lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20231013204340.1112036-4-john.ogness@linutronix.de>
Date:   Fri, 13 Oct 2023 22:49:39 +0206
From:   John Ogness <john.ogness@...utronix.de>
To:     Petr Mladek <pmladek@...e.com>
Cc:     Sergey Senozhatsky <senozhatsky@...omium.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        linux-kernel@...r.kernel.org
Subject: [PATCH printk v2 3/4] printk: Skip unfinalized records in panic

Normally a reader will stop once reaching an unfinalized record.
However, when a panic happens, writers from other CPUs (or an
interrupted context on the panic CPU) may have been writing a
record and were unable to finalize it. The panic CPU will
reserve/commit/finalize its panic records, but these will be
located after the unfinalized records. This results in panic()
not flushing the panic messages.

Add a special case to printk_get_next_message() to allow
skipping over unfinalized records if on the panic CPU.

Also refine the documentation of the ringbuffer reading
functions to clarify that failure may also be due to an
unfinalized record.

Fixes: 896fbe20b4e2 ("printk: use the lockless ringbuffer")
Signed-off-by: John Ogness <john.ogness@...utronix.de>
---
 kernel/printk/printk.c            | 15 +++++++++++++--
 kernel/printk/printk_ringbuffer.c |  9 +++++----
 2 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 19b752880879..56d9b4acbbf2 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -2813,8 +2813,19 @@ static bool printk_get_next_message(struct printk_message *pmsg, u64 seq,
 	else
 		prb_rec_init_rd(&r, &info, outbuf, outbuf_sz);
 
-	if (!prb_read_valid(prb, seq, &r))
-		return false;
+	while (!prb_read_valid(prb, seq, &r)) {
+		if (this_cpu_in_panic() && seq < prb_next_seq(prb)) {
+			/*
+			 * The record @seq is not finalized and there may be
+			 * more records in the ringbuffer. Since this is the
+			 * panic CPU, skip over the unfinalized record and
+			 * try to read a finalized record that may follow.
+			 */
+			seq++;
+		} else {
+			return false;
+		}
+	}
 
 	pmsg->seq = r.info->seq;
 	pmsg->dropped = r.info->seq - seq;
diff --git a/kernel/printk/printk_ringbuffer.c b/kernel/printk/printk_ringbuffer.c
index 2dc4d5a1f1ff..1bbc008109ef 100644
--- a/kernel/printk/printk_ringbuffer.c
+++ b/kernel/printk/printk_ringbuffer.c
@@ -1876,8 +1876,9 @@ static u64 prb_first_seq(struct printk_ringbuffer *rb)
 }
 
 /*
- * Non-blocking read of a record. Updates @seq to the last finalized record
- * (which may have no data available).
+ * Non-blocking read of a record. Updates @seq to the record that was read
+ * (which may have no data available) or was attempted to be read (in case
+ * it was unfinalized or non-existent).
  *
  * See the description of prb_read_valid() and prb_read_valid_info()
  * for details.
@@ -1932,7 +1933,7 @@ static bool _prb_read_valid(struct printk_ringbuffer *rb, u64 *seq,
  * On success, the reader must check r->info.seq to see which record was
  * actually read. This allows the reader to detect dropped records.
  *
- * Failure means @seq refers to a not yet written record.
+ * Failure means @seq refers to a not yet finalized or non-existing record.
  */
 bool prb_read_valid(struct printk_ringbuffer *rb, u64 seq,
 		    struct printk_record *r)
@@ -1962,7 +1963,7 @@ bool prb_read_valid(struct printk_ringbuffer *rb, u64 seq,
  * On success, the reader must check info->seq to see which record meta data
  * was actually read. This allows the reader to detect dropped records.
  *
- * Failure means @seq refers to a not yet written record.
+ * Failure means @seq refers to a not yet finalized or non-existing record.
  */
 bool prb_read_valid_info(struct printk_ringbuffer *rb, u64 seq,
 			 struct printk_info *info, unsigned int *line_count)
-- 
2.39.2

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ