RPI4 Install r357606 test 09.02.2020 failed
Mark Millard
marklmi at yahoo.com
Tue Feb 11 01:41:11 UTC 2020
On 2020-Feb-10, at 17:14, bob prohaska <fbsd at www.zefox.net> wrote:
> On Tue, Feb 11, 2020 at 12:18:42AM +0100, Klaus K??chemann wrote:
>>
>>
>>
>> Hi Bob, old proud RPI4-owner ;-) ,
>>
> [for pedantry's sake, this is a Pi3....]
>
>
>> according to your request I???ve updated the Wiki (at the bottom) with details
>> How you can get back to previous kernel while debugging .
>> https://wiki.freebsd.org/arm/Raspberry%20Pi#RPI4
>>
>
> Looks as if the loader path defaults are a little trickier than I realized:
>
> OK
> OK load kernel
> /boot/kernel/kernel text=0x987ddc data=0x1905a8 data=0x0+0x79d260 syms=[0x8+0x10bd20+0x8+0x130451]
> OK unload kernel
> OK load disk0s2:/boot/kernel.spare
> don't know how to load module 'disk0s2:/boot/kernel.spare'
> OK
>
> However,
> OK load disk0s2:/boot/kernel.spare/kernel
> disk0s2:/boot/kernel.spare/kernel text=0x9877bc data=0x1905a8 data=0x0+0x79d260 syms=[0x8+0x10bc60+0x8+0x130430]
> worked. Alas, that kernel stops as before, with:
> OK load disk0s2:/boot/kernel.spare/kernel
> disk0s2:/boot/kernel.spare/kernel text=0x9877bc data=0x1905a8 data=0x0+0x79d260 syms=[0x8+0x10bc60+0x8+0x130430]
> OK boot
> Using DTB provided by EFI at 0x7ef6000.
> EFI framebuffer information:
> addr, size 0x3eaf0000, 0x10a800
> dimensions 656 x 416
> stride 656
> masks 0x00ff0000, 0x0000ff00, 0x000000ff, 0xff000000
>
> Progress stops at this point.
>
> Looks like the problem is something other than the kernel.
> The power supply pins are at 4.88 volts when boot hangs.
> That would be within spec for USB (5.0 plus or minus .25
> volt) _except_ that the Pi3 is spec'd for 5.1 volts...
>
> That puts me within meter error of undervoltage.
>
If you recently followed the instructions to use:
2. https://github.com/raspberrypi/firmware/tree/master/boot/overlays -> overwrite overlays-folder on uSD:
3. https://github.com/raspberrypi/firmware/tree/master/boot -> copy the following files to uSD:
instead of what is in sysutils/rpi-firmware and sysutils/u-boot-rpi4
then you likely picked up bad materials.
https://github.com/raspberrypi/firmware/issues
shows people reporting that things quit working for
what appear to be (at times) power problems. This
seems to have been going on for days.
I do not recommend using https://github.com/raspberrypi/firmware/tree/
materials unless you know specific good versions to pick (say
a specific Tag selection).
I also do not recommend copying only some of the files from
https://github.com/raspberrypi/firmware/tree/master/boot/ and mixing
them with files of other vintages. They are not designed to mix and
match arbitrarily as far as I can tell.
I have stuck with sysutils/rpi-firmware and sysutils/u-boot-rpi4 materials
of a fixed vintage. In part is this so that the kernel explorations are
what vary, since -r356767 works and -r356776 does not for the same set
of other materials.
It is already complicated enough trying to find a way to isolate
what is leading to PSCI version 0 instead of 2 based on switching
kernels (and only the kernels).
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
More information about the freebsd-arm
mailing list