[PATCH] dim: fix the repo for drm-next

[Jani Nikula]

On Wed, 09 May 2018, Daniel Vetter <daniel at ffwll.ch> wrote:
> On Tue, May 08, 2018 at 11:50:37PM +0300, Jani Nikula wrote:
>> Since drm-rerere commit 589f83df3e5a ("Revert "Revert "move drm-next to
>> drm.git for experimentation""") the drm-next branch is expected to be in
>> the drm repo instead of drm-upstream.
>> 
>> This fixes dim status and a bunch of pull request and other commands.
>> 
>> Cc: Daniel Vetter <daniel.vetter at ffwll.ch>
>> Signed-off-by: Jani Nikula <jani.nikula at intel.com>
>> ---
>>  dim | 16 ++++++----------
>>  1 file changed, 6 insertions(+), 10 deletions(-)
>> 
>> diff --git a/dim b/dim
>> index 6b684ba5308c..a582765c3da9 100755
>> --- a/dim
>> +++ b/dim
>> @@ -1758,7 +1758,7 @@ function dim_update_next
>>  	$DRY git add drivers/gpu/drm/i915/i915_drv.h
>>  	git commit $DRY_RUN -sm "drm/i915: Update DRIVER_DATE to $driver_date"
>>  
>> -	gitk drm-intel-next-queued ^$(repo_to_remote drm-upstream)/drm-next &
>> +	gitk drm-intel-next-queued ^$(repo_to_remote drm)/drm-next &
>
> Since we defacto hard-code the assumption that a given branch name is
> unique among all the repos maintained by dim should we instead use the
> branch_to_remote helper here and all other places?

Annoyingly, branch_to_remote looks at $branch@{upstream} which will fail
for any branch not checked out. This needs to do branch_to_repo followed
by repo_to_remote. *sigh*.

J.

>
> Otherwise we're going to have the same fun again when/if we move
> drm-fixes. Or when we move drm-next back.
>>  
>>  	# try to push dinq first in case someone raced
>>  	FORCE=1 dim_push_queued
>> @@ -1876,7 +1876,7 @@ function dim_pull_request
>>  
>>  function dim_pull_request_next
>>  {
>> -	upstream=${1:-$(repo_to_remote drm-upstream)/drm-next}
>> +	upstream=${1:-$(repo_to_remote drm)/drm-next}
>>  	dim_pull_request drm-intel-next $upstream
>>  }
>>  
>> @@ -1888,22 +1888,18 @@ function dim_pull_request_fixes
>>  
>>  function dim_pull_request_next_fixes
>>  {
>> -	upstream=${1:-$(repo_to_remote drm-upstream)/drm-next}
>> +	upstream=${1:-$(repo_to_remote drm)/drm-next}
>>  	dim_pull_request drm-intel-next-fixes $upstream
>>  }
>>  
>>  # Note: used by bash completion
>>  function dim_list_upstreams
>>  {
>> -	local dim_drm_upstream_remote
>> -
>>  	cd $DIM_PREFIX/$DIM_REPO
>>  
>> -	dim_drm_upstream_remote=$(repo_to_remote drm-upstream)
>> -
>>  	echo origin/master
>> -	echo $dim_drm_upstream_remote/drm-next
>> -	echo $dim_drm_upstream_remote/drm-fixes
>> +	echo $(repo_to_remote drm)/drm-next
>> +	echo $(repo_to_remote drm-upstream)/drm-fixes
>
> Similar here, we could just directly brancht_to_remote for both drm-fixes
> and drm-next.
>
>>  }
>>  
>>  # Note: used by bash completion
>> @@ -1964,7 +1960,7 @@ function dim_status
>>  
>>  	fetch_all
>>  
>> -	drm_remote=$(repo_to_remote drm-upstream)
>> +	drm_remote=$(repo_to_remote drm)
>
> And here we'd need to track them separately as both drm-fixes and drm-next
> remotes.
> -Daniel
>
>>  
>>  	for branch in $dim_branches ; do
>>  		repo=$(branch_to_repo $branch)
>> -- 
>> 2.11.0
>> 

-- 
Jani Nikula, Intel Open Source Technology Center