On Thu, 24 Jul 2014 18:39:59 +0200
Lucas Stach <l.stach@xxxxxxxxxxxxxx> wrote:
> Am Donnerstag, den 24.07.2014, 20:46 +0400 schrieb Antony Pavlov:
> > On Thu, 24 Jul 2014 18:08:33 +0200
> > Lucas Stach <l.stach@xxxxxxxxxxxxxx> wrote:
> >
> > > Am Donnerstag, den 24.07.2014, 20:05 +0400 schrieb Antony Pavlov:
> > > > Hi!
> > > >
> > > > I have just tested this commit
> > > >
> > > > commit d8bdd537220dd83912949f19ba67029bd1f11c58
> > > > Author: Lucas Stach <dev@xxxxxxxxxx>
> > > > Date: Fri Apr 12 12:28:14 2013 +0200
> > > >
> > > > tegra: switch to proper CPU type
> > > >
> > > > Tegras main CPUs are all ARMv7.
> > > >
> > > > Signed-off-by: Lucas Stach <dev@xxxxxxxxxx>
> > > > Tested-by: Antony Pavlov <antonynpavlov@xxxxxxxxx>
> > > > Signed-off-by: Sascha Hauer <s.hauer@xxxxxxxxxxxxxx>
> > > >
> > > > diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> > > > index bb9b47b..dfb7e24 100644
> > > > --- a/arch/arm/Kconfig
> > > > +++ b/arch/arm/Kconfig
> > > > @@ -137,7 +137,7 @@ config ARCH_VEXPRESS
> > > >
> > > > config ARCH_TEGRA
> > > > bool "Nvidia Tegra-based boards"
> > > > - select CPU_ARM926T
> > > > + select CPU_V7
> > > > select HAS_DEBUG_LL
> > > >
> > > > config ARCH_ZYNQ
> > > >
> > > > This commit breaks 'usb' command --- it does not scan devices anymore, just hangs.
> > > >
> > > > I have used codesourcery arm-2013.11 toolchain.
> > > >
> > > I haven't tested USB myself, as I'm not using second stage barebox after
> > > u-boot and there's still things missing for USB to work on first stage.
> > > Currently only SDMMMC is fully supported.
> >
> > How I can run barebox from barebox (u-boot) on AC100?
> >
> Um, how did you test things before? I personally haven't used second
I have AC100 with u-boot burned into the flash.
Here is my AC100 test seance:
1. reset AC100 (activate reset pin for 1 second and deactivate it;
i use USB-relay board for this task at the moment);
2. wait for u-boot prompt;
3. load barebox.bin via y-modem;
4. use 'go' commmand for running barebox.
You can see that I always use barebox second stage image.
> stage for a long time. For u-boot you need to copy the barebox image
> into the right spot (0x108000 for T20) and issue 'go' command. For
> barebox a 'bootm /path/to/image' should do, but I won't guarantee that I
> didn't break anything for second stage.
>
> > > Are you testing exactly this commit? If so you might need to apply
> > > 431d8a247dfa63f817cda137a81dbb2244d413dc "arm: properly init alignment
> > > trap bit", as USB might trigger an unaligned access.
> >
> > Thanks!
> > I'll try to apply 431d8a.. tomorrow.
> >
> > > > I'll try to check latest barebox from next, but just now I have no means to burn
> > > > barebox-tegra20-toshiba-ac100.img into AC100's ROM.
> > > >
> > > You'll notice that a newer barebox won't build images for AC100 that are
> > > suitable to burn into a boot media, because there are no BCTs publicly
> > > available for this device. The resulting AC100 image is only useful for
> > > starting with TegraRCM.
> >
> > I work with my AC100 remotely so I can't press Ctrl-ESC on every boot for TegraRCM recovery mode.
> > How I can run barebox from barebox (u-boot) on AC100?
> >
> New barebox has a 'tegrarcm' command that will drop you right into
> recovery mode, but obviously won't help if the machine is hanging.
Hmm, I have to see it :)
--
Best regards,
Antony Pavlov
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
