September 2013 - coreboot - mail.coreboot.org

Re: [coreboot] [liberationtech] Fwd: Firefox OS with built in support for OpenPGP encryption
by Eugen Leitl Sept. 20, 2013

Sept. 20, 2013

----- Forwarded message from Blibbet <blibbet(a)gmail.com> ----- Date: Thu, 19 Sep 2013 11:01:43 -0700 From: Blibbet <blibbet(a)gmail.com> To: liberationtech <liberationtech(a)lists.stanford.edu> Subject: Re: [liberationtech] Fwd: Firefox OS with built in support for OpenPGP encryption User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.5; rv:16.0) Gecko/20121005 Thunderbird/16.0 Reply-To: liberationtech <liberationtech(a)lists.stanford.edu> > I don't think so -- unless you have a laptop flashed with a free > software BIOS / boot firmware that you can inspect and modify. There are > a handful of dated possibilities out there like that (Thinkpad x60 > models that support coreboot, Lemote Yeelongs), but not the vast > majority of laptops. The situation with laptops actually seems pretty > analogous to the worst category of smartphones (those where the baseband > firmware shares the main CPU and RAM). Unfortunately, Coreboot is still not mainstream, netiher is Lemote. UEFI is the current mainstream system firmware, for Intel systems, and ramping up quickly for ARM (though not yet on Android-based devices yet, AFAIK). Intel's Tunnel Mountain UEFI dev platform lets you flash your own UEFI-based firmware, and mainly targets firmware IHVs. http://www.tunnelmountain.net/ http://uefidk.intel.com/develop/workstation-development-kit The Intel Atom-based MinnowBoard is a new UEFI dev platform, and it's Linux-based, and targets hackers; it uses Intel's definition of "Open Hardware", mainly meaning no NDAs involved. It is much cheaper and smaller than the above box. http://minnowboard.org/ http://uefidk.intel.com/content/minnowboard-uefi-firmware Both of these boxes let you reflash your system firmware with your custom build of BSD-licensed TianoCore UEFI. For Linaro's Linux/ARM device targets, Linaro also has a fork of Tianocore, and ships their own images. You can use the Linaro targets and modify this forked Tianocore firmware, as well, but it requires a bit more ARM-centric eLinux skills. https://wiki.linaro.org/ARM/UEFI <soapbox> There is a large OEM/ODM/IBV/IHV/ISV ecosystem that currently runs the hardware, and it is UEFI-centric. IMO, focusing only on fringe Lemote/Coreboot technology is not a good bet. Personally, I wish EFF/FSF and other open/free tech groups would form a Linaro-like firmware group and produce their own UEFI firmware image, as an option for OEMs. There needs to be some Free Boot alternative to Secure Boot, with certs from EFF/FSF/etc and the open source distro vendors, not just OEMs/MSFT in the firmware, and it needs to target booting from a handful of main open source distros, not just 1 commercial OS. Else, UEFI will turn Personal Computers into Windows PCs, ending the era of General Purpose computing. </soapbox> -- Liberationtech is public & archives are searchable on Google. Violations of list guidelines will get you moderated: https://mailman.stanford.edu/mailman/listinfo/liberationtech. Unsubscribe, change to digest, or change password by emailing moderator at companys(a)stanford.edu. ----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://ativel.com http://postbiota.org AC894EC5: 38A5 5F46 A4FF 59B8 336B 47EE F46E 3489 AC89 4EC5

3 2

Re: [coreboot] Freescale i.MX6
by default＠rlf.de Sept. 20, 2013

Sept. 20, 2013

On Fri, 20 Sep 2013 12:53:07 +0200 Christian Gmeiner <christian.gmeiner(a)gmail.com> wrote: Hi Guys. > > thank you for your quick response. > > > > If there is actually no active effort to support i.MX6, I’m interested in > > leading that project. > > > > But I cannot start before the end of the year because I have to wait for our > > hardware and to bring-up the board first with u-boot. > > > > I think that I could spend 2-3 days to look into initial support for > the I.MX6 during the > next 2-3 weeks. > well I can give you a git-link for an U-Boot already running and kicking on i.MX6 (D&Q), it's DT/FDT capable. If you want to cut this short. > > Another point is to think about hardware donation. > > > > I have access to some I.MX6D and one I.MX6Q boards.. also I own a ARM > jtag. So I have > all I need :) > I also have access to i.MX6D/Q boards, no JTAG/OLIMEX though. -- regards Oliver Wendt

2 1

new PC Engines APU system in the pipeline
by Eugen Leitl Sept. 20, 2013

Sept. 20, 2013

http://www.pcengines.ch/apu.htm apu platform The PC Engines apu system board is a big step up in performance and capacity from the popular ALIX series. Applications: Routers, firewalls, VOIP, dedicated servers, special purpose network plumbing, education tools... CPU: AMD G series T40N or T40E APU, 1 GHz dual core (Bobcat core) with 64 bit support, 32K data + 32K instruction + 512KB L2 cache per core DRAM: 2 GB DDR3-1066 DRAM Storage: Boot from SD card (connected through USB), external USB or m-SATA SSD. 1 SATA data + power connector. Power: About 6 to 12W of 12V DC power depending on CPU load. Expansion: 2 miniPCI express (one with SIM socket for 3G modem), LPC bus, GPIO header, I2C bus, COM2 (3.3V RXD/TXD). Connectivity: 3 Gigabit Ethernet (Realtek RTL8111E), 1 DB9 serial port (console). Firmware: CoreBoot open source system BIOS with support for iPXE and USB boot. Form factor: 6"x6" (152.4 x 152.4 mm), fits in our case1d2*u enclosures. Cooling: Conductive cooling from the CPU and south bridge to the enclosure using a 3 mm alu heat spreader. Please contact us for advice if you want to integrate this board in your own enclosure. Status: Prototypes alive and well. Pilot run / customer samples planned for fall 2013. Production expected for early 2014. OEM options: 4 GB DDR3 DRAM, various build options, different CPU speed grades. Roadmap: We will upgrade to the newer AMD Kabini / Jaguar core based SOCs, up to 2 GHz quad core (Opteron X1150), up to 4 GB ECC DRAM.

1 0

Re: [coreboot] Freescale i.MX6
by Christian Gmeiner Sept. 20, 2013

Sept. 20, 2013

Hi Wolfgang > > thank you for your quick response. > > If there is actually no active effort to support i.MX6, I’m interested in > leading that project. > > But I cannot start before the end of the year because I have to wait for our > hardware and to bring-up the board first with u-boot. > I think that I could spend 2-3 days to look into initial support for the I.MX6 during the next 2-3 weeks. > Another point is to think about hardware donation. > I have access to some I.MX6D and one I.MX6Q boards.. also I own a ARM jtag. So I have all I need :) greets -- Christian Gmeiner, MSc

1 0

Re: [coreboot] Freescale i.MX6
by Wolfgang Kamp - datakamp Sept. 20, 2013

Sept. 20, 2013

Hi David, Hi Christian, thank you for your quick response. If there is actually no active effort to support i.MX6, I’m interested in leading that project. But I cannot start before the end of the year because I have to wait for our hardware and to bring-up the board first with u-boot. Another point is to think about hardware donation. Wolfgang Von: coreboot-bounces(a)coreboot.org [mailto:coreboot-bounces@coreboot.org] Im Auftrag von David Hendricks Gesendet: Mittwoch, 18. September 2013 00:22 An: Wolfgang Kamp - datakamp Cc: coreboot(a)coreboot.org Betreff: Re: [coreboot] Freescale i.MX6 Hi Wolfgang, I do not know of any active efforts to support i.MX6 in coreboot. However, if you are interested in leading the porting effort then we can help. I think that ARM-specific portions will be simple. The hard parts are DRAM and peripheral init. If we can help you boot to the point where UART works, will you feel comfortable doing the remaining work? As far as I know, the focus of ARM development in coreboot is on Exynos5 (Cortex-A15). We are currently booting Exynos5250 and Exynos5420 systems. TI AM335X (Cortex-A8) support is still somewhat active, but is a side-project. We can initialize the CPU and get serial console output so future development will be much easier.

1 0

Re: [coreboot] crossgcc build fails
by Patrick Georgi Sept. 19, 2013

Sept. 19, 2013

Am Donnerstag, den 19.09.2013, 16:52 +0200 schrieb Marcello Stanisci: > $ make crossgcc > > the i386-elf part worked fine but the armv7a-eabi part fails when > tries to compile > GCC 4.8.1, giving the following error message: Two things: First, unless you want to build coreboot for ARM, the cross compiler is sufficiently set up now, so you could just ignore the issue. Second, http://review.coreboot.org/#/c/3930/ should fix it, so if you have the time, please test if that fixes the ARM issue for you, too (it does for me). Thanks, Patrick

1 0

crossgcc build fails
by Marcello Stanisci Sept. 19, 2013

Sept. 19, 2013

Hi, I'm trying to compile the cross compiler with version 1.22 of coreboot cross toolchain builder. I just did: $ make crossgcc the i386-elf part worked fine but the armv7a-eabi part fails when tries to compile GCC 4.8.1, giving the following error message: Building GCC 4.8.1 ... failed make[1]: *** [build-without-gdb] Error 1 make: *** [crossgcc] Error 2 I'm doing this on Debian 6 (squeeze).. help..

1 0

crossgcc build fails
by Marcello Stanisci Sept. 19, 2013

Sept. 19, 2013

Hi, I'm trying to compile the cross compiler with version 1.22 of coreboot cross toolchain builder. I just did: $ make crossgcc the i386-elf part worked fine but the armv7a-eabi part fails when tries to compile GCC 4.8.1, giving the following error message: Building GCC 4.8.1 ... failed make[1]: *** [build-without-gdb] Error 1 make: *** [crossgcc] Error 2 I'm doing this on Debian 6 (squeeze).. help..

1 0

Re: [coreboot] Tyan s2912-E reset loop, Please help
by Marc Jones Sept. 18, 2013

Sept. 18, 2013

On Wed, Sep 18, 2013 at 7:39 AM, Mark Mc <markymcd13(a)gmail.com> wrote: > Hi Marc, > > I think you are correct and that something is going wrong in romstage.c > where "wants_reset = mcp55_early_setup_x()" which seems to execute that > function in early_setup_car.c just before the warm reset however the code > within this function is beyond my skill set and the only way I can see > myself correcting this is if I cound find an older release of coreboot v4 > (late 2010) as there are a number of posts in the mail archive of this board > working correctly at that time. Do you of anyway to get this? You can do a git bisect, but going back that far in history can be difficult (toolchain and architecture changes). You should just keep drilling down to why it fails. You don't have to understand too much to dissect down to the failure point. Then we can look at changes in that area and figure out why. Regards, Marc -- http://se-eng.com

1 0

missing files ?
by jacek burghardt Sept. 18, 2013

Sept. 18, 2013

1 0