Merge tag 'printk-for-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux

Pull printk updates from Petr Mladek:

 - Finally allow parallel writes and reads into/from the lockless
   ringbuffer. But it is not a complete solution. Readers are still
   serialized against each other. And nested writes are still prevented
   by printk_safe per-CPU buffers.

 - Use ttynull as the ultimate fallback for /dev/console.

 - Officially allow disabling console output by using console="" or
   console=null

 - A few code cleanups

* tag 'printk-for-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux:
  printk: remove logbuf_lock writer-protection of ringbuffer
  printk: inline log_output(),log_store() in vprintk_store()
  printk: remove obsolete dead assignment
  printk/console: Allow to disable console output by using console="" or console=null
  init/console: Use ttynull as a fallback when there is no console
  printk: ringbuffer: Reference text_data_ring directly in callees.

drivers/tty/Kconfig

@@ -401,20 +401,6 @@ config MIPS_EJTAG_FDC_KGDB_CHAN
 	help
 	  FDC channel number to use for KGDB.
 
-config NULL_TTY
-	tristate "NULL TTY driver"
-	help
-	  Say Y here if you want a NULL TTY which simply discards messages.
-
-	  This is useful to allow userspace applications which expect a console
-	  device to work without modifications even when no console is
-	  available or desired.
-
-	  In order to use this driver, you should redirect the console to this
-	  TTY, or boot the kernel with console=ttynull.
-
-	  If unsure, say N.
-
 config TRACE_ROUTER
 	tristate "Trace data router for MIPI P1149.7 cJTAG standard"
 	depends on TRACE_SINK

drivers/tty/Makefile

@@ -2,7 +2,7 @@
 obj-$(CONFIG_TTY)		+= tty_io.o n_tty.o tty_ioctl.o tty_ldisc.o \
 				   tty_buffer.o tty_port.o tty_mutex.o \
 				   tty_ldsem.o tty_baudrate.o tty_jobctrl.o \
-				   n_null.o
+				   n_null.o ttynull.o
 obj-$(CONFIG_LEGACY_PTYS)	+= pty.o
 obj-$(CONFIG_UNIX98_PTYS)	+= pty.o
 obj-$(CONFIG_AUDIT)		+= tty_audit.o
@@ -25,7 +25,6 @@ obj-$(CONFIG_ISI)		+= isicom.o
 obj-$(CONFIG_MOXA_INTELLIO)	+= moxa.o
 obj-$(CONFIG_MOXA_SMARTIO)	+= mxser.o
 obj-$(CONFIG_NOZOMI)		+= nozomi.o
-obj-$(CONFIG_NULL_TTY)	        += ttynull.o
 obj-$(CONFIG_ROCKETPORT)	+= rocket.o
 obj-$(CONFIG_SYNCLINK_GT)	+= synclink_gt.o
 obj-$(CONFIG_PPC_EPAPR_HV_BYTECHAN) += ehv_bytechan.o

drivers/tty/ttynull.c

@@ -2,6 +2,13 @@
 /*
  * Copyright (C) 2019 Axis Communications AB
  *
+ * The console is useful for userspace applications which expect a console
+ * device to work without modifications even when no console is available
+ * or desired.
+ *
+ * In order to use this driver, you should redirect the console to this
+ * TTY, or boot the kernel with console=ttynull.
+ *
  * Based on ttyprintk.c:
  *  Copyright (C) 2010 Samo Pogacnik
  */
@@ -59,6 +66,17 @@ static struct console ttynull_console = {
 	.device = ttynull_device,
 };
 
+void __init register_ttynull_console(void)
+{
+	if (!ttynull_driver)
+		return;
+
+	if (add_preferred_console(ttynull_console.name, 0, NULL))
+		return;
+
+	register_console(&ttynull_console);
+}
+
 static int __init ttynull_init(void)
 {
 	struct tty_driver *driver;

include/linux/console.h

@@ -186,9 +186,12 @@ extern int braille_register_console(struct console *, int index,
 extern int braille_unregister_console(struct console *);
 #ifdef CONFIG_TTY
 extern void console_sysfs_notify(void);
+extern void register_ttynull_console(void);
 #else
 static inline void console_sysfs_notify(void)
 { }
+static inline void register_ttynull_console(void)
+{ }
 #endif
 extern bool console_suspend_enabled;
 

init/main.c

@@ -1480,9 +1480,15 @@ void __init console_on_rootfs(void)
 	struct file *file = filp_open("/dev/console", O_RDWR, 0);
 
 	if (IS_ERR(file)) {
-		pr_err("Warning: unable to open an initial console.\n");
+		pr_err("Warning: unable to open an initial console. Fallback to ttynull.\n");
+		register_ttynull_console();
+
+		file = filp_open("/dev/console", O_RDWR, 0);
+		if (IS_ERR(file)) {
+			pr_err("Warning: Failed to add ttynull console. No stdin, stdout, and stderr for the init process!\n");
 			return;
 		}
+	}
 	init_dup(file);
 	init_dup(file);
 	init_dup(file);

kernel/printk/printk.c

@@ -491,52 +491,6 @@ static void truncate_msg(u16 *text_len, u16 *trunc_msg_len)
 		*trunc_msg_len = 0;
 }
 
-/* insert record into the buffer, discard old ones, update heads */
-static int log_store(u32 caller_id, int facility, int level,
-		     enum log_flags flags, u64 ts_nsec,
-		     const struct dev_printk_info *dev_info,
-		     const char *text, u16 text_len)
-{
-	struct prb_reserved_entry e;
-	struct printk_record r;
-	u16 trunc_msg_len = 0;
-
-	prb_rec_init_wr(&r, text_len);
-
-	if (!prb_reserve(&e, prb, &r)) {
-		/* truncate the message if it is too long for empty buffer */
-		truncate_msg(&text_len, &trunc_msg_len);
-		prb_rec_init_wr(&r, text_len + trunc_msg_len);
-		/* survive when the log buffer is too small for trunc_msg */
-		if (!prb_reserve(&e, prb, &r))
-			return 0;
-	}
-
-	/* fill message */
-	memcpy(&r.text_buf[0], text, text_len);
-	if (trunc_msg_len)
-		memcpy(&r.text_buf[text_len], trunc_msg, trunc_msg_len);
-	r.info->text_len = text_len + trunc_msg_len;
-	r.info->facility = facility;
-	r.info->level = level & 7;
-	r.info->flags = flags & 0x1f;
-	if (ts_nsec > 0)
-		r.info->ts_nsec = ts_nsec;
-	else
-		r.info->ts_nsec = local_clock();
-	r.info->caller_id = caller_id;
-	if (dev_info)
-		memcpy(&r.info->dev_info, dev_info, sizeof(r.info->dev_info));
-
-	/* A message without a trailing newline can be continued. */
-	if (!(flags & LOG_NEWLINE))
-		prb_commit(&e);
-	else
-		prb_final_commit(&e);
-
-	return (text_len + trunc_msg_len);
-}
-
 int dmesg_restrict = IS_ENABLED(CONFIG_SECURITY_DMESG_RESTRICT);
 
 static int syslog_action_restricted(int type)
@@ -741,7 +695,6 @@ static ssize_t devkmsg_write(struct kiocb *iocb, struct iov_iter *from)
 			if (LOG_FACILITY(u) != 0)
 				facility = LOG_FACILITY(u);
 			endp++;
-			len -= endp - line;
 			line = endp;
 		}
 	}
@@ -1172,7 +1125,7 @@ void __init setup_log_buf(int early)
 		 new_descs, ilog2(new_descs_count),
 		 new_infos);
 
-	logbuf_lock_irqsave(flags);
+	printk_safe_enter_irqsave(flags);
 
 	log_buf_len = new_log_buf_len;
 	log_buf = new_log_buf;
@@ -1189,7 +1142,7 @@ void __init setup_log_buf(int early)
 	 */
 	prb = &printk_rb_dynamic;
 
-	logbuf_unlock_irqrestore(flags);
+	printk_safe_exit_irqrestore(flags);
 
 	if (seq != prb_next_seq(&printk_rb_static)) {
 		pr_err("dropped %llu messages\n",
@@ -1907,83 +1860,177 @@ static inline u32 printk_caller_id(void)
 		0x80000000 + raw_smp_processor_id();
 }
 
-static size_t log_output(int facility, int level, enum log_flags lflags,
-			 const struct dev_printk_info *dev_info,
-			 char *text, size_t text_len)
+/**
+ * parse_prefix - Parse level and control flags.
+ *
+ * @text:     The terminated text message.
+ * @level:    A pointer to the current level value, will be updated.
+ * @lflags:   A pointer to the current log flags, will be updated.
+ *
+ * @level may be NULL if the caller is not interested in the parsed value.
+ * Otherwise the variable pointed to by @level must be set to
+ * LOGLEVEL_DEFAULT in order to be updated with the parsed value.
+ *
+ * @lflags may be NULL if the caller is not interested in the parsed value.
+ * Otherwise the variable pointed to by @lflags will be OR'd with the parsed
+ * value.
+ *
+ * Return: The length of the parsed level and control flags.
+ */
+static u16 parse_prefix(char *text, int *level, enum log_flags *lflags)
 {
-	const u32 caller_id = printk_caller_id();
+	u16 prefix_len = 0;
+	int kern_level;
 
-	if (lflags & LOG_CONT) {
-		struct prb_reserved_entry e;
-		struct printk_record r;
+	while (*text) {
+		kern_level = printk_get_level(text);
+		if (!kern_level)
+			break;
 
-		prb_rec_init_wr(&r, text_len);
-		if (prb_reserve_in_last(&e, prb, &r, caller_id, LOG_LINE_MAX)) {
-			memcpy(&r.text_buf[r.info->text_len], text, text_len);
-			r.info->text_len += text_len;
-			if (lflags & LOG_NEWLINE) {
-				r.info->flags |= LOG_NEWLINE;
-				prb_final_commit(&e);
-			} else {
-				prb_commit(&e);
+		switch (kern_level) {
+		case '0' ... '7':
+			if (level && *level == LOGLEVEL_DEFAULT)
+				*level = kern_level - '0';
+			break;
+		case 'c':	/* KERN_CONT */
+			if (lflags)
+				*lflags |= LOG_CONT;
 		}
-			return text_len;
+
+		prefix_len += 2;
+		text += 2;
 	}
+
+	return prefix_len;
+}
+
+static u16 printk_sprint(char *text, u16 size, int facility, enum log_flags *lflags,
+			 const char *fmt, va_list args)
+{
+	u16 text_len;
+
+	text_len = vscnprintf(text, size, fmt, args);
+
+	/* Mark and strip a trailing newline. */
+	if (text_len && text[text_len - 1] == '\n') {
+		text_len--;
+		*lflags |= LOG_NEWLINE;
 	}
 
-	/* Store it in the record log */
-	return log_store(caller_id, facility, level, lflags, 0,
-			 dev_info, text, text_len);
+	/* Strip log level and control flags. */
+	if (facility == 0) {
+		u16 prefix_len;
+
+		prefix_len = parse_prefix(text, NULL, NULL);
+		if (prefix_len) {
+			text_len -= prefix_len;
+			memmove(text, text + prefix_len, text_len);
+		}
+	}
+
+	return text_len;
 }
 
-/* Must be called under logbuf_lock. */
+__printf(4, 0)
 int vprintk_store(int facility, int level,
 		  const struct dev_printk_info *dev_info,
 		  const char *fmt, va_list args)
 {
-	static char textbuf[LOG_LINE_MAX];
-	char *text = textbuf;
-	size_t text_len;
+	const u32 caller_id = printk_caller_id();
+	struct prb_reserved_entry e;
 	enum log_flags lflags = 0;
+	struct printk_record r;
+	u16 trunc_msg_len = 0;
+	char prefix_buf[8];
+	u16 reserve_size;
+	va_list args2;
+	u16 text_len;
+	u64 ts_nsec;
 
 	/*
-	 * The printf needs to come first; we need the syslog
-	 * prefix which might be passed-in as a parameter.
+	 * Since the duration of printk() can vary depending on the message
+	 * and state of the ringbuffer, grab the timestamp now so that it is
+	 * close to the call of printk(). This provides a more deterministic
+	 * timestamp with respect to the caller.
 	 */
-	text_len = vscnprintf(text, sizeof(textbuf), fmt, args);
+	ts_nsec = local_clock();
 
-	/* mark and strip a trailing newline */
-	if (text_len && text[text_len-1] == '\n') {
-		text_len--;
-		lflags |= LOG_NEWLINE;
-	}
+	/*
+	 * The sprintf needs to come first since the syslog prefix might be
+	 * passed in as a parameter. An extra byte must be reserved so that
+	 * later the vscnprintf() into the reserved buffer has room for the
+	 * terminating '\0', which is not counted by vsnprintf().
+	 */
+	va_copy(args2, args);
+	reserve_size = vsnprintf(&prefix_buf[0], sizeof(prefix_buf), fmt, args2) + 1;
+	va_end(args2);
 
-	/* strip kernel syslog prefix and extract log level or control flags */
-	if (facility == 0) {
-		int kern_level;
+	if (reserve_size > LOG_LINE_MAX)
+		reserve_size = LOG_LINE_MAX;
+
+	/* Extract log level or control flags. */
+	if (facility == 0)
+		parse_prefix(&prefix_buf[0], &level, &lflags);
 
-		while ((kern_level = printk_get_level(text)) != 0) {
-			switch (kern_level) {
-			case '0' ... '7':
 	if (level == LOGLEVEL_DEFAULT)
-					level = kern_level - '0';
-				break;
-			case 'c':	/* KERN_CONT */
-				lflags |= LOG_CONT;
+		level = default_message_loglevel;
+
+	if (dev_info)
+		lflags |= LOG_NEWLINE;
+
+	if (lflags & LOG_CONT) {
+		prb_rec_init_wr(&r, reserve_size);
+		if (prb_reserve_in_last(&e, prb, &r, caller_id, LOG_LINE_MAX)) {
+			text_len = printk_sprint(&r.text_buf[r.info->text_len], reserve_size,
+						 facility, &lflags, fmt, args);
+			r.info->text_len += text_len;
+
+			if (lflags & LOG_NEWLINE) {
+				r.info->flags |= LOG_NEWLINE;
+				prb_final_commit(&e);
+			} else {
+				prb_commit(&e);
 			}
 
-			text_len -= 2;
-			text += 2;
+			return text_len;
 		}
 	}
 
-	if (level == LOGLEVEL_DEFAULT)
-		level = default_message_loglevel;
+	/*
+	 * Explicitly initialize the record before every prb_reserve() call.
+	 * prb_reserve_in_last() and prb_reserve() purposely invalidate the
+	 * structure when they fail.
+	 */
+	prb_rec_init_wr(&r, reserve_size);
+	if (!prb_reserve(&e, prb, &r)) {
+		/* truncate the message if it is too long for empty buffer */
+		truncate_msg(&reserve_size, &trunc_msg_len);
+
+		prb_rec_init_wr(&r, reserve_size + trunc_msg_len);
+		if (!prb_reserve(&e, prb, &r))
+			return 0;
+	}
 
+	/* fill message */
+	text_len = printk_sprint(&r.text_buf[0], reserve_size, facility, &lflags, fmt, args);
+	if (trunc_msg_len)
+		memcpy(&r.text_buf[text_len], trunc_msg, trunc_msg_len);
+	r.info->text_len = text_len + trunc_msg_len;
+	r.info->facility = facility;
+	r.info->level = level & 7;
+	r.info->flags = lflags & 0x1f;
+	r.info->ts_nsec = ts_nsec;
+	r.info->caller_id = caller_id;
 	if (dev_info)
-		lflags |= LOG_NEWLINE;
+		memcpy(&r.info->dev_info, dev_info, sizeof(r.info->dev_info));
+
+	/* A message without a trailing newline can be continued. */
+	if (!(lflags & LOG_NEWLINE))
+		prb_commit(&e);
+	else
+		prb_final_commit(&e);
 
-	return log_output(facility, level, lflags, dev_info, text, text_len);
+	return (text_len + trunc_msg_len);
 }
 
 asmlinkage int vprintk_emit(int facility, int level,
@@ -2006,10 +2053,9 @@ asmlinkage int vprintk_emit(int facility, int level,
 	boot_delay_msec(level);
 	printk_delay();
 
-	/* This stops the holder of console_sem just where we want him */
-	logbuf_lock_irqsave(flags);
+	printk_safe_enter_irqsave(flags);
 	printed_len = vprintk_store(facility, level, dev_info, fmt, args);
-	logbuf_unlock_irqrestore(flags);
+	printk_safe_exit_irqrestore(flags);
 
 	/* If called from the scheduler, we can not call up(). */
 	if (!in_sched) {
@@ -2189,8 +2235,15 @@ static int __init console_setup(char *str)
 	char *s, *options, *brl_options = NULL;
 	int idx;
 
-	if (str[0] == 0)
+	/*
+	 * console="" or console=null have been suggested as a way to
+	 * disable console output. Use ttynull that has been created
+	 * for exacly this purpose.
+	 */
+	if (str[0] == 0 || strcmp(str, "null") == 0) {
+		__add_preferred_console("ttynull", 0, NULL, NULL, true);
 		return 1;
+	}
 
 	if (_braille_console_setup(&str, &brl_options))
 		return 1;

kernel/printk/printk_ringbuffer.c

@@ -559,11 +559,12 @@ static void desc_make_reusable(struct prb_desc_ring *desc_ring,
  * on error the caller can re-load the tail lpos to determine the situation.
  */
 static bool data_make_reusable(struct printk_ringbuffer *rb,
-			       struct prb_data_ring *data_ring,
 			       unsigned long lpos_begin,
 			       unsigned long lpos_end,
 			       unsigned long *lpos_out)
 {
+
+	struct prb_data_ring *data_ring = &rb->text_data_ring;
 	struct prb_desc_ring *desc_ring = &rb->desc_ring;
 	struct prb_data_block *blk;
 	enum desc_state d_state;
@@ -625,10 +626,9 @@ static bool data_make_reusable(struct printk_ringbuffer *rb,
  * descriptors into the reusable state if the tail is pushed beyond
  * their associated data block.
  */
-static bool data_push_tail(struct printk_ringbuffer *rb,
-			   struct prb_data_ring *data_ring,
-			   unsigned long lpos)
+static bool data_push_tail(struct printk_ringbuffer *rb, unsigned long lpos)
 {
+	struct prb_data_ring *data_ring = &rb->text_data_ring;
 	unsigned long tail_lpos_new;
 	unsigned long tail_lpos;
 	unsigned long next_lpos;
@@ -669,8 +669,7 @@ static bool data_push_tail(struct printk_ringbuffer *rb,
 		 * Make all descriptors reusable that are associated with
 		 * data blocks before @lpos.
 		 */
-		if (!data_make_reusable(rb, data_ring, tail_lpos, lpos,
-					&next_lpos)) {
+		if (!data_make_reusable(rb, tail_lpos, lpos, &next_lpos)) {
 			/*
 			 * 1. Guarantee the block ID loaded in
 			 *    data_make_reusable() is performed before
@@ -807,7 +806,7 @@ static bool desc_push_tail(struct printk_ringbuffer *rb,
 	 * data blocks once their associated descriptor is gone.
 	 */
 
-	if (!data_push_tail(rb, &rb->text_data_ring, desc.text_blk_lpos.next))
+	if (!data_push_tail(rb, desc.text_blk_lpos.next))
 		return false;
 
 	/*
@@ -1019,10 +1018,10 @@ static unsigned long get_next_lpos(struct prb_data_ring *data_ring,
  * if necessary. This function also associates the data block with
  * a specified descriptor.
  */
-static char *data_alloc(struct printk_ringbuffer *rb,
-			struct prb_data_ring *data_ring, unsigned int size,
+static char *data_alloc(struct printk_ringbuffer *rb, unsigned int size,
 			struct prb_data_blk_lpos *blk_lpos, unsigned long id)
 {
+	struct prb_data_ring *data_ring = &rb->text_data_ring;
 	struct prb_data_block *blk;
 	unsigned long begin_lpos;
 	unsigned long next_lpos;
@@ -1041,7 +1040,7 @@ static char *data_alloc(struct printk_ringbuffer *rb,
 	do {
 		next_lpos = get_next_lpos(data_ring, begin_lpos, size);
 
-		if (!data_push_tail(rb, data_ring, next_lpos - DATA_SIZE(data_ring))) {
+		if (!data_push_tail(rb, next_lpos - DATA_SIZE(data_ring))) {
 			/* Failed to allocate, specify a data-less block. */
 			blk_lpos->begin = FAILED_LPOS;
 			blk_lpos->next = FAILED_LPOS;
@@ -1100,10 +1099,10 @@ static char *data_alloc(struct printk_ringbuffer *rb,
  * Return a pointer to the beginning of the entire data buffer or NULL on
  * failure.
  */
-static char *data_realloc(struct printk_ringbuffer *rb,
-			  struct prb_data_ring *data_ring, unsigned int size,
+static char *data_realloc(struct printk_ringbuffer *rb, unsigned int size,
 			  struct prb_data_blk_lpos *blk_lpos, unsigned long id)
 {
+	struct prb_data_ring *data_ring = &rb->text_data_ring;
 	struct prb_data_block *blk;
 	unsigned long head_lpos;
 	unsigned long next_lpos;
@@ -1130,7 +1129,7 @@ static char *data_realloc(struct printk_ringbuffer *rb,
 		return &blk->data[0];
 	}
 
-	if (!data_push_tail(rb, data_ring, next_lpos - DATA_SIZE(data_ring)))
+	if (!data_push_tail(rb, next_lpos - DATA_SIZE(data_ring)))
 		return NULL;
 
 	/* The memory barrier involvement is the same as data_alloc:A. */
@@ -1395,7 +1394,7 @@ bool prb_reserve_in_last(struct prb_reserved_entry *e, struct printk_ringbuffer
 		if (r->text_buf_size > max_size)
 			goto fail;
 
-		r->text_buf = data_alloc(rb, &rb->text_data_ring, r->text_buf_size,
+		r->text_buf = data_alloc(rb, r->text_buf_size,
 					 &d->text_blk_lpos, id);
 	} else {
 		if (!get_data(&rb->text_data_ring, &d->text_blk_lpos, &data_size))
@@ -1419,7 +1418,7 @@ bool prb_reserve_in_last(struct prb_reserved_entry *e, struct printk_ringbuffer
 		if (r->text_buf_size > max_size)
 			goto fail;
 
-		r->text_buf = data_realloc(rb, &rb->text_data_ring, r->text_buf_size,
+		r->text_buf = data_realloc(rb, r->text_buf_size,
 					   &d->text_blk_lpos, id);
 	}
 	if (r->text_buf_size && !r->text_buf)
@@ -1547,8 +1546,7 @@ bool prb_reserve(struct prb_reserved_entry *e, struct printk_ringbuffer *rb,
 	if (info->seq > 0)
 		desc_make_final(desc_ring, DESC_ID(id - 1));
 
-	r->text_buf = data_alloc(rb, &rb->text_data_ring, r->text_buf_size,
-				 &d->text_blk_lpos, id);
+	r->text_buf = data_alloc(rb, r->text_buf_size, &d->text_blk_lpos, id);
 	/* If text data allocation fails, a data-less record is committed. */
 	if (r->text_buf_size && !r->text_buf) {
 		prb_commit(e);