Merge: Scheduler updates for 9.7

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/6398

JIRA: https://issues.redhat.com/browse/RHEL-78821

Proactive fixes and minor updates for scheduler related
code. This includes needed commits up to v6.14-rc1. There
are not as many since there are a few features upstream
which we are not taking into rhel9 at this point.

Signed-off-by: Phil Auld <[email protected]>

Approved-by: Waiman Long <[email protected]>
Approved-by: Herton R. Krzesinski <[email protected]>
Approved-by: Tony Camuso <[email protected]>
Approved-by: Juri Lelli <[email protected]>
Approved-by: Rafael Aquini <[email protected]>
Approved-by: CKI KWF Bot <[email protected]>

Merged-by: Augusto Caringi <[email protected]>

Author: caringi

Documentation/admin-guide/kernel-parameters.txt

@@ -2321,7 +2321,9 @@
 			specified in the flag list (default: domain):
 
 			nohz
-			  Disable the tick when a single task runs.
+			  Disable the tick when a single task runs as well as
+			  disabling other kernel noises like having RCU callbacks
+			  offloaded. This is equivalent to the nohz_full parameter.
 
 			  A residual 1Hz tick is offloaded to workqueues, which you
 			  need to affine to housekeeping through the global

Documentation/scheduler/sched-bwc.rst

@@ -74,7 +74,6 @@ Quota, period and burst are managed within the cpu subsystem via cgroupfs.
    to cgroup v1. For cgroup v2, see
    :ref:`Documentation/admin-guide/cgroup-v2.rst <cgroup-v2-cpu>`.
 
-- cpu.cfs_quota_us: the total available run-time within a period (in
 - cpu.cfs_quota_us: run-time replenished within a period (in microseconds)
 - cpu.cfs_period_us: the length of a period (in microseconds)
 - cpu.stat: exports throttling statistics [explained further below]
@@ -135,7 +134,7 @@ cpu.stat:
   of the group have been throttled.
 - nr_bursts: Number of periods burst occurs.
 - burst_time: Cumulative wall-time (in nanoseconds) that any CPUs has used
-  above quota in respective periods
+  above quota in respective periods.
 
 This interface is read-only.
 
@@ -238,7 +237,7 @@ Examples
    additionally, in case accumulation has been done.
 
    With 50ms period, 20ms quota will be equivalent to 40% of 1 CPU.
-   And 10ms burst will be equivalent to 20% of 1 CPU.
+   And 10ms burst will be equivalent to 20% of 1 CPU::
 
 	# echo 20000 > cpu.cfs_quota_us /* quota = 20ms */
 	# echo 50000 > cpu.cfs_period_us /* period = 50ms */

drivers/base/topology.c

@@ -27,19 +27,35 @@ static ssize_t name##_read(struct file *file, struct kobject *kobj,		\
 			   loff_t off, size_t count)				\
 {										\
 	struct device *dev = kobj_to_dev(kobj);                                 \
+	cpumask_var_t mask;							\
+	ssize_t n;								\
 										\
-	return cpumap_print_bitmask_to_buf(buf, topology_##mask(dev->id),	\
-					   off, count);                         \
+	if (!alloc_cpumask_var(&mask, GFP_KERNEL))				\
+		return -ENOMEM;							\
+										\
+	cpumask_copy(mask, topology_##mask(dev->id));				\
+	n = cpumap_print_bitmask_to_buf(buf, mask, off, count);			\
+	free_cpumask_var(mask);							\
+										\
+	return n;								\
 }										\
 										\
 static ssize_t name##_list_read(struct file *file, struct kobject *kobj,	\
 				struct bin_attribute *attr, char *buf,		\
 				loff_t off, size_t count)			\
 {										\
 	struct device *dev = kobj_to_dev(kobj);					\
+	cpumask_var_t mask;							\
+	ssize_t n;								\
+										\
+	if (!alloc_cpumask_var(&mask, GFP_KERNEL))				\
+		return -ENOMEM;							\
+										\
+	cpumask_copy(mask, topology_##mask(dev->id));				\
+	n = cpumap_print_list_to_buf(buf, mask, off, count);			\
+	free_cpumask_var(mask);							\
 										\
-	return cpumap_print_list_to_buf(buf, topology_##mask(dev->id),		\
-					off, count);				\
+	return n;								\
 }
 
 define_id_show_func(physical_package_id, "%d");

include/linux/sched.h

@@ -631,6 +631,8 @@ struct sched_dl_entity {
 	 *
 	 * @dl_overrun tells if the task asked to be informed about runtime
 	 * overruns.
+	 *
+	 * @dl_server tells if this is a server entity.
 	 */
 	unsigned int			dl_throttled      : 1;
 	unsigned int			dl_yielded        : 1;

include/linux/sched/isolation.h

@@ -7,16 +7,21 @@
 #include <linux/tick.h>
 
 enum hk_type {
-	HK_TYPE_TIMER,
-	HK_TYPE_RCU,
-	HK_TYPE_MISC,
-	HK_TYPE_SCHED,
-	HK_TYPE_TICK,
 	HK_TYPE_DOMAIN,
-	HK_TYPE_WQ,
 	HK_TYPE_MANAGED_IRQ,
-	HK_TYPE_KTHREAD,
-	HK_TYPE_MAX
+	HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_MAX,
+
+	/*
+	 * The following housekeeping types are only set by the nohz_full
+	 * boot commandline option. So they can share the same value.
+	 */
+	HK_TYPE_TICK    = HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_TIMER   = HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_RCU     = HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_MISC    = HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_WQ      = HK_TYPE_KERNEL_NOISE,
+	HK_TYPE_KTHREAD = HK_TYPE_KERNEL_NOISE
 };
 
 #ifdef CONFIG_CPU_ISOLATION

include/linux/uaccess.h

@@ -372,6 +372,103 @@ copy_struct_from_user(void *dst, size_t ksize, const void __user *src,
 	return 0;
 }
 
+/**
+ * copy_struct_to_user: copy a struct to userspace
+ * @dst:   Destination address, in userspace. This buffer must be @ksize
+ *         bytes long.
+ * @usize: (Alleged) size of @dst struct.
+ * @src:   Source address, in kernel space.
+ * @ksize: Size of @src struct.
+ * @ignored_trailing: Set to %true if there was a non-zero byte in @src that
+ * userspace cannot see because they are using an smaller struct.
+ *
+ * Copies a struct from kernel space to userspace, in a way that guarantees
+ * backwards-compatibility for struct syscall arguments (as long as future
+ * struct extensions are made such that all new fields are *appended* to the
+ * old struct, and zeroed-out new fields have the same meaning as the old
+ * struct).
+ *
+ * Some syscalls may wish to make sure that userspace knows about everything in
+ * the struct, and if there is a non-zero value that userspce doesn't know
+ * about, they want to return an error (such as -EMSGSIZE) or have some other
+ * fallback (such as adding a "you're missing some information" flag). If
+ * @ignored_trailing is non-%NULL, it will be set to %true if there was a
+ * non-zero byte that could not be copied to userspace (ie. was past @usize).
+ *
+ * While unconditionally returning an error in this case is the simplest
+ * solution, for maximum backward compatibility you should try to only return
+ * -EMSGSIZE if the user explicitly requested the data that couldn't be copied.
+ * Note that structure sizes can change due to header changes and simple
+ * recompilations without code changes(!), so if you care about
+ * @ignored_trailing you probably want to make sure that any new field data is
+ * associated with a flag. Otherwise you might assume that a program knows
+ * about data it does not.
+ *
+ * @ksize is just sizeof(*src), and @usize should've been passed by userspace.
+ * The recommended usage is something like the following:
+ *
+ *   SYSCALL_DEFINE2(foobar, struct foo __user *, uarg, size_t, usize)
+ *   {
+ *      int err;
+ *      bool ignored_trailing;
+ *      struct foo karg = {};
+ *
+ *      if (usize > PAGE_SIZE)
+ *		return -E2BIG;
+ *      if (usize < FOO_SIZE_VER0)
+ *		return -EINVAL;
+ *
+ *      // ... modify karg somehow ...
+ *
+ *      err = copy_struct_to_user(uarg, usize, &karg, sizeof(karg),
+ *				  &ignored_trailing);
+ *      if (err)
+ *		return err;
+ *      if (ignored_trailing)
+ *		return -EMSGSIZE:
+ *
+ *      // ...
+ *   }
+ *
+ * There are three cases to consider:
+ *  * If @usize == @ksize, then it's copied verbatim.
+ *  * If @usize < @ksize, then the kernel is trying to pass userspace a newer
+ *    struct than it supports. Thus we only copy the interoperable portions
+ *    (@usize) and ignore the rest (but @ignored_trailing is set to %true if
+ *    any of the trailing (@ksize - @usize) bytes are non-zero).
+ *  * If @usize > @ksize, then the kernel is trying to pass userspace an older
+ *    struct than userspace supports. In order to make sure the
+ *    unknown-to-the-kernel fields don't contain garbage values, we zero the
+ *    trailing (@usize - @ksize) bytes.
+ *
+ * Returns (in all cases, some data may have been copied):
+ *  * -EFAULT: access to userspace failed.
+ */
+static __always_inline __must_check int
+copy_struct_to_user(void __user *dst, size_t usize, const void *src,
+		    size_t ksize, bool *ignored_trailing)
+{
+	size_t size = min(ksize, usize);
+	size_t rest = max(ksize, usize) - size;
+
+	/* Double check if ksize is larger than a known object size. */
+	if (WARN_ON_ONCE(ksize > __builtin_object_size(src, 1)))
+		return -E2BIG;
+
+	/* Deal with trailing bytes. */
+	if (usize > ksize) {
+		if (clear_user(dst + size, rest))
+			return -EFAULT;
+	}
+	if (ignored_trailing)
+		*ignored_trailing = ksize < usize &&
+			memchr_inv(src + size, 0, rest) != NULL;
+	/* Copy the interoperable parts of the struct. */
+	if (copy_to_user(dst, src, size))
+		return -EFAULT;
+	return 0;
+}
+
 bool copy_from_kernel_nofault_allowed(const void *unsafe_src, size_t size);
 
 long copy_from_kernel_nofault(void *dst, const void *src, size_t size);

include/linux/wait_bit.h

@@ -49,7 +49,6 @@ int wake_bit_function(struct wait_queue_entry *wq_entry, unsigned mode, int sync
 extern int bit_wait(struct wait_bit_key *key, int mode);
 extern int bit_wait_io(struct wait_bit_key *key, int mode);
 extern int bit_wait_timeout(struct wait_bit_key *key, int mode);
-extern int bit_wait_io_timeout(struct wait_bit_key *key, int mode);
 
 /**
  * wait_on_bit - wait for a bit to be cleared