[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20231120122708.UxA5Wf4b@linutronix.de>
Date: Mon, 20 Nov 2023 13:27:08 +0100
From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To: Francesco Dolcini <francesco@...cini.it>
Cc: John Ogness <john.ogness@...utronix.de>,
Thomas Gleixner <tglx@...utronix.de>,
LKML <linux-kernel@...r.kernel.org>,
linux-rt-users@...r.kernel.org,
Steven Rostedt <rostedt@...dmis.org>,
francesco.dolcini@...adex.com
Subject: Re: [ANNOUNCE] v6.6-rt13
On 2023-11-13 20:07:51 [+0100], Francesco Dolcini wrote:
> Hello Sebastian, hello John
Hi,
> During our weekly CI tests of rt-devel branch we noticed that commit
> 78f4b59c6faa ("printk: Update the printk series.") breaks booting on
> multiple NXP i.MX platforms (arm and arm64), likely related to
> drivers/tty/serial/imx.c.
>
> We are not planning any additional investigation at the moment and
> we have no debug outputs to share, just wanted to give you a head-up.
Thank you for the report.
As it turns out, the new bits never saw the lights of a 32bit machine.
The following hunk should fix it:
diff --git a/kernel/printk/printk_ringbuffer.c b/kernel/printk/printk_ringbuffer.c
index a949d02fbd174..b6a8306bfda89 100644
--- a/kernel/printk/printk_ringbuffer.c
+++ b/kernel/printk/printk_ringbuffer.c
@@ -1477,7 +1477,10 @@ static inline u64 __ulseq_to_u64seq(u32 ulseq)
* sequence. It needs to be expanded to 64bit. Get the first sequence
* number from the ringbuffer and fold it.
*/
- seq = rb_first_seq - ((u32)rb_first_seq - ulseq);
+ if (rb_first_seq == 0)
+ seq = ulseq;
+ else
+ seq = rb_first_seq - ((u32)rb_first_seq - ulseq);
return seq;
}
> Francesco
Sebastian
Powered by blists - more mailing lists