Raspberry Pi - Page 4

ET_ · 2016-10-27, 11:11

Quote:

Originally Posted by ldesnogu

That page is obsolete, look at the date: "Last modified: 2014-08-22".

64-bit is now supported:
https://gmplib.org/repo/gmp/file/tip/mpn/arm64

OTOH it seems Cortex-A53 is slower than A15:
https://gmplib.org/gmpbench.html

From a Raspberry point of view, A53 looks faster

And if I read the bench correctly, it looks like the A53 scores as half Athlon 1090...

Thank you for the links!

ET_ · 2016-12-03, 19:11

BUMP

I found the following 64bit images for out PI 3:

https://www.kraxel.org/blog/2016/06/...ages-uploaded/

http://www.cnx-software.com/2016/03/...-instructions/

Did anybody test them? I'm wondering if they are worth the time neded to study and optimize them (especially Fedora 24) from our computational point of view.

I suppose Paul is happily ECMing at 32 bit at the moment...

xilman · 2016-12-03, 19:31

Quote:

Originally Posted by ET_

I suppose Paul is happily ECMing at 32 bit at the moment...

Yes. My current stock of round tuits are all being used on image processing.

Perhaps I'll have a go at getting Gentoo running on the Pi next year.

ET_ · 2016-12-03, 19:42

Quote:

Originally Posted by xilman

Yes. My current stock of round tuits are all being used on image processing.

Perhaps I'll have a go at getting Gentoo running on the Pi next year.

I hope, by then, to have news about the Fedora image. I will keep posting here, we will exchange our experiences.

EdH · 2016-12-19, 03:50

I have a Pi 3 that I'm using to assign ecm tasks to several machines and run poly gathering while the others run their ecm.

For a while from initial powerup, everything works fine and assignments are quick. However, after a while I find the assignments have slowed down and many of the clients are sitting waiting. As I troubleshoot further I find that the scp calls are taking 5-10+ seconds to complete. If I try to go in via ssh or remote vnc, they take a while to connect as well. This is independent of whether the machine is running any poly gathering.

If I reboot, all is fine again for a while, but then slows again.

Any suggestions as to why/where the slowdown may be?

Thanks...

Mark Rose · 2016-12-19, 11:23

What is the memory usage?

cat /proc/meminfo

EdH · 2016-12-19, 14:17

Quote:

Originally Posted by Mark Rose

What is the memory usage?

cat /proc/meminfo

Code:

MemTotal:         948012 kB
MemFree:          126800 kB
MemAvailable:     542820 kB
Buffers:           89284 kB
Cached:           359292 kB
SwapCached:            0 kB
Active:           512980 kB
Inactive:         216648 kB
Active(anon):     282380 kB
Inactive(anon):    59636 kB
Active(file):     230600 kB
Inactive(file):   157012 kB
Unevictable:           0 kB
Mlocked:               0 kB
SwapTotal:             0 kB
SwapFree:              0 kB
Dirty:                48 kB
Writeback:             0 kB
AnonPages:        281116 kB
Mapped:            49772 kB
Shmem:             60908 kB
Slab:              62948 kB
SReclaimable:      43000 kB
SUnreclaim:        19948 kB
KernelStack:        3720 kB
PageTables:         9704 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:      474004 kB
Committed_AS:    2098684 kB
VmallocTotal:    1105920 kB
VmallocUsed:        1768 kB
VmallocChunk:     877984 kB
CmaTotal:           8192 kB
CmaFree:            1672 kB

It is currently idle, waiting for a command (via inotifywait) to start issuing new assignments. In this "idle" state, it took 28 seconds to respond to an ssh connect. After the ssh connect, the new memory usage shows:

Code:

MemTotal:         948012 kB
MemFree:          123580 kB
MemAvailable:     540652 kB
Buffers:           89724 kB
Cached:           359840 kB
SwapCached:            0 kB
Active:           515820 kB
Inactive:         216980 kB
Active(anon):     284512 kB
Inactive(anon):    59640 kB
Active(file):     231308 kB
Inactive(file):   157340 kB
Unevictable:           0 kB
Mlocked:               0 kB
SwapTotal:             0 kB
SwapFree:              0 kB
Dirty:                24 kB
Writeback:             0 kB
AnonPages:        283236 kB
Mapped:            50884 kB
Shmem:             60916 kB
Slab:              62936 kB
SReclaimable:      43016 kB
SUnreclaim:        19920 kB
KernelStack:        3744 kB
PageTables:         9860 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:      474004 kB
Committed_AS:    2104412 kB
VmallocTotal:    1105920 kB
VmallocUsed:        1800 kB
VmallocChunk:     877952 kB
CmaTotal:           8192 kB
CmaFree:            1672 kB

EdH · 2016-12-19, 15:42

So, after a fresh start, I get:

Code:

MemTotal:         948012 kB
MemFree:          413644 kB
MemAvailable:     662528 kB
Buffers:           34004 kB
Cached:           272300 kB
SwapCached:            0 kB
Active:           320696 kB
Inactive:         163936 kB
Active(anon):     179376 kB
Inactive(anon):    56652 kB
Active(file):     141320 kB
Inactive(file):   107284 kB
Unevictable:           0 kB
Mlocked:               0 kB
SwapTotal:             0 kB
SwapFree:              0 kB
Dirty:               116 kB
Writeback:             0 kB
AnonPages:        178464 kB
Mapped:           103248 kB
Shmem:             57704 kB
Slab:              28156 kB
SReclaimable:      14872 kB
SUnreclaim:        13284 kB
KernelStack:        2776 kB
PageTables:         5724 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:      474004 kB
Committed_AS:    1889796 kB
VmallocTotal:    1105920 kB
VmallocUsed:        1768 kB
VmallocChunk:     891712 kB
CmaTotal:           8192 kB
CmaFree:            3736 kB

I guess that's showing me that there isn't enough memory to run well and I have no way up adding memory to the RP3. I guess that means either accepting the lag or using a different machine...

Thanks!

ldesnogu · 2016-12-19, 15:46

Before drawing conclusions, you'd rather run the 'top' command to see what processes are consuming CPU time.

Mark Rose · 2016-12-19, 17:31

Quote:

Originally Posted by EdH

I guess that's showing me that there isn't enough memory to run well and I have no way up adding memory to the RP3. I guess that means either accepting the lag or using a different machine...

Thanks!

Actually, it shows you have tons of memory available. That's not the problem. I have in the past seen systems that behave similarly when memory is low and the kernel has to constantly page programs back into RAM, but that's not happening here.

My next guesses would be the following:

1. CPU overheating from a process using 100% all the time.
2. A bug causing massive interrupts

To check for the first, run `top` (obviously)

To check for the second, run `watch -d=cumulative cat /proc/interrupts` and see if there's anything in particular that sticks out. Compare the difference in output between a fresh boot and when the problem has returned.

EdH · 2016-12-19, 20:55

This will probably be long winded (fingered):

I did some "top" runs and watches and saw something I hadn't noticed before. All of the first three are at idle waiting to start assignments. The fist and second tops snapshots were showing up frequently:

Code:

Tasks: 327 total,  22 running, 305 sleeping,   0 stopped,   0 zombie
%Cpu(s): 41.1 us, 11.1 sy,  0.0 ni, 47.7 id,  0.0 wa,  0.0 hi,  0.1 si,  0.0 st
KiB Mem :   948012 total,   270724 free,   299220 used,   378068 buff/cache
KiB Swap:        0 total,        0 free,        0 used.   557892 avail Mem 

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND     
16983 root      20   0    9464   7504   3868 R   9.9  0.8   0:00.30 lsb_release 
16948 root      20   0    9492   7448   3812 R   9.6  0.8   0:00.29 lsb_release 
17038 root      20   0    9496   7408   3768 R   9.6  0.8   0:00.29 lsb_release 
16999 root      20   0    9492   7408   3768 R   9.2  0.8   0:00.28 lsb_release 
17059 root      20   0    9496   7448   3804 R   9.2  0.8   0:00.28 lsb_release 
17046 root      20   0    9496   7372   3732 R   8.3  0.8   0:00.25 lsb_release 
17070 root      20   0    9460   7404   3772 R   8.3  0.8   0:00.25 lsb_release 
16987 root      20   0    9404   7220   3744 R   7.9  0.8   0:00.24 lsb_release 
17042 root      20   0    9412   7260   3780 R   7.9  0.8   0:00.24 lsb_release 
17067 root      20   0    9412   7284   3808 R   7.9  0.8   0:00.24 lsb_release 
17025 root      20   0    9404   7284   3812 R   7.6  0.8   0:00.23 lsb_release 
17062 root      20   0    9404   7356   3884 R   7.6  0.8   0:00.23 lsb_release 
17050 root      20   0    9372   7048   3748 R   7.3  0.7   0:00.22 lsb_release 
17017 root      20   0    9116   7036   3736 R   6.9  0.7   0:00.21 lsb_release 
17024 root      20   0    9156   6968   3752 R   6.9  0.7   0:00.21 lsb_release 
17032 root      20   0    9148   7040   3832 R   6.9  0.7   0:00.21 lsb_release 
17034 root      20   0    9156   6976   3760 R   6.9  0.7   0:00.21 lsb_release

Code:

Tasks: 203 total,   1 running, 202 sleeping,   0 stopped,   0 zombie
%Cpu(s): 11.1 us,  3.8 sy,  0.0 ni, 85.0 id,  0.0 wa,  0.0 hi,  0.2 si,  0.0 st
KiB Mem :   948012 total,   353200 free,   216728 used,   378084 buff/cache
KiB Swap:        0 total,        0 free,        0 used.   640668 avail Mem 

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND     
  685 root      20   0  122692  55468  29292 S   3.3  5.9  12:22.96 Xorg        
 3922 mathpi    20   0   29336  19920  13036 S   3.3  2.1   6:45.80 x11vnc      
22462 root      20   0   13496   5036   4220 S   3.0  0.5   0:00.09 sshd        
22464 root      20   0   13496   4972   4156 S   2.6  0.5   0:00.08 sshd        
22466 root      20   0   13496   4944   4128 S   2.3  0.5   0:00.07 sshd        
22488 root      20   0   13496   5024   4208 S   1.6  0.5   0:00.05 sshd        
22489 root      20   0   13496   4976   4160 S   1.6  0.5   0:00.05 sshd        
22490 root      20   0   13496   4988   4172 S   1.6  0.5   0:00.05 sshd        
22497 root      20   0   13496   4992   4176 S   1.6  0.5   0:00.05 sshd        
22498 root      20   0   13496   5068   4252 S   1.6  0.5   0:00.05 sshd        
22499 root      20   0   13496   5044   4228 S   1.6  0.5   0:00.05 sshd        
22500 root      20   0   13496   5004   4188 S   1.6  0.5   0:00.05 sshd        
22502 root      20   0   13496   4980   4164 S   1.6  0.5   0:00.05 sshd        
22503 root      20   0   13496   5060   4244 S   1.6  0.5   0:00.05 sshd        
22504 root      20   0   13496   5024   4208 S   1.6  0.5   0:00.05 sshd        
22521 root      20   0   13496   5020   4204 S   1.6  0.5   0:00.05 sshd        
22486 root      20   0   13496   5000   4184 S   1.3  0.5   0:00.04 sshd

This third snapshot is with the machine freshly started and sitting at idle:

Code:

Tasks: 183 total,   7 running, 175 sleeping,   0 stopped,   1 zombie
%Cpu(s): 33.5 us,  8.0 sy,  0.0 ni, 58.3 id,  0.1 wa,  0.0 hi,  0.1 si,  0.0 st
KiB Mem :   948012 total,   437740 free,   196088 used,   314184 buff/cache
KiB Swap:        0 total,        0 free,        0 used.   684412 avail Mem 

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND     
 2852 mathpi    20   0   45320   5240   4648 S  18.3  0.6   0:01.13 gnome-keyr+ 
  654 root      20   0   36180   6068   4004 R  17.9  0.6   2:04.02 polkitd     
  516 message+  20   0    5884   3432   2308 R  10.3  0.4   1:23.09 dbus-daemon 
  548 root      20   0   84264  11776  10092 R   7.0  1.2   0:48.60 NetworkMan+ 
28302 root      20   0    8696   6228   3524 R   5.3  0.7   0:00.16 lsb_release 
  691 root      20   0  118096  49700  27608 S   4.3  5.2   0:10.18 Xorg        
    1 root      20   0   22964   3908   2644 S   3.0  0.4   0:33.22 systemd     
 3584 mathpi    20   0   28948  17808  11532 S   3.0  1.9   0:05.20 x11vnc      
28242 root      20   0   13496   5112   4292 S   2.7  0.5   0:00.08 sshd        
28276 root      20   0   13496   5056   4240 S   1.7  0.5   0:00.05 sshd        
28301 root      20   0   13496   5056   4240 S   1.7  0.5   0:00.05 sshd        
   96 root      20   0    9684   3436   3044 S   1.0  0.4   0:09.99 systemd-jo+ 
  553 root      20   0    3420   2144   1820 S   1.0  0.2   0:07.43 systemd-lo+ 
 3566 mathpi    20   0   40008  11544  10204 S   1.0  1.2   0:06.33 polkit-mat+ 
 4834 mathpi    20   0   93276  18508  15244 S   1.0  2.0   0:02.97 clock-appl+ 
 3628 mathpi    20   0  183032  41676  34856 S   0.7  4.4   0:06.12 nm-applet   
 3754 root      20   0   76572   7020   6208 S   0.7  0.7   0:01.69 upowerd

And, this fourth snapshot is with the assignments running as well as msieve doing poly searching:

Code:

top - 15:27:13 up 10 min,  4 users,  load average: 4.74, 5.13, 3.08
Tasks: 173 total,   4 running, 169 sleeping,   0 stopped,   0 zombie
%Cpu(s): 26.7 us,  0.4 sy,  0.0 ni, 72.5 id,  0.3 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem :   948012 total,   427984 free,   191180 used,   328848 buff/cache
KiB Swap:        0 total,        0 free,        0 used.   676056 avail Mem 

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND     
12173 mathpi    20   0    4316   2792   2152 R 100.0  0.3   1:00.65 msieve      
  691 root      20   0  118128  50920  28796 R   4.3  5.4   0:19.37 Xorg        
 3577 mathpi    20   0  229220  23340  18420 S   2.0  2.5   0:05.96 mate-termi+ 
   96 root      20   0    9684   2120   1728 S   0.3  0.2   0:13.19 systemd-jo+ 
 3332 mathpi    20   0    7188   2676   2148 R   0.3  0.3   0:00.90 top         
 4834 mathpi    20   0   93276  18508  15244 R   0.3  2.0   0:04.00 clock-appl+ 
11701 mathpi    20   0    4376   2024   1836 S   0.3  0.2   0:00.12 bash        
    1 root      20   0   22964   3908   2644 S   0.0  0.4   0:43.28 systemd     
    2 root      20   0       0      0      0 S   0.0  0.0   0:00.00 kthreadd    
    3 root      20   0       0      0      0 S   0.0  0.0   0:00.75 ksoftirqd/0 
    4 root      20   0       0      0      0 S   0.0  0.0   0:00.01 kworker/0:0 
    5 root       0 -20       0      0      0 S   0.0  0.0   0:00.00 kworker/0:+ 
    6 root      20   0       0      0      0 S   0.0  0.0   0:00.30 kworker/u8+ 
    7 root      20   0       0      0      0 S   0.0  0.0   0:01.56 rcu_sched   
    8 root      20   0       0      0      0 S   0.0  0.0   0:00.00 rcu_bh      
    9 root      rt   0       0      0      0 S   0.0  0.0   0:00.30 migration/0 
   10 root      rt   0       0      0      0 S   0.0  0.0   0:00.30 migration/1

The two attachments are the watches. T1 is with the delay and T2, without. I see a lot of big numbers in T1, but don't know how to interpret them.

In thinking about my scripts, I have them calling each other. Is this possibly a result of calling scripts not ending when they call the next one? Maybe I have a circular script problem... Off to try something...

Thanks...

2016-12-19, 03:50	#38
EdH "Ed Hall" Dec 2009 Adirondack Mtns 2²×13×107 Posts	Network Slowdown for ssh and scp I have a Pi 3 that I'm using to assign ecm tasks to several machines and run poly gathering while the others run their ecm. For a while from initial powerup, everything works fine and assignments are quick. However, after a while I find the assignments have slowed down and many of the clients are sitting waiting. As I troubleshoot further I find that the scp calls are taking 5-10+ seconds to complete. If I try to go in via ssh or remote vnc, they take a while to connect as well. This is independent of whether the machine is running any poly gathering. If I reboot, all is fine again for a while, but then slows again. Any suggestions as to why/where the slowdown may be? Thanks...

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
raspberry pi3 and srxsieve speed	pepi37	Hardware	6	2018-03-20 18:01
Runs Prime95 on Raspberry Pi	primeawesome	Hardware	6	2018-02-14 08:19
Which SIMD flag to use for Raspberry Pi	BrainStone	Mlucas	14	2017-11-19 00:59
Raspberry Pi	lavalamp	Hobbies	10	2017-08-16 00:37
Raspberry Pi	sloppyonefoot	Software	1	2017-07-02 08:48

2016-12-03, 19:11	#35
ET_ Banned "Luigi" Aug 2002 Team Italia 5·7·139 Posts	BUMP I found the following 64bit images for out PI 3: https://www.kraxel.org/blog/2016/06/...ages-uploaded/ http://www.cnx-software.com/2016/03/...-instructions/ Did anybody test them? I'm wondering if they are worth the time neded to study and optimize them (especially Fedora 24) from our computational point of view. I suppose Paul is happily ECMing at 32 bit at the moment...

2016-12-19, 11:23	#39
Mark Rose "/X\(‘-‘)/X\" Jan 2013 https://pedan.tech/ 2⁴·199 Posts	What is the memory usage? cat /proc/meminfo

2016-12-19, 15:46	#42
ldesnogu Jan 2008 France 255₁₆ Posts	Before drawing conclusions, you'd rather run the 'top' command to see what processes are consuming CPU time.