Boot method declaration

[Stefan Kalkowski]

Hi,

just to clarify things:

In general, it's useful to produce u-boot images on ARM platforms where
we use u-boot as bootloader. Normally u-boot isn't configured to provide
the ability to boot elf images directly (bootelf command). Therefore, we
always had to replace the given bootloader with our own binary.
Moreover, if you want to boot scenarios over network, which reach a
certain size it's handy to transfer compressed images.
Especially on the i.MX53 QSB I detected problems with u-boot loading
large images. That's why I always had to produce an gzipped u-boot image
for that platform anyway. After I've noticed that Alex built pulsar
support into the run tool, I thought I'll do the same for producing
u-boot images. The pulsar support is realized by using an additional
SPEC variable, namely 'boot_pxe'. Therefore, I've used a SPEC variable
too, to evaluate whether an u-boot image should be built or not. Apart
from that fact, I didn't copied anything, just tried to do it in a
convenient way ;-)

regards
Stefan

On 12/17/2012 04:02 PM, Martin Stein wrote:
> Hi Norman,
> 
> I'm ok with that. I've already discussed this with Stefan. He'd
> introduced the SPEC-way in base-hw and said that he simply copied it
> from another case (in base-nova i think). Why this spec is needed wasn't
> that interesting for me, i think he said that the imx53 uBoot behaves a
> little strange (images must get extra-processed to work on it) and thus
> requires distinction. I only wanted to know wich way should be targeted
> as final solution.
> 
> Cheers,
> Martin
> 
> On 12/17/2012 03:35 PM, Norman Feske wrote:
>> Hi Martin,
>>
>>> By now when doing a build there are two ways of defining the boot
>>> method. In 'base-mb' one writes 'RUN_OPT="--target jtag" make demo' or
>>> 'RUN_OPT="--target qemu" make demo' while qemu is set by default. At
>>> least in base-hw with imx53 one must enrich the SPECS e.g. in
>>> etc/specs.mk by the 'uboot' spec to get an appropriate Uboot image.
>>>
>>> Are there pros or cons for one of these methods, because i would prefer
>>> to use only one them.
>>
>> admittedly, I am not familiar with the rationale behind the 'uboot' spec
>> value. As far as I know, it hasn't entered genodelabs/master yet. Maybe
>> Stefan can clarify?
>>
>> Because the selection of the boot method is outside of the scope of the
>> build system but in the scope of the run tool, I regard the 'RUN_OPT'
>> variable as the most natural way to steer the boot method.
>>
>> I'd like to keep the purposes of both tools (build system and run tool
>> respectively) well separated. By adding spec values for the boot method
>> just to be picked up by the run tool but not the build system, we would
>> blur the boundaries between those tools.
>>
>> Cheers
>> Norman
>>
> 
> 
> ------------------------------------------------------------------------------
> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
> Remotely access PCs and mobile devices and provide instant support
> Improve your efficiency, and focus on delivering more value-add services
> Discover what IT Professionals Know. Rescue delivers
> http://p.sf.net/sfu/logmein_12329d2d
> _______________________________________________
> Genode-main mailing list
> Genode-main at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/genode-main
> 

-- 
Stefan Kalkowski
Genode Labs

http://www.genode-labs.com/ · http://genode.org/