sched_ext: Maintain an online cid mask in the scheduler arenagrafted

Schedulers on the default cid mapping treat [0, nr_online_cids) as the
online set and restart on hotplug. Schedulers that install their own mapping
with scx_bpf_cid_override() have no way to learn which cids are online: the
count no longer identifies members and the CPU-form cpumask is unusable from
cid programs. This is an obvious hole in the cid API.

Add scx_bpf_online_cmask(), a kernel-maintained cmask in the scheduler's
arena, allocated alongside the per-CPU scratch masks and populated after the
cid mapping is finalized and before ops.init(), for child schedulers too.
The pointer stays valid through ops.exit() with no reference to take. It is
the arena offset as a void pointer, the same form struct_ops arena arguments
arrive in. The verifier types the void return as a scalar for the program's
arena cast.

The mask follows the SCX hotplug notifications: seeded from cpu_active_mask
and updated before ops.cid_online/offline() runs, so it lags cpu_online_mask
only inside a hotplug transition. Updates walk the scheduler list under the
lock that also serializes unlinking. Reads are live, not atomic snapshots.
Root initialization excludes hotplug.

v2: Reworded the getter kerneldoc (Andrea Righi).

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>

1 files changed, 148 insertions, 0 deletions

diff --git a/fs/fuse/req_timeout.c b/fs/fuse/req_timeout.c
new file mode 100644
index 000000000..6cc6fc491
--- /dev/null
+++ b/fs/fuse/req_timeout.c
@@ -0,0 +1,148 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "dev.h"
+#include "sysctl.h"
+#include "fuse_dev_i.h"
+#include "dev_uring_i.h"
+
+/* Frequency (in seconds) of request timeout checks, if opted into */
+#define FUSE_TIMEOUT_TIMER_FREQ 15
+
+/* Frequency (in jiffies) of request timeout checks, if opted into */
+static const unsigned long fuse_timeout_timer_freq =
+	secs_to_jiffies(FUSE_TIMEOUT_TIMER_FREQ);
+
+/*
+ * Default timeout (in seconds) for the server to reply to a request
+ * before the connection is aborted, if no timeout was specified on mount.
+ *
+ * Exported via sysctl
+ */
+unsigned int fuse_default_req_timeout;
+
+/*
+ * Max timeout (in seconds) for the server to reply to a request before
+ * the connection is aborted.
+ *
+ * Exported via sysctl
+ */
+unsigned int fuse_max_req_timeout;
+
+bool fuse_request_expired(struct fuse_chan *fch, struct list_head *list)
+{
+	struct fuse_req *req;
+
+	req = list_first_entry_or_null(list, struct fuse_req, list);
+	if (!req)
+		return false;
+	return time_is_before_jiffies(req->create_time + fch->timeout.req_timeout);
+}
+
+static bool fuse_fpq_processing_expired(struct fuse_chan *fch, struct list_head *processing)
+{
+	int i;
+
+	for (i = 0; i < FUSE_PQ_HASH_SIZE; i++)
+		if (fuse_request_expired(fch, &processing[i]))
+			return true;
+
+	return false;
+}
+
+/*
+ * Check if any requests aren't being completed by the time the request timeout
+ * elapses. To do so, we:
+ * - check the fiq pending list
+ * - check the bg queue
+ * - check the fpq io and processing lists
+ *
+ * To make this fast, we only check against the head request on each list since
+ * these are generally queued in order of creation time (eg newer requests get
+ * queued to the tail). We might miss a few edge cases (eg requests transitioning
+ * between lists, re-sent requests at the head of the pending list having a
+ * later creation time than other requests on that list, etc.) but that is fine
+ * since if the request never gets fulfilled, it will eventually be caught.
+ */
+static void fuse_check_timeout(struct work_struct *work)
+{
+	struct delayed_work *dwork = to_delayed_work(work);
+	struct fuse_chan *fch = container_of(dwork, struct fuse_chan, timeout.work);
+	struct fuse_iqueue *fiq = &fch->iq;
+	struct fuse_dev *fud;
+	struct fuse_pqueue *fpq;
+	bool expired = false;
+
+	if (!atomic_read(&fch->num_waiting))
+		goto out;
+
+	spin_lock(&fiq->lock);
+	expired = fuse_request_expired(fch, &fiq->pending);
+	spin_unlock(&fiq->lock);
+	if (expired)
+		goto chan_abort;
+
+	spin_lock(&fch->bg_lock);
+	expired = fuse_request_expired(fch, &fch->bg_queue);
+	spin_unlock(&fch->bg_lock);
+	if (expired)
+		goto chan_abort;
+
+	spin_lock(&fch->lock);
+	if (!fch->connected) {
+		spin_unlock(&fch->lock);
+		return;
+	}
+	list_for_each_entry(fud, &fch->devices, entry) {
+		fpq = &fud->pq;
+		spin_lock(&fpq->lock);
+		if (fuse_request_expired(fch, &fpq->io) ||
+		    fuse_fpq_processing_expired(fch, fpq->processing)) {
+			spin_unlock(&fpq->lock);
+			spin_unlock(&fch->lock);
+			goto chan_abort;
+		}
+
+		spin_unlock(&fpq->lock);
+	}
+	spin_unlock(&fch->lock);
+
+	if (fuse_uring_request_expired(fch))
+		goto chan_abort;
+
+out:
+	queue_delayed_work(system_percpu_wq, &fch->timeout.work,
+			   fuse_timeout_timer_freq);
+	return;
+
+chan_abort:
+	fuse_chan_abort(fch, false);
+}
+
+static void set_request_timeout(struct fuse_chan *fch, unsigned int timeout)
+{
+	fch->timeout.req_timeout = secs_to_jiffies(timeout);
+	INIT_DELAYED_WORK(&fch->timeout.work, fuse_check_timeout);
+	queue_delayed_work(system_percpu_wq, &fch->timeout.work,
+			   fuse_timeout_timer_freq);
+}
+
+void fuse_init_server_timeout(struct fuse_chan *fch, unsigned int timeout)
+{
+	if (!timeout && !fuse_max_req_timeout && !fuse_default_req_timeout)
+		return;
+
+	if (!timeout)
+		timeout = fuse_default_req_timeout;
+
+	if (fuse_max_req_timeout) {
+		if (timeout)
+			timeout = min(fuse_max_req_timeout, timeout);
+		else
+			timeout = fuse_max_req_timeout;
+	}
+
+	timeout = max(FUSE_TIMEOUT_TIMER_FREQ, timeout);
+
+	set_request_timeout(fch, timeout);
+}
+