kubuntu patches and backtraces

[Harald Sitter]

On Thu, Dec 12, 2013 at 4:25 PM, Scott Kitterman <ubuntu at kitterman.com> wrote:
> But only for long term, not upstreamable changes, right?

The upstreamable ones are per the standing policy (since Lucid?)
expected to go upstream first anyway.

While applying the naming to such patches would in fact be necessary
to carry the thought through and ensure backtraces always show changes
to upstream code I don't think it viable for a patch that may be
around for a month or two. Also it would then need to be applied to
*all* patches which would cause *substantial* creation and maintenance
overhead. For example if we backport a fix from upstream's 4.11 git
branch (targetting 4.11.4) to our 4.11.3 packages that patch would
then need to be modified to reflect it's patchyness in a backtrace. I
at least am ETOOLAZY for that :P

The primary motivation for the function prefixes is to enable everyone
to identify *our distro specific code* crashing.

HS