* [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
@ 2020-08-04 12:30 Colin King
2020-08-04 23:06 ` David Miller
2020-08-05 0:54 ` Eric Dumazet
0 siblings, 2 replies; 7+ messages in thread
From: Colin King @ 2020-08-04 12:30 UTC (permalink / raw)
To: David S . Miller, Shuah Khan, netdev, linux-kselftest
Cc: kernel-janitors, linux-kernel
From: Colin Ian King <colin.king@canonical.com>
The current test will exit with a failure if it cannot set affinity on
specific CPUs which is problematic when running this on single CPU
systems. Add a check for the number of CPUs and skip the test if
the CPU requirement is not met.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
---
tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
index 825ffec85cea..97bc527e1297 100755
--- a/tools/testing/selftests/net/msg_zerocopy.sh
+++ b/tools/testing/selftests/net/msg_zerocopy.sh
@@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
readonly path_sysctl_mem="net.core.optmem_max"
+if [[ $(nproc) -lt 4 ]]; then
+ echo "SKIP: test requires at least 4 CPUs"
+ exit 4
+fi
+
# No arguments: automated test
if [[ "$#" -eq "0" ]]; then
$0 4 tcp -t 1
--
2.27.0
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-04 12:30 [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs Colin King
@ 2020-08-04 23:06 ` David Miller
2020-08-05 0:54 ` Eric Dumazet
1 sibling, 0 replies; 7+ messages in thread
From: David Miller @ 2020-08-04 23:06 UTC (permalink / raw)
To: colin.king, willemb
Cc: shuah, netdev, linux-kselftest, kernel-janitors, linux-kernel
From: Colin King <colin.king@canonical.com>
Date: Tue, 4 Aug 2020 13:30:12 +0100
> From: Colin Ian King <colin.king@canonical.com>
>
> The current test will exit with a failure if it cannot set affinity on
> specific CPUs which is problematic when running this on single CPU
> systems. Add a check for the number of CPUs and skip the test if
> the CPU requirement is not met.
>
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
Willem, please review.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-04 12:30 [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs Colin King
2020-08-04 23:06 ` David Miller
@ 2020-08-05 0:54 ` Eric Dumazet
2020-08-05 8:06 ` Willem de Bruijn
1 sibling, 1 reply; 7+ messages in thread
From: Eric Dumazet @ 2020-08-05 0:54 UTC (permalink / raw)
To: Colin King, David S . Miller, Shuah Khan, netdev, linux-kselftest
Cc: kernel-janitors, linux-kernel
On 8/4/20 5:30 AM, Colin King wrote:
> From: Colin Ian King <colin.king@canonical.com>
>
> The current test will exit with a failure if it cannot set affinity on
> specific CPUs which is problematic when running this on single CPU
> systems. Add a check for the number of CPUs and skip the test if
> the CPU requirement is not met.
>
> Signed-off-by: Colin Ian King <colin.king@canonical.com>
> ---
> tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
> index 825ffec85cea..97bc527e1297 100755
> --- a/tools/testing/selftests/net/msg_zerocopy.sh
> +++ b/tools/testing/selftests/net/msg_zerocopy.sh
> @@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
>
> readonly path_sysctl_mem="net.core.optmem_max"
>
> +if [[ $(nproc) -lt 4 ]]; then
> + echo "SKIP: test requires at least 4 CPUs"
> + exit 4
> +fi
> +
> # No arguments: automated test
> if [[ "$#" -eq "0" ]]; then
> $0 4 tcp -t 1
>
Test explicitly uses CPU 2 and 3, right ?
nproc could be 500, yet cpu 2 or 3 could be offline
# cat /sys/devices/system/cpu/cpu3/online
0
# echo $(nproc)
71
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-05 0:54 ` Eric Dumazet
@ 2020-08-05 8:06 ` Willem de Bruijn
2020-08-05 8:22 ` Colin Ian King
0 siblings, 1 reply; 7+ messages in thread
From: Willem de Bruijn @ 2020-08-05 8:06 UTC (permalink / raw)
To: Eric Dumazet
Cc: Colin King, David S . Miller, Shuah Khan, Network Development,
open list:KERNEL SELFTEST FRAMEWORK, kernel-janitors,
linux-kernel
On Wed, Aug 5, 2020 at 2:54 AM Eric Dumazet <eric.dumazet@gmail.com> wrote:
>
>
>
> On 8/4/20 5:30 AM, Colin King wrote:
> > From: Colin Ian King <colin.king@canonical.com>
> >
> > The current test will exit with a failure if it cannot set affinity on
> > specific CPUs which is problematic when running this on single CPU
> > systems. Add a check for the number of CPUs and skip the test if
> > the CPU requirement is not met.
> >
> > Signed-off-by: Colin Ian King <colin.king@canonical.com>
> > ---
> > tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
> > index 825ffec85cea..97bc527e1297 100755
> > --- a/tools/testing/selftests/net/msg_zerocopy.sh
> > +++ b/tools/testing/selftests/net/msg_zerocopy.sh
> > @@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
> >
> > readonly path_sysctl_mem="net.core.optmem_max"
> >
> > +if [[ $(nproc) -lt 4 ]]; then
> > + echo "SKIP: test requires at least 4 CPUs"
> > + exit 4
> > +fi
> > +
> > # No arguments: automated test
> > if [[ "$#" -eq "0" ]]; then
> > $0 4 tcp -t 1
> >
>
> Test explicitly uses CPU 2 and 3, right ?
>
> nproc could be 500, yet cpu 2 or 3 could be offline
>
> # cat /sys/devices/system/cpu/cpu3/online
> 0
> # echo $(nproc)
> 71
The cpu affinity is only set to bring some stability across runs.
The test does not actually verify that a run with zerocopy is some
factor faster than without, as that factor is hard to choose across
all platforms. As a result the automated run mainly gives code coverage.
It's preferable to always run. And on sched_setaffinity failure log a
message about possible jitter and continue. I can send that patch, if
the approach sounds good.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-05 8:06 ` Willem de Bruijn
@ 2020-08-05 8:22 ` Colin Ian King
2020-08-05 8:44 ` Willem de Bruijn
0 siblings, 1 reply; 7+ messages in thread
From: Colin Ian King @ 2020-08-05 8:22 UTC (permalink / raw)
To: Willem de Bruijn, Eric Dumazet
Cc: David S . Miller, Shuah Khan, Network Development,
open list:KERNEL SELFTEST FRAMEWORK, kernel-janitors,
linux-kernel
On 05/08/2020 09:06, Willem de Bruijn wrote:
> On Wed, Aug 5, 2020 at 2:54 AM Eric Dumazet <eric.dumazet@gmail.com> wrote:
>>
>>
>>
>> On 8/4/20 5:30 AM, Colin King wrote:
>>> From: Colin Ian King <colin.king@canonical.com>
>>>
>>> The current test will exit with a failure if it cannot set affinity on
>>> specific CPUs which is problematic when running this on single CPU
>>> systems. Add a check for the number of CPUs and skip the test if
>>> the CPU requirement is not met.
>>>
>>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
>>> ---
>>> tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
>>> 1 file changed, 5 insertions(+)
>>>
>>> diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
>>> index 825ffec85cea..97bc527e1297 100755
>>> --- a/tools/testing/selftests/net/msg_zerocopy.sh
>>> +++ b/tools/testing/selftests/net/msg_zerocopy.sh
>>> @@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
>>>
>>> readonly path_sysctl_mem="net.core.optmem_max"
>>>
>>> +if [[ $(nproc) -lt 4 ]]; then
>>> + echo "SKIP: test requires at least 4 CPUs"
>>> + exit 4
>>> +fi
>>> +
>>> # No arguments: automated test
>>> if [[ "$#" -eq "0" ]]; then
>>> $0 4 tcp -t 1
>>>
>>
>> Test explicitly uses CPU 2 and 3, right ?
>>
>> nproc could be 500, yet cpu 2 or 3 could be offline
>>
>> # cat /sys/devices/system/cpu/cpu3/online
>> 0
>> # echo $(nproc)
>> 71
>
> The cpu affinity is only set to bring some stability across runs.
>
> The test does not actually verify that a run with zerocopy is some
> factor faster than without, as that factor is hard to choose across
> all platforms. As a result the automated run mainly gives code coverage.
>
> It's preferable to always run. And on sched_setaffinity failure log a
> message about possible jitter and continue. I can send that patch, if
> the approach sounds good.
>
That's sounds preferable to my bad fix for sure :-)
Colin
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-05 8:22 ` Colin Ian King
@ 2020-08-05 8:44 ` Willem de Bruijn
2020-08-05 8:48 ` Colin Ian King
0 siblings, 1 reply; 7+ messages in thread
From: Willem de Bruijn @ 2020-08-05 8:44 UTC (permalink / raw)
To: Colin Ian King
Cc: Willem de Bruijn, Eric Dumazet, David S . Miller, Shuah Khan,
Network Development, open list:KERNEL SELFTEST FRAMEWORK,
kernel-janitors, linux-kernel
On Wed, Aug 5, 2020 at 10:22 AM Colin Ian King <colin.king@canonical.com> wrote:
>
> On 05/08/2020 09:06, Willem de Bruijn wrote:
> > On Wed, Aug 5, 2020 at 2:54 AM Eric Dumazet <eric.dumazet@gmail.com> wrote:
> >>
> >>
> >>
> >> On 8/4/20 5:30 AM, Colin King wrote:
> >>> From: Colin Ian King <colin.king@canonical.com>
> >>>
> >>> The current test will exit with a failure if it cannot set affinity on
> >>> specific CPUs which is problematic when running this on single CPU
> >>> systems. Add a check for the number of CPUs and skip the test if
> >>> the CPU requirement is not met.
> >>>
> >>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
> >>> ---
> >>> tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
> >>> 1 file changed, 5 insertions(+)
> >>>
> >>> diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
> >>> index 825ffec85cea..97bc527e1297 100755
> >>> --- a/tools/testing/selftests/net/msg_zerocopy.sh
> >>> +++ b/tools/testing/selftests/net/msg_zerocopy.sh
> >>> @@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
> >>>
> >>> readonly path_sysctl_mem="net.core.optmem_max"
> >>>
> >>> +if [[ $(nproc) -lt 4 ]]; then
> >>> + echo "SKIP: test requires at least 4 CPUs"
> >>> + exit 4
> >>> +fi
> >>> +
> >>> # No arguments: automated test
> >>> if [[ "$#" -eq "0" ]]; then
> >>> $0 4 tcp -t 1
> >>>
> >>
> >> Test explicitly uses CPU 2 and 3, right ?
> >>
> >> nproc could be 500, yet cpu 2 or 3 could be offline
> >>
> >> # cat /sys/devices/system/cpu/cpu3/online
> >> 0
> >> # echo $(nproc)
> >> 71
> >
> > The cpu affinity is only set to bring some stability across runs.
> >
> > The test does not actually verify that a run with zerocopy is some
> > factor faster than without, as that factor is hard to choose across
> > all platforms. As a result the automated run mainly gives code coverage.
> >
> > It's preferable to always run. And on sched_setaffinity failure log a
> > message about possible jitter and continue. I can send that patch, if
> > the approach sounds good.
> >
> That's sounds preferable to my bad fix for sure :-)
Certainly not a bad fix! Thanks for addressing the issue. Alternative
approach at
http://patchwork.ozlabs.org/project/netdev/patch/20200805084045.1549492-1-willemdebruijn.kernel@gmail.com/
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs
2020-08-05 8:44 ` Willem de Bruijn
@ 2020-08-05 8:48 ` Colin Ian King
0 siblings, 0 replies; 7+ messages in thread
From: Colin Ian King @ 2020-08-05 8:48 UTC (permalink / raw)
To: Willem de Bruijn
Cc: Eric Dumazet, David S . Miller, Shuah Khan, Network Development,
open list:KERNEL SELFTEST FRAMEWORK, kernel-janitors,
linux-kernel
On 05/08/2020 09:44, Willem de Bruijn wrote:
> On Wed, Aug 5, 2020 at 10:22 AM Colin Ian King <colin.king@canonical.com> wrote:
>>
>> On 05/08/2020 09:06, Willem de Bruijn wrote:
>>> On Wed, Aug 5, 2020 at 2:54 AM Eric Dumazet <eric.dumazet@gmail.com> wrote:
>>>>
>>>>
>>>>
>>>> On 8/4/20 5:30 AM, Colin King wrote:
>>>>> From: Colin Ian King <colin.king@canonical.com>
>>>>>
>>>>> The current test will exit with a failure if it cannot set affinity on
>>>>> specific CPUs which is problematic when running this on single CPU
>>>>> systems. Add a check for the number of CPUs and skip the test if
>>>>> the CPU requirement is not met.
>>>>>
>>>>> Signed-off-by: Colin Ian King <colin.king@canonical.com>
>>>>> ---
>>>>> tools/testing/selftests/net/msg_zerocopy.sh | 5 +++++
>>>>> 1 file changed, 5 insertions(+)
>>>>>
>>>>> diff --git a/tools/testing/selftests/net/msg_zerocopy.sh b/tools/testing/selftests/net/msg_zerocopy.sh
>>>>> index 825ffec85cea..97bc527e1297 100755
>>>>> --- a/tools/testing/selftests/net/msg_zerocopy.sh
>>>>> +++ b/tools/testing/selftests/net/msg_zerocopy.sh
>>>>> @@ -21,6 +21,11 @@ readonly DADDR6='fd::2'
>>>>>
>>>>> readonly path_sysctl_mem="net.core.optmem_max"
>>>>>
>>>>> +if [[ $(nproc) -lt 4 ]]; then
>>>>> + echo "SKIP: test requires at least 4 CPUs"
>>>>> + exit 4
>>>>> +fi
>>>>> +
>>>>> # No arguments: automated test
>>>>> if [[ "$#" -eq "0" ]]; then
>>>>> $0 4 tcp -t 1
>>>>>
>>>>
>>>> Test explicitly uses CPU 2 and 3, right ?
>>>>
>>>> nproc could be 500, yet cpu 2 or 3 could be offline
>>>>
>>>> # cat /sys/devices/system/cpu/cpu3/online
>>>> 0
>>>> # echo $(nproc)
>>>> 71
>>>
>>> The cpu affinity is only set to bring some stability across runs.
>>>
>>> The test does not actually verify that a run with zerocopy is some
>>> factor faster than without, as that factor is hard to choose across
>>> all platforms. As a result the automated run mainly gives code coverage.
>>>
>>> It's preferable to always run. And on sched_setaffinity failure log a
>>> message about possible jitter and continue. I can send that patch, if
>>> the approach sounds good.
>>>
>> That's sounds preferable to my bad fix for sure :-)
>
> Certainly not a bad fix! Thanks for addressing the issue. Alternative
> approach at
>
> http://patchwork.ozlabs.org/project/netdev/patch/20200805084045.1549492-1-willemdebruijn.kernel@gmail.com/
>
Thanks, you solution is good for my testing requirements
Acked-by: Colin Ian King <colin.king@canonical.com>
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2020-08-05 8:48 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-04 12:30 [PATCH] selftests/net: skip msg_zerocopy test if we have less than 4 CPUs Colin King
2020-08-04 23:06 ` David Miller
2020-08-05 0:54 ` Eric Dumazet
2020-08-05 8:06 ` Willem de Bruijn
2020-08-05 8:22 ` Colin Ian King
2020-08-05 8:44 ` Willem de Bruijn
2020-08-05 8:48 ` Colin Ian King
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).