sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling buggrafted

alb_break_llc() decides whether to break LLC preference during active
load balance. It does so by testing that every runnable fair task on the
source rq prefers its LLC:

	env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable

But the two counters cover different sets. nr_pref_llc_running is updated
in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued,
so it follows queued tasks. h_nr_runnable is updated in set_delayed()/
clear_delayed() and drops delay-dequeued tasks.

So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted
in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks,
alb_break_llc() returns false, and active balance is free to pull a task
off its preferred LLC. Active balance only moves runnable tasks, and this
is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB
skips the per-task test in can_migrate_task(). The runnable set is the one
we want.

Fix it on the counter side. A task should be counted in
nr_pref_llc_running exactly while it is both queued on its preferred LLC
(pref_llc_queued) and runnable (!sched_delayed). Define that membership
once in task_pref_llc_runnable(), and adjust the counter only through
pref_llc_running_inc()/pref_llc_running_dec() from the four sites that
change either input: account_llc_enqueue(), account_llc_dequeue(),
set_delayed() and clear_delayed(). Gating every update on the same
predicate keeps the delay, wake and dequeue paths from double-counting
or underflowing; see the comments at those sites for the ordering.

nr_llc_running and sd->llc_counts are not touched and stay on queued
semantics.

Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/
Reported-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Suggested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kayra Cizmeci <kayracizmeci@gmail.com>
Cc: <stable@kernel.org> # v7.2.x
Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com

1 files changed, 443 insertions, 0 deletions

diff --git a/mm/kmsan/hooks.c b/mm/kmsan/hooks.c
new file mode 100644
index 000000000..5f1b8053f
--- /dev/null
+++ b/mm/kmsan/hooks.c
@@ -0,0 +1,443 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KMSAN hooks for kernel subsystems.
+ *
+ * These functions handle creation of KMSAN metadata for memory allocations.
+ *
+ * Copyright (C) 2018-2022 Google LLC
+ * Author: Alexander Potapenko <glider@google.com>
+ *
+ */
+
+#include <linux/cacheflush.h>
+#include <linux/dma-direction.h>
+#include <linux/gfp.h>
+#include <linux/kmsan.h>
+#include <linux/mm.h>
+#include <linux/mm_types.h>
+#include <linux/scatterlist.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/usb.h>
+
+#include "../internal.h"
+#include "../vmalloc.h"
+#include "../slab.h"
+#include "kmsan.h"
+
+/*
+ * Instrumented functions shouldn't be called under
+ * kmsan_enter_runtime()/kmsan_leave_runtime(), because this will lead to
+ * skipping effects of functions like memset() inside instrumented code.
+ */
+
+void kmsan_task_create(struct task_struct *task)
+{
+	kmsan_enter_runtime();
+	kmsan_internal_task_create(task);
+	kmsan_leave_runtime();
+}
+
+void kmsan_task_exit(struct task_struct *task)
+{
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+
+	kmsan_disable_current();
+}
+
+void kmsan_slab_alloc(struct kmem_cache *s, void *object, gfp_t flags)
+{
+	if (unlikely(object == NULL))
+		return;
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+	/*
+	 * There's a ctor or this is an RCU cache - do nothing. The memory
+	 * status hasn't changed since last use.
+	 */
+	if (s->ctor || (s->flags & SLAB_TYPESAFE_BY_RCU))
+		return;
+
+	kmsan_enter_runtime();
+	if (flags & __GFP_ZERO)
+		kmsan_internal_unpoison_memory(object, s->object_size,
+					       KMSAN_POISON_CHECK);
+	else
+		kmsan_internal_poison_memory(object, s->object_size, flags,
+					     KMSAN_POISON_CHECK);
+	kmsan_leave_runtime();
+}
+
+void kmsan_slab_free(struct kmem_cache *s, void *object)
+{
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+
+	/* RCU slabs could be legally used after free within the RCU period */
+	if (unlikely(s->flags & SLAB_TYPESAFE_BY_RCU))
+		return;
+	/*
+	 * If there's a constructor, freed memory must remain in the same state
+	 * until the next allocation. We cannot save its state to detect
+	 * use-after-free bugs, instead we just keep it unpoisoned.
+	 */
+	if (s->ctor)
+		return;
+	kmsan_enter_runtime();
+	kmsan_internal_poison_memory(object, s->object_size,
+				     GFP_KERNEL & ~(__GFP_RECLAIM),
+				     KMSAN_POISON_CHECK | KMSAN_POISON_FREE);
+	kmsan_leave_runtime();
+}
+
+void kmsan_kmalloc_large(const void *ptr, size_t size, gfp_t flags)
+{
+	if (unlikely(ptr == NULL))
+		return;
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+	kmsan_enter_runtime();
+	if (flags & __GFP_ZERO)
+		kmsan_internal_unpoison_memory((void *)ptr, size,
+					       /*checked*/ true);
+	else
+		kmsan_internal_poison_memory((void *)ptr, size, flags,
+					     KMSAN_POISON_CHECK);
+	kmsan_leave_runtime();
+}
+
+void kmsan_kfree_large(const void *ptr)
+{
+	struct page *page;
+
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+	kmsan_enter_runtime();
+	page = virt_to_head_page((void *)ptr);
+	KMSAN_WARN_ON(ptr != page_address(page));
+	kmsan_internal_poison_memory((void *)ptr, page_size(page),
+				     GFP_KERNEL & ~(__GFP_RECLAIM),
+				     KMSAN_POISON_CHECK | KMSAN_POISON_FREE);
+	kmsan_leave_runtime();
+}
+
+static unsigned long vmalloc_shadow(unsigned long addr)
+{
+	return (unsigned long)kmsan_get_metadata((void *)addr,
+						 KMSAN_META_SHADOW);
+}
+
+static unsigned long vmalloc_origin(unsigned long addr)
+{
+	return (unsigned long)kmsan_get_metadata((void *)addr,
+						 KMSAN_META_ORIGIN);
+}
+
+void kmsan_vunmap_range_noflush(unsigned long start, unsigned long end)
+{
+	__vunmap_range_noflush(vmalloc_shadow(start), vmalloc_shadow(end));
+	__vunmap_range_noflush(vmalloc_origin(start), vmalloc_origin(end));
+	flush_cache_vmap(vmalloc_shadow(start), vmalloc_shadow(end));
+	flush_cache_vmap(vmalloc_origin(start), vmalloc_origin(end));
+}
+
+/*
+ * This function creates new shadow/origin pages for the physical pages mapped
+ * into the virtual memory. If those physical pages already had shadow/origin,
+ * those are ignored.
+ */
+int kmsan_ioremap_page_range(unsigned long start, unsigned long end,
+			     phys_addr_t phys_addr, pgprot_t prot,
+			     unsigned int page_shift)
+{
+	gfp_t gfp_mask = GFP_KERNEL | __GFP_ZERO;
+	struct page *shadow, *origin;
+	unsigned long off = 0;
+	int nr, err = 0, clean = 0, mapped;
+
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return 0;
+
+	nr = (end - start) / PAGE_SIZE;
+	kmsan_enter_runtime();
+	for (int i = 0; i < nr; i++, off += PAGE_SIZE, clean = i) {
+		shadow = alloc_pages(gfp_mask, 1);
+		origin = alloc_pages(gfp_mask, 1);
+		if (!shadow || !origin) {
+			err = -ENOMEM;
+			goto ret;
+		}
+		mapped = __vmap_pages_range_noflush(
+			vmalloc_shadow(start + off),
+			vmalloc_shadow(start + off + PAGE_SIZE), prot, &shadow,
+			PAGE_SHIFT);
+		if (mapped) {
+			err = mapped;
+			goto ret;
+		}
+		shadow = NULL;
+		mapped = __vmap_pages_range_noflush(
+			vmalloc_origin(start + off),
+			vmalloc_origin(start + off + PAGE_SIZE), prot, &origin,
+			PAGE_SHIFT);
+		if (mapped) {
+			__vunmap_range_noflush(
+				vmalloc_shadow(start + off),
+				vmalloc_shadow(start + off + PAGE_SIZE));
+			err = mapped;
+			goto ret;
+		}
+		origin = NULL;
+	}
+	/* Page mapping loop finished normally, nothing to clean up. */
+	clean = 0;
+
+ret:
+	if (clean > 0) {
+		/*
+		 * Something went wrong. Clean up shadow/origin pages allocated
+		 * on the last loop iteration, then delete mappings created
+		 * during the previous iterations.
+		 */
+		if (shadow)
+			__free_pages(shadow, 1);
+		if (origin)
+			__free_pages(origin, 1);
+		__vunmap_range_noflush(
+			vmalloc_shadow(start),
+			vmalloc_shadow(start + clean * PAGE_SIZE));
+		__vunmap_range_noflush(
+			vmalloc_origin(start),
+			vmalloc_origin(start + clean * PAGE_SIZE));
+	}
+	flush_cache_vmap(vmalloc_shadow(start), vmalloc_shadow(end));
+	flush_cache_vmap(vmalloc_origin(start), vmalloc_origin(end));
+	kmsan_leave_runtime();
+	return err;
+}
+
+void kmsan_iounmap_page_range(unsigned long start, unsigned long end)
+{
+	unsigned long v_shadow, v_origin;
+	struct page *shadow, *origin;
+	int nr;
+
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+
+	nr = (end - start) / PAGE_SIZE;
+	kmsan_enter_runtime();
+	v_shadow = (unsigned long)vmalloc_shadow(start);
+	v_origin = (unsigned long)vmalloc_origin(start);
+	for (int i = 0; i < nr;
+	     i++, v_shadow += PAGE_SIZE, v_origin += PAGE_SIZE) {
+		shadow = kmsan_vmalloc_to_page_or_null((void *)v_shadow);
+		origin = kmsan_vmalloc_to_page_or_null((void *)v_origin);
+		__vunmap_range_noflush(v_shadow, vmalloc_shadow(end));
+		__vunmap_range_noflush(v_origin, vmalloc_origin(end));
+		if (shadow)
+			__free_pages(shadow, 1);
+		if (origin)
+			__free_pages(origin, 1);
+	}
+	flush_cache_vmap(vmalloc_shadow(start), vmalloc_shadow(end));
+	flush_cache_vmap(vmalloc_origin(start), vmalloc_origin(end));
+	kmsan_leave_runtime();
+}
+
+void kmsan_copy_to_user(void __user *to, const void *from, size_t to_copy,
+			size_t left)
+{
+	unsigned long ua_flags;
+
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+	/*
+	 * At this point we've copied the memory already. It's hard to check it
+	 * before copying, as the size of actually copied buffer is unknown.
+	 */
+
+	/* copy_to_user() may copy zero bytes. No need to check. */
+	if (!to_copy)
+		return;
+	/* Or maybe copy_to_user() failed to copy anything. */
+	if (to_copy <= left)
+		return;
+
+	ua_flags = user_access_save();
+	if (!IS_ENABLED(CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE) ||
+	    (u64)to < TASK_SIZE) {
+		/* This is a user memory access, check it. */
+		kmsan_internal_check_memory((void *)from, to_copy - left, to,
+					    REASON_COPY_TO_USER);
+	} else {
+		/* Otherwise this is a kernel memory access. This happens when a
+		 * compat syscall passes an argument allocated on the kernel
+		 * stack to a real syscall.
+		 * Don't check anything, just copy the shadow of the copied
+		 * bytes.
+		 */
+		kmsan_enter_runtime();
+		kmsan_internal_memmove_metadata((void *)to, (void *)from,
+						to_copy - left);
+		kmsan_leave_runtime();
+	}
+	user_access_restore(ua_flags);
+}
+EXPORT_SYMBOL(kmsan_copy_to_user);
+
+void kmsan_memmove(void *to, const void *from, size_t size)
+{
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+
+	kmsan_enter_runtime();
+	kmsan_internal_memmove_metadata(to, (void *)from, size);
+	kmsan_leave_runtime();
+}
+EXPORT_SYMBOL(kmsan_memmove);
+
+/* Helper function to check an URB. */
+void kmsan_handle_urb(const struct urb *urb, bool is_out)
+{
+	if (!urb)
+		return;
+	if (is_out)
+		kmsan_internal_check_memory(urb->transfer_buffer,
+					    urb->transfer_buffer_length,
+					    /*user_addr*/ NULL,
+					    REASON_SUBMIT_URB);
+	else
+		kmsan_internal_unpoison_memory(urb->transfer_buffer,
+					       urb->transfer_buffer_length,
+					       /*checked*/ false);
+}
+EXPORT_SYMBOL_GPL(kmsan_handle_urb);
+
+static void kmsan_handle_dma_page(const void *addr, size_t size,
+				  enum dma_data_direction dir)
+{
+	switch (dir) {
+	case DMA_BIDIRECTIONAL:
+		kmsan_internal_check_memory((void *)addr, size,
+					    /*user_addr*/ NULL, REASON_ANY);
+		kmsan_internal_unpoison_memory((void *)addr, size,
+					       /*checked*/ false);
+		break;
+	case DMA_TO_DEVICE:
+		kmsan_internal_check_memory((void *)addr, size,
+					    /*user_addr*/ NULL, REASON_ANY);
+		break;
+	case DMA_FROM_DEVICE:
+		kmsan_internal_unpoison_memory((void *)addr, size,
+					       /*checked*/ false);
+		break;
+	case DMA_NONE:
+		break;
+	}
+}
+
+/* Helper function to handle DMA data transfers. */
+void kmsan_handle_dma(phys_addr_t phys, size_t size,
+		      enum dma_data_direction dir)
+{
+	u64 page_offset, to_go;
+	void *addr;
+
+	if (PhysHighMem(phys))
+		return;
+	addr = phys_to_virt(phys);
+	/*
+	 * The kernel may occasionally give us adjacent DMA pages not belonging
+	 * to the same allocation. Process them separately to avoid triggering
+	 * internal KMSAN checks.
+	 */
+	while (size > 0) {
+		page_offset = offset_in_page(addr);
+		to_go = min(PAGE_SIZE - page_offset, (u64)size);
+		kmsan_handle_dma_page((void *)addr, to_go, dir);
+		addr += to_go;
+		size -= to_go;
+	}
+}
+EXPORT_SYMBOL_GPL(kmsan_handle_dma);
+
+void kmsan_handle_dma_sg(struct scatterlist *sg, int nents,
+			 enum dma_data_direction dir)
+{
+	struct scatterlist *item;
+	int i;
+
+	for_each_sg(sg, item, nents, i)
+		kmsan_handle_dma(sg_phys(item), item->length, dir);
+}
+
+/* Functions from kmsan-checks.h follow. */
+
+/*
+ * To create an origin, kmsan_poison_memory() unwinds the stacks and stores it
+ * into the stack depot. This may cause deadlocks if done from within KMSAN
+ * runtime, therefore we bail out if kmsan_in_runtime().
+ */
+void kmsan_poison_memory(const void *address, size_t size, gfp_t flags)
+{
+	if (!kmsan_enabled || kmsan_in_runtime())
+		return;
+	kmsan_enter_runtime();
+	/* The users may want to poison/unpoison random memory. */
+	kmsan_internal_poison_memory((void *)address, size, flags,
+				     KMSAN_POISON_NOCHECK);
+	kmsan_leave_runtime();
+}
+EXPORT_SYMBOL(kmsan_poison_memory);
+
+/*
+ * Unlike kmsan_poison_memory(), this function can be used from within KMSAN
+ * runtime, because it does not trigger allocations or call instrumented code.
+ */
+void kmsan_unpoison_memory(const void *address, size_t size)
+{
+	unsigned long ua_flags;
+
+	if (!kmsan_enabled)
+		return;
+
+	ua_flags = user_access_save();
+	/* The users may want to poison/unpoison random memory. */
+	kmsan_internal_unpoison_memory((void *)address, size,
+				       KMSAN_POISON_NOCHECK);
+	user_access_restore(ua_flags);
+}
+EXPORT_SYMBOL(kmsan_unpoison_memory);
+
+/*
+ * Version of kmsan_unpoison_memory() called from IRQ entry functions.
+ */
+void kmsan_unpoison_entry_regs(const struct pt_regs *regs)
+{
+	kmsan_unpoison_memory((void *)regs, sizeof(*regs));
+}
+
+void kmsan_check_memory(const void *addr, size_t size)
+{
+	if (!kmsan_enabled)
+		return;
+	return kmsan_internal_check_memory((void *)addr, size,
+					   /*user_addr*/ NULL, REASON_ANY);
+}
+EXPORT_SYMBOL(kmsan_check_memory);
+
+void kmsan_enable_current(void)
+{
+	KMSAN_WARN_ON(current->kmsan_ctx.depth == 0);
+	current->kmsan_ctx.depth--;
+}
+EXPORT_SYMBOL(kmsan_enable_current);
+
+void kmsan_disable_current(void)
+{
+	current->kmsan_ctx.depth++;
+	KMSAN_WARN_ON(current->kmsan_ctx.depth == 0);
+}
+EXPORT_SYMBOL(kmsan_disable_current);