Received-SPF: pass (google.com: domain of linux-kernel+bounces-199823-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: John Ogness <john.ogness@linutronix.de>
To: Petr Mladek <pmladek@suse.com>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	linux-kernel@vger.kernel.org,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: [PATCH printk v2 04/18] printk: nbcon: Introduce printing kthreads
Date: Tue,  4 Jun 2024 01:30:39 +0206
Message-Id: <20240603232453.33992-5-john.ogness@linutronix.de>
In-Reply-To: <20240603232453.33992-1-john.ogness@linutronix.de>
References: <20240603232453.33992-1-john.ogness@linutronix.de>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Thomas Gleixner <tglx@linutronix.de>

Provide the main implementation for running a printer kthread
per nbcon console that is takeover/handover aware.

Co-developed-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Thomas Gleixner (Intel) <tglx@linutronix.de>
---
 include/linux/console.h  |  26 ++++++
 kernel/printk/internal.h |  26 ++++++
 kernel/printk/nbcon.c    | 196 +++++++++++++++++++++++++++++++++++++--
 kernel/printk/printk.c   |  34 +++++++
 4 files changed, 275 insertions(+), 7 deletions(-)

diff --git a/include/linux/console.h b/include/linux/console.h
index 4aaf053840ee..4de42ec1527c 100644
--- a/include/linux/console.h
+++ b/include/linux/console.h
@@ -17,6 +17,7 @@
 #include <linux/atomic.h>
 #include <linux/bits.h>
 #include <linux/rculist.h>
+#include <linux/rcuwait.h>
 #include <linux/types.h>
 #include <linux/vesa.h>
 
@@ -324,6 +325,8 @@ struct nbcon_write_context {
  * @nbcon_seq:		Sequence number of the next record for nbcon to print
  * @nbcon_device_ctxt:	Context available for non-printing operations
  * @pbufs:		Pointer to nbcon private buffer
+ * @kthread:		Printer kthread for this console
+ * @rcuwait:		RCU-safe wait object for @kthread waking
  */
 struct console {
 	char			name[16];
@@ -373,6 +376,27 @@ struct console {
 	 */
 	void (*write_atomic)(struct console *con, struct nbcon_write_context *wctxt);
 
+	/**
+	 * @write_thread:
+	 *
+	 * NBCON callback to write out text in task context.
+	 *
+	 * This callback is called after device_lock() and with the nbcon
+	 * console acquired. Any necessary driver synchronization should have
+	 * been performed by the device_lock() callback.
+	 *
+	 * This callback is always called from task context but with migration
+	 * disabled.
+	 *
+	 * The same criteria for console ownership verification and unsafe
+	 * sections applies as with write_atomic(). The difference between
+	 * this callback and write_atomic() is that this callback is used
+	 * during normal operation and is always called from task context.
+	 * This allows drivers to operate in their own locking context for
+	 * synchronizing output to the hardware.
+	 */
+	void (*write_thread)(struct console *con, struct nbcon_write_context *wctxt);
+
 	/**
 	 * @device_lock:
 	 *
@@ -420,6 +444,8 @@ struct console {
 	atomic_long_t		__private nbcon_seq;
 	struct nbcon_context	__private nbcon_device_ctxt;
 	struct printk_buffers	*pbufs;
+	struct task_struct	*kthread;
+	struct rcuwait		rcuwait;
 };
 
 #ifdef CONFIG_LOCKDEP
diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h
index 0439cf2fdc22..38680c6b2b39 100644
--- a/kernel/printk/internal.h
+++ b/kernel/printk/internal.h
@@ -92,6 +92,7 @@ enum nbcon_prio nbcon_get_default_prio(void);
 void nbcon_atomic_flush_pending(void);
 bool nbcon_legacy_emit_next_record(struct console *con, bool *handover,
 				   int cookie);
+void nbcon_kthread_create(struct console *con);
 
 /*
  * Check if the given console is currently capable and allowed to print
@@ -110,6 +111,8 @@ static inline bool console_is_usable(struct console *con, short flags)
 	if (flags & CON_NBCON) {
 		if (!con->write_atomic)
 			return false;
+		if (!con->write_thread)
+			return false;
 	} else {
 		if (!con->write)
 			return false;
@@ -126,12 +129,35 @@ static inline bool console_is_usable(struct console *con, short flags)
 	return true;
 }
 
+/**
+ * nbcon_kthread_wake - Wake up a printk thread
+ * @con:        Console to operate on
+ */
+static inline void nbcon_kthread_wake(struct console *con)
+{
+	/*
+	 * Guarantee any new records can be seen by tasks preparing to wait
+	 * before this context checks if the rcuwait is empty.
+	 *
+	 * The full memory barrier in rcuwait_wake_up() pairs with the full
+	 * memory barrier within set_current_state() of
+	 * ___rcuwait_wait_event(), which is called after prepare_to_rcuwait()
+	 * adds the waiter but before it has checked the wait condition.
+	 *
+	 * This pairs with nbcon_kthread_func:A.
+	 */
+	rcuwait_wake_up(&con->rcuwait); /* LMM(nbcon_kthread_wake:A) */
+}
+
 #else
 
 #define PRINTK_PREFIX_MAX	0
 #define PRINTK_MESSAGE_MAX	0
 #define PRINTKRB_RECORD_MAX	0
 
+static inline void nbcon_kthread_wake(struct console *con) { }
+static inline void nbcon_kthread_create(struct console *con) { }
+
 /*
  * In !PRINTK builds we still export console_sem
  * semaphore and some of console functions (console_unlock()/etc.), so
diff --git a/kernel/printk/nbcon.c b/kernel/printk/nbcon.c
index 6e9e24aa0a7f..89b340ca303c 100644
--- a/kernel/printk/nbcon.c
+++ b/kernel/printk/nbcon.c
@@ -10,6 +10,7 @@
 #include <linux/export.h>
 #include <linux/init.h>
 #include <linux/irqflags.h>
+#include <linux/kthread.h>
 #include <linux/minmax.h>
 #include <linux/percpu.h>
 #include <linux/preempt.h>
@@ -837,6 +838,7 @@ EXPORT_SYMBOL_GPL(nbcon_exit_unsafe);
 /**
  * nbcon_emit_next_record - Emit a record in the acquired context
  * @wctxt:	The write context that will be handed to the write function
+ * @use_atomic:	True if the write_atomic() callback is to be used
  *
  * Return:	True if this context still owns the console. False if
  *		ownership was handed over or taken.
@@ -850,7 +852,7 @@ EXPORT_SYMBOL_GPL(nbcon_exit_unsafe);
  * When true is returned, @wctxt->ctxt.backlog indicates whether there are
  * still records pending in the ringbuffer,
  */
-static bool nbcon_emit_next_record(struct nbcon_write_context *wctxt)
+static bool nbcon_emit_next_record(struct nbcon_write_context *wctxt, bool use_atomic)
 {
 	struct nbcon_context *ctxt = &ACCESS_PRIVATE(wctxt, ctxt);
 	struct console *con = ctxt->console;
@@ -899,8 +901,14 @@ static bool nbcon_emit_next_record(struct nbcon_write_context *wctxt)
 	nbcon_state_read(con, &cur);
 	wctxt->unsafe_takeover = cur.unsafe_takeover;
 
-	if (con->write_atomic) {
+	if (use_atomic &&
+	    con->write_atomic) {
 		con->write_atomic(con, wctxt);
+
+	} else if (!use_atomic &&
+		   con->write_thread) {
+		con->write_thread(con, wctxt);
+
 	} else {
 		/*
 		 * This function should never be called for legacy consoles.
@@ -936,6 +944,120 @@ static bool nbcon_emit_next_record(struct nbcon_write_context *wctxt)
 	return nbcon_context_exit_unsafe(ctxt);
 }
 
+/**
+ * nbcon_kthread_should_wakeup - Check whether a printer thread should wakeup
+ * @con:	Console to operate on
+ * @ctxt:	The nbcon context from nbcon_context_try_acquire()
+ *
+ * Return:	True if the thread should shutdown or if the console is
+ *		allowed to print and a record is available. False otherwise.
+ *
+ * After the thread wakes up, it must first check if it should shutdown before
+ * attempting any printing.
+ */
+static bool nbcon_kthread_should_wakeup(struct console *con, struct nbcon_context *ctxt)
+{
+	bool ret = false;
+	short flags;
+	int cookie;
+
+	if (kthread_should_stop())
+		return true;
+
+	cookie = console_srcu_read_lock();
+
+	flags = console_srcu_read_flags(con);
+	if (console_is_usable(con, flags)) {
+		/* Bring the sequence in @ctxt up to date */
+		ctxt->seq = nbcon_seq_read(con);
+
+		ret = prb_read_valid(prb, ctxt->seq, NULL);
+	}
+
+	console_srcu_read_unlock(cookie);
+	return ret;
+}
+
+/**
+ * nbcon_kthread_func - The printer thread function
+ * @__console:	Console to operate on
+ *
+ * Return:	0
+ */
+static int nbcon_kthread_func(void *__console)
+{
+	struct console *con = __console;
+	struct nbcon_write_context wctxt = {
+		.ctxt.console	= con,
+		.ctxt.prio	= NBCON_PRIO_NORMAL,
+	};
+	struct nbcon_context *ctxt = &ACCESS_PRIVATE(&wctxt, ctxt);
+	short con_flags;
+	bool backlog;
+	int cookie;
+	int ret;
+
+wait_for_event:
+	/*
+	 * Guarantee this task is visible on the rcuwait before
+	 * checking the wake condition.
+	 *
+	 * The full memory barrier within set_current_state() of
+	 * ___rcuwait_wait_event() pairs with the full memory
+	 * barrier within rcuwait_has_sleeper().
+	 *
+	 * This pairs with rcuwait_has_sleeper:A and nbcon_kthread_wake:A.
+	 */
+	ret = rcuwait_wait_event(&con->rcuwait,
+				 nbcon_kthread_should_wakeup(con, ctxt),
+				 TASK_INTERRUPTIBLE); /* LMM(nbcon_kthread_func:A) */
+
+	if (kthread_should_stop())
+		return 0;
+
+	/* Wait was interrupted by a spurious signal, go back to sleep. */
+	if (ret)
+		goto wait_for_event;
+
+	do {
+		backlog = false;
+
+		cookie = console_srcu_read_lock();
+
+		con_flags = console_srcu_read_flags(con);
+
+		if (console_is_usable(con, con_flags)) {
+			unsigned long lock_flags;
+
+			con->device_lock(con, &lock_flags);
+
+			/*
+			 * Ensure this stays on the CPU to make handover and
+			 * takeover possible.
+			 */
+			cant_migrate();
+
+			if (nbcon_context_try_acquire(ctxt)) {
+				/*
+				 * If the emit fails, this context is no
+				 * longer the owner.
+				 */
+				if (nbcon_emit_next_record(&wctxt, false)) {
+					nbcon_context_release(ctxt);
+					backlog = ctxt->backlog;
+				}
+			}
+
+			con->device_unlock(con, lock_flags);
+		}
+
+		console_srcu_read_unlock(cookie);
+
+	} while (backlog);
+
+	goto wait_for_event;
+}
+
 /* Track the nbcon emergency nesting per CPU. */
 static DEFINE_PER_CPU(unsigned int, nbcon_pcpu_emergency_nesting);
 static unsigned int early_nbcon_pcpu_emergency_nesting __initdata;
@@ -1012,7 +1134,7 @@ static bool nbcon_atomic_emit_one(struct nbcon_write_context *wctxt)
 	 * The higher priority printing context takes over responsibility
 	 * to print the pending records.
 	 */
-	if (!nbcon_emit_next_record(wctxt))
+	if (!nbcon_emit_next_record(wctxt, true))
 		return false;
 
 	nbcon_context_release(ctxt);
@@ -1113,7 +1235,7 @@ static int __nbcon_atomic_flush_pending_con(struct console *con, u64 stop_seq,
 		 * handed over or taken over. In both cases the context is no
 		 * longer valid.
 		 */
-		if (!nbcon_emit_next_record(&wctxt))
+		if (!nbcon_emit_next_record(&wctxt, true))
 			return -EAGAIN;
 
 		if (!ctxt->backlog) {
@@ -1172,10 +1294,10 @@ static void nbcon_atomic_flush_pending_con(struct console *con, u64 stop_seq,
 
 	/*
 	 * If flushing was successful but more records are available, this
-	 * context must flush those remaining records because there is no
-	 * other context that will do it.
+	 * context must flush those remaining records if the printer thread
+	 * is not available do it.
 	 */
-	if (prb_read_valid(prb, nbcon_seq_read(con), NULL)) {
+	if (!con->kthread && prb_read_valid(prb, nbcon_seq_read(con), NULL)) {
 		stop_seq = prb_next_reserve_seq(prb);
 		goto again;
 	}
@@ -1332,6 +1454,63 @@ void nbcon_cpu_emergency_flush(void)
 	}
 }
 
+/*
+ * nbcon_kthread_stop - Stop a printer thread
+ * @con:	Console to operate on
+ */
+static void nbcon_kthread_stop(struct console *con)
+{
+	lockdep_assert_console_list_lock_held();
+
+	if (!con->kthread)
+		return;
+
+	kthread_stop(con->kthread);
+	con->kthread = NULL;
+}
+
+/**
+ * nbcon_kthread_create - Create a printer thread
+ * @con:	Console to operate on
+ *
+ * If it fails, let the console proceed. The atomic part might
+ * be usable and useful.
+ */
+void nbcon_kthread_create(struct console *con)
+{
+	struct task_struct *kt;
+
+	lockdep_assert_console_list_lock_held();
+
+	if (!(con->flags & CON_NBCON) || !con->write_thread)
+		return;
+
+	if (con->kthread)
+		return;
+
+	/*
+	 * Printer threads cannot be started as long as any boot console is
+	 * registered because there is no way to synchronize the hardware
+	 * registers between boot console code and regular console code.
+	 */
+	if (have_boot_console)
+		return;
+
+	kt = kthread_run(nbcon_kthread_func, con, "pr/%s%d", con->name, con->index);
+	if (IS_ERR(kt)) {
+		con_printk(KERN_ERR, con, "failed to start printing thread\n");
+		return;
+	}
+
+	con->kthread = kt;
+
+	/*
+	 * It is important that console printing threads are scheduled
+	 * shortly after a printk call and with generous runtime budgets.
+	 */
+	sched_set_normal(con->kthread, -20);
+}
+
 /**
  * nbcon_alloc - Allocate buffers needed by the nbcon console
  * @con:	Console to allocate buffers for
@@ -1377,6 +1556,7 @@ void nbcon_init(struct console *con, u64 init_seq)
 	/* nbcon_alloc() must have been called and successful! */
 	BUG_ON(!con->pbufs);
 
+	rcuwait_init(&con->rcuwait);
 	nbcon_seq_force(con, init_seq);
 	nbcon_state_set(con, &state);
 }
@@ -1389,6 +1569,7 @@ void nbcon_free(struct console *con)
 {
 	struct nbcon_state state = { };
 
+	nbcon_kthread_stop(con);
 	nbcon_state_set(con, &state);
 
 	/* Boot consoles share global printk buffers. */
@@ -1458,6 +1639,7 @@ void nbcon_device_release(struct console *con)
 	 */
 	cookie = console_srcu_read_lock();
 	if (console_is_usable(con, console_srcu_read_flags(con)) &&
+	    !con->kthread &&
 	    prb_read_valid(prb, nbcon_seq_read(con), NULL)) {
 		__nbcon_atomic_flush_pending_con(con, prb_next_reserve_seq(prb), false);
 	}
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 5dcc05e1aa56..137bd9a721c4 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -2685,6 +2685,8 @@ void suspend_console(void)
 void resume_console(void)
 {
 	struct console *con;
+	short flags;
+	int cookie;
 
 	if (!console_suspend_enabled)
 		return;
@@ -2701,6 +2703,14 @@ void resume_console(void)
 	 */
 	synchronize_srcu(&console_srcu);
 
+	cookie = console_srcu_read_lock();
+	for_each_console_srcu(con) {
+		flags = console_srcu_read_flags(con);
+		if (flags & CON_NBCON)
+			nbcon_kthread_wake(con);
+	}
+	console_srcu_read_unlock(cookie);
+
 	pr_flush(1000, true);
 }
 
@@ -3021,6 +3031,13 @@ static bool console_flush_all(bool do_cond_resched, u64 *next_seq, bool *handove
 			u64 printk_seq;
 			bool progress;
 
+			/*
+			 * console_flush_all() is only for legacy consoles,
+			 * unless the nbcon console has no kthread printer.
+			 */
+			if ((flags & CON_NBCON) && con->kthread)
+				continue;
+
 			if (!console_is_usable(con, flags))
 				continue;
 			any_usable = true;
@@ -3326,9 +3343,26 @@ EXPORT_SYMBOL(console_stop);
 
 void console_start(struct console *console)
 {
+	short flags;
+	int cookie;
+
 	console_list_lock();
 	console_srcu_write_flags(console, console->flags | CON_ENABLED);
 	console_list_unlock();
+
+	/*
+	 * Ensure that all SRCU list walks have completed. The related
+	 * printing context must be able to see it is enabled so that
+	 * it is guaranteed to wake up and resume printing.
+	 */
+	synchronize_srcu(&console_srcu);
+
+	cookie = console_srcu_read_lock();
+	flags = console_srcu_read_flags(console);
+	if (flags & CON_NBCON)
+		nbcon_kthread_wake(console);
+	console_srcu_read_unlock(cookie);
+
 	__pr_flush(console, 1000, true);
 }
 EXPORT_SYMBOL(console_start);
-- 
2.39.2