linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Petr Mladek <pmladek@suse.com>
To: John Ogness <john.ogness@linutronix.de>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>,
	Steven Rostedt <rostedt@goodmis.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	linux-kernel@vger.kernel.org
Subject: [PATCH] printk: nbcon_atomic_flush_pending() is safe only when there is no boot console
Date: Thu, 13 Jun 2024 14:52:20 +0200	[thread overview]
Message-ID: <ZmrrhCf5GRUdGwh9@pathway.suse.cz> (raw)
In-Reply-To: <ZmrpxWxUb8k4swrA@pathway.suse.cz>

Boot consoles are not serialized with the nbcon consoles via the nbcon
console context or con->device_lock(). The serialization is possible only
via the legacy console_lock().

The decision whether nbcon_atomic_flush_pending() should and can be
called safely is similar and closely related to the decision
whether the legacy loop should be used.

Define printing_via_context_safe symmetrically with printing_via_unlock.
Allow to call nbcon_atomic_flush_pending() only when it is needed and safe.

Signed-off-by: Petr Mladek <pmladek@suse.com>
---
 kernel/printk/internal.h | 10 ++++++++++
 kernel/printk/nbcon.c    |  9 +++++++--
 kernel/printk/printk.c   |  7 ++++---
 3 files changed, 21 insertions(+), 5 deletions(-)

diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h
index 38680c6b2b39..bafec0a27da3 100644
--- a/kernel/printk/internal.h
+++ b/kernel/printk/internal.h
@@ -182,6 +182,7 @@ static inline bool console_is_usable(struct console *con, short flags) { return
 #endif /* CONFIG_PRINTK */
 
 extern bool have_boot_console;
+extern bool have_nbcon_console;
 extern bool have_legacy_console;
 
 /*
@@ -192,6 +193,15 @@ extern bool have_legacy_console;
  */
 #define printing_via_unlock (have_legacy_console || have_boot_console)
 
+/*
+ * Specifies if printing on nbcon consoles is needed and also safe
+ * when serialized only by the nbcon context. If @have_boot_console
+ * is true, the nbcon consoles must be serialized with the boot
+ * consoles using the legacy console_lock().
+ */
+#define printing_via_context_safe (have_nbcon_console || !have_boot_console)
+
+
 extern struct printk_buffers printk_shared_pbufs;
 
 /**
diff --git a/kernel/printk/nbcon.c b/kernel/printk/nbcon.c
index 89b340ca303c..9b2df848718c 100644
--- a/kernel/printk/nbcon.c
+++ b/kernel/printk/nbcon.c
@@ -1404,7 +1404,8 @@ void nbcon_cpu_emergency_exit(void)
 	 * for the emergency messages is NBCON_PRIO_EMERGENCY.
 	 */
 	if (*cpu_emergency_nesting == 1) {
-		nbcon_atomic_flush_pending();
+		if (printing_via_context_safe)
+			nbcon_atomic_flush_pending();
 
 		/*
 		 * Safely attempt to flush the legacy consoles in this
@@ -1446,7 +1447,8 @@ void nbcon_cpu_emergency_flush(void)
 	if (*(nbcon_get_cpu_emergency_nesting()) == 0)
 		return;
 
-	nbcon_atomic_flush_pending();
+	if (printing_via_context_safe)
+		nbcon_atomic_flush_pending();
 
 	if (printing_via_unlock && !is_printk_deferred()) {
 		if (console_trylock())
@@ -1637,6 +1639,9 @@ void nbcon_device_release(struct console *con)
 	 * was locked. The console_srcu_read_lock must be taken to ensure
 	 * the console is usable throughout flushing.
 	 */
+	if (!printing_via_context_safe)
+		return;
+
 	cookie = console_srcu_read_lock();
 	if (console_is_usable(con, console_srcu_read_flags(con)) &&
 	    !con->kthread &&
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 137bd9a721c4..3183db5b4180 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -475,7 +475,7 @@ bool have_legacy_console;
  * synchronous printing of legacy consoles will not occur during panic until
  * the backtrace has been stored to the ringbuffer.
  */
-static bool have_nbcon_console;
+bool have_nbcon_console;
 
 /*
  * Specifies if a boot console is registered. If boot consoles are present,
@@ -2370,7 +2370,7 @@ asmlinkage int vprintk_emit(int facility, int level,
 
 	printed_len = vprintk_store(facility, level, dev_info, fmt, args);
 
-	if (have_nbcon_console && !have_boot_console) {
+	if (printing_via_context_safe) {
 		bool is_panic_context = this_cpu_in_panic();
 
 		/*
@@ -3283,7 +3283,8 @@ void console_flush_on_panic(enum con_flush_mode mode)
 	if (mode == CONSOLE_REPLAY_ALL)
 		__console_rewind_all();
 
-	nbcon_atomic_flush_pending();
+	if (printing_via_context_safe)
+		nbcon_atomic_flush_pending();
 
 	if (printing_via_unlock)
 		console_flush_all(false, &next_seq, &handover);
-- 
2.45.2


  reply	other threads:[~2024-06-13 12:52 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-03 23:24 [PATCH printk v2 00/18] add threaded printing + the rest John Ogness
2024-06-03 23:24 ` [PATCH printk v2 01/18] printk: Add function to replay kernel log on consoles John Ogness
2024-06-03 23:24 ` [PATCH printk v2 02/18] tty/sysrq: Replay kernel log messages on consoles via sysrq John Ogness
2024-06-03 23:24 ` [PATCH printk v2 03/18] printk: Rename console_replay_all() and update context John Ogness
2024-06-03 23:24 ` [PATCH printk v2 04/18] printk: nbcon: Introduce printing kthreads John Ogness
2024-06-07 13:17   ` Petr Mladek
2024-06-10 12:09     ` John Ogness
2024-06-11 14:51       ` Petr Mladek
2024-06-12  8:51         ` John Ogness
2024-06-12  9:24           ` Petr Mladek
2024-06-12 11:18             ` John Ogness
2024-06-12 11:33               ` Petr Mladek
2024-06-13 15:21           ` John Ogness
2024-06-14  7:40             ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 05/18] printk: Atomic print in printk context on shutdown John Ogness
2024-06-13 12:32   ` Petr Mladek
2024-06-13 12:44   ` atomic_flush vs boot consoles - was: " Petr Mladek
2024-06-13 12:52     ` Petr Mladek [this message]
2024-06-13 15:10       ` [PATCH] printk: nbcon_atomic_flush_pending() is safe only when there is no boot console Petr Mladek
2024-06-25 20:53       ` John Ogness
2024-06-28 15:51         ` how to flush consoles: was: " Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 06/18] printk: nbcon: Add context to console_is_usable() John Ogness
2024-06-13 13:22   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 07/18] printk: nbcon: Add printer thread wakeups John Ogness
2024-06-13 15:08   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 08/18] printk: nbcon: Stop threads on shutdown/reboot John Ogness
2024-06-17 15:21   ` Petr Mladek
2024-06-25 19:56     ` John Ogness
2024-06-26 12:14       ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 09/18] printk: nbcon: Start printing threads John Ogness
2024-06-18 15:34   ` Petr Mladek
2024-06-19 15:13     ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 10/18] printk: Provide helper for message prepending John Ogness
2024-06-20 10:28   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 11/18] printk: nbcon: Show replay message on takeover John Ogness
2024-06-20 13:02   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 12/18] printk: Add kthread for all legacy consoles John Ogness
2024-06-28 11:46   ` Petr Mladek
2024-06-28 12:22     ` John Ogness
2024-06-28 13:32       ` Petr Mladek
2024-06-28 14:11         ` John Ogness
2024-06-28 15:56           ` John Ogness
2024-07-01 15:33             ` Petr Mladek
2024-07-01 21:01               ` John Ogness
2024-07-02  9:11                 ` Petr Mladek
2024-07-01 14:50           ` Petr Mladek
2024-07-02  9:30             ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 13/18] proc: consoles: Add notation to c_start/c_stop John Ogness
2024-07-01 15:43   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 14/18] proc: Add nbcon support for /proc/consoles John Ogness
2024-07-01 15:47   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 15/18] tty: sysfs: Add nbcon support for 'active' John Ogness
2024-06-04 11:48   ` Greg Kroah-Hartman
2024-07-01 15:50   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 16/18] printk: Provide threadprintk boot argument John Ogness
2024-07-02 12:12   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 17/18] printk: Avoid false positive lockdep report for legacy printing John Ogness
2024-07-02 13:26   ` Petr Mladek
2024-06-03 23:24 ` [PATCH printk v2 18/18] printk: nbcon: Add function for printers to reacquire ownership John Ogness
2024-07-02 14:31   ` Petr Mladek
2024-06-04 13:31 ` [PATCH printk v2 00/18] add threaded printing + the rest Juri Lelli
2024-06-05  8:09   ` John Ogness
2024-06-05  9:32     ` Juri Lelli

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ZmrrhCf5GRUdGwh9@pathway.suse.cz \
    --to=pmladek@suse.com \
    --cc=john.ogness@linutronix.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=senozhatsky@chromium.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).