Merge lp://staging/~dave-long/linaro-image-tools/fix-panda-u-boot into lp://staging/linaro-image-tools/11.11

Proposed by David Long
Status: Merged
Merged at revision: 403
Proposed branch: lp://staging/~dave-long/linaro-image-tools/fix-panda-u-boot
Merge into: lp://staging/linaro-image-tools/11.11
Diff against target: 25 lines (+4/-0)
2 files modified
linaro_image_tools/media_create/boards.py (+2/-0)
linaro_image_tools/media_create/tests/test_media_create.py (+2/-0)
To merge this branch: bzr merge lp://staging/~dave-long/linaro-image-tools/fix-panda-u-boot
Reviewer Review Type Date Requested Status
Ricardo Salveti (community) Approve
James Tunnicliffe (community) Approve
Review via email: mp+67554@code.staging.launchpad.net

Description of the change

Add lines to boot.scr for omap4 which will set the "initrd_high" and the new "fdt_high" environment variables to the magic value 0xffffffff, which instructs u-boot not to copy the initial ramdisk and flattened device tree to the end of SDRAM. This is necessary on platforms with more than about 3/4G of RAM as the Linux kernel does not seem to like finding the ramdisk or fdt in HIGHMEM.

To post a comment you must log in.
Revision history for this message
James Tunnicliffe (dooferlad) wrote :

Looks good to me.

review: Approve
Revision history for this message
James Westby (james-w) wrote :

On Mon, 11 Jul 2011 14:02:55 -0000, David Long <email address hidden> wrote:
> David Long has proposed merging lp:~dave-long/linaro-image-tools/fix-panda-u-boot into lp:linaro-image-tools.
>
> Requested reviews:
> linaro-image-tools maintainers (linaro-image-tools)
>
> For more details, see:
> https://code.launchpad.net/~dave-long/linaro-image-tools/fix-panda-u-boot/+merge/67554
>
> Add lines to boot.scr for omap4 which will set the "initrd_high" and
> the new "fdt_high" environment variables to the magic value
> 0xffffffff, which instructs u-boot not to copy the initial ramdisk and
> flattened device tree to the end of SDRAM. This is necessary on
> platforms with more than about 3/4G of RAM as the Linux kernel does
> not seem to like finding the ramdisk or fdt in HIGHMEM.

Hi,

What will happen if the newer u-boot isn't used with this env?

Should this wait on the patches being accepted upstream?

Thanks,

James

Revision history for this message
David Long (dave-long) wrote :

On Mon, 2011-07-11 at 11:33 -0400, James Westby wrote:

> Hi,
>
> What will happen if the newer u-boot isn't used with this env?
>
> Should this wait on the patches being accepted upstream?

The fdt_tree environment variable will be ignored, and the initial
ramdisk will run in a different place. The system boots just fine this
way. It's a hybrid of what happens with and with the device tree
present at boot.

The change that really does need to come last is the one to
arch/arm/boot/dts/omap4-panda.dts, which I have not sent out a patch for
yet. That change will actually make use of highmem.

-dl

Revision history for this message
David Long (dave-long) wrote :

The fdt_tree environment variable will be ignored, and the initial ramdisk will run in a different place. The system boots just fine this way. It's a hybrid of what happens with and with the device tree present at boot.

The change that really does need to come last is the one to arch/arm/boot/dts/omap4-panda.dts, which I have not sent out a patch for yet. That change will actually make use of highmem.

Revision history for this message
Ricardo Salveti (rsalveti) wrote :

Seems this change is safe, as older u-boot will just ignore the new variable, and setting initrd_high to 0xffffffff shouldn't change the boot behavior.

James, can you merge this now that 11.07 is done? Then we can make sure everything gets well tested during 11.08 cycle.

review: Approve

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
The diff is not available at this time. You can reload the page or download it.

Subscribers

People subscribed via source and target branches