_bits);
 }
 
-/*
- * cpumask_var_t: struct cpumask for stack usage.
- *
- * Oh, the wicked games we play!  In order to make kernel coding a
- * little more difficult, we typedef cpumask_var_t to an array or a
- * pointer: doing &mask on an array is a noop, so it still works.
- *
- * i.e.
- *	cpumask_var_t tmpmask;
- *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
- *		return -ENOMEM;
- *
- *	  ... use 'tmpmask' like a normal struct cpumask * ...
- *
- *	free_cpumask_var(tmpmask);
- *
- *
- * However, one notable exception is there. alloc_cpumask_var() allocates
- * only nr_cpumask_bits bits (in the other hand, real cpumask_t always has
- * NR_CPUS bits). Therefore you don't have to dereference cpumask_var_t.
- *
- *	cpumask_var_t tmpmask;
- *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
- *		return -ENOMEM;
- *
- *	var = *tmpmask;
- *
- * This code makes NR_CPUS length memcopy and brings to a memory corruption.
- * cpumask_copy() provide safe copy functionality.
- *
- * Note that there is another evil here: If you define a cpumask_var_t
- * as a percpu variable then the way to obtain the address of the cpumask
- * structure differently influences what this_cpu_* operation needs to be
- * used. Please use this_cpu_cpumask_var_t in those cases. The direct use
- * of this_cpu_ptr() or this_cpu_read() will lead to failures when the
- * other type of cpumask_var_t implementation is configured.
- *
- * Please also note that __cpumask_var_read_mostly can be used to declare
- * a cpumask_var_t variable itself (not its content) as read mostly.
- */
 #ifdef CONFIG_CPUMASK_OFFSTACK
-typedef struct cpumask *cpumask_var_t;
 
 #define this_cpu_cpumask_var_ptr(x)	this_cpu_read(x)
 #define __cpumask_var_read_mostly	__read_mostly
@@ -1013,7 +960,6 @@ static inline bool cpumask_available(cpumask_var_t mask)
 }
 
 #else
-typedef struct cpumask cpumask_var_t[1];
 
 #define this_cpu_cpumask_var_ptr(x) this_cpu_ptr(x)
 #define __cpumask_var_read_mostly
diff --git a/include/linux/cpumask_types.h b/include/linux/cpumask_types.h
new file mode 100644
index 000000000000..461ed1b6bcdb
--- /dev/null
+++ b/include/linux/cpumask_types.h
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __LINUX_CPUMASK_TYPES_H
+#define __LINUX_CPUMASK_TYPES_H
+
+#include <linux/bitops.h>
+#include <linux/threads.h>
+
+/* Don't assign or return these: may not be this big! */
+typedef struct cpumask { DECLARE_BITMAP(bits, NR_CPUS); } cpumask_t;
+
+/**
+ * cpumask_bits - get the bits in a cpumask
+ * @maskp: the struct cpumask *
+ *
+ * You should only assume nr_cpu_ids bits of this mask are valid.  This is
+ * a macro so it's const-correct.
+ */
+#define cpumask_bits(maskp) ((maskp)->bits)
+
+/*
+ * cpumask_var_t: struct cpumask for stack usage.
+ *
+ * Oh, the wicked games we play!  In order to make kernel coding a
+ * little more difficult, we typedef cpumask_var_t to an array or a
+ * pointer: doing &mask on an array is a noop, so it still works.
+ *
+ * i.e.
+ *	cpumask_var_t tmpmask;
+ *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ *		return -ENOMEM;
+ *
+ *	  ... use 'tmpmask' like a normal struct cpumask * ...
+ *
+ *	free_cpumask_var(tmpmask);
+ *
+ *
+ * However, one notable exception is there. alloc_cpumask_var() allocates
+ * only nr_cpumask_bits bits (in the other hand, real cpumask_t always has
+ * NR_CPUS bits). Therefore you don't have to dereference cpumask_var_t.
+ *
+ *	cpumask_var_t tmpmask;
+ *	if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ *		return -ENOMEM;
+ *
+ *	var = *tmpmask;
+ *
+ * This code makes NR_CPUS length memcopy and brings to a memory corruption.
+ * cpumask_copy() provide safe copy functionality.
+ *
+ * Note that there is another evil here: If you define a cpumask_var_t
+ * as a percpu variable then the way to obtain the address of the cpumask
+ * structure differently influences what this_cpu_* operation needs to be
+ * used. Please use this_cpu_cpumask_var_t in those cases. The direct use
+ * of this_cpu_ptr() or this_cpu_read() will lead to failures when the
+ * other type of cpumask_var_t implementation is configured.
+ *
+ * Please also note that __cpumask_var_read_mostly can be used to declare
+ * a cpumask_var_t variable itself (not its content) as read mostly.
+ */
+#ifdef CONFIG_CPUMASK_OFFSTACK
+typedef struct cpumask *cpumask_var_t;
+#else
+typedef struct cpumask cpumask_var_t[1];
+#endif /* CONFIG_CPUMASK_OFFSTACK */
+
+#endif /* __LINUX_CPUMASK_TYPES_H */
diff --git a/include/linux/init.h b/include/linux/init.h
index 58cef4c2e59a..ee1309473bc6 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -84,11 +84,15 @@
 
 #define __exit          __section(".exit.text") __exitused __cold notrace
 
-/* Used for MEMORY_HOTPLUG */
-#define __meminit        __section(".meminit.text") __cold notrace \
-						  __latent_entropy
-#define __meminitdata    __section(".meminit.data")
-#define __meminitconst   __section(".meminit.rodata")
+#ifdef CONFIG_MEMORY_HOTPLUG
+#define __meminit
+#define __meminitdata
+#define __meminitconst
+#else
+#define __meminit	__init
+#define __meminitdata	__initdata
+#define __meminitconst	__initconst
+#endif
 
 /* For assembly routines */
 #define __HEAD		.section	".head.text","ax"
@@ -99,10 +103,6 @@
 #define __INITRODATA	.section	".init.rodata","a",%progbits
 #define __FINITDATA	.previous
 
-#define __MEMINIT        .section	".meminit.text", "ax"
-#define __MEMINITDATA    .section	".meminit.data", "aw"
-#define __MEMINITRODATA  .section	".meminit.rodata", "a"
-
 /* silence warnings when references are OK */
 #define __REF            .section       ".ref.text", "ax"
 #define __REFDATA        .section       ".ref.data", "aw"
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index 3a36e64119c8..bea39a0292eb 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -6,13 +6,13 @@
 #include <linux/kernel.h>
 #include <linux/bitops.h>
 #include <linux/cleanup.h>
-#include <linux/cpumask.h>
 #include <linux/irqreturn.h>
 #include <linux/irqnr.h>
 #include <linux/hardirq.h>
 #include <linux/irqflags.h>
 #include <linux/hrtimer.h>
 #include <linux/kref.h>
+#include <linux/cpumask_types.h>
 #include <linux/workqueue.h>
 #include <linux/jump_label.h>
 
diff --git a/include/linux/irqchip/irq-partition-percpu.h b/include/linux/irqchip/irq-partition-percpu.h
index 2f6ae7551748..b35ee22c278f 100644
--- a/include/linux/irqchip/irq-partition-percpu.h
+++ b/include/linux/irqchip/irq-partition-percpu.h
@@ -8,7 +8,7 @@
 #define __LINUX_IRQCHIP_IRQ_PARTITION_PERCPU_H
 
 #include <linux/fwnode.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/irqdomain.h>
 
 struct partition_affinity {
diff --git a/include/linux/jhash.h b/include/linux/jhash.h
index ab7f8c152b89..fa26a2dd3b52 100644
--- a/include/linux/jhash.h
+++ b/include/linux/jhash.h
@@ -31,7 +31,7 @@
 /* Mask the hash value, i.e (value & jhash_mask(n)) instead of (value % n) */
 #define jhash_mask(n)   (jhash_size(n)-1)
 
-/* __jhash_mix -- mix 3 32-bit values reversibly. */
+/* __jhash_mix - mix 3 32-bit values reversibly. */
 #define __jhash_mix(a, b, c)			\
 {						\
 	a -= c;  a ^= rol32(c, 4);  c += b;	\
@@ -60,7 +60,7 @@
 /* jhash - hash an arbitrary key
  * @k: sequence of bytes as key
  * @length: the length of the key
- * @initval: the previous hash, or an arbitray value
+ * @initval: the previous hash, or an arbitrary value
  *
  * The generic version, hashes an arbitrary sequence of bytes.
  * No alignment or length assumptions are made about the input key.
@@ -110,7 +110,7 @@ static inline u32 jhash(const void *key, u32 length, u32 initval)
 /* jhash2 - hash an array of u32's
  * @k: the key which must be an array of u32's
  * @length: the number of u32's in the key
- * @initval: the previous hash, or an arbitray value
+ * @initval: the previous hash, or an arbitrary value
  *
  * Returns the hash value of the key.
  */
diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h
index 9c042c6384bb..b97ce2df376f 100644
--- a/include/linux/kernel_stat.h
+++ b/include/linux/kernel_stat.h
@@ -5,7 +5,6 @@
 #include <linux/smp.h>
 #include <linux/threads.h>
 #include <linux/percpu.h>
-#include <linux/cpumask.h>
 #include <linux/interrupt.h>
 #include <linux/sched.h>
 #include <linux/vtime.h>
diff --git a/include/linux/min_heap.h b/include/linux/min_heap.h
index d52daf45861b..43a7b9dcf15e 100644
--- a/include/linux/min_heap.h
+++ b/include/linux/min_heap.h
@@ -7,46 +7,89 @@
 #include <linux/types.h>
 
 /**
- * struct min_heap - Data structure to hold a min-heap.
- * @data: Start of array holding the heap elements.
+ * Data structure to hold a min-heap.
  * @nr: Number of elements currently in the heap.
  * @size: Maximum number of elements that can be held in current storage.
+ * @data: Pointer to the start of array holding the heap elements.
+ * @preallocated: Start of the static preallocated array holding the heap elements.
  */
-struct min_heap {
-	void *data;
-	int nr;
-	int size;
-};
+#define MIN_HEAP_PREALLOCATED(_type, _name, _nr)	\
+struct _name {	\
+	int nr;	\
+	int size;	\
+	_type *data;	\
+	_type preallocated[_nr];	\
+}
+
+#define DEFINE_MIN_HEAP(_type, _name) MIN_HEAP_PREALLOCATED(_type, _name, 0)
+
+typedef DEFINE_MIN_HEAP(char, min_heap_char) min_heap_char;
+
+#define __minheap_cast(_heap)		(typeof((_heap)->data[0]) *)
+#define __minheap_obj_size(_heap)	sizeof((_heap)->data[0])
 
 /**
  * struct min_heap_callbacks - Data/functions to customise the min_heap.
- * @elem_size: The nr of each element in bytes.
  * @less: Partial order function for this heap.
  * @swp: Swap elements function.
  */
 struct min_heap_callbacks {
-	int elem_size;
-	bool (*less)(const void *lhs, const void *rhs);
-	void (*swp)(void *lhs, void *rhs);
+	bool (*less)(const void *lhs, const void *rhs, void *args);
+	void (*swp)(void *lhs, void *rhs, void *args);
 };
 
+/* Initialize a min-heap. */
+static __always_inline
+void __min_heap_init(min_heap_char *heap, void *data, int size)
+{
+	heap->nr = 0;
+	heap->size = size;
+	if (data)
+		heap->data = data;
+	else
+		heap->data = heap->preallocated;
+}
+
+#define min_heap_init(_heap, _data, _size)	\
+	__min_heap_init((min_heap_char *)_heap, _data, _size)
+
+/* Get the minimum element from the heap. */
+static __always_inline
+void *__min_heap_peek(struct min_heap_char *heap)
+{
+	return heap->nr ? heap->data : NULL;
+}
+
+#define min_heap_peek(_heap)	\
+	(__minheap_cast(_heap) __min_heap_peek((min_heap_char *)_heap))
+
+/* Check if the heap is full. */
+static __always_inline
+bool __min_heap_full(min_heap_char *heap)
+{
+	return heap->nr == heap->size;
+}
+
+#define min_heap_full(_heap)	\
+	__min_heap_full((min_heap_char *)_heap)
+
 /* Sift the element at pos down the heap. */
 static __always_inline
-void min_heapify(struct min_heap *heap, int pos,
-		const struct min_heap_callbacks *func)
+void __min_heap_sift_down(min_heap_char *heap, int pos, size_t elem_size,
+		const struct min_heap_callbacks *func, void *args)
 {
 	void *left, *right;
 	void *data = heap->data;
-	void *root = data + pos * func->elem_size;
+	void *root = data + pos * elem_size;
 	int i = pos, j;
 
 	/* Find the sift-down path all the way to the leaves. */
 	for (;;) {
 		if (i * 2 + 2 >= heap->nr)
 			break;
-		left = data + (i * 2 + 1) * func->elem_size;
-		right = data + (i * 2 + 2) * func->elem_size;
-		i = func->less(left, right) ? i * 2 + 1 : i * 2 + 2;
+		left = data + (i * 2 + 1) * elem_size;
+		right = data + (i * 2 + 2) * elem_size;
+		i = func->less(left, right, args) ? i * 2 + 1 : i * 2 + 2;
 	}
 
 	/* Special case for the last leaf with no sibling. */
@@ -54,83 +97,140 @@ void min_heapify(struct min_heap *heap, int pos,
 		i = i * 2 + 1;
 
 	/* Backtrack to the correct location. */
-	while (i != pos && func->less(root, data + i * func->elem_size))
+	while (i != pos && func->less(root, data + i * elem_size, args))
 		i = (i - 1) / 2;
 
 	/* Shift the element into its correct place. */
 	j = i;
 	while (i != pos) {
 		i = (i - 1) / 2;
-		func->swp(data + i * func->elem_size, data + j * func->elem_size);
+		func->swp(data + i * elem_size, data + j * elem_size, args);
+	}
+}
+
+#define min_heap_sift_down(_heap, _pos, _func, _args)	\
+	__min_heap_sift_down((min_heap_char *)_heap, _pos, __minheap_obj_size(_heap), _func, _args)
+
+/* Sift up ith element from the heap, O(log2(nr)). */
+static __always_inline
+void __min_heap_sift_up(min_heap_char *heap, size_t elem_size, size_t idx,
+		const struct min_heap_callbacks *func, void *args)
+{
+	void *data = heap->data;
+	size_t parent;
+
+	while (idx) {
+		parent = (idx - 1) / 2;
+		if (func->less(data + parent * elem_size, data + idx * elem_size, args))
+			break;
+		func->swp(data + parent * elem_size, data + idx * elem_size, args);
+		idx = parent;
 	}
 }
 
+#define min_heap_sift_up(_heap, _idx, _func, _args)	\
+	__min_heap_sift_up((min_heap_char *)_heap, __minheap_obj_size(_heap), _idx, _func, _args)
+
 /* Floyd's approach to heapification that is O(nr). */
 static __always_inline
-void min_heapify_all(struct min_heap *heap,
-		const struct min_heap_callbacks *func)
+void __min_heapify_all(min_heap_char *heap, size_t elem_size,
+		const struct min_heap_callbacks *func, void *args)
 {
 	int i;
 
 	for (i = heap->nr / 2 - 1; i >= 0; i--)
-		min_heapify(heap, i, func);
+		__min_heap_sift_down(heap, i, elem_size, func, args);
 }
 
+#define min_heapify_all(_heap, _func, _args)	\
+	__min_heapify_all((min_heap_char *)_heap, __minheap_obj_size(_heap), _func, _args)
+
 /* Remove minimum element from the heap, O(log2(nr)). */
 static __always_inline
-void min_heap_pop(struct min_heap *heap,
-		const struct min_heap_callbacks *func)
+bool __min_heap_pop(min_heap_char *heap, size_t elem_size,
+		const struct min_heap_callbacks *func, void *args)
 {
 	void *data = heap->data;
 
 	if (WARN_ONCE(heap->nr <= 0, "Popping an empty heap"))
-		return;
+		return false;
 
 	/* Place last element at the root (position 0) and then sift down. */
 	heap->nr--;
-	memcpy(data, data + (heap->nr * func->elem_size), func->elem_size);
-	min_heapify(heap, 0, func);
+	memcpy(data, data + (heap->nr * elem_size), elem_size);
+	__min_heap_sift_down(heap, 0, elem_size, func, args);
+
+	return true;
 }
 
+#define min_heap_pop(_heap, _func, _args)	\
+	__min_heap_pop((min_heap_char *)_heap, __minheap_obj_size(_heap), _func, _args)
+
 /*
  * Remove the minimum element and then push the given element. The
  * implementation performs 1 sift (O(log2(nr))) and is therefore more
  * efficient than a pop followed by a push that does 2.
  */
 static __always_inline
-void min_heap_pop_push(struct min_heap *heap,
-		const void *element,
-		const struct min_heap_callbacks *func)
+void __min_heap_pop_push(min_heap_char *heap,
+		const void *element, size_t elem_size,
+		const struct min_heap_callbacks *func,
+		void *args)
 {
-	memcpy(heap->data, element, func->elem_size);
-	min_heapify(heap, 0, func);
+	memcpy(heap->data, element, elem_size);
+	__min_heap_sift_down(heap, 0, elem_size, func, args);
 }
 
+#define min_heap_pop_push(_heap, _element, _func, _args)	\
+	__min_heap_pop_push((min_heap_char *)_heap, _element, __minheap_obj_size(_heap), _func, _args)
+
 /* Push an element on to the heap, O(log2(nr)). */
 static __always_inline
-void min_heap_push(struct min_heap *heap, const void *element,
-		const struct min_heap_callbacks *func)
+bool __min_heap_push(min_heap_char *heap, const void *element, size_t elem_size,
+		const struct min_heap_callbacks *func, void *args)
 {
 	void *data = heap->data;
-	void *child, *parent;
 	int pos;
 
 	if (WARN_ONCE(heap->nr >= heap->size, "Pushing on a full heap"))
-		return;
+		return false;
 
 	/* Place at the end of data. */
 	pos = heap->nr;
-	memcpy(data + (pos * func->elem_size), element, func->elem_size);
+	memcpy(data + (pos * elem_size), element, elem_size);
 	heap->nr++;
 
 	/* Sift child at pos up. */
-	for (; pos > 0; pos = (pos - 1) / 2) {
-		child = data + (pos * func->elem_size);
-		parent = data + ((pos - 1) / 2) * func->elem_size;
-		if (func->less(parent, child))
-			break;
-		func->swp(parent, child);
-	}
+	__min_heap_sift_up(heap, elem_size, pos, func, args);
+
+	return true;
 }
 
+#define min_heap_push(_heap, _element, _func, _args)	\
+	__min_heap_push((min_heap_char *)_heap, _element, __minheap_obj_size(_heap), _func, _args)
+
+/* Remove ith element from the heap, O(log2(nr)). */
+static __always_inline
+bool __min_heap_del(min_heap_char *heap, size_t elem_size, size_t idx,
+		const struct min_heap_callbacks *func, void *args)
+{
+	void *data = heap->data;
+
+	if (WARN_ONCE(heap->nr <= 0, "Popping an empty heap"))
+		return false;
+
+	/* Place last element at the root (position 0) and then sift down. */
+	heap->nr--;
+	if (idx == heap->nr)
+		return true;
+	func->swp(data + (idx * elem_size), data + (heap->nr * elem_size), args);
+	__min_heap_sift_up(heap, elem_size, idx, func, args);
+	__min_heap_sift_down(heap, idx, elem_size, func, args);
+
+	return true;
+}
+
+#define min_heap_del(_heap, _idx, _func, _args)	\
+	__min_heap_del((min_heap_char *)_heap, __minheap_obj_size(_heap), _idx, _func, _args)
+
 #endif /* _LINUX_MIN_HEAP_H */
diff --git a/include/linux/msi.h b/include/linux/msi.h
index dc27cf3903d5..26588da88bdd 100644
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -19,7 +19,7 @@
  */
 
 #include <linux/irqdomain_defs.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/msi_api.h>
 #include <linux/xarray.h>
 #include <linux/mutex.h>
diff --git a/include/linux/node.h b/include/linux/node.h
index dfc004e4bee7..9a881c2208b3 100644
--- a/include/linux/node.h
+++ b/include/linux/node.h
@@ -16,7 +16,6 @@
 #define _LINUX_NODE_H_
 
 #include <linux/device.h>
-#include <linux/cpumask.h>
 #include <linux/list.h>
 
 /**
diff --git a/include/linux/nvme-fc-driver.h b/include/linux/nvme-fc-driver.h
index 89ea1ebd975a..9f6acadfe0c8 100644
--- a/include/linux/nvme-fc-driver.h
+++ b/include/linux/nvme-fc-driver.h
@@ -620,7 +620,7 @@ enum {
  *
  * Structure used between LLDD and nvmet-fc layer to represent the exchange
  * context for a FC-NVME FCP I/O operation (e.g. a nvme sqe, the sqe-related
- * memory transfers, and its assocated cqe transfer).
+ * memory transfers, and its associated cqe transfer).
  *
  * The structure is allocated by the LLDD whenever a FCP CMD IU is received
  * from the FC link. The address of the structure is passed to the nvmet-fc
diff --git a/include/linux/panic.h b/include/linux/panic.h
index 6717b15e798c..3130e0b5116b 100644
--- a/include/linux/panic.h
+++ b/include/linux/panic.h
@@ -77,9 +77,10 @@ static inline void set_arch_panic_timeout(int timeout, int arch_default_timeout)
 #define TAINT_FLAGS_MAX			((1UL << TAINT_FLAGS_COUNT) - 1)
 
 struct taint_flag {
-	char c_true;	/* character printed when tainted */
-	char c_false;	/* character printed when not tainted */
-	bool module;	/* also show as a per-module taint flag */
+	char c_true;		/* character printed when tainted */
+	char c_false;		/* character printed when not tainted */
+	bool module;		/* also show as a per-module taint flag */
+	const char *desc;	/* verbose description of the set taint flag */
 };
 
 extern const struct taint_flag taint_flags[TAINT_FLAGS_COUNT];
@@ -90,6 +91,7 @@ enum lockdep_ok {
 };
 
 extern const char *print_tainted(void);
+extern const char *print_tainted_verbose(void);
 extern void add_taint(unsigned flag, enum lockdep_ok);
 extern int test_taint(unsigned flag);
 extern unsigned long get_taint(void);
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 03053de557cf..4b2047b78b67 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -6,7 +6,6 @@
 #include <linux/mmdebug.h>
 #include <linux/preempt.h>
 #include <linux/smp.h>
-#include <linux/cpumask.h>
 #include <linux/pfn.h>
 #include <linux/init.h>
 #include <linux/cleanup.h>
diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h
index 015751b64746..858c8e7851fb 100644
--- a/include/linux/pm_domain.h
+++ b/include/linux/pm_domain.h
@@ -16,7 +16,7 @@
 #include <linux/of.h>
 #include <linux/notifier.h>
 #include <linux/spinlock.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/time64.h>
 
 /*
diff --git a/include/linux/poison.h b/include/linux/poison.h
index 9c1a035af97c..331a9a996fa8 100644
--- a/include/linux/poison.h
+++ b/include/linux/poison.h
@@ -49,12 +49,6 @@
 /********** arch/$ARCH/mm/init.c **********/
 #define POISON_FREE_INITMEM	0xcc
 
-/********** arch/ia64/hp/common/sba_iommu.c **********/
-/*
- * arch/ia64/hp/common/sba_iommu.c uses a 16-byte poison string with a
- * value of "SBAIOMMU POISON\0" for spill-over poisoning.
- */
-
 /********** fs/jbd/journal.c **********/
 #define JBD_POISON_FREE		0x5b
 #define JBD2_POISON_FREE	0x5c
diff --git a/include/linux/profile.h b/include/linux/profile.h
index 04ae5ebcb637..2fb487f61d12 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -4,7 +4,6 @@
 
 #include <linux/kernel.h>
 #include <linux/init.h>
-#include <linux/cpumask.h>
 #include <linux/cache.h>
 
 #include <asm/errno.h>
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index be450a3477be..13f6f00aecf9 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -29,7 +29,6 @@
 #include <linux/lockdep.h>
 #include <linux/cleanup.h>
 #include <asm/processor.h>
-#include <linux/cpumask.h>
 #include <linux/context_tracking_irq.h>
 
 #define ULONG_CMP_GE(a, b)	(ULONG_MAX / 2 >= (a) - (b))
diff --git a/include/linux/sched.h b/include/linux/sched.h
index a898d9bde8ca..f8d150343d42 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -13,7 +13,7 @@
 #include <asm/processor.h>
 #include <linux/thread_info.h>
 #include <linux/preempt.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 
 #include <linux/cache.h>
 #include <linux/irqflags_types.h>
@@ -1618,7 +1618,7 @@ static inline char task_index_to_char(unsigned int state)
 {
 	static const char state_char[] = "RSDTtXZPI";
 
-	BUILD_BUG_ON(1 + ilog2(TASK_REPORT_MAX) != sizeof(state_char) - 1);
+	BUILD_BUG_ON(TASK_REPORT_MAX * 2 != 1 << (sizeof(state_char) - 1));
 
 	return state_char[state];
 }
@@ -1792,7 +1792,8 @@ static inline void do_set_cpus_allowed(struct task_struct *p, const struct cpuma
 }
 static inline int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask)
 {
-	if (!cpumask_test_cpu(0, new_mask))
+	/* Opencoded cpumask_test_cpu(0, new_mask) to avoid dependency on cpumask.h */
+	if ((*cpumask_bits(new_mask) & 1) == 0)
 		return -EINVAL;
 	return 0;
 }
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 8bd4fda6e027..2fb266ea69fa 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -7,7 +7,6 @@
 #include <linux/string_helpers.h>
 #include <linux/bug.h>
 #include <linux/mutex.h>
-#include <linux/cpumask.h>
 #include <linux/nodemask.h>
 #include <linux/fs.h>
 #include <linux/cred.h>
diff --git a/include/linux/soc/apple/rtkit.h b/include/linux/soc/apple/rtkit.h
index 8c9ca857ccf6..c06d17599ae7 100644
--- a/include/linux/soc/apple/rtkit.h
+++ b/include/linux/soc/apple/rtkit.h
@@ -69,7 +69,7 @@ struct apple_rtkit;
  * Initializes the internal state required to handle RTKit. This
  * should usually be called within _probe.
  *
- * @dev:         Pointer to the device node this coprocessor is assocated with
+ * @dev:         Pointer to the device node this coprocessor is associated with
  * @cookie:      opaque cookie passed to all functions defined in rtkit_ops
  * @mbox_name:   mailbox name used to communicate with the co-processor
  * @mbox_idx:    mailbox index to be used if mbox_name is NULL
@@ -83,7 +83,7 @@ struct apple_rtkit *devm_apple_rtkit_init(struct device *dev, void *cookie,
  * Non-devm version of devm_apple_rtkit_init. Must be freed with
  * apple_rtkit_free.
  *
- * @dev:         Pointer to the device node this coprocessor is assocated with
+ * @dev:         Pointer to the device node this coprocessor is associated with
  * @cookie:      opaque cookie passed to all functions defined in rtkit_ops
  * @mbox_name:   mailbox name used to communicate with the co-processor
  * @mbox_idx:    mailbox index to be used if mbox_name is NULL
diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h
index ea7a74ea7389..3132262a404d 100644
--- a/include/linux/stop_machine.h
+++ b/include/linux/stop_machine.h
@@ -3,7 +3,7 @@
 #define _LINUX_STOP_MACHINE
 
 #include <linux/cpu.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/smp.h>
 #include <linux/list.h>
 
diff --git a/include/linux/torture.h b/include/linux/torture.h
index 1541454da03e..c2e979f82f8d 100644
--- a/include/linux/torture.h
+++ b/include/linux/torture.h
@@ -14,7 +14,7 @@
 #include <linux/cache.h>
 #include <linux/spinlock.h>
 #include <linux/threads.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/seqlock.h>
 #include <linux/lockdep.h>
 #include <linux/completion.h>
diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index 689b6d71590e..6be396bb4297 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -16,7 +16,6 @@
 #include <linux/srcu.h>
 #include <linux/errno.h>
 #include <linux/types.h>
-#include <linux/cpumask.h>
 #include <linux/rcupdate.h>
 #include <linux/tracepoint-defs.h>
 #include <linux/static_call.h>
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index d9968bfc8eac..4eb8f9563136 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -12,7 +12,7 @@
 #include <linux/lockdep.h>
 #include <linux/threads.h>
 #include <linux/atomic.h>
-#include <linux/cpumask.h>
+#include <linux/cpumask_types.h>
 #include <linux/rcupdate.h>
 #include <linux/workqueue_types.h>