On Friday, September 10, 2010 14:53:10 Wolfgang Denk wrote:

Mike Frysinger wrote:

...

not sure if you wanted to include these for 2010.09, or wait until after: 11.08.2010 [PATCH] Makefile: restore support for board shortcut targets 19.08.2010 [PATCH v2] MAKEALL: cut down on duplication of targets 15.08.2010 [PATCH 1/4] tools: enable img2srec for "all" target 15.08.2010 [PATCH 2/4] tools: update .gitignore 15.08.2010 [PATCH 3/4] unify toplevel tools targets 15.08.2010 [PATCH 4/4] tools/env: use host build flags

I've seen these on my list for "next". Should these go into 2010.09?

if you want to keep the tree settled, then keeping them in "next" is OK by me. not that there should be anything wrong with these changes ... -mike

Dear Ben Gardiner,

In message AANLkTinFDGngQiuLvVEX2d3+=fUDwa4uwf4Y9t9hOpxR@mail.gmail.com you wrote:

Tested on da850evm with NAND enabled; the env.OOB marker was recognized, the environment was loaded and I was able to load and boot a uImage kernel from NAND.

Thanks a lot.

Best regards,

Wolfgang Denk

Dear Dirk Behme,

In message 4C8B1A02.4070301@googlemail.com you wrote:

Dear Wolfgang,

On 10.09.2010 19:54, Wolfgang Denk wrote:

...

... check if all your relevant patches have been included.

What's about

http://lists.denx.de/pipermail/u-boot/2010-July/074542.html

Doesn't apply:

Applying: Remove unused variable bi_env fatal: sha1 information is lacking or useless (common/cmd_bdinfo.c). Repository lacks necessary blobs to fall back on 3-way merge. Cannot fall back to three-way merge. Patch failed at 0001 Remove unused variable bi_env When you have resolved this problem run "git am --resolved". If you would prefer to skip this patch, instead run "git am --skip". To restore the original branch and stop patching run "git am --abort".

? Or if you don't like that, this

http://lists.denx.de/pipermail/u-boot/2010-August/075303.html

This one has been applied to the "next" branch.

Best regards,

Wolfgang Denk

On Sunday, September 12, 2010 08:58:29 Reinhard Meyer wrote:

1110f276e9785d0270cfc9ec316dd475145f95d0 SPI: cmd_spi.c: add options to specify bus and mode - has been posted 26.08.2010

i already merged into my sf branch and will push post 2010.09 if Woflgang doesnt pick it up himself

7c7e2813318b84fc4fbec57b23e2ad68ff000564 NET: move legacy enc28j60.c to sidetrack - Ben took that one to net/next on 31.08.2010

that's Ben's call to move from net to mainline ... and unless there's a pressing reason for this to be in mainline (i cant see any), keeping it for post release is OK ... -mike

Dear Mike Frysinger,

In message 201009121518.13750.vapier@gentoo.org you wrote:

On Sunday, September 12, 2010 08:58:29 Reinhard Meyer wrote:

...

1110f276e9785d0270cfc9ec316dd475145f95d0 SPI: cmd_spi.c: add options to specify bus and mode - has been posted 26.08.2010

i already merged into my sf branch and will push post 2010.09 if Woflgang doesnt pick it up himself

I see this for "next" - pull requests welcome.

["next" will be out RSN.]

Best regards,

Wolfgang Denk

Dear Reinhard Meyer,

In message 4C8CCE75.1000505@emk-elektronik.de you wrote:

Dear Wolfgang Denk,

...

Please help testing, and check if all your relevant patches have been included.

Rebasing worked without manual intervention, u-boot still compiles and works for my board (at91sam9xe-based). Code size increased 0x20 bytes.

The following (published) patches are left ahead of master here:

3d70dcf2d8ea7e7a50d5d5d5a0429341db87e38f display_buffer: fix misaligned buffer

• urgent!

Is in "master".

1110f276e9785d0270cfc9ec316dd475145f95d0 SPI: cmd_spi.c: add options to specify bus and mode

• has been posted 26.08.2010

Scheduled for next.

7c7e2813318b84fc4fbec57b23e2ad68ff000564 NET: move legacy enc28j60.c to sidetrack

• Ben took that one to net/next on 31.08.2010

I'm waiting for his pull req.

Best regards,

Wolfgang Denk

On Monday, September 13, 2010, Clint Adams schizo@debian.org wrote:

On Fri, Sep 10, 2010 at 07:54:40PM +0200, Wolfgang Denk wrote:

...

Please help testing, and check if all your relevant patches have been included.

In arch/i386/lib/board.c, the declaration of do_go_exec() has changed, thus leading to the following build failure[0]:

board.c: In function 'do_go_exec': board.c:440: error: assignment of read-only location '*(argv + 0xfffffffffffffffffffffffffffffffcu)' cc1: warnings being treated as errors board.c:442: error: passing argument 2 of 'entry' discards qualifiers from pointer target type board.c:442: note: expected 'char **' but argument is of type 'char * const*' make[2]: *** [board.o] Error 1

I have a fix for this in my offline tree. I'm ok with missing this merge window - I'll upload my patches for the next one.

Thanks for catching this though

Regards,

Graeme

Wolfgang,

On 13/09/10 05:04, Wolfgang Denk wrote:

Dear Graeme Russ,

In message AANLkTi=iHWNM2hkPU58qOh_FojdXYc1LQbUKEopFcWpu@mail.gmail.com you wrote:

...

I have a fix for this in my offline tree. I'm ok with missing this merge window - I'll upload my patches for the next one.

Fixes should go in _now_. This is what this phase of the development cycle is for.

Pull request sent.

Wolfgang, I was waiting for feedback for some other patches in the series, but have cut the series short to at least get the fixes in.

This is my first ever git push / git pull request and I stuffed up and included one extra patch (removal of unmaintained boards) which is not a fix, but would have gone in next merge - sorry about that

Regards,

Graeme

Dear Steve Sakoman,

I did a bit of v2010.09-rc1 testing on OMAP3 this evening and things looked good until I tried a simple memory display!

The output froze after the memory address was displayed:

Overo # md 82000000 10 82000000:

That _might_ be the unaligned access to the buffer. Try

[PATCH v2] display_buffer: fix misaligned buffer

from 31.08.2010. If that helps its even more important that patch gets mainline asap ;)

Reinhard

On Mon, Sep 13, 2010 at 12:24 AM, Stefano Babic sbabic@denx.de wrote:

Steve Sakoman wrote:

...

On Fri, Sep 10, 2010 at 10:54 AM, Wolfgang Denk wd@denx.de wrote:

...

Hello everybody:

• U-Boot v2010.09-rc1 was released on Fri, 10 Sep 2010.

• Release "v2010.09" is scheduled in 18 days - on September 28, 2010,

i. e. we have a delay against the original plan of about 2 weeks.

For details please see http://www.denx.de/wiki/U-Boot/ReleaseCycle

Please help testing, and check if all your relevant patches have been included.

I did a bit of v2010.09-rc1 testing on OMAP3 this evening and things looked good until I tried a simple memory display!

The output froze after the memory address was displayed:

Overo # md 82000000 10 82000000:

Hi Steve,

do not worry, this is a known problem. As reported by Rheinard, we need the patch to fix misaligned access. It is was not merged up now, but we need soon before the release:

"3d70dcf2d8ea7e7a50d5d5d5a0429341db87e38f display_buffer: fix misaligned buffer

• urgent!"

I tried this patch and it does indeed fix the issue.

I see that Wolfgang has already pushed it to master, so I'll get back to testing the rc-1 on OMAP3 and OMAP4.

Thanks for the quick response!

Steve

Dear Stefano Babic,

In message 4C8DE7F0.3020903@denx.de you wrote:

a recent post pointed out to the following patch:

[PATCH] [PATCH v6]Save environment data to mmc.

http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/78772

I see no urgency, but I have not seen any opened issues with this patch and first version was sent some months ago. Any chance to get it in the release or should we only mark it for the "next" branch ?

Did you test this? Is this to be considered a formal ACK?

Best regards,

Wolfgang Denk