Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1287485 > unrolled thread

[PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

Started byPetr Mladek <pmladek@suse.com>
First post2015-12-09 14:30 +0100
Last post2015-12-12 00:00 +0100
Articles 6 — 4 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable Petr Mladek <pmladek@suse.com> - 2015-12-09 14:30 +0100
    Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and  make it configurable Geert Uytterhoeven <geert@linux-m68k.org> - 2015-12-11 12:20 +0100
      Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable Arnd Bergmann <arnd@arndb.de> - 2015-12-11 13:50 +0100
      Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and  make it configurable Petr Mladek <pmladek@suse.com> - 2015-12-11 13:50 +0100
        Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and  make it configurable Geert Uytterhoeven <geert@linux-m68k.org> - 2015-12-11 14:00 +0100
        Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and  make it configurable Andrew Morton <akpm@linux-foundation.org> - 2015-12-12 00:00 +0100

#1287485 — [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

FromPetr Mladek <pmladek@suse.com>
Date2015-12-09 14:30 +0100
Subject[PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable
Message-ID<qDMYG-u3-11@gated-at.bofh.it>
Testing has shown that the backtrace sometimes does not fit
into the 4kB temporary buffer that is used in NMI context.
The warnings are gone when I double the temporary buffer size.

This patch doubles the buffer size and makes it configurable.

Note that this problem existed even in the x86-specific
implementation that was added by the commit a9edc8809328
("x86/nmi: Perform a safe NMI stack trace on all CPUs").
Nobody noticed it because it did not print any warnings.

Signed-off-by: Petr Mladek <pmladek@suse.com>
---
 init/Kconfig        | 22 ++++++++++++++++++++++
 kernel/printk/nmi.c |  3 ++-
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/init/Kconfig b/init/Kconfig
index c1c0b6a2d712..efcff25a112d 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT
 		     13 =>   8 KB for each CPU
 		     12 =>   4 KB for each CPU
 
+config NMI_LOG_BUF_SHIFT
+	int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
+	range 10 21
+	default 13
+	depends on PRINTK && HAVE_NMI
+	help
+	  Select the size of a per-CPU buffer where NMI messages are temporary
+	  stored. They are copied to the main log buffer in a safe context
+	  to avoid a deadlock. The value defines the size as a power of 2.
+
+	  NMI messages are rare and limited. The largest one is when
+	  a backtrace is printed. It usually fits into 4KB. Select
+	  8KB if you want to be on the safe side.
+
+	  Examples:
+		     17 => 128 KB for each CPU
+		     16 =>  64 KB for each CPU
+		     15 =>  32 KB for each CPU
+		     14 =>  16 KB for each CPU
+		     13 =>   8 KB for each CPU
+		     12 =>   4 KB for each CPU
+
 #
 # Architectures with an unreliable sched_clock() should select this:
 #
diff --git a/kernel/printk/nmi.c b/kernel/printk/nmi.c
index 5465230b75ec..78c07d441b4e 100644
--- a/kernel/printk/nmi.c
+++ b/kernel/printk/nmi.c
@@ -41,7 +41,8 @@ DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default;
 static int printk_nmi_irq_ready;
 atomic_t nmi_message_lost;
 
-#define NMI_LOG_BUF_LEN (4096 - sizeof(atomic_t) - sizeof(struct irq_work))
+#define NMI_LOG_BUF_LEN ((1 << CONFIG_NMI_LOG_BUF_SHIFT) -		\
+			 sizeof(atomic_t) - sizeof(struct irq_work))
 
 struct nmi_seq_buf {
 	atomic_t		len;	/* length of written data */
-- 
1.8.5.6

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1289433 — Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

FromGeert Uytterhoeven <geert@linux-m68k.org>
Date2015-12-11 12:20 +0100
SubjectRe: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable
Message-ID<qEtTY-3va-19@gated-at.bofh.it>
In reply to#1287485
On Wed, Dec 9, 2015 at 2:21 PM, Petr Mladek <pmladek@suse.com> wrote:
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT
>                      13 =>   8 KB for each CPU
>                      12 =>   4 KB for each CPU
>
> +config NMI_LOG_BUF_SHIFT
> +       int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
> +       range 10 21
> +       default 13
> +       depends on PRINTK && HAVE_NMI

Symbol NMI_LOG_BUF_SHIFT does not exist if its dependencies are not met.

> +       help
> +         Select the size of a per-CPU buffer where NMI messages are temporary
> +         stored. They are copied to the main log buffer in a safe context
> +         to avoid a deadlock. The value defines the size as a power of 2.
> +
> +         NMI messages are rare and limited. The largest one is when
> +         a backtrace is printed. It usually fits into 4KB. Select
> +         8KB if you want to be on the safe side.
> +
> +         Examples:
> +                    17 => 128 KB for each CPU
> +                    16 =>  64 KB for each CPU
> +                    15 =>  32 KB for each CPU
> +                    14 =>  16 KB for each CPU
> +                    13 =>   8 KB for each CPU
> +                    12 =>   4 KB for each CPU
> +
>  #
>  # Architectures with an unreliable sched_clock() should select this:
>  #
> diff --git a/kernel/printk/nmi.c b/kernel/printk/nmi.c
> index 5465230b75ec..78c07d441b4e 100644
> --- a/kernel/printk/nmi.c
> +++ b/kernel/printk/nmi.c
> @@ -41,7 +41,8 @@ DEFINE_PER_CPU(printk_func_t, printk_func) = vprintk_default;
>  static int printk_nmi_irq_ready;
>  atomic_t nmi_message_lost;
>
> -#define NMI_LOG_BUF_LEN (4096 - sizeof(atomic_t) - sizeof(struct irq_work))
> +#define NMI_LOG_BUF_LEN ((1 << CONFIG_NMI_LOG_BUF_SHIFT) -             \
> +                        sizeof(atomic_t) - sizeof(struct irq_work))

kernel/printk/nmi.c:50:24: error: 'CONFIG_NMI_LOG_BUF_SHIFT'
undeclared here (not in a function)

E.g. efm32_defconfig
http://kisskb.ellerman.id.au/kisskb/buildresult/12565754/

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1289525

FromArnd Bergmann <arnd@arndb.de>
Date2015-12-11 13:50 +0100
Message-ID<qEvj4-4o4-1@gated-at.bofh.it>
In reply to#1289433
On Friday 11 December 2015 13:41:59 Petr Mladek wrote:
> diff --git a/init/Kconfig b/init/Kconfig
> index efcff25a112d..61cfd96a3c96 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -870,7 +870,7 @@ config NMI_LOG_BUF_SHIFT
>         int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
>         range 10 21
>         default 13
> -       depends on PRINTK && HAVE_NMI
> +       depends on PRINTK_NMI
>         help
>           Select the size of a per-CPU buffer where NMI messages are temporary
>           stored. They are copied to the main log buffer in a safe context
> 
> 

Acked-by: Arnd Bergmann <arnd@arndb.de>

I found this on linux-next as well today and came to the same conclusion.

	Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1289529 — Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

FromPetr Mladek <pmladek@suse.com>
Date2015-12-11 13:50 +0100
SubjectRe: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable
Message-ID<qEvj4-4o4-3@gated-at.bofh.it>
In reply to#1289433
On Fri 2015-12-11 12:10:02, Geert Uytterhoeven wrote:
> On Wed, Dec 9, 2015 at 2:21 PM, Petr Mladek <pmladek@suse.com> wrote:
> > --- a/init/Kconfig
> > +++ b/init/Kconfig
> > @@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT
> >                      13 =>   8 KB for each CPU
> >                      12 =>   4 KB for each CPU
> >
> > +config NMI_LOG_BUF_SHIFT
> > +       int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
> > +       range 10 21
> > +       default 13
> > +       depends on PRINTK && HAVE_NMI
> 
> Symbol NMI_LOG_BUF_SHIFT does not exist if its dependencies are not met.

Åh, the NMI buffer is enabled on arm via NEED_PRINTK_NMI.

The buffer is compiled when CONFIG_PRINTK_NMI is defined. I am going
to fix it the following way:


diff --git a/init/Kconfig b/init/Kconfig
index efcff25a112d..61cfd96a3c96 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -870,7 +870,7 @@ config NMI_LOG_BUF_SHIFT
 	int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
 	range 10 21
 	default 13
-	depends on PRINTK && HAVE_NMI
+	depends on PRINTK_NMI
 	help
 	  Select the size of a per-CPU buffer where NMI messages are temporary
 	  stored. They are copied to the main log buffer in a safe context


Thanks a lot for report,
Petr
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1289538 — Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

FromGeert Uytterhoeven <geert@linux-m68k.org>
Date2015-12-11 14:00 +0100
SubjectRe: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable
Message-ID<qEvsK-4sJ-21@gated-at.bofh.it>
In reply to#1289529
On Fri, Dec 11, 2015 at 1:41 PM, Petr Mladek <pmladek@suse.com> wrote:
> On Fri 2015-12-11 12:10:02, Geert Uytterhoeven wrote:
>> On Wed, Dec 9, 2015 at 2:21 PM, Petr Mladek <pmladek@suse.com> wrote:
>> > --- a/init/Kconfig
>> > +++ b/init/Kconfig
>> > @@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT
>> >                      13 =>   8 KB for each CPU
>> >                      12 =>   4 KB for each CPU
>> >
>> > +config NMI_LOG_BUF_SHIFT
>> > +       int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
>> > +       range 10 21
>> > +       default 13
>> > +       depends on PRINTK && HAVE_NMI
>>
>> Symbol NMI_LOG_BUF_SHIFT does not exist if its dependencies are not met.
>
> Åh, the NMI buffer is enabled on arm via NEED_PRINTK_NMI.
>
> The buffer is compiled when CONFIG_PRINTK_NMI is defined. I am going
> to fix it the following way:
>
>
> diff --git a/init/Kconfig b/init/Kconfig
> index efcff25a112d..61cfd96a3c96 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -870,7 +870,7 @@ config NMI_LOG_BUF_SHIFT
>         int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
>         range 10 21
>         default 13
> -       depends on PRINTK && HAVE_NMI
> +       depends on PRINTK_NMI
>         help
>           Select the size of a per-CPU buffer where NMI messages are temporary
>           stored. They are copied to the main log buffer in a safe context

Makes sense, as kernel/printk/nmi.c is compiled if PRINTK_NMI is set.

Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1289990 — Re: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable

FromAndrew Morton <akpm@linux-foundation.org>
Date2015-12-12 00:00 +0100
SubjectRe: [PATCH v3 4/4] printk/nmi: Increase the size of NMI buffer and make it configurable
Message-ID<qEEPp-2db-29@gated-at.bofh.it>
In reply to#1289529
On Fri, 11 Dec 2015 13:41:59 +0100 Petr Mladek <pmladek@suse.com> wrote:

> On Fri 2015-12-11 12:10:02, Geert Uytterhoeven wrote:
> > On Wed, Dec 9, 2015 at 2:21 PM, Petr Mladek <pmladek@suse.com> wrote:
> > > --- a/init/Kconfig
> > > +++ b/init/Kconfig
> > > @@ -866,6 +866,28 @@ config LOG_CPU_MAX_BUF_SHIFT
> > >                      13 =>   8 KB for each CPU
> > >                      12 =>   4 KB for each CPU
> > >
> > > +config NMI_LOG_BUF_SHIFT
> > > +       int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
> > > +       range 10 21
> > > +       default 13
> > > +       depends on PRINTK && HAVE_NMI
> > 
> > Symbol NMI_LOG_BUF_SHIFT does not exist if its dependencies are not met.
> 
> __h, the NMI buffer is enabled on arm via NEED_PRINTK_NMI.
> 
> The buffer is compiled when CONFIG_PRINTK_NMI is defined. I am going
> to fix it the following way:
> 
> 
> diff --git a/init/Kconfig b/init/Kconfig
> index efcff25a112d..61cfd96a3c96 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -870,7 +870,7 @@ config NMI_LOG_BUF_SHIFT
>  	int "Temporary per-CPU NMI log buffer size (12 => 4KB, 13 => 8KB)"
>  	range 10 21
>  	default 13
> -	depends on PRINTK && HAVE_NMI
> +	depends on PRINTK_NMI
>  	help
>  	  Select the size of a per-CPU buffer where NMI messages are temporary
>  	  stored. They are copied to the main log buffer in a safe context

I'm wondering why we're building kernel/printk/nmi.o if HAVE_NMI is not
set.

	obj-$(CONFIG_PRINTK_NMI)		+= nmi.o

and

	config PRINTK_NMI
		def_bool y
		depends on PRINTK
		depends on HAVE_NMI || NEED_PRINTK_NMI

This is a bit messy.  NEED_PRINTK_NMI is an added-on hack for one
particular arm variant.  From the changelog:

   "One exception is arm where the deferred printing is used for
    printing backtraces even without NMI.  For this purpose, we define
    NEED_PRINTK_NMI Kconfig flag.  The alternative printk_func is
    explicitly set when IPI_CPU_BACKTRACE is handled."


- why does arm needs deferred printing for backtraces?

- why is this specific to CONFIG_CPU_V7M?

- can this Kconfig logic be cleaned up a bit?

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web