From: Stephen Warren <swarren@wwwdotorg.org>
To: u-boot@lists.denx.de
Subject: [U-Boot] pull request for u-boot-tegra/master
Date: Fri, 08 Jun 2012 16:52:26 -0600 [thread overview]
Message-ID: <4FD2822A.7050703@wwwdotorg.org> (raw)
In-Reply-To: <4FD27E87.1070305@wwwdotorg.org>
On 06/08/2012 04:36 PM, Stephen Warren wrote:
> On 06/08/2012 04:00 PM, Stephen Warren wrote:
>> On 06/08/2012 03:44 PM, Albert ARIBAUD wrote:
>>> Hi Tom,
>>>
>>> Le 31/05/2012 18:47, Tom Warren a ?crit :
>>>> Albert,
>>>>
>>>> Please pull u-boot-tegra/master into ARM master. Thanks!
>>>>
>>>> The following changes since commit
>>>> 2ca4a209a5b961ad1be8782c68dabe326d77dfaf:
>>>> SRICHARAN R (1):
>>>> OMAP4/5: Change omap4_sdp, omap4_panda, omap5_evm maintainer
>>>>
>>>> are available in the git repository at:
>>>>
>>>> git://git.denx.de/u-boot-tegra master
>>>
>>> Trying trimslice with the stock Linaro toolchain from Xubuntu:
>>>
>>> uboot at lilith:~/src/u-boot-arm$ LC_ALL=C ./MAKEALL trimslice
>>> Configuring for trimslice board...
>>> text data bss dec hex filename
>>> 234794 4716 276460 515970 7df82 ./u-boot
>>> /bin/sh: line 1: exit: too many arguments
>>> make[1]: *** [dt.dtb] Error 1
>>> make: *** [u-boot.dtb] Error 2
>>>
>>> Are there special needs for building trimslice?
>>
>> There should be nothing special about TrimSlice vs. any other Tegra board.
>>
>> This is somehow related to the dtc (device tree compiler) version that's
>> picked up by the build process. I can reproduce it by placing Ubuntu
>> Lucid's dtc in my PATH, but not with another version of dtc I had in my
>> path (require to build our downstream U-Boot device trees).
>>
>> I'll investigate, unless anyone else wants to.
>
> In order to see the real problem with dtc, you will need:
>
> http://lists.denx.de/pipermail/u-boot/2012-June/125967.html
>
> The problem is as follows:
>
>> /home/swarren/shared/git_wa/u-boot/arch/arm/dts/tegra20.dtsi:3 syntax error
>> FATAL ERROR: Couldn't read input tree
>> make[1]: *** [dt.dtb] Error 1
>
> This is because tegra-trimslice.dts contains:
>
>> /include/ ARCH_CPU_DTS
>
> which is run through cpp to substitute the correct absolute path:
>
>> /include/ "/home/swarren/shared/git_wa/u-boot/arch/arm/dts/tegra20.dtsi"
>
> However, some versions of dtc appear to have a problem parsing this
> path; perhaps it's the leading /?
Uggh. The problem is that older versions of dtc would only accept a
single definition of the root node; they don't allow another copy to be
specified, which "overlays" or is "merged with" it. Put another way, dtc
can't compile a simple:
> /dts-v1/;
>
> / {
> foo = "bar";
> };
>
> / {
> foo = "bar";
> };
That means we can't separate the device tree source into separate
skeleton.dtsi, tegra20.dtsi, and tegra2-$board.dts.
So, we have basically no choice but to require people to upgrade to a
recent dtc. If we do that, we can also get rid of the cpp hacks, since
the latest dtc has a -i option that can be used to set an include path.
next prev parent reply other threads:[~2012-06-08 22:52 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-05-31 16:47 [U-Boot] pull request for u-boot-tegra/master Tom Warren
2012-06-07 16:02 ` Tom Warren
2012-06-08 19:36 ` Albert ARIBAUD
2012-06-08 19:41 ` Albert ARIBAUD
2012-06-08 21:44 ` Albert ARIBAUD
2012-06-08 22:00 ` Stephen Warren
2012-06-08 22:36 ` Stephen Warren
2012-06-08 22:52 ` Stephen Warren [this message]
2012-06-09 1:53 ` Simon Glass
2012-06-11 21:59 ` Stephen Warren
2012-06-10 8:22 ` Albert ARIBAUD
-- strict thread matches above, loose matches on Subject: below --
2012-08-31 21:28 Tom Warren
2012-09-01 6:56 ` Albert ARIBAUD
2012-08-17 23:25 Tom Warren
[not found] ` <20120818212141.3e4f5a3a@lilith>
[not found] ` <CA+m5__J7O=YYWEwTGqTBqiagTL+YU3BR-TyE5M2W_xdesVWqmA@mail.gmail.com>
[not found] ` <5032684A.2040703@ti.com>
[not found] ` <CA+m5__+prjk_GgvK5uZf1m23t3nsSiYu-nYRcci4my59q7T6-w@mail.gmail.com>
[not found] ` <CA+m5__Kiq0-ZYjarTXZRFZTC71wWV=c0fmzCJW=-Dt5fgK+dxA@mail.gmail.com>
[not found] ` <20120825103410.7fc86952@lilith>
[not found] ` <20120828165448.GD9587@nvidia.com>
[not found] ` <20120830083325.031e31d4@lilith>
[not found] ` <CA+m5__+gNxyB1nVSe-8DTE=seqBvOQTz2NsPTh=h3kTOCY9rCw@mail.gmail.com>
[not found] ` <20120830224742.678423bc@lilith>
[not found] ` <5FBF8E85CA34454794F0F7ECBA79798F379E005FB3@HQMAIL04.nvidia.com>
[not found] ` <20120830234733.1aea4c7e@lilith>
[not found] ` <5FBF8E85CA34454794F0F7ECBA79798F379E006051@HQMAIL04.nvidia.com>
2012-08-31 6:32 ` Albert ARIBAUD
2012-07-30 20:07 Tom Warren
2012-08-03 23:35 ` Tom Warren
2012-08-05 19:46 ` Albert ARIBAUD
2012-08-05 20:01 ` Albert ARIBAUD
2012-08-06 15:53 ` Tom Warren
2012-08-06 17:37 ` Allen Martin
2012-08-06 22:46 ` Allen Martin
2012-08-07 6:53 ` Albert ARIBAUD
2012-08-07 16:03 ` Tom Warren
2012-08-07 17:15 ` Allen Martin
2012-08-06 15:18 ` Tom Warren
2012-08-07 16:27 ` Tom Warren
2012-07-09 17:49 Tom Warren
2012-07-09 20:42 ` Albert ARIBAUD
2012-05-03 15:56 Tom Warren
2012-05-04 7:30 ` Albert ARIBAUD
2012-05-04 15:36 ` Tom Warren
2012-04-26 21:14 Tom Warren
2012-03-28 21:48 Tom Warren
2012-03-29 6:13 ` Albert ARIBAUD
2012-03-09 22:39 Tom Warren
2012-03-10 8:25 ` Wolfgang Denk
2012-03-10 19:34 ` Simon Glass
2012-03-12 19:30 ` Tom Warren
2012-03-12 20:24 ` Simon Glass
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4FD2822A.7050703@wwwdotorg.org \
--to=swarren@wwwdotorg.org \
--cc=u-boot@lists.denx.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.