On Mon, 08 May 2017 09:59:50 +0100 "mike.valk@gmail.com" mike.valk@gmail.com wrote:
2017-05-08 3:34 GMT+02:00 doark@mail.com:
I apologize for DOS'ing the list, I can only get online about once a week.
On Thu, 4 May 2017 17:13:23 +0200 "mike.valk@gmail.com" mike.valk@gmail.com wrote:
2017-05-04 9:04 GMT+02:00 John Luke Gibson eaterjolly@gmail.com:
Since it seems like a trivially simple task that for some reason no one has taken up, I would like to take the opportunity to exercise a learning experience and simultaneously benefit the community, by liberating PocketCHIP by deblobbing the source and re-compiling.
The PocketCHIP is powered by their SoM: http://linux-sunxi.org/NextThingCo_CHIP
That is apparently a Allwinner R8 pared with an external rtl8723bs Wifi/BT chip.
The R8 is a rebranded A13.
What? I own one of those and I'm almost certain that the CPU is an A7. Let's boot the PocketCHIP up... The processor is detected as an A7.
It should be a Cortex-A8. http://linux-sunxi.org/Allwinner_SoC_Family
The new chip GR8, which is specific SoC for NextThingCo, seems to be an "sun5i" as well. Also a slightly modified A13 I guess.
The're seems to but mainline support for it. Icenowy Zheng has addid it I think. But the'res no wiki page fo it. http://linux-sunxi.org/Linux_mainlining_effort http://linux-sunxi.org/GR8
Sorry, I (foolishly) though that ARMv7 == A7 (i.e. a contraction).
I'll attach the output, it would probably be interesting to see all of it... Done, it's compressed bzip2 since it's ~300KiB decompressed which is large for an email.
Use pastebin or sorts. Or just cut out the specifica part.
Sorry, I thought the rest of it might be useful to look at.
<snip>
You're not giving us enough details. Who is Verhaegen? What did he burn out on? When I first considered purchasing a PocketCHiP I read about the GPU not having 3D capabilities because of a binary blob. So, the CHIP folks hired (I think it was an extended goal of the kickstarter campaign), a kernel dev to add support to the Linux kernel for the GPU.
That did not happen. The're is no Opensource linux driver for MALI. NTC hardly involves itself with the linux-sunxi community.
Their website is hardly obvious to the software needs of running their hardware.
How hard can it be....
"To use our hardware you have two options: Our BSP which has closed source drivers, but you have full utilization of the hardware. Or use the mainline kernel with some restrictions"
Where is that quote from?
And state your involvement in freeing the hardware or not.
NTC website is just one big selling machine.
Prior to purchasing the Pocket CHIP I read their docs and kickstarter page. See this (their kickstarter page says similar): https://docs.getchip.com/chip.html#is-chip-open-source-where-are-the-docs Are they flat out lying?
Thanks, David
On Mon, May 29, 2017 at 10:37 PM, David Niklas doark@mail.com wrote:
Prior to purchasing the Pocket CHIP I read their docs and kickstarter page. See this (their kickstarter page says similar): https://docs.getchip.com/chip.html#is-chip-open-source-where-are-the-docs Are they flat out lying?
absolutely not. absolutely everything (with the exception of MALI) has been available for the A13 core for... like... 4 years. getting this through to people's thick skulls, thanks to the high noise-to-signal ratio, is getting frankly a little tiresome, i don't mind admitting.
l.
On Tue, 30 May 2017 03:27:19 +0100 Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
On Mon, May 29, 2017 at 10:37 PM, David Niklas doark@mail.com wrote:
Prior to purchasing the Pocket CHIP I read their docs and kickstarter page. See this (their kickstarter page says similar): https://docs.getchip.com/chip.html#is-chip-open-source-where-are-the-docs Are they flat out lying?
absolutely not. absolutely everything (with the exception of MALI) has been available for the A13 core for... like... 4 years.
The wifi requires a binary blob according to this thread (unless I'm remembering wrongly).
getting this through to people's thick skulls, thanks to the high noise-to-signal ratio, is getting frankly a little tiresome, i don't mind admitting.
l.
Well, when you say "opensource hardware" most people, including myself, think "Oh, a completely opensource down to the last transistor machine!!!" When i first read the slashdot page on the C.H.I.P. I really thought that that is what they meant... Same story with eoma. Which only makes sense, since, taking spamassasin as an example, if you get an "opensource" program you expect not only every line available for viewing but also the docs and mass-check rules to be "opensource", which they are.
No offence intended luke, but when I see people trying to "Opensource" something I am often times disappointed. I know that you and others work very hard but it looks, to me, to be a half done job almost all the time and I mean with software too, because it tends to lack tests to verify it's functionality and docs to train the coming generations, myself numbering among them.
Thanks, David
2017-05-29 23:37 GMT+02:00 David Niklas doark@mail.com:
On Mon, 08 May 2017 09:59:50 +0100 "mike.valk@gmail.com" mike.valk@gmail.com wrote:
2017-05-08 3:34 GMT+02:00 doark@mail.com:
Sorry, I (foolishly) though that ARMv7 == A7 (i.e. a contraction).
Don't worry that is common one.
I'll attach the output, it would probably be interesting to see all of it... Done, it's compressed bzip2 since it's ~300KiB decompressed which is large for an email.
Use pastebin or sorts. Or just cut out the specifica part.
Sorry, I thought the rest of it might be useful to look at.
It might be. That's why I suggested pastebin et al. Those are better for public mailinglists than zipped files.
<snip> > > You're not giving us enough details. Who is Verhaegen? What did he > > burn out on? > > When I first considered purchasing a PocketCHiP I read about the GPU > > not having 3D capabilities because of a binary blob. So, the CHIP > > folks hired (I think it was an extended goal of the kickstarter > > campaign), a kernel dev to add support to the Linux kernel for the > > GPU. > > > That did not happen. The're is no Opensource linux driver for MALI. NTC > hardly involves itself with the linux-sunxi community. > > Their website is hardly obvious to the software needs of running their > hardware. > > How hard can it be.... > > "To use our hardware you have two options: Our BSP which has closed > source drivers, but you have full utilization of the hardware. Or use > the mainline kernel with some restrictions" Where is that quote from?
No ware. It was suggestion from me for them. That's what I would like to see on all those sites selling this type of s*ht.
Be honest, be open. Don't façade the truth. It will come out and it will bite you.
I don't believe the quote above would scare any potential buyer. The fact they don't mention it while I know it is a reason I wouldn't buy from them.
It's like selling a car of which they have painted over rust and rewinded the odo-meter. On first glance it looks terrific but when you find the truth it will leave you angry and you'll go and tell everyone you know not to buy from them.
So whenever I see a fancy site trying to sell a product of which I know the limitations and they hide it: They become unreliable to me and I'll move one and suggest to everyone that want's to listen to do the same.
Sadly that's the state of all vendors today. So everybody buy a crappy painted over car with hardly any km/miles on it. ;-)
arm-netbook@lists.phcomp.co.uk
-
David Niklas -
doark＠mail.com
-
Luke Kenneth Casson Leighton -
mike.valk＠gmail.com