Add two new subcommands for cherry-picking fixes from dinq to
drm-intel-fixes and drm-intel-next-fixes. The only difference in the
subcommands is the assert branch check to ensure the user is on the
right branch.

The commands scan dinq for commits Cc'd to stable or drm-intel-fixes,
checks whether they've already been backported, attempts cherry-pick,
and asks for directions on failed cherry-pick.

It's still rough around the edges and slow as molasses due to unlimited
scan for backports (should just check a release or two back at most),
and an aborted run just starts over next time (though it should take the
freshly backported commits into account).

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Show usage also with the -h option, and improve output on unknown
options.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

I find displaying the man page on every typo a bit distracting. Now that
we can list all the supported subcommands (by way of listing all the
dim_ prefixed functions), add usage to show quick help. Also make it a
subcommand of its own.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

dim help being the last subcommand handled by the switch case,
obliterate it as well, finishing the job we set out to do. Now all
subcommands are handled via dim_ prefixed functions.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Helpful convenience.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

If there's a dim_ prefixed function matching the subcommand, call it to
handle the subcommand without going through the big switch case.

Drop the cases which already match the convention. Once all the cases
are converted to functions, the handling of subcommands becomes much
cleaner.

This also allows extending of dim by the user via functions declared
outside of dim. The current aliases could also be made the user's
responsibility to define as the user wishes (the user could define dim_
prefixed functions in the dimrc).

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Allow specifying dim subcommand aliases using variables of the format
dim_alias_<ALIAS>. The value of the variable should be the name of the
actual subcommand. The user can specify aliases in the dimrc.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

More abstraction is better, also guards against invalid internal use.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Add dim_ prefix to functions corresponding to subcommands.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the subcommand name.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Add dim_ prefix to functions corresponding to subcommands.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the subcommand.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the subcommand. Also add checkpatch subcommand (without
the hyphen) to match the spelling of the tool.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename as there's no corresponding subcommand (the check-patch
subcommand does a slightly different thing).

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Drop the dim_ prefix as there's no corresponding subcommand.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the (long) subcommand name. Move error checking to the
function too.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the subcommand name.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Rename to match the subcommand name. Move error checking to the function
too.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Yeah dim apply-patch could probably autofudge this, but it's too close
to w/e time for me to figure out how to do that.

Noticed since a few folks forgot this when pushing other people's
patches.

v2: Adjust to Tvrtko's patch and also update todo.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

It is very easy to forget that commiters should sign-off
patches not authored by themselves.

v2: Missed to make sob variable local and align with dim
    coding style. (Daniel Vetter)

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>

Currently CI servers still push test results to internal servers, and
that's why we right now can't document this all properly in public.
Meanwhile just add a basic tl;dr section as a placeholder.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Also try to polish the formatting a bit.

v2: Fixup $DIM_PREFIX (Jani)

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

- Add links to for all the trees that get merged into -nightly (Yann
  Argotti asked about some of those details), including sound trees.

v2: Don't explain -internal.

Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Emabarrassing.

Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>