Welcome! Log In Create A New Profile

Advanced

Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?

Posted by dkyyz 
Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 09:14AM
Hello World,

Recently I just repurposed my Pogoplug Mobile. It has been running asterisk for years without problems. And for a fresh start, I reload the SD card with Debian 8 Jessie...

[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] Initializing cgroup subsys cpuset
[    0.000000] Initializing cgroup subsys cpu
[    0.000000] Initializing cgroup subsys cpuacct
[    0.000000] Linux version 3.14.0-kirkwood-tld-1 (root@tldDebian) (gcc version 4.6.3 (Debian 4.6.3-14) ) #1 PREEMPT Tue Apr 1 22:54:58 PDT 2014
[    0.000000] CPU: Feroceon 88FR131 [56251311] revision 1 (ARMv5TE), cr=00053977
[    0.000000] CPU: VIVT data cache, VIVT instruction cache
[    0.000000] Machine: Marvell SheevaPlug Reference Board
[    0.000000] Memory policy: Data cache writeback
[    0.000000] On node 0 totalpages: 32768
[    0.000000] free_area_init_node: node 0, pgdat c06d7f18, node_mem_map c7efb000
[    0.000000]   DMA zone: 256 pages used for memmap
[    0.000000]   DMA zone: 0 pages reserved
[    0.000000]   DMA zone: 32768 pages, LIFO batch:7
[    0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
[    0.000000] pcpu-alloc: [0] 0 
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.  Total pages: 32512
[    0.000000] Kernel command line: console=ttyS0,115200 root=LABEL=rootfs rootdelay=10 mtdparts=orion_nand:1M(u-boot),4M(uImage),32M(rootfs),-(data)
[    0.000000] PID hash table entries: 512 (order: -1, 2048 bytes)
[    0.000000] Dentry cache hash table entries: 16384 (order: 4, 65536 bytes)
[    0.000000] Inode-cache hash table entries: 8192 (order: 3, 32768 bytes)
[    0.000000] Memory: 116244K/131072K available (4779K kernel code, 352K rwdata, 1644K rodata, 213K init, 424K bss, 14828K reserved)
[    0.000000] Virtual kernel memory layout:
[    0.000000]     vector  : 0xffff0000 - 0xffff1000   (   4 kB)
[    0.000000]     fixmap  : 0xfff00000 - 0xfffe0000   ( 896 kB)
[    0.000000]     vmalloc : 0xc8800000 - 0xff000000   ( 872 MB)
[    0.000000]     lowmem  : 0xc0000000 - 0xc8000000   ( 128 MB)
[    0.000000]     modules : 0xbf000000 - 0xc0000000   (  16 MB)
[    0.000000]       .text : 0xc0008000 - 0xc064df14   (6424 kB)
[    0.000000]       .init : 0xc064e000 - 0xc0683444   ( 214 kB)
[    0.000000]       .data : 0xc0684000 - 0xc06dc3c4   ( 353 kB)
[    0.000000]        .bss : 0xc06dc3c4 - 0xc074665c   ( 425 kB)
[    0.000000] Preemptible hierarchical RCU implementation.
[    0.000000] NR_IRQS:114
[    0.000018] sched_clock: 32 bits at 166MHz, resolution 6ns, wraps every 25769803770ns
[   22.806516] Console: colour dummy device 80x30
[   22.806564] Calibrating delay loop... 795.44 BogoMIPS (lpj=3977216)

Notice the BogoMIPS was ~795

but I am not sure since when (I did updated UBoot as well) since I rarely pay much attention to it, now I am running Debian 11 Bullseye, the BogoMIPS dropped to 333.33!

[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] Linux version 5.13.6-kirkwood-tld-1 (root@tldDebian) (gcc (Debian 8.3.0-6) 8.3.0, GNU ld (GNU Binutils for Debian) 2.31.1) #1.0 PREEMPT Sat Jul 31 22:10:39 PDT 2021
[    0.000000] CPU: Feroceon 88FR131 [56251311] revision 1 (ARMv5TE), cr=0005397f
[    0.000000] CPU: VIVT data cache, VIVT instruction cache
[    0.000000] OF: fdt: Machine model: Pogoplug v4
[    0.000000] Memory policy: Data cache writeback
[    0.000000] Zone ranges:
[    0.000000]   Normal   [mem 0x0000000000000000-0x0000000007ffffff]
[    0.000000] Movable zone start for each node
[    0.000000] Early memory node ranges
[    0.000000]   node   0: [mem 0x0000000000000000-0x0000000007ffffff]
[    0.000000] Initmem setup node 0 [mem 0x0000000000000000-0x0000000007ffffff]
[    0.000000] On node 0 totalpages: 32768
[    0.000000]   Normal zone: 256 pages used for memmap
[    0.000000]   Normal zone: 0 pages reserved
[    0.000000]   Normal zone: 32768 pages, LIFO batch:7
[    0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768
[    0.000000] pcpu-alloc: [0] 0 
[    0.000000] Built 1 zonelists, mobility grouping on.  Total pages: 32512
[    0.000000] Kernel command line: console=ttyS0,115200 root=LABEL=rootfs rootdelay=10 mtdparts=orion_nand:1M(u-boot),4M(uImage),32M(rootfs),-(data)
[    0.000000] Dentry cache hash table entries: 16384 (order: 4, 65536 bytes, linear)
[    0.000000] Inode-cache hash table entries: 8192 (order: 3, 32768 bytes, linear)
[    0.000000] mem auto-init: stack:off, heap alloc:on, heap free:off
[    0.000000] Memory: 101108K/131072K available (11264K kernel code, 1553K rwdata, 4272K rodata, 1024K init, 309K bss, 29964K reserved, 0K cma-reserved)
[    0.000000] random: get_random_u32 called from ____cache_alloc+0x408/0x7b4 with crng_init=0
[    0.000000] trace event string verifier disabled
[    0.000000] rcu: Preemptible hierarchical RCU implementation.
[    0.000000] 	Trampoline variant of Tasks RCU enabled.
[    0.000000] 	Tracing variant of Tasks RCU enabled.
[    0.000000] rcu: RCU calculated value of scheduler-enlistment delay is 10 jiffies.
[    0.000000] NR_IRQS: 16, nr_irqs: 16, preallocated irqs: 16
[    0.000000] clocksource: orion_clocksource: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 11467562657 ns
[    0.000003] sched_clock: 32 bits at 166MHz, resolution 6ns, wraps every 12884901885ns
[    0.000090] Switching to timer-based delay loop, resolution 6ns
[    0.001398] Console: colour dummy device 80x30
[    0.001513] Calibrating delay loop (skipped), value calculated using timer frequency.. 333.33 BogoMIPS (lpj=1666666)

So I did a bit of Googling since I thought it was because a different version of Debian, but from the BogoMIPS FAQ
Quote

No, the BogoMips algorithm, contrarily to popular believe, did NOT change over the various kernel versions. Furthermore, it is intrinsically the same for all CPUs of all makes.

Now, this really puzzles me... And some users even said they get ~ 1000-1500 BogoMIPS with the same processor as PogoPlug Mobile. Given there are a few versions, if I remember correctly were 1.0, 1.2 & 1.5GHz (read it last night but can't find the website now) at 333.33 BogoMIPS is really running much slower than it supposes to be.

Also notice the boot messages appear differently in dmesg between version, specifically in Jessie, line 2-4 seems to be doing some kind of init and there is no such messages on Bullseye.

Any ideas? Suggestions?

Care to share your Pogoplug mobile BogoMIPS and Linux version info here? Thx
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 09:15AM
And a follow up, just run
lscpu
Architecture:                    armv5tel
Byte Order:                      Little Endian
CPU(s):                          1
On-line CPU(s) list:             0
Thread(s) per core:              1
Core(s) per socket:              1
Socket(s):                       1
Vendor ID:                       Marvell
Model:                           1
Model name:                      Feroceon 88FR131
Stepping:                        0x2
CPU max MHz:                     800.0000
CPU min MHz:                     200.0000
BogoMIPS:                        333.33
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Not affected
Vulnerability Spectre v1:        Not affected
Vulnerability Spectre v2:        Not affected
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected
Flags:                           swp half thumb fastmult edsp

cpufreq-info
cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009
Report errors and bugs to cpufreq@vger.kernel.org, please.
analyzing CPU 0:
  driver: kirkwood-cpufreq
  CPUs which run at the same hardware frequency: 0
  CPUs which need to have their frequency coordinated by software: 0
  maximum transition latency: 5.0 us.
  hardware limits: 200 MHz - 800 MHz
  available frequency steps: 800 MHz, 200 MHz
  available cpufreq governors: conservative, userspace, ondemand, powersave, performance
  current policy: frequency should be within 200 MHz and 800 MHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.
  current CPU frequency is 800 MHz (asserted by call to hardware).



Edited 1 time(s). Last edit at 06/13/2022 10:36AM by dkyyz.
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 10:28AM
Another noticeable different between the two dmesg log

Jessie
[    0.000000] NR_IRQS:114
[    0.000018] sched_clock: 32 bits at 166MHz, resolution 6ns, wraps every 25769803770ns

Bullseye
[    0.000000] NR_IRQS: 16, nr_irqs: 16, preallocated irqs: 16
[    0.000000] clocksource: orion_clocksource: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 11467562657 ns
[    0.000003] sched_clock: 32 bits at 166MHz, resolution 6ns, wraps every 12884901885ns

What's NR_IRQS? It was changed from 114 to 16?? Also the "sched_clock" line, "wraps every" value was shorten by half in Bullseye.
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 10:56AM
Just don't care about it... Pretend that it is no use. That's my opinion. :-)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 04:47PM
dkyyz,

Don't worry about it. BogoMIPS has been misleading since after Debian Jessie (at one point, Linus even wanted to remove BogoMIPS from the log).

The reason is the calculated BogoMIPS is oftenly wrong on multicore processors.

And also as you might have known, cpufreq-info will only output the current frequency at the time you run it. With the governor "ondemand", this box might slow down to 200 MHz to save electricity usage. And when you do something, it will be pumped up to full speed. So we don't know that has messed up the BogoMIPS calculation during boot.

I had a BogoMIPS calculation tool (with SMP taken into account), not sure where it is in my boxes. If I can find it I will post here.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 05:02PM
Quote

And some users even said they get ~ 1000-1500 BogoMIPS with the same processor as PogoPlug Mobile. Given there are a few versions, if I remember correctly were 1.0, 1.2 & 1.5GHz (read it last night but can't find the website now)

And 1.0, 1.2 & 1.5GHz are just wrong numbers. There is no Pogo Mobile faster than 800 MHz. Unless they overclocked it (possible to do).

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 07:36PM
The Smartenit Harmony P2 which was a re-labeled Pogoplug Mobile with a Jennic Zigbee card attached to the serial pins was overclocked to 1 GHz.
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 09:11PM
bodhi Wrote:
-------------------------------------------------------
>
Quote

And some users even said they get ~
> 1000-1500 BogoMIPS with the same processor as
> PogoPlug Mobile. Given there are a few versions,
> if I remember correctly were 1.0, 1.2 & 1.5GHz
> (read it last night but can't find the website
> now)
>
> And 1.0, 1.2 & 1.5GHz are just wrong numbers.
> There is no Pogo Mobile faster than 800 MHz.
> Unless they overclocked it (possible to do).

Oh, those faster ones (esp 1.2 & 1.5GHz) were not Pogoplug but in some NAS. It's the same processor but running at a higher speed. I wish I could find that info now.

Anyways, it sounds like this is a 'known' issue (and I did find other posts with dmesg showing 333.33 BogoMIPS on newer kernel and 700+ BogoMIPS on older kernel/older posts, 900+ on the 'overclocked' pogoplug under older kernel as well)

I guess it's all good, just a false alarm. :)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 09:48PM
> Oh, those faster ones (esp 1.2 & 1.5GHz) were not
> Pogoplug but in some NAS. It's the same processor
> but running at a higher speed. I wish I could find
> that info now.
>

It is rare to find a 88f6192 SoC faster than 1Ghz (88f6192 is the SoC in the Pogo V4 and Mobile). I think I only have seen 1 board at 1Ghz, but cant recall which.

Those 1.2Ghz or 1.5Ghz are probably 88f6281 SoC like the Dockstar and many other Kirkwoods NAS.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 13, 2022 11:05PM
> The
> Smartenit
> Harmony P2
which was a re-labeled Pogoplug
> Mobile with a Jennic Zigbee card attached to the
> serial pins was overclocked to 1 GHz.

Thanks Ray! I did not know such as a box exists. Are they really overclocked? They might have a better batch of the 6192 SoC chips from Marvell.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 15, 2022 11:54PM
They really are just Pogoplug Mobile devices with the Zigbee module soldered to the serial pins. I believe the overclocking to 1 GHz was done by jumping some pins on the serial. I have a box full of these devices, if you'd like one just PM me.

Ray
Re: Pogoplug Mobile BogoMIPS dropped in half from Jessie -> Bullseye?
June 16, 2022 02:11AM
> They really are just Pogoplug Mobile devices with
> the Zigbee module soldered to the serial pins. I
> believe the overclocking to 1 GHz was done by
> jumping some pins on the serial.

Interesting! yeah, the overclocking is done after market. I'm just not aware of anybody is actually doing that and selling a product.


> I have a box
> full of these devices, if you'd like one just PM
> me.
>

Thanks Ray! But I do have a couple Pogo Mobiles in my farm.

-bodhi
===========================
Forum Wiki
bodhi's corner (buy bodhi a beer)
Author:

Subject:


Spam prevention:
Please, enter the code that you see below in the input field. This is for blocking bots that try to post this form automatically. If the code is hard to read, then just try to guess it right. If you enter the wrong code, a new image is created and you get another chance to enter it right.
Message: