Ville Syrjälä authored 8 years ago and Jani Nikula committed 8 years ago

Add a command for extracting various tags (eg. Reviwed-by:) from
emails. You can give the comamnd a rangeish to add the tags from
the same email to multiple already applied patches.

The regexp used to pick up tags is purposefully quite broad. People
tend to typo these things, or add extra whitespace etc. However the
broad regexp does mean this occasionally picks up stuff that isn't
a tag. So manually amending the commit is probably a wise idea,
and so I simply decided to also leave a '--- extracted tags ---'
separator in the commit message just before the extracted tags,
which can be cleaned up manually when verifying that the tags look
correct.

v2: Drop the aliases
    Remove the temp file
    Clean up locals
    Other codingstyle cleanups
    Use '# *** ... ***' for the separator

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Helper to rebase a branch onto drm-tip.

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

This is what I currently use for cherry-picking fixes. Completely
non-interactive and stateless.

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

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

I pushed the patch a bit fast ...

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

Boris Brezillon authored 8 years ago and Simona Vetter committed 8 years ago

Some people are specifying the username directly in the url with the
following syntax: ssh://<username>@git.freedesktop.org/git/<repo>, but
dim expects something like ssh://git.freedesktop.org/git/<repo

>.

Update the documentation to document how to assign a default username
when connecting to git.freedesktop.org through SSH.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
[danvet: Apply suggestion from Daniel Stone, and markup-annotate.]
Reviewed-by: Daniel Stone <daniels@collabora.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

Let's make sure that review doesn't go through needless cycles. Diff
doesn't show this, but this is only for the "small drivers" part of
drm.

Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

My normalization hack fell over a topic branch :(

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

I'm typing this too often on irc.

Acked-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Kinda no reason only to use them for drm-intel-next ...

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

For the experiement we have right now Eric (with vc4) and Sean Paul
(with rockchip and zte) volunteering, and Gerd (entire pile of qemu
drivers) and Boris (atmel) are also considering to participate. I
think that's enough to get started and figure things out as we go.

I tried to summarize the main points from the rfc discussions into a
short chapter.

v2: Spelling fixes (Anholt).

v3: Clarify which branches to use (Sean).

Cc: Boris Brezillon <boris.brezillon@free-electrons.com>
Cc: Eric Anholt <eric@anholt.net>
Cc: Sean Paul <seanpaul@chromium.org>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Mark Yao <mark.yao@rock-chips.com>
Cc: Shawn Guo <shawnguo@kernel.org>
Acked-by: Sean Paul <seanpaul@chromium.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Acked-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Cc: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

The declare -n option was added in bash 4.3. Handle missing -n
gracefully, at the expense of losing dim alias functionality.

Cc: Gerd Hoffmann <kraxel@redhat.com>
Reported-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Chris ran into this, and I guess worth documenting best practices
since it happens sooooooo rarely.

v2: Clarify stuff a bit.

v3: Also normalize branch names in conflict markers. Without this we
can't use fixup patches to fixup up conflicts when git rerere fails to
store/reapply the resolution.

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

Avoids sprawnling branches everywhere, and makes the script faster for
folks who don't need everything. We already skip remotes, so this is
in line with what's there.

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

And roll it out.

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

And use it everywhere. Found a few more places that don't use the
usual style.

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

Maybe there's only a read-only remote available, and with the
drm-misc/-intel split that's perfectly fine.

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

Laurent got mightily confused about this, and assumed he's getting
volunteered as an intel maintainer because dim asks for an intel
remote.

With the branch autodetection we can postpone all the remote checks to
when we need them. This is similar to the changes to update-branches
to only update branches that exist, but for remotes.

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

- Remove branch overview, instead link to drm-intel and drm-misc
  pages.

- Move quickstart to the top, to make it easier to find.

- Make quickstart generic, we use dim for other stuff than drm-intel
  now.

- s/drm-intel-rerere/drm-rerere/

- Remove the section about resolving conflicts, that's now explained
  in detail in the process pages.

v2: Also eradicate dinq (Laurent).

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

Set $remote before using it.

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

Old versions of git fetch exited with 128 when there was nothing to
fetch. But ignoring all return values is bad as Chris recently found
out, since then rebuilding drm-tip might go back to older branch
versions if the fetch fails for some reason.

I also noticed that current git (I have 2.11 here) don't exhibit this
anymore.

v2: s/exit/return/ to give callers a chance to do something with this.
We use set -e, so same result (Jani).

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

It's been years since drm-intel had only one maintainer, and the
primary use-case for dim ub was only to sync between machines. Update
the docs to reflect this, to avoid misleading committers. Also update
the maintainer-tools branch, which is something Paulo's script also
does.

Motivated because Paulo typed an actualize.sh script which does
exactly what dim ub is meant to do.

v2: Also update quickstart guide.

v3: And spelling fixes.

v4: Fix title and only update maintainer-tools when not a private
branch is checkout out (Jani).

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

When a branch can be fast-forwarded, try it first before rebasing.
This will prevent a whole lot of editor windows opening with 'noop'
when running dim ub.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>

We insert the tag right below, and recent git versions started to
strongly suggested that tags should read like commit messages, i.e.
summary, blank line, extended content. That means I don't need a
heading of my own to make it look spiffy.

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

Without this, my non-git-worktree setup will fail to push (or it takes
forever) if the local drm-tip is not up to date.

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

Fixes: 6552af2d ("dim: fix update-branches to fetch all available nightly repos")
Cc: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

Useful with our branch proliferation to make sure nothing is stuck (we
now also have drm-misc-next/-next-fixes/-fixes).

v2: Gracefully handle if some remotes arent' there.

Acked-by:  <seanpaul@chromium.org>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>

Update all nightly repos that have a corresponding local remote.

Cc: Archit Taneja <architt@codeaurora.org>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>

I don't have user.email set, and instead rely on git's fallback to
EMAIL. Implement that. Couldn't apply patches anymore without this.

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

No need to read the config etc. twice. Add a check to prevent us from
adding new ones.

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

Noticed by Ville.

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

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

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

'dim list-upstreams' is needed by bash completion and is used outside of
git repos, also within git repos without drm upstream, so don't fail if
it can be avoided.

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

This lets us handle url_to_remote failures gracefully where they're not
fatal. Our use of 'set -e' guarantees this will bail out otherwise.

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

We rely on 'set -e' to bail out on errors. We also do a lot of local
variable assignments with command substitution like this:

	local foo=$(bar)

However, that masks the return value of the command, failing to bail out
on errors. Split up local variable declarations from assignments to
avoid the problems.

There are no issues with literal assignments like

	local foo="bar"

but split those out too for more uniform code, preferring to have local
declarations in the beginning of functions or nested blocks.

Details at
http://stackoverflow.com/questions/4421257/why-does-local-sweep-the-return-code-of-a-command



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

Reference drm-tip, use $integration_config, or just drop old/stale
comments.

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

Keep update-branches as an alias.

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

Noticed by Archit, but probably a good idea in general.

v2: Fetch before printing (Jani). While at it make them all less
noisy, it's kinda confusing to have random git output interleaved with
dim messages.

v3: Don't silence stderr (Jani)

v4: Use git fetch -q (Abdiel).

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