From: "Luis R. Rodriguez" <[email protected]>
The default size of the ring buffer is too small for machines
with a large amount of CPUs under heavy load. What ends up
happening when debugging is the ring buffer overlaps and chews
up old messages making debugging impossible unless the size is
passed as a kernel parameter. An idle system upon boot up will
on average spew out only about one or two extra lines but where
this really matters is on heavy load and that will vary widely
depending on the system and environment.
There are mechanisms to help increase the kernel ring buffer
for tracing through debugfs, and those interfaces even allow growing
the kernel ring buffer per CPU. We also have a static value which
can be passed upon boot. Relying on debugfs however is not ideal
for production, and relying on the value passed upon bootup is
can only used *after* an issue has creeped up. Instead of being
reactive this adds a proactive measure which lets you scale the
amount of contributions you'd expect to the kernel ring buffer
under load by each CPU in the worst case scenario.
We use num_possible_cpus() to avoid complexities which could be
introduced by dynamically changing the ring buffer size at run
time, num_possible_cpus() lets us use the upper limit on possible
number of CPUs therefore avoiding having to deal with hotplugging
CPUs on and off. This introduces the kernel configuration option
LOG_CPU_MIN_BUF_SHIFT which is used to specify the maximum amount
of contributions to the kernel ring buffer in the worst case before
the kernel ring buffer flips over, the size is specified as a power
of 2. The total amount of contributions made by each CPU must be
greater than half of the default kernel ring buffer size
(1 << LOG_BUF_SHIFT bytes) in order to trigger an increase upon
bootup. For example if LOG_BUF_SHIFT is 18 (256 KB) you'd require at
least 128 KB contributions by other CPUs in order to trigger an
increase. With a LOG_CPU_BUF_SHIFT of 12 (4 KB) you'd require at
least anything over > 64 possible CPUs to trigger an increase. If
you had 128 possible CPUs your kernel buffer size would be:
((1 << 18) + ((128 - 1) * (1 << 12))) / 1024 = 764 KB
This value is ignored when "log_buf_len" kernel parameter is used
as it forces the exact size of the ring buffer to an expected value.
Cc: Michal Hocko <[email protected]>
Cc: Petr Mladek <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Joe Perches <[email protected]>
Cc: Arun KS <[email protected]>
Cc: Kees Cook <[email protected]>
Cc: Davidlohr Bueso <[email protected]>
Cc: Chris Metcalf <[email protected]>
Cc: [email protected]
Signed-off-by: Luis R. Rodriguez <[email protected]>
---
Documentation/kernel-parameters.txt | 8 +++++--
init/Kconfig | 45 ++++++++++++++++++++++++++++++++++++-
kernel/printk/printk.c | 12 ++++++++++
3 files changed, 62 insertions(+), 3 deletions(-)
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 30a8ad0d..23c50eb 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1646,8 +1646,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
7 (KERN_DEBUG) debug-level messages
log_buf_len=n[KMG] Sets the size of the printk ring buffer,
- in bytes. n must be a power of two. The default
- size is set in the kernel config file.
+ in bytes. n must be a power of two and greater
+ than the minimal size. The minimal size is defined
+ by LOG_BUF_SHIFT kernel config parameter. There is
+ also CONFIG_LOG_CPU_MIN_BUF_SHIFT config parameter
+ that allows to increase the default size depending on
+ the number of CPUs. See init/Kconfig for more details.
logo.nologo [FB] Disables display of the built-in Linux logo.
This may be used to provide more screen space for
diff --git a/init/Kconfig b/init/Kconfig
index 9d3585b..7e4da9a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -797,7 +797,11 @@ config LOG_BUF_SHIFT
range 12 21
default 17
help
- Select kernel log buffer size as a power of 2.
+ Select the minimal kernel log buffer size as a power of 2.
+ The final size is affected by LOG_CPU_MIN_BUF_SHIFT config
+ parameter, see below. Any higher size also might be forced
+ by "log_buf_len" boot parameter.
+
Examples:
17 => 128 KB
16 => 64 KB
@@ -806,6 +810,45 @@ config LOG_BUF_SHIFT
13 => 8 KB
12 => 4 KB
+config LOG_CPU_MIN_BUF_SHIFT
+ int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
+ range 0 21
+ default 12
+ depends on SMP
+ depends on !BASE_SMALL
+ help
+ The kernel ring buffer will get additional data logged onto it
+ when multiple CPUs are supported. Typically the contributions are
+ only a few lines when idle however under under load this can vary
+ and in the worst case it can mean losing logging information. You
+ can use this to set the maximum expected mount of amount of logging
+ contribution under load by each CPU in the worst case scenario, as
+ a power of 2. The total amount of contributions made by each CPU
+ must be greater than half of the default kernel ring buffer size
+ (1 << LOG_BUF_SHIFT bytes) in order to trigger an increase upon
+ bootup. For example if LOG_BUF_SHIFT is 18 (256 KB) you're require
+ at least 128 KB contributions by other CPUs in order to trigger
+ an increase. With a LOG_CPU_BUF_SHIFT of 12 (4 KB) you'd require
+ at least anything over > 64 possible CPUs to trigger an increase.
+ If you had 128 possible CPUs your kernel buffer size would be:
+
+ ((1 << 18) + ((128 - 1) * (1 << 12))) / 1024 = 764 KB
+
+ This value is ignored when "log_buf_len" kernel parameter is used
+ as it forces the exact size of the ring buffer to an expected value.
+
+ The number of possible CPUs is used for this computation ignoring
+ hotplugging making the compuation optimal for the the worst case
+ scenerio while allowing a simple algorithm to be used from bootup.
+
+ Examples shift values and their meaning:
+ 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/printk.c b/kernel/printk/printk.c
index 7228258..3f3356b 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -246,6 +246,7 @@ static u32 clear_idx;
#define LOG_ALIGN __alignof__(struct printk_log)
#endif
#define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
+#define __LOG_CPU_MIN_BUF_LEN (1 << CONFIG_LOG_CPU_MIN_BUF_SHIFT)
static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
static char *log_buf = __log_buf;
static u32 log_buf_len = __LOG_BUF_LEN;
@@ -752,6 +753,17 @@ void __init setup_log_buf(int early)
unsigned long flags;
char *new_log_buf;
int free;
+ int cpu_extra = (num_possible_cpus() - 1) * __LOG_CPU_MIN_BUF_LEN;
+
+ /*
+ * If you set log_buf_len=n kernel parameter LOG_CPU_MIN_BUF_SHIFT will
+ * be ignored. LOG_CPU_MIN_BUF_SHIFT is a proactive measure for large
+ * systems. With a LOG_BUF_SHIFT of 18 and LOG_CPU_MIN_BUF_SHIFT 12 at
+ * we'd require more than 64 CPUs to trigger an increase from the
+ * default.
+ */
+ if (!new_log_buf_len && (cpu_extra > __LOG_BUF_LEN / 2))
+ new_log_buf_len = __LOG_BUF_LEN + cpu_extra;
if (!new_log_buf_len)
return;
--
2.0.0.rc3.18.g00a5b79
On Fri, 2014-06-13 at 11:28 -0700, Luis R. Rodriguez wrote:
> From: "Luis R. Rodriguez" <[email protected]>
>
> The default size of the ring buffer is too small for machines
> with a large amount of CPUs under heavy load. What ends up
> happening when debugging is the ring buffer overlaps and chews
> up old messages making debugging impossible unless the size is
> passed as a kernel parameter. An idle system upon boot up will
> on average spew out only about one or two extra lines but where
> this really matters is on heavy load and that will vary widely
> depending on the system and environment.
>
> There are mechanisms to help increase the kernel ring buffer
> for tracing through debugfs, and those interfaces even allow growing
> the kernel ring buffer per CPU. We also have a static value which
> can be passed upon boot. Relying on debugfs however is not ideal
> for production, and relying on the value passed upon bootup is
> can only used *after* an issue has creeped up. Instead of being
> reactive this adds a proactive measure which lets you scale the
> amount of contributions you'd expect to the kernel ring buffer
> under load by each CPU in the worst case scenario.
>
> We use num_possible_cpus() to avoid complexities which could be
> introduced by dynamically changing the ring buffer size at run
> time, num_possible_cpus() lets us use the upper limit on possible
> number of CPUs therefore avoiding having to deal with hotplugging
> CPUs on and off. This introduces the kernel configuration option
> LOG_CPU_MIN_BUF_SHIFT which is used to specify the maximum amount
> of contributions to the kernel ring buffer in the worst case before
> the kernel ring buffer flips over, the size is specified as a power
> of 2. The total amount of contributions made by each CPU must be
> greater than half of the default kernel ring buffer size
> (1 << LOG_BUF_SHIFT bytes) in order to trigger an increase upon
> bootup. For example if LOG_BUF_SHIFT is 18 (256 KB) you'd require at
> least 128 KB contributions by other CPUs in order to trigger an
> increase. With a LOG_CPU_BUF_SHIFT of 12 (4 KB) you'd require at
> least anything over > 64 possible CPUs to trigger an increase. If
> you had 128 possible CPUs your kernel buffer size would be:
>
> ((1 << 18) + ((128 - 1) * (1 << 12))) / 1024 = 764 KB
>
> This value is ignored when "log_buf_len" kernel parameter is used
> as it forces the exact size of the ring buffer to an expected value.
>
> Cc: Michal Hocko <[email protected]>
> Cc: Petr Mladek <[email protected]>
> Cc: Andrew Morton <[email protected]>
> Cc: Joe Perches <[email protected]>
> Cc: Arun KS <[email protected]>
> Cc: Kees Cook <[email protected]>
> Cc: Davidlohr Bueso <[email protected]>
> Cc: Chris Metcalf <[email protected]>
> Cc: [email protected]
> Signed-off-by: Luis R. Rodriguez <[email protected]>
> ---
> Documentation/kernel-parameters.txt | 8 +++++--
> init/Kconfig | 45 ++++++++++++++++++++++++++++++++++++-
> kernel/printk/printk.c | 12 ++++++++++
> 3 files changed, 62 insertions(+), 3 deletions(-)
>
> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
> index 30a8ad0d..23c50eb 100644
> --- a/Documentation/kernel-parameters.txt
> +++ b/Documentation/kernel-parameters.txt
> @@ -1646,8 +1646,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
> 7 (KERN_DEBUG) debug-level messages
>
> log_buf_len=n[KMG] Sets the size of the printk ring buffer,
> - in bytes. n must be a power of two. The default
> - size is set in the kernel config file.
> + in bytes. n must be a power of two and greater
> + than the minimal size. The minimal size is defined
> + by LOG_BUF_SHIFT kernel config parameter. There is
> + also CONFIG_LOG_CPU_MIN_BUF_SHIFT config parameter
> + that allows to increase the default size depending on
> + the number of CPUs. See init/Kconfig for more details.
>
> logo.nologo [FB] Disables display of the built-in Linux logo.
> This may be used to provide more screen space for
> diff --git a/init/Kconfig b/init/Kconfig
> index 9d3585b..7e4da9a 100644
> --- a/init/Kconfig
> +++ b/init/Kconfig
> @@ -797,7 +797,11 @@ config LOG_BUF_SHIFT
> range 12 21
> default 17
> help
> - Select kernel log buffer size as a power of 2.
> + Select the minimal kernel log buffer size as a power of 2.
> + The final size is affected by LOG_CPU_MIN_BUF_SHIFT config
> + parameter, see below. Any higher size also might be forced
> + by "log_buf_len" boot parameter.
> +
> Examples:
> 17 => 128 KB
> 16 => 64 KB
> @@ -806,6 +810,45 @@ config LOG_BUF_SHIFT
> 13 => 8 KB
> 12 => 4 KB
>
> +config LOG_CPU_MIN_BUF_SHIFT
> + int "CPU kernel log buffer size contribution (13 => 8 KB, 17 => 128KB)"
> + range 0 21
> + default 12
> + depends on SMP
> + depends on !BASE_SMALL
> + help
> + The kernel ring buffer will get additional data logged onto it
> + when multiple CPUs are supported. Typically the contributions are
> + only a few lines when idle however under under load this can vary
> + and in the worst case it can mean losing logging information. You
> + can use this to set the maximum expected mount of amount of logging
> + contribution under load by each CPU in the worst case scenario, as
> + a power of 2. The total amount of contributions made by each CPU
> + must be greater than half of the default kernel ring buffer size
> + (1 << LOG_BUF_SHIFT bytes) in order to trigger an increase upon
> + bootup. For example if LOG_BUF_SHIFT is 18 (256 KB) you're require
> + at least 128 KB contributions by other CPUs in order to trigger
> + an increase. With a LOG_CPU_BUF_SHIFT of 12 (4 KB) you'd require
> + at least anything over > 64 possible CPUs to trigger an increase.
> + If you had 128 possible CPUs your kernel buffer size would be:
> +
> + ((1 << 18) + ((128 - 1) * (1 << 12))) / 1024 = 764 KB
> +
> + This value is ignored when "log_buf_len" kernel parameter is used
> + as it forces the exact size of the ring buffer to an expected value.
> +
> + The number of possible CPUs is used for this computation ignoring
> + hotplugging making the compuation optimal for the the worst case
> + scenerio while allowing a simple algorithm to be used from bootup.
> +
> + Examples shift values and their meaning:
> + 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/printk.c b/kernel/printk/printk.c
> index 7228258..3f3356b 100644
> --- a/kernel/printk/printk.c
> +++ b/kernel/printk/printk.c
> @@ -246,6 +246,7 @@ static u32 clear_idx;
> #define LOG_ALIGN __alignof__(struct printk_log)
> #endif
> #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
> +#define __LOG_CPU_MIN_BUF_LEN (1 << CONFIG_LOG_CPU_MIN_BUF_SHIFT)
> static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
> static char *log_buf = __log_buf;
> static u32 log_buf_len = __LOG_BUF_LEN;
> @@ -752,6 +753,17 @@ void __init setup_log_buf(int early)
> unsigned long flags;
> char *new_log_buf;
> int free;
> + int cpu_extra = (num_possible_cpus() - 1) * __LOG_CPU_MIN_BUF_LEN;
I think you forgot to drop the - 1 here.
On Fri, Jun 13, 2014 at 1:36 PM, Davidlohr Bueso <[email protected]> wrote:
> On Fri, 2014-06-13 at 11:28 -0700, Luis R. Rodriguez wrote:
>> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
>> index 7228258..3f3356b 100644
>> --- a/kernel/printk/printk.c
>> +++ b/kernel/printk/printk.c
>> @@ -246,6 +246,7 @@ static u32 clear_idx;
>> #define LOG_ALIGN __alignof__(struct printk_log)
>> #endif
>> #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
>> +#define __LOG_CPU_MIN_BUF_LEN (1 << CONFIG_LOG_CPU_MIN_BUF_SHIFT)
>> static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
>> static char *log_buf = __log_buf;
>> static u32 log_buf_len = __LOG_BUF_LEN;
>> @@ -752,6 +753,17 @@ void __init setup_log_buf(int early)
>> unsigned long flags;
>> char *new_log_buf;
>> int free;
>> + int cpu_extra = (num_possible_cpus() - 1) * __LOG_CPU_MIN_BUF_LEN;
>
> I think you forgot to drop the - 1 here.
Actually that was on purpose as I had noticed in your suggestion on
the other thread to fix the minimum number of CPUs required for this
patch to be > 64 rather than >= 64 (with the defaults of 18 for log
shift, and 12 for min cpu log shift). Minor difference, but it was
intentional. Even though we do require SMP now the -1 can take effect,
and it seems safer for now to only require this for > 64 CPU
configurations for the defaults set. I'm happy to remove that as well,
but bumping the CPU configuration mark for this to > 64 might be
better for starters unless we do know for sure we also need this for
64.
Luis
On Fri, 2014-06-13 at 13:44 -0700, Luis R. Rodriguez wrote:
> On Fri, Jun 13, 2014 at 1:36 PM, Davidlohr Bueso <[email protected]> wrote:
> > On Fri, 2014-06-13 at 11:28 -0700, Luis R. Rodriguez wrote:
> >> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
> >> index 7228258..3f3356b 100644
> >> --- a/kernel/printk/printk.c
> >> +++ b/kernel/printk/printk.c
> >> @@ -246,6 +246,7 @@ static u32 clear_idx;
> >> #define LOG_ALIGN __alignof__(struct printk_log)
> >> #endif
> >> #define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
> >> +#define __LOG_CPU_MIN_BUF_LEN (1 << CONFIG_LOG_CPU_MIN_BUF_SHIFT)
> >> static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN);
> >> static char *log_buf = __log_buf;
> >> static u32 log_buf_len = __LOG_BUF_LEN;
> >> @@ -752,6 +753,17 @@ void __init setup_log_buf(int early)
> >> unsigned long flags;
> >> char *new_log_buf;
> >> int free;
> >> + int cpu_extra = (num_possible_cpus() - 1) * __LOG_CPU_MIN_BUF_LEN;
> >
> > I think you forgot to drop the - 1 here.
>
> Actually that was on purpose as I had noticed in your suggestion on
> the other thread to fix the minimum number of CPUs required for this
> patch to be > 64 rather than >= 64 (with the defaults of 18 for log
> shift, and 12 for min cpu log shift). Minor difference, but it was
> intentional. Even though we do require SMP now the -1 can take effect,
> and it seems safer for now to only require this for > 64 CPU
> configurations for the defaults set. I'm happy to remove that as well,
> but bumping the CPU configuration mark for this to > 64 might be
> better for starters unless we do know for sure we also need this for
> 64.
Ah, ok I had missed that. Makes sense now.
Thanks,
Davidlohr
On Fri, 2014-06-13 at 11:28 -0700, Luis R. Rodriguez wrote:
> + /*
> + * If you set log_buf_len=n kernel parameter LOG_CPU_MIN_BUF_SHIFT will
> + * be ignored. LOG_CPU_MIN_BUF_SHIFT is a proactive measure for large
> + * systems. With a LOG_BUF_SHIFT of 18 and LOG_CPU_MIN_BUF_SHIFT 12 at
> + * we'd require more than 64 CPUs to trigger an increase from the
> + * default.
> + */
> + if (!new_log_buf_len && (cpu_extra > __LOG_BUF_LEN / 2))
^ that ! looks wrong. We should be checking for log_buf_len set instead.
> + new_log_buf_len = __LOG_BUF_LEN + cpu_extra;
You could also move the whole thing below the return statement, that way
we can avoid double checking new_log_buf_len. Otherwise looks kinda
weird.
>
> if (!new_log_buf_len)
> return;
Thanks,
Davidlohr
On Fri, Jun 13, 2014 at 2:41 PM, Davidlohr Bueso <[email protected]> wrote:
> On Fri, 2014-06-13 at 11:28 -0700, Luis R. Rodriguez wrote:
>> + /*
>> + * If you set log_buf_len=n kernel parameter LOG_CPU_MIN_BUF_SHIFT will
>> + * be ignored. LOG_CPU_MIN_BUF_SHIFT is a proactive measure for large
>> + * systems. With a LOG_BUF_SHIFT of 18 and LOG_CPU_MIN_BUF_SHIFT 12 at
>> + * we'd require more than 64 CPUs to trigger an increase from the
>> + * default.
>> + */
>> + if (!new_log_buf_len && (cpu_extra > __LOG_BUF_LEN / 2))
> ^ that ! looks wrong.
That check is there so that we ignore the cpu_extra stuff if the
kernel parameter was passed, given that in that case new_log_buf_len
would be set.
> We should be checking for log_buf_len set instead.
When log_buf_len=n is set as a kernel parameter log_buf_len_setup()
will set new_log_buf_len to something, the sanity test to not update
the ring buffer unless the value passed is greater than the default
value is checked by log_buf_len_setup().
>> + new_log_buf_len = __LOG_BUF_LEN + cpu_extra;
>
> You could also move the whole thing below the return statement, that way
> we can avoid double checking new_log_buf_len. Otherwise looks kinda
> weird.
If we did we'd be forcing the kernel parameter to be used to enable
this functionality, but we don't want that.
Luis
I am seeing an issue with a small kernel ring buffer on a small system
with this for some reason though. Please hold off on merging this for
now.
Luis
It turns out we can't use num_possible_cpus() that early on boot, as
it hasn't yet been properly set, nor other helpers, a late setup for
big CPUs is therefore needed in order to use num_possible_cpus().
That also means we can use kzalloc() instead. I'll send a v4 and
I think we should test this a bit so will send as RFT.
Luis