Tests showed, that under certain conditions, the summary number of jiffies
spent on softirq/idle, which are counted by system statistics can be even
below 10% of expected value, resulting in false load presentation.
The issue was observed on the quad-core Marvell Armada 8k SoC, whose two
10G ports were bound into L2 bridge. Load was controlled by bidirectional
UDP traffic, produced by a packet generator. Under such condition,
the dominant load is softirq. With 100% single CPU occupation or without
any activity (all CPUs 100% idle), total number of jiffies is 10000 (2500
per each core) in 10s interval. Also with other kind of load this was
true.
However below a saturation threshold it was observed, that with CPU which
was occupied almost by softirqs only, the statistic were awkward. See
the mpstat output:
CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle
all 0.00 0.00 0.13 0.00 0.00 0.55 0.00 0.00 0.00 99.32
0 0.00 0.00 0.00 0.00 0.00 23.08 0.00 0.00 0.00 76.92
1 0.00 0.00 0.40 0.00 0.00 0.00 0.00 0.00 0.00 99.60
2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
Above would mean basically no total load, debug CPU0 occupied in 25%.
Raw statistics, printed every 10s from /proc/stat unveiled a root
cause - summary idle/softirq jiffies on loaded CPU were below 200,
i.e. over 90% samples lost. All problems were gone after enabling
fine granulity IRQ time accounting.
This patch fixes possible wrong statistics processing by enabling
CONFIG_IRQ_TIME_ACCOUNTING for arm64 platfroms, which is by
default done on other architectures, e.g. x86 and arm. Tests
showed no noticeable performance penalty, nor stability impact.
Signed-off-by: Marcin Wojtas <[email protected]>
---
arch/arm64/configs/defconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
index 44423e6..ed51ac6 100644
--- a/arch/arm64/configs/defconfig
+++ b/arch/arm64/configs/defconfig
@@ -3,6 +3,7 @@ CONFIG_POSIX_MQUEUE=y
CONFIG_AUDIT=y
CONFIG_NO_HZ_IDLE=y
CONFIG_HIGH_RES_TIMERS=y
+CONFIG_IRQ_TIME_ACCOUNTING=y
CONFIG_BSD_PROCESS_ACCT=y
CONFIG_BSD_PROCESS_ACCT_V3=y
CONFIG_TASKSTATS=y
--
1.8.3.1
Hi,
(Adding Arnd and Olof)
On lun., juil. 31 2017, Marcin Wojtas <[email protected]> wrote:
> Tests showed, that under certain conditions, the summary number of jiffies
> spent on softirq/idle, which are counted by system statistics can be even
> below 10% of expected value, resulting in false load presentation.
>
> The issue was observed on the quad-core Marvell Armada 8k SoC, whose two
> 10G ports were bound into L2 bridge. Load was controlled by bidirectional
> UDP traffic, produced by a packet generator. Under such condition,
> the dominant load is softirq. With 100% single CPU occupation or without
> any activity (all CPUs 100% idle), total number of jiffies is 10000 (2500
> per each core) in 10s interval. Also with other kind of load this was
> true.
>
> However below a saturation threshold it was observed, that with CPU which
> was occupied almost by softirqs only, the statistic were awkward. See
> the mpstat output:
>
> CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle
> all 0.00 0.00 0.13 0.00 0.00 0.55 0.00 0.00 0.00 99.32
> 0 0.00 0.00 0.00 0.00 0.00 23.08 0.00 0.00 0.00 76.92
> 1 0.00 0.00 0.40 0.00 0.00 0.00 0.00 0.00 0.00 99.60
> 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
> 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
>
> Above would mean basically no total load, debug CPU0 occupied in 25%.
> Raw statistics, printed every 10s from /proc/stat unveiled a root
> cause - summary idle/softirq jiffies on loaded CPU were below 200,
> i.e. over 90% samples lost. All problems were gone after enabling
> fine granulity IRQ time accounting.
>
> This patch fixes possible wrong statistics processing by enabling
> CONFIG_IRQ_TIME_ACCOUNTING for arm64 platfroms, which is by
> default done on other architectures, e.g. x86 and arm. Tests
> showed no noticeable performance penalty, nor stability impact.
Who should take this patch?
I think that all the defconfig under arm64 are merged through the
arm-soc subsystem, but this one is not really specific to a
SoC. However, as it was experimented on an mvebu SoC, if you agree I can
take it.
Thanks,
Gregory
>
> Signed-off-by: Marcin Wojtas <[email protected]>
> ---
> arch/arm64/configs/defconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
> index 44423e6..ed51ac6 100644
> --- a/arch/arm64/configs/defconfig
> +++ b/arch/arm64/configs/defconfig
> @@ -3,6 +3,7 @@ CONFIG_POSIX_MQUEUE=y
> CONFIG_AUDIT=y
> CONFIG_NO_HZ_IDLE=y
> CONFIG_HIGH_RES_TIMERS=y
> +CONFIG_IRQ_TIME_ACCOUNTING=y
> CONFIG_BSD_PROCESS_ACCT=y
> CONFIG_BSD_PROCESS_ACCT_V3=y
> CONFIG_TASKSTATS=y
> --
> 1.8.3.1
>
--
Gregory Clement, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com
On Wed, Aug 02, 2017 at 03:11:43PM +0200, Gregory CLEMENT wrote:
> On lun., juil. 31 2017, Marcin Wojtas <[email protected]> wrote:
> > Tests showed, that under certain conditions, the summary number of jiffies
> > spent on softirq/idle, which are counted by system statistics can be even
> > below 10% of expected value, resulting in false load presentation.
> >
> > The issue was observed on the quad-core Marvell Armada 8k SoC, whose two
> > 10G ports were bound into L2 bridge. Load was controlled by bidirectional
> > UDP traffic, produced by a packet generator. Under such condition,
> > the dominant load is softirq. With 100% single CPU occupation or without
> > any activity (all CPUs 100% idle), total number of jiffies is 10000 (2500
> > per each core) in 10s interval. Also with other kind of load this was
> > true.
> >
> > However below a saturation threshold it was observed, that with CPU which
> > was occupied almost by softirqs only, the statistic were awkward. See
> > the mpstat output:
> >
> > CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle
> > all 0.00 0.00 0.13 0.00 0.00 0.55 0.00 0.00 0.00 99.32
> > 0 0.00 0.00 0.00 0.00 0.00 23.08 0.00 0.00 0.00 76.92
> > 1 0.00 0.00 0.40 0.00 0.00 0.00 0.00 0.00 0.00 99.60
> > 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
> > 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
> >
> > Above would mean basically no total load, debug CPU0 occupied in 25%.
> > Raw statistics, printed every 10s from /proc/stat unveiled a root
> > cause - summary idle/softirq jiffies on loaded CPU were below 200,
> > i.e. over 90% samples lost. All problems were gone after enabling
> > fine granulity IRQ time accounting.
> >
> > This patch fixes possible wrong statistics processing by enabling
> > CONFIG_IRQ_TIME_ACCOUNTING for arm64 platfroms, which is by
> > default done on other architectures, e.g. x86 and arm. Tests
> > showed no noticeable performance penalty, nor stability impact.
>
> Who should take this patch?
>
> I think that all the defconfig under arm64 are merged through the
> arm-soc subsystem, but this one is not really specific to a
> SoC. However, as it was experimented on an mvebu SoC, if you agree I
> can take it.
It's fine by me to go via arm-soc.
--
Catalin
Hi Marcin,
On lun., juil. 31 2017, Marcin Wojtas <[email protected]> wrote:
> Tests showed, that under certain conditions, the summary number of jiffies
> spent on softirq/idle, which are counted by system statistics can be even
> below 10% of expected value, resulting in false load presentation.
>
> The issue was observed on the quad-core Marvell Armada 8k SoC, whose two
> 10G ports were bound into L2 bridge. Load was controlled by bidirectional
> UDP traffic, produced by a packet generator. Under such condition,
> the dominant load is softirq. With 100% single CPU occupation or without
> any activity (all CPUs 100% idle), total number of jiffies is 10000 (2500
> per each core) in 10s interval. Also with other kind of load this was
> true.
>
> However below a saturation threshold it was observed, that with CPU which
> was occupied almost by softirqs only, the statistic were awkward. See
> the mpstat output:
>
> CPU %usr %nice %sys %iowait %irq %soft %steal %guest %gnice %idle
> all 0.00 0.00 0.13 0.00 0.00 0.55 0.00 0.00 0.00 99.32
> 0 0.00 0.00 0.00 0.00 0.00 23.08 0.00 0.00 0.00 76.92
> 1 0.00 0.00 0.40 0.00 0.00 0.00 0.00 0.00 0.00 99.60
> 2 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
> 3 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
>
> Above would mean basically no total load, debug CPU0 occupied in 25%.
> Raw statistics, printed every 10s from /proc/stat unveiled a root
> cause - summary idle/softirq jiffies on loaded CPU were below 200,
> i.e. over 90% samples lost. All problems were gone after enabling
> fine granulity IRQ time accounting.
>
> This patch fixes possible wrong statistics processing by enabling
> CONFIG_IRQ_TIME_ACCOUNTING for arm64 platfroms, which is by
> default done on other architectures, e.g. x86 and arm. Tests
> showed no noticeable performance penalty, nor stability impact.
>
> Signed-off-by: Marcin Wojtas <[email protected]>
Applied on mvebu/arm64
Thanks,
Gregory
> ---
> arch/arm64/configs/defconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig
> index 44423e6..ed51ac6 100644
> --- a/arch/arm64/configs/defconfig
> +++ b/arch/arm64/configs/defconfig
> @@ -3,6 +3,7 @@ CONFIG_POSIX_MQUEUE=y
> CONFIG_AUDIT=y
> CONFIG_NO_HZ_IDLE=y
> CONFIG_HIGH_RES_TIMERS=y
> +CONFIG_IRQ_TIME_ACCOUNTING=y
> CONFIG_BSD_PROCESS_ACCT=y
> CONFIG_BSD_PROCESS_ACCT_V3=y
> CONFIG_TASKSTATS=y
> --
> 1.8.3.1
>
--
Gregory Clement, Free Electrons
Kernel, drivers, real-time and embedded Linux
development, consulting, training and support.
http://free-electrons.com