Skip to contentSkip to conversation

lubbock patch: ram, flash, ethernet

U-Boot-Users David Brownell 4 participants 5 messages Mar 11, 2005 – Sep 25, 2005

lore
Share & tools
lore
#0 lore raw

U-Boot support for Lubbock still needs some work, but this fixes a few of the more blatant problems. I look forward to replacing that crufty old "BLOB" with a shiny new U-Boot. :)

I saw some email archived from about 10 months back saying that the TFTP timeout/retry problems went away given PXA timer updates, but like the ethernet patch that was also needed, a patch wasn't posted or merged. Maybe someone who has unmerged PXA updates has the relevant code ready to merge?

- Dave
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lubbock-0310.patch
Type: text/x-diff
Size: 3802 bytes
Desc: not available
Url : http://lists.denx.de/pipermail/u-boot/attachments/20050310/9dcbe500/attachment.patch

#1 lore raw
Quoted 5 lines · David Brownell

I saw some email archived from about 10 months back saying that the TFTP timeout/retry problems went away given PXA timer updates, but like the ethernet patch that was also needed, a patch wasn't posted or merged. Maybe someone who has unmerged PXA updates has the relevant code ready to merge?

There's a gumstix patch against 1.1.2 which fixes the timer problems which lead to TFTP timeouts on PXA.

http://localhost/svn/gumstix-buildroot/trunk/sources/u-boot-
get_timer.patch

I haven't submitted the patch yet, because I have a handful of others too which need to be cleaned of stuff which doesn't want to be in the mainline before I submit them.

That one though, is pretty clean and nicely separate from anything that doesn't apply to all PXA platforms.

C

#2 lore raw
Quoted 10 lines · Craig Hughes
Quoted 5 lines · David Brownell

I saw some email archived from about 10 months back saying that the TFTP timeout/retry problems went away given PXA timer updates, but like the ethernet patch that was also needed, a patch wasn't posted or merged. Maybe someone who has unmerged PXA updates has the relevant code ready to merge?

There's a gumstix patch against 1.1.2 which fixes the timer problems which lead to TFTP timeouts on PXA.

Which, to save anyone else tracking down the Gumstix subversion info, I've attached. Adding this over my patch (and a few other tweaks, only one of which might be important) gives me a U-Boot that can download a kernel starting from DHCP! Worth merging.

Thanks, Craig. :)

- Dave

p.s. Those MMC patches looked rather interesting too. A fully
featured Lubbock U-Boot would talk Ethernet, MMC, and CF; we're not there yet, but given Ethernet I'd call it very much usable.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: u-boot-get_timer.patch
Type: text/x-diff
Size: 494 bytes
Desc: not available
Url : http://lists.denx.de/pipermail/u-boot/attachments/20050311/a88152fd/attachment.patch

#3 lore raw
Quoted 4 lines · David Brownell

p.s. Those MMC patches looked rather interesting too. A fully
featured Lubbock U-Boot would talk Ethernet, MMC, and CF; we're not there yet, but given Ethernet I'd call it very much usable.

The MMC patches are part of what's not really complete there at the moment. There's some SD support I want to fold back in, and some ugly debugging stuff which needs to be cleaned up.

If anyone is curious in the meantime, feel free to browse/take from

http://svn.rungie.com/svn/gumstix-buildroot/trunk/sources/

The u-boot* files are the ones of interest here.

C

#4 lore raw
In reply to David Brownell
Quoted 9 lines · In message <[email protected]> you wrote:

This patch has three fixes to the Lubbock (PXA 25x) support:

* Resolve the FIXME with respect to saving the u-boot environment. It uses the second sector; default bootcommand now uses the third.

* Make the default load address land in real memory: 32KB past start of RAM, as recommended by Linux 2.6 ARM booting docs.

Quoted 8 lines

* Fix lan91c96 SMC_{in,out}{b,w,l}() macros:

- The address fixups (SMC A0..A9 == PXA A2..A11) were wrong for bytewise access on Lubbock ... and inappropriate on most other platforms. This uses a new SMC_IO_SHIFT #define, just like the Linux 2.6 smc91x driver.

- Evil unparenthesized references to macro parameters are gone.

Added, thanks. But please provide a proper CHANGELOG entry next time.

Best regards,

Wolfgang Denk

Signature · 4 lines
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
You can love it, change it, or leave it.    There is NO other option.
But do not complain - it is your own choice...                  -- wd