gkreidl
Posts: 5327
Joined: Thu Jan 26, 2012 1:07 pm
Location: Germany

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 9:25 am

From the official RPi 2 launch statement:
At launch, we are using the same ARMv6 Raspbian userland on both Raspberry Pi 1 and 2; over the next few months we will investigate whether we can obtain higher performance from regular ARMv7 Debian, or whether we can selectively replace a small number of libraries to get the best of both worlds
Minimal Kiosk Browser (kweb)
Slim, fast webkit browser with support for audio+video+playlists+youtube+pdf+download
Optional fullscreen kiosk mode and command interface for embedded applications
Includes omxplayerGUI, an X front end for omxplayer

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 18211
Joined: Sat Jul 30, 2011 7:41 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 9:58 am

[IDC]Dragon wrote:I have a question about the Pi2 (probably it has been raised elsewhere):

I'm surprised that the existing binaries run on it. Isn't it that the Pi1 with its ARMv6 hard float was a rather unusual beast, requiring a specially compiled distro like Raspian? Well, obviously the Pi2 is somewhat downwards compatible, I wonder how and why.
I noticed the kernel is now split into both ABI flavours. Shoudn't the userland be, too?
Could things run better if the binaries are compiled for the newer ARMv7 architecture, perhaps ordering the instructions for superscalar execution where possible?

Jörg
Userland is mainly shims over GPU calls, so a lot of it will see no benefit from being compiled for A7, there's simply not enough work being done. Some of the demo apps might improve, the raw camera aps for example may have higher performance. But it's all something that will come to light in the next few months.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Please direct all questions to the forum, I do not do support via PM.

pico
Posts: 16
Joined: Sun Sep 16, 2012 2:49 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 1:25 pm

Has my warranty bit been set on RPi2?

rfx@raspberrypi ~/tmp $ cat /proc/cpuinfo |grep Rev
Revision : 1a21041

User avatar
joan
Posts: 12819
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 1:41 pm

pico wrote:Has my warranty bit been set on RPi2?

rfx@raspberrypi ~/tmp $ cat /proc/cpuinfo |grep Rev
Revision : 1a21041
Probably?

May I ask where you bought the Pi 2? It doesn't appear to have been manufactured by Sony.

jdb
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 1704
Joined: Thu Jul 11, 2013 2:37 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 1:42 pm

pico wrote:Has my warranty bit been set on RPi2?

rfx@raspberrypi ~/tmp $ cat /proc/cpuinfo |grep Rev
Revision : 1a21041
Can you post your full /boot/config.txt
Rockets are loud.
https://astro-pi.org

rahlquist
Posts: 146
Joined: Tue Jan 21, 2014 1:02 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:27 pm

joan wrote:
pico wrote:Has my warranty bit been set on RPi2?

rfx@raspberrypi ~/tmp $ cat /proc/cpuinfo |grep Rev
Revision : 1a21041
Probably?

May I ask where you bought the Pi 2? It doesn't appear to have been manufactured by Sony.
How can you discern that?

User avatar
joan
Posts: 12819
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:35 pm

My understanding is that the final six hex digits have the following interpretation.

Code: Select all

/proc/cpuinfo revision field

SRRR MMMM PPPP TTTT TTTT VVVV

S scheme (0=old, 1=new)
R RAM (0=256, 1=512, 2=1024)
M manufacturer (0='SONY',1='EGOMAN',2='EMBEST',3='UNKNOWN',4='EMBEST')
P processor (0=2835, 1=2836)
T type (0='A', 1='B', 2='A+', 3='B+', 4='Pi 2 B', 5='Alpha', 6='Compute Module')
V revision (0-15)
Your 1a21041 give a bit pattern of

Code: Select all

1010 0010 0001 0000 0100 0001
SRRR MMMM PPPP TTTT TTTT VVVV

S=1 new scheme
R=2 1024 MB
M=2 EMBEST
P=1 2836
T=4 Pi 2 B
V=1 1

pico
Posts: 16
Joined: Sun Sep 16, 2012 2:49 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:37 pm

From reading elsewhere on the forum, it appears that

force_turbo=1

will set the warranty bit.

Has that always been the case? Or did the criteria change somewhere along the line?

Anyway, here's the config.txt:

Code: Select all

# uncomment to force a console size. By default it will be display's size minus
# overscan.
#framebuffer_width=1280
#framebuffer_height=720

# uncomment if hdmi display is not detected and composite is being output
#hdmi_force_hotplug=1

# uncomment to force a specific HDMI mode (this will force VGA)
#hdmi_group=1
#hdmi_mode=1

# uncomment to force a HDMI mode rather than DVI. This can make audio work in
# DMT (computer monitor) modes
#hdmi_drive=2

# uncomment to increase signal to HDMI, if you have interference, blanking, or
# no display
#config_hdmi_boost=4

# uncomment for composite PAL
#sdtv_mode=2

#uncomment to overclock the arm. 700 MHz is the default.
arm_freq=840

# for more options see http://elinux.org/RPi_config.txt
gpu_mem=16
core_freq=250
sdram_freq=420
over_voltage=0
force_turbo=1

device_tree=
##dtparam=i2c_arm=on
##dtparam=i2c_vc=on
#dtparam=i2c=on
##dtparam=i2c0=on
##dtparam=i2c1=on
#dtparam=spi=on


pico
Posts: 16
Joined: Sun Sep 16, 2012 2:49 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:40 pm

joan wrote:My understanding is that the final six hex digits have the following interpretation.

Code: Select all

/proc/cpuinfo revision field

SRRR MMMM PPPP TTTT TTTT VVVV

S scheme (0=old, 1=new)
[/quote]

Thanks for this -- very interesting.

So is the "scheme bit" above the warranty bit, or does it encode something else?

User avatar
ric96
Posts: 1254
Joined: Sun Mar 17, 2013 6:03 am
Location: NOIDA, India
Contact: Website

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:42 pm

Does the non-PoP arrangement offer better cpu temps?
My apologies for shameless YouTube Plugs...
youtube.com/sahajsarup
twitter @sahajsarup
skype srics1996
e-mail: sahajsarup@gmail.com
Blog: http://www.geektillithertz.com/wordpress
Web: http://www.geektillithertz.com

User avatar
joan
Posts: 12819
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:45 pm

The warranty bit may be encoded in that leading 1 you have. My Pi 2 shows a revision of a01041.

The bit structure was derived from software linked to by gadgetoid (I think).

User avatar
mikerr
Posts: 2520
Joined: Thu Jan 12, 2012 12:46 pm
Location: UK
Contact: Website

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 2:48 pm

My pi 2 also shows a01041
Android app - Raspi Card Imager - download and image SD cards - No PC required !

LeBigMacke
Posts: 2
Joined: Sat Feb 07, 2015 1:07 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 3:36 pm

Hello,

there is a new option in the overclock menue called "Pi2". In this the standard configuratoin for the Pi2 or what?

Iam confused, please help!
Thanks a lot!

dom
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 5106
Joined: Wed Aug 17, 2011 7:41 pm
Location: Cambridge

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 3:44 pm

LeBigMacke wrote: there is a new option in the overclock menue called "Pi2". In this the standard configuratoin for the Pi2 or what?
Yes, as the Pi2 is now 900MHz by default, the lower Pi1 overclocks don't actually make sense, so the Pi2 setting is the most appropriate.
A subsequent update may show a different list depending on whether you are running on Pi1 or Pi2 which may make things clearer.

LeBigMacke
Posts: 2
Joined: Sat Feb 07, 2015 1:07 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 5:07 pm

dom wrote: A subsequent update may show a different list depending on whether you are running on Pi1 or Pi2 which may make things clearer.
Thank you for the answer, that would be great!
But when i am using the Pi2 setting my Pi2 freeze when i run one of my own written programs.
The programs are written in Python using OpenCV, Numpy and the V4l2-driver for the Camera Module.
So I'm going to stay in the Medium setting.

tvjon
Posts: 560
Joined: Mon Jan 07, 2013 9:11 am

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 5:24 pm

joan wrote:The warranty bit may be encoded in that leading 1 you have. My Pi 2 shows a revision of a01041.

The bit structure was derived from software linked to by gadgetoid (I think).
I wondered about that too, as I'd inadvertently upgraded a Kodi SD card from a B+ which had a 6 overvolt to RPi2, which has a much less aggressive overvolt, yet my revision still says a01041.

Has your Pi2 arrived yet? I have a small project to use PIGPIO, so I checked github to see if you'd updated it.

Do you have a link to the rev' settings please?

User avatar
joan
Posts: 12819
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 6:01 pm

tvjon wrote:
joan wrote:The warranty bit may be encoded in that leading 1 you have. My Pi 2 shows a revision of a01041.

The bit structure was derived from software linked to by gadgetoid (I think).
I wondered about that too, as I'd inadvertently upgraded a Kodi SD card from a B+ which had a 6 overvolt to RPi2, which has a much less aggressive overvolt, yet my revision still says a01041.

Has your Pi2 arrived yet? I have a small project to use PIGPIO, so I checked github to see if you'd updated it.

Do you have a link to the rev' settings please?
gadgetoid posted earlier at http://www.raspberrypi.org/forums/viewt ... 34#p684334 linking to their Python script at https://github.com/pimoroni/RPi.version. I'm not sure of the original source of the information.

I now have a Pi 2 and have made limited progress with pigpio. See http://www.raspberrypi.org/forums/viewt ... 31&t=98955

It works once successfully initialised but I have to do strange things to get it to initialise. I think I'm running into the same problems I originally had in getting userland DMA working, i.e. ensuring consistency between the DMA and user view of memory (cache coherency). There are a few other oddities which I can not explain but seem to be nothing to do with DMA.

If you want I can host a work in progress version of pigpio that you can play with.

tvjon
Posts: 560
Joined: Mon Jan 07, 2013 9:11 am

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 6:07 pm

joan wrote:...
If you want I can host a work in progress version of pigpio that you can play with.
Yes please.

User avatar
joan
Posts: 12819
Joined: Thu Jul 05, 2012 5:09 pm
Location: UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 6:24 pm

tvjon wrote:
joan wrote:...
If you want I can host a work in progress version of pigpio that you can play with.
Yes please.
Link now removed.

If you can, access the library through the pigpio daemon rather than linking the library with your own code. Don't ask me why it is more reliable, it just is.
Last edited by joan on Sun Feb 08, 2015 9:07 am, edited 1 time in total.

dom
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 5106
Joined: Wed Aug 17, 2011 7:41 pm
Location: Cambridge

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 8:11 pm

mpnico wrote:I thought that force_turbo with no over_voltage doesn't void the warranty. Is this not correct for the RPi2 ?
You are correct, but Pi2 runs at a higher voltage by default which meant that force_turbo was incorrectly setting the "warranty" bit. That was unintended.

So, what we've agreed to do is to no longer treat bit 24 of the board revision as the warranty warranty bit on Pi 2.
The latest rpi-update will now set bit 25 when a real warranty condition arises (e.g. force_turbo=1 *and* over_voltage > 0)

If bit 24 is currently set, then don't worry, it will be ignored. Bit 25 is the new Pi 2 warranty bit.

Pi1 will remain as it did with bit 24 being the warranty bit.

mpnico
Posts: 2
Joined: Fri Feb 06, 2015 10:03 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 8:22 pm

dom wrote:
mpnico wrote:I thought that force_turbo with no over_voltage doesn't void the warranty. Is this not correct for the RPi2 ?
You are correct, but Pi2 runs at a higher voltage by default which meant that force_turbo was incorrectly setting the "warranty" bit. That was unintended.

So, what we've agreed to do is to no longer treat bit 24 of the board revision as the warranty warranty bit on Pi 2.
The latest rpi-update will now set bit 25 when a real warranty condition arises (e.g. force_turbo=1 *and* over_voltage > 0)

If bit 24 is currently set, then don't worry, it will be ignored. Bit 25 is the new Pi 2 warranty bit.

Pi1 will remain as it did with bit 24 being the warranty bit.
Thank you for answering me, I was really surprised seeing the warranty bit set on mine. I hope my post will be usefull for other people.

PS : really good job with this Pi 2 !

User avatar
davidcoton
Posts: 2446
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 8:22 pm

Is there still a potential problem with cards being moved between Pi1 and Pi2 when overclocking is set? Surely some conditions that are not warranty breakers on one, will break it on the other? Most likely where the more tolerant version (whichever that is) has been set to just below the level where the bit sets, and the card is transferred to the other version.

Possible solution: keep the overclock settings separate for Pi1 and Pi2, so that transferring to a different version will not activate the wrong set of settings :?
"If it ain't broke, fix it until it is."
“Raspberry Pi is a trademark of the Raspberry Pi Foundation”
The Pink Unicorn is a symbol of commitment to sensible discussion of future versions of the Raspberry Pi ™

dom
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 5106
Joined: Wed Aug 17, 2011 7:41 pm
Location: Cambridge

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 8:43 pm

davidcoton wrote:Is there still a potential problem with cards being moved between Pi1 and Pi2 when overclocking is set? Surely some conditions that are not warranty breakers on one, will break it on the other?
No, I believe anything that sets warranty bit on Pi1 will set it on Pi2 and vice-versa.

rahlquist
Posts: 146
Joined: Tue Jan 21, 2014 1:02 pm

Re: The Raspberry Pi 2 Q&A thread

Sat Feb 07, 2015 11:38 pm

ric96 wrote:Does the non-PoP arrangement offer better cpu temps?
I am finding things appear to run about 10deg higher.

CPU Temp: 45.4ºC
GPU Temp: 45.5ºC

The increase could be a change in sensor location? Not sure.

pico
Posts: 16
Joined: Sun Sep 16, 2012 2:49 pm

Re: The Raspberry Pi 2 Q&A thread

Sun Feb 08, 2015 1:25 am

dom wrote: You are correct, but Pi2 runs at a higher voltage by default which meant that force_turbo was incorrectly setting the "warranty" bit. That was unintended.

So, what we've agreed to do is to no longer treat bit 24 of the board revision as the warranty warranty bit on Pi 2.
The latest rpi-update will now set bit 25 when a real warranty condition arises (e.g. force_turbo=1 *and* over_voltage > 0)

If bit 24 is currently set, then don't worry, it will be ignored. Bit 25 is the new Pi 2 warranty bit.

Pi1 will remain as it did with bit 24 being the warranty bit.
Thanks for the clarification on this. I was confused about what appeared to be the warranty bit appearing on the Pi2 when I hadn't seen this occur on my model Bs.

I have one further question. I want to have a card image that will boot and work on either a model B or a Pi2. i have an image that I've been working on that has the following config.txt:

Code: Select all

#uncomment to overclock the arm. 700 MHz is the default.
arm_freq=840

# for more options see http://elinux.org/RPi_config.txt
gpu_mem=16
core_freq=250
sdram_freq=420
over_voltage=0
force_turbo=1

device_tree=


Will this have any other unexpected consequences, apart from setting bit 24 on the Pi2 (as discussed)? Will the arm_freq of 840 simply be ignored by the Pi2, for example, and it will continue to run at the default 900MHz?

Also, i've discovered that loading the peripheral modules for spi etc using the device tree is not equivalent to loading the old way... the drivers loaded using device tree seem to lead to some performance issues compared to the old drivers. Haven't investigated closely enough to figure out what is happening, but the take home is that the device_tree drivers may not perform as expected when compared to the no device-tree drivers. So in the meantime i've simply disabled the device_tree loading.

Return to “General discussion”

Who is online

Users browsing this forum: Gadgetguy, itimpi, user879 and 31 guests