[PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 1608 bytes --]

If we need to increase the number of huge pages, drop caches first
to reduce fragmentation and then check that we actually allocated
as many as we wanted.  Retry once if that doesn't work.

Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
The test always fails for me in a 1 GB VM without this.

Ben.

 tools/testing/selftests/vm/run_vmtests | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests
index 9179ce8..97ed1b2 100755
--- a/tools/testing/selftests/vm/run_vmtests
+++ b/tools/testing/selftests/vm/run_vmtests
@@ -20,13 +20,26 @@ done < /proc/meminfo
 if [ -n "$freepgs" ] && [ -n "$pgsize" ]; then
 	nr_hugepgs=`cat /proc/sys/vm/nr_hugepages`
 	needpgs=`expr $needmem / $pgsize`
-	if [ $freepgs -lt $needpgs ]; then
+	tries=2
+	while [ $tries -gt 0 ] && [ $freepgs -lt $needpgs ]; do
 		lackpgs=$(( $needpgs - $freepgs ))
+		echo 3 > /proc/sys/vm/drop_caches
 		echo $(( $lackpgs + $nr_hugepgs )) > /proc/sys/vm/nr_hugepages
 		if [ $? -ne 0 ]; then
 			echo "Please run this test as root"
 			exit 1
 		fi
+		while read name size unit; do
+			if [ "$name" = "HugePages_Free:" ]; then
+				freepgs=$size
+			fi
+		done < /proc/meminfo
+		tries=$((tries - 1))
+	done
+	if [ $freepgs -lt $needpgs ]; then
+		printf "Not enough huge pages available (%d < %d)\n" \
+		       $freepgs $needpgs
+		exit 1
 	fi
 else
 	echo "no hugetlbfs support in kernel?"

-- 
Ben Hutchings
All extremists should be taken out and shot.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]

Re: [PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[David Rientjes]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1902 bytes --]

On Sat, 31 Oct 2015, Ben Hutchings wrote:

> If we need to increase the number of huge pages, drop caches first
> to reduce fragmentation and then check that we actually allocated
> as many as we wanted.A A Retry once if that doesn't work.
> 
> Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
> ---
> The test always fails for me in a 1 GB VM without this.
> 
> Ben.
> 
> A tools/testing/selftests/vm/run_vmtests | 15 ++++++++++++++-
> A 1 file changed, 14 insertions(+), 1 deletion(-)
> 
> diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests
> index 9179ce8..97ed1b2 100755
> --- a/tools/testing/selftests/vm/run_vmtests
> +++ b/tools/testing/selftests/vm/run_vmtests
> @@ -20,13 +20,26 @@ done < /proc/meminfo
> A if [ -n "$freepgs" ] && [ -n "$pgsize" ]; then
> A 	nr_hugepgs=`cat /proc/sys/vm/nr_hugepages`
> A 	needpgs=`expr $needmem / $pgsize`
> -	if [ $freepgs -lt $needpgs ]; then
> +	tries=2
> +	while [ $tries -gt 0 ] && [ $freepgs -lt $needpgs ]; do
> A 		lackpgs=$(( $needpgs - $freepgs ))
> +		echo 3 > /proc/sys/vm/drop_caches
> A 		echo $(( $lackpgs + $nr_hugepgs )) > /proc/sys/vm/nr_hugepages
> A 		if [ $? -ne 0 ]; then
> A 			echo "Please run this test as root"
> A 			exit 1
> A 		fi
> +		while read name size unit; do
> +			if [ "$name" = "HugePages_Free:" ]; then
> +				freepgs=$size
> +			fi
> +		done < /proc/meminfo
> +		tries=$((tries - 1))
> +	done
> +	if [ $freepgs -lt $needpgs ]; then
> +		printf "Not enough huge pages available (%d < %d)\n" \
> +		A A A A A A A $freepgs $needpgs
> +		exit 1
> A 	fi
> A else
> A 	echo "no hugetlbfs support in kernel?"
> 

I know this patch is in -mm and hasn't been merged by Linus yet, but I'm 
wondering why the multiple /proc/sys/vm/drop_caches is helping?  Would it 
simply suffice to put a sleep in there instead or is drop_caches actually 
doing something useful a second time around?

Re: [PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[Shuah Khan]

On 11/10/2015 01:01 PM, David Rientjes wrote:
> On Sat, 31 Oct 2015, Ben Hutchings wrote:
> 
>> If we need to increase the number of huge pages, drop caches first
>> to reduce fragmentation and then check that we actually allocated
>> as many as we wanted.  Retry once if that doesn't work.
>>
>> Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
>> ---
>> The test always fails for me in a 1 GB VM without this.
>>
>> Ben.
>>
>>  tools/testing/selftests/vm/run_vmtests | 15 ++++++++++++++-
>>  1 file changed, 14 insertions(+), 1 deletion(-)
>>
>> diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests
>> index 9179ce8..97ed1b2 100755
>> --- a/tools/testing/selftests/vm/run_vmtests
>> +++ b/tools/testing/selftests/vm/run_vmtests
>> @@ -20,13 +20,26 @@ done < /proc/meminfo
>>  if [ -n "$freepgs" ] && [ -n "$pgsize" ]; then
>>  	nr_hugepgs=`cat /proc/sys/vm/nr_hugepages`
>>  	needpgs=`expr $needmem / $pgsize`
>> -	if [ $freepgs -lt $needpgs ]; then
>> +	tries=2
>> +	while [ $tries -gt 0 ] && [ $freepgs -lt $needpgs ]; do
>>  		lackpgs=$(( $needpgs - $freepgs ))
>> +		echo 3 > /proc/sys/vm/drop_caches
>>  		echo $(( $lackpgs + $nr_hugepgs )) > /proc/sys/vm/nr_hugepages
>>  		if [ $? -ne 0 ]; then
>>  			echo "Please run this test as root"
>>  			exit 1
>>  		fi
>> +		while read name size unit; do
>> +			if [ "$name" = "HugePages_Free:" ]; then
>> +				freepgs=$size
>> +			fi
>> +		done < /proc/meminfo
>> +		tries=$((tries - 1))
>> +	done
>> +	if [ $freepgs -lt $needpgs ]; then
>> +		printf "Not enough huge pages available (%d < %d)\n" \
>> +		       $freepgs $needpgs
>> +		exit 1
>>  	fi
>>  else
>>  	echo "no hugetlbfs support in kernel?"
>>
> 
> I know this patch is in -mm and hasn't been merged by Linus yet, but I'm 
> wondering why the multiple /proc/sys/vm/drop_caches is helping?  Would it 
> simply suffice to put a sleep in there instead or is drop_caches actually 
> doing something useful a second time around?
> 

I sent this up for merge in my pull request. Adding sleep would increase
test run-time. Something to keep in mind.

thanks,
-- Shuah

-- 
Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
shuahkh@osg.samsung.com | (970) 217-8978

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 2978 bytes --]

On Tue, 2015-11-10 at 12:01 -0800, David Rientjes wrote:
> On Sat, 31 Oct 2015, Ben Hutchings wrote:
> 
> > If we need to increase the number of huge pages, drop caches first
> > to reduce fragmentation and then check that we actually allocated
> > as many as we wanted.  Retry once if that doesn't work.
> > 
> > Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
> > ---
> > The test always fails for me in a 1 GB VM without this.
> > 
> > Ben.
> > 
> >  tools/testing/selftests/vm/run_vmtests | 15 ++++++++++++++-
> >  1 file changed, 14 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tools/testing/selftests/vm/run_vmtests b/tools/testing/selftests/vm/run_vmtests
> > index 9179ce8..97ed1b2 100755
> > --- a/tools/testing/selftests/vm/run_vmtests
> > +++ b/tools/testing/selftests/vm/run_vmtests
> > @@ -20,13 +20,26 @@ done < /proc/meminfo
> >  if [ -n "$freepgs" ] && [ -n "$pgsize" ]; then
> >       nr_hugepgs=`cat /proc/sys/vm/nr_hugepages`
> >       needpgs=`expr $needmem / $pgsize`
> > -     if [ $freepgs -lt $needpgs ]; then
> > +     tries=2
> > +     while [ $tries -gt 0 ] && [ $freepgs -lt $needpgs ]; do
> >               lackpgs=$(( $needpgs - $freepgs ))
> > +             echo 3 > /proc/sys/vm/drop_caches
> >               echo $(( $lackpgs + $nr_hugepgs )) > /proc/sys/vm/nr_hugepages
> >               if [ $? -ne 0 ]; then
> >                       echo "Please run this test as root"
> >                       exit 1
> >               fi
> > +             while read name size unit; do
> > +                     if [ "$name" = "HugePages_Free:" ]; then
> > +                             freepgs=$size
> > +                     fi
> > +             done < /proc/meminfo
> > +             tries=$((tries - 1))
> > +     done
> > +     if [ $freepgs -lt $needpgs ]; then
> > +             printf "Not enough huge pages available (%d < %d)\n" \
> > +                    $freepgs $needpgs
> > +             exit 1
> >       fi
> >  else
> >       echo "no hugetlbfs support in kernel?"
> > 
> 
> I know this patch is in -mm and hasn't been merged by Linus yet, but I'm 
> wondering why the multiple /proc/sys/vm/drop_caches is helping?  Would it 
> simply suffice to put a sleep in there instead or is drop_caches actually 
> doing something useful a second time around?

Initially I just retried setting nr_hugepages up to 10 times, which
wasn't sufficient.  Then I added the drop_caches, and after that
setting nr_hugepages tended to worked first time so I reduced the retry
count.  It might not be necessary to retry at all.

Ben.

-- 
Ben Hutchings
All the simple programs have been written, and all the good names taken.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]

Re: [PATCH selftests 5/6] selftests: vm: Try harder to allocate huge pages

[David Rientjes]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 768 bytes --]

On Tue, 10 Nov 2015, Ben Hutchings wrote:

> > I know this patch is in -mm and hasn't been merged by Linus yet, but I'm 
> > wondering why the multiple /proc/sys/vm/drop_caches is helping?A  Would it 
> > simply suffice to put a sleep in there instead or is drop_caches actually 
> > doing something useful a second time around?
> 
> Initially I just retried setting nr_hugepages up to 10 times, which
> wasn't sufficient. A Then I added the drop_caches, and after that
> setting nr_hugepages tended to worked first time so I reduced the retry
> count. A It might not be necessary to retry at all.
> 

Ok, thanks.  I was just trying to make sure that the additional 
drop_caches wasn't actually required for the test, in which case we would 
have to fix drop_caches :)