* [U-Boot] Tegra board problems
@ 2014-11-25 4:03 Simon Glass
2014-11-25 16:00 ` Stephen Warren
2014-11-25 16:11 ` Stephen Warren
0 siblings, 2 replies; 6+ messages in thread
From: Simon Glass @ 2014-11-25 4:03 UTC (permalink / raw)
To: u-boot
Hi,
I'm a bit messed up with my testing now.
1. Jetson-TK1: While I2C testing I unfortunately erased part of the
I2C board ID EEPROM, and now it will not boot. It says:
Dummy read for TPS65913
Board id = 0x1ff
and then dies. I can probably change the board id back, but I'm not
sure what it should be! It will boot U-Boot but not the AVP boot
loader. If someone could list out the contents of their EEPROM (bus 1,
address 56) that might help.
2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
'part list mmc 0' just prints an error:
## Unknown partition table
It is getting a partition ID of 0. Is there something wrong with
Beaver's mmc support in mainline?
The old 2011 U-Boot on the board works though.
Also a question - are there any Jetson-like Tegra 132 boards available yet?
Regards,
Simon
^ permalink raw reply [flat|nested] 6+ messages in thread
* [U-Boot] Tegra board problems
2014-11-25 4:03 [U-Boot] Tegra board problems Simon Glass
@ 2014-11-25 16:00 ` Stephen Warren
2014-11-25 16:11 ` Stephen Warren
1 sibling, 0 replies; 6+ messages in thread
From: Stephen Warren @ 2014-11-25 16:00 UTC (permalink / raw)
To: u-boot
On 11/24/2014 09:03 PM, Simon Glass wrote:
> Hi,
>
> I'm a bit messed up with my testing now.
>
> 1. Jetson-TK1: While I2C testing I unfortunately erased part of the
> I2C board ID EEPROM, and now it will not boot. It says:
>
> Dummy read for TPS65913
> Board id = 0x1ff
>
> and then dies. I can probably change the board id back, but I'm not
> sure what it should be! It will boot U-Boot but not the AVP boot
> loader. If someone could list out the contents of their EEPROM (bus 1,
> address 56) that might help.
"AVP boot loader" isn't quite the right term; all bootloaders on Tegra
start out execution on the AVP, and typically transition to the main CPU
(CCPLEX). "fastboot" is unfortunately (since it's also a protocol name)
the term we typically use for the Android/L4T/binary bootloader on this
chip.
Here's my ID EEPROM dump; I don't recall if there's a serial number in
here, or just board type identification. It probably doesn't matter
either way though.
(I2C bus IDs are different in Linux I guess)
# i2cdump -f -y 0 0x56
No size specified (using byte-data access)
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: 01 00 07 00 77 01 00 00 03 42 00 ff ff ff ff ff ?.?.w?..?B......
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
20: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
> 'part list mmc 0' just prints an error:
>
> ## Unknown partition table
>
> It is getting a partition ID of 0. Is there something wrong with
> Beaver's mmc support in mainline?
>
> The old 2011 U-Boot on the board works though.
Hmm. I haven't tried very recent U-Boot on Beaver; let me check.
> Also a question - are there any Jetson-like Tegra 132 boards available yet?
Not that I know of. Nexus 9 seems to be the best bet, although it's not
useful for bootloader work.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [U-Boot] Tegra board problems
2014-11-25 4:03 [U-Boot] Tegra board problems Simon Glass
2014-11-25 16:00 ` Stephen Warren
@ 2014-11-25 16:11 ` Stephen Warren
2014-11-25 16:16 ` Simon Glass
1 sibling, 1 reply; 6+ messages in thread
From: Stephen Warren @ 2014-11-25 16:11 UTC (permalink / raw)
To: u-boot
On 11/24/2014 09:03 PM, Simon Glass wrote:
> Hi,
>
> I'm a bit messed up with my testing now.
...
> 2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
> 'part list mmc 0' just prints an error:
>
> ## Unknown partition table
>
> It is getting a partition ID of 0. Is there something wrong with
> Beaver's mmc support in mainline?
I just built commit 2a82ec77d27e "Prepare v2015.01-rc2" and both eMMC
and SD appear to work.
Do you have a valid partition table on the eMMC? If you installed L4T
there rather than partitioning it yourself, note that L4T uses a
non-standard "tegra PT" partition table format. Some newer versions of
L4T also include a backup GPT that makes some of the partitions
available in a more standard way, but IIUC there's never a primary GPT.
I don't recall if the version of L4T for Beaver was new enough to do
even that though. Your best bet is to use standard tools to install a
complete GPT and then install a Linux distro on top of that.
Tegra30 (Beaver) # part list mmc 0
Partition Map for MMC device 0 -- Partition Type: EFI
Part Start LBA End LBA Name
Attributes
Type GUID
Partition GUID
1 0x00000800 0x01da97ff ""
attrs: 0x0000000000000000
type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
guid: ac3c736e-a29b-43d4-9819-4cf4fd84c3b0
Tegra30 (Beaver) # part list mmc 1
Partition Map for MMC device 1 -- Partition Type: EFI
Part Start LBA End LBA Name
Attributes
Type GUID
Partition GUID
1 0x00000022 0x003b9132 ""
attrs: 0x0000000000000000
type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
guid: b2f82cda-2535-4779-b467-094a210fbae7
^ permalink raw reply [flat|nested] 6+ messages in thread
* [U-Boot] Tegra board problems
2014-11-25 16:11 ` Stephen Warren
@ 2014-11-25 16:16 ` Simon Glass
2014-11-25 16:25 ` Stephen Warren
0 siblings, 1 reply; 6+ messages in thread
From: Simon Glass @ 2014-11-25 16:16 UTC (permalink / raw)
To: u-boot
Hi Stephen,
On 25 November 2014 at 09:11, Stephen Warren <swarren@wwwdotorg.org> wrote:
> On 11/24/2014 09:03 PM, Simon Glass wrote:
>>
>> Hi,
>>
>> I'm a bit messed up with my testing now.
>
> ...
>>
>> 2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
>> 'part list mmc 0' just prints an error:
>>
>> ## Unknown partition table
>>
>> It is getting a partition ID of 0. Is there something wrong with
>> Beaver's mmc support in mainline?
>
>
> I just built commit 2a82ec77d27e "Prepare v2015.01-rc2" and both eMMC and SD
> appear to work.
>
> Do you have a valid partition table on the eMMC? If you installed L4T there
> rather than partitioning it yourself, note that L4T uses a non-standard
> "tegra PT" partition table format. Some newer versions of L4T also include a
> backup GPT that makes some of the partitions available in a more standard
> way, but IIUC there's never a primary GPT. I don't recall if the version of
> L4T for Beaver was new enough to do even that though. Your best bet is to
> use standard tools to install a complete GPT and then install a Linux distro
> on top of that.
>
> Tegra30 (Beaver) # part list mmc 0
>
> Partition Map for MMC device 0 -- Partition Type: EFI
>
> Part Start LBA End LBA Name
> Attributes
> Type GUID
> Partition GUID
> 1 0x00000800 0x01da97ff ""
> attrs: 0x0000000000000000
> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
> guid: ac3c736e-a29b-43d4-9819-4cf4fd84c3b0
>
> Tegra30 (Beaver) # part list mmc 1
>
> Partition Map for MMC device 1 -- Partition Type: EFI
>
> Part Start LBA End LBA Name
> Attributes
> Type GUID
> Partition GUID
> 1 0x00000022 0x003b9132 ""
> attrs: 0x0000000000000000
> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
> guid: b2f82cda-2535-4779-b467-094a210fbae7
>
OK thank you, that could be my problem. I'll try installing a new GPT.
Will try out the Jetson and see if I can get it running again. So does
that mean as shipped it runs an Android boot loader and then jumps to
U-Boot?
Regards,
Simon
^ permalink raw reply [flat|nested] 6+ messages in thread
* [U-Boot] Tegra board problems
2014-11-25 16:16 ` Simon Glass
@ 2014-11-25 16:25 ` Stephen Warren
2014-11-26 15:39 ` Simon Glass
0 siblings, 1 reply; 6+ messages in thread
From: Stephen Warren @ 2014-11-25 16:25 UTC (permalink / raw)
To: u-boot
On 11/25/2014 09:16 AM, Simon Glass wrote:
> Hi Stephen,
>
> On 25 November 2014 at 09:11, Stephen Warren <swarren@wwwdotorg.org> wrote:
>> On 11/24/2014 09:03 PM, Simon Glass wrote:
>>>
>>> Hi,
>>>
>>> I'm a bit messed up with my testing now.
>>
>> ...
>>>
>>> 2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
>>> 'part list mmc 0' just prints an error:
>>>
>>> ## Unknown partition table
>>>
>>> It is getting a partition ID of 0. Is there something wrong with
>>> Beaver's mmc support in mainline?
>>
>>
>> I just built commit 2a82ec77d27e "Prepare v2015.01-rc2" and both eMMC and SD
>> appear to work.
>>
>> Do you have a valid partition table on the eMMC? If you installed L4T there
>> rather than partitioning it yourself, note that L4T uses a non-standard
>> "tegra PT" partition table format. Some newer versions of L4T also include a
>> backup GPT that makes some of the partitions available in a more standard
>> way, but IIUC there's never a primary GPT. I don't recall if the version of
>> L4T for Beaver was new enough to do even that though. Your best bet is to
>> use standard tools to install a complete GPT and then install a Linux distro
>> on top of that.
>>
>> Tegra30 (Beaver) # part list mmc 0
>>
>> Partition Map for MMC device 0 -- Partition Type: EFI
>>
>> Part Start LBA End LBA Name
>> Attributes
>> Type GUID
>> Partition GUID
>> 1 0x00000800 0x01da97ff ""
>> attrs: 0x0000000000000000
>> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
>> guid: ac3c736e-a29b-43d4-9819-4cf4fd84c3b0
>>
>> Tegra30 (Beaver) # part list mmc 1
>>
>> Partition Map for MMC device 1 -- Partition Type: EFI
>>
>> Part Start LBA End LBA Name
>> Attributes
>> Type GUID
>> Partition GUID
>> 1 0x00000022 0x003b9132 ""
>> attrs: 0x0000000000000000
>> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
>> guid: b2f82cda-2535-4779-b467-094a210fbae7
>>
>
> OK thank you, that could be my problem. I'll try installing a new GPT.
>
> Will try out the Jetson and see if I can get it running again. So does
> that mean as shipped it runs an Android boot loader and then jumps to
> U-Boot?
It probably depends on which L4T release you have.
I think the following is accurate:
The first L4T releases for Jetson may have only supported the Android
boot loader.
A later release certainly supported either the Android boot loader or
U-Boot. You had to choose which to install at flashing time. The Android
boot loader is/was the default, and is/was what is installed at the
factory. In either case, there's no chain-loading of boot loaders,
simply a choice of which to put into flash.
I believe the latest release only contains U-Boot, or perhaps supports
both with U-Boot as the default, but I haven't validated that, nor
whether it's actually shipped yet.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [U-Boot] Tegra board problems
2014-11-25 16:25 ` Stephen Warren
@ 2014-11-26 15:39 ` Simon Glass
0 siblings, 0 replies; 6+ messages in thread
From: Simon Glass @ 2014-11-26 15:39 UTC (permalink / raw)
To: u-boot
Hi Stephen,
On 25 November 2014 at 09:25, Stephen Warren <swarren@wwwdotorg.org> wrote:
>
> On 11/25/2014 09:16 AM, Simon Glass wrote:
>>
>> Hi Stephen,
>>
>> On 25 November 2014 at 09:11, Stephen Warren <swarren@wwwdotorg.org> wrote:
>>>
>>> On 11/24/2014 09:03 PM, Simon Glass wrote:
>>>>
>>>>
>>>> Hi,
>>>>
>>>> I'm a bit messed up with my testing now.
>>>
>>>
>>> ...
>>>>
>>>>
>>>> 2. beaver: I cannot get mainline U-Boot to boot a kernel. In fact
>>>> 'part list mmc 0' just prints an error:
>>>>
>>>> ## Unknown partition table
>>>>
>>>> It is getting a partition ID of 0. Is there something wrong with
>>>> Beaver's mmc support in mainline?
>>>
>>>
>>>
>>> I just built commit 2a82ec77d27e "Prepare v2015.01-rc2" and both eMMC and SD
>>> appear to work.
>>>
>>> Do you have a valid partition table on the eMMC? If you installed L4T there
>>> rather than partitioning it yourself, note that L4T uses a non-standard
>>> "tegra PT" partition table format. Some newer versions of L4T also include a
>>> backup GPT that makes some of the partitions available in a more standard
>>> way, but IIUC there's never a primary GPT. I don't recall if the version of
>>> L4T for Beaver was new enough to do even that though. Your best bet is to
>>> use standard tools to install a complete GPT and then install a Linux distro
>>> on top of that.
>>>
>>> Tegra30 (Beaver) # part list mmc 0
>>>
>>> Partition Map for MMC device 0 -- Partition Type: EFI
>>>
>>> Part Start LBA End LBA Name
>>> Attributes
>>> Type GUID
>>> Partition GUID
>>> 1 0x00000800 0x01da97ff ""
>>> attrs: 0x0000000000000000
>>> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
>>> guid: ac3c736e-a29b-43d4-9819-4cf4fd84c3b0
>>>
>>> Tegra30 (Beaver) # part list mmc 1
>>>
>>> Partition Map for MMC device 1 -- Partition Type: EFI
>>>
>>> Part Start LBA End LBA Name
>>> Attributes
>>> Type GUID
>>> Partition GUID
>>> 1 0x00000022 0x003b9132 ""
>>> attrs: 0x0000000000000000
>>> type: ebd0a0a2-b9e5-4433-87c0-68b6b72699c7
>>> guid: b2f82cda-2535-4779-b467-094a210fbae7
>>>
>>
>> OK thank you, that could be my problem. I'll try installing a new GPT.
>>
>> Will try out the Jetson and see if I can get it running again. So does
>> that mean as shipped it runs an Android boot loader and then jumps to
>> U-Boot?
>
>
> It probably depends on which L4T release you have.
>
> I think the following is accurate:
>
> The first L4T releases for Jetson may have only supported the Android boot loader.
>
> A later release certainly supported either the Android boot loader or U-Boot. You had to choose which to install at flashing time. The Android boot loader is/was the default, and is/was what is installed at the factory. In either case, there's no chain-loading of boot loaders, simply a choice of which to put into flash.
>
> I believe the latest release only contains U-Boot, or perhaps supports both with U-Boot as the default, but I haven't validated that, nor whether it's actually shipped yet.
Thanks for the info, very helpful.
Regards,
Simon
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2014-11-26 15:39 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2014-11-25 4:03 [U-Boot] Tegra board problems Simon Glass
2014-11-25 16:00 ` Stephen Warren
2014-11-25 16:11 ` Stephen Warren
2014-11-25 16:16 ` Simon Glass
2014-11-25 16:25 ` Stephen Warren
2014-11-26 15:39 ` Simon Glass
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox