Hi Wolfgang,
On Tue, Mar 20, 2012 at 12:03 PM, Stephen Warren swarren@wwwdotorg.org wrote:
On 03/19/2012 10:59 PM, Simon Glass wrote:
This reverts commit 295d3942b806552503243f5cfb36aec6f1b5a9bf.
It turns that this really doesn't work very nicely. Instead we should have a pre-console panic function so that we know that further execution is impossible and we don't need to worry about trampling on UARTs, etc.
The series:
Tested-by: Stephen Warren swarren@wwwdotorg.org
If you have concerns with this new approach please sing out.
For reference, I took u-boot-tegra/master, removed the top 4 commits that were the previous putc changes, then applied the 5 commits in this series.
Tom (assuming everyone else is OK with this series), when you apply this series, can you put it /before/ the USB/device-tree patches in your branch, so that there are no points in git history where the USB/DT code is active, but this code is not. In other words, apply these 5 commits, then use "git rebase -i" to re-order the patches so this series is before all the USB/DT changes. Thanks.
Regards, Simon