Re: [PATCH] drm-misc: Another check in the bugfix flow

[Daniel Vetter]

On Tue, May 24, 2022 at 06:16:11PM +0200, Marek Vasut wrote:
> On 5/24/22 17:17, Daniel Vetter wrote:
> > Even in feature freeze it can happen that the broken patch is only in
> > drm-misc-next, and then applying to drm-misc-next-fixes will
> > rightfully stumble over dim's Fixes: validation since the sha1 wont be
> > an ancestor.
> > 
> > Fix that by adding another check.
> > 
> > v2: Less confusing wording (Marek)
> > 
> > Cc: Marek Vasut 
> > Signed-off-by: Daniel Vetter 
> 
> Reviewed-by: Marek Vasut 

Thanks for your review, patch pushed.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

[PATCH] drm-misc: Another check in the bugfix flow

[Daniel Vetter]

Even in feature freeze it can happen that the broken patch is only in
drm-misc-next, and then applying to drm-misc-next-fixes will
rightfully stumble over dim's Fixes: validation since the sha1 wont be
an ancestor.

Fix that by adding another check.

v2: Less confusing wording (Marek)

Cc: Marek Vasut 
Signed-off-by: Daniel Vetter 
---
 drm-misc-commit-flow.dot | 8 +++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drm-misc-commit-flow.dot b/drm-misc-commit-flow.dot
index 04afab6f4074..f1062d2ff8a8 100644
--- a/drm-misc-commit-flow.dot
+++ b/drm-misc-commit-flow.dot
@@ -1,6 +1,7 @@
 digraph {
next_0[shape=box, style=rounded, color=blue, label="drm-misc-next"];
next_1[shape=box, style=rounded, color=blue, label="drm-misc-next"];
+   next_2[shape=box, style=rounded, color=blue, label="drm-misc-next"];
fixes_0[shape=box, style=rounded, color=blue, label="drm-misc-fixes"];
next_fixes_0[shape=box, style=rounded, color=blue,
 label="drm-misc-next-fixes"];
@@ -10,6 +11,8 @@ digraph {
  label="Is the bug in the current rc?"];
is_late[shape=diamond, color=red,
label="Is drm in feature freeze?\n(occurs after -rc6)"];
+   is_next[shape=diamond, color=red,
+   label="Is the bug both in drm-next\nand in drm-misc-next?"];
 
is_fix -> next_0[label="no"];
is_fix -> in_origin[label="yes"];
@@ -18,5 +21,8 @@ digraph {
in_origin -> is_late[label="no"];
 
is_late -> next_1[label="no"]
-   is_late -> next_fixes_0[label="yes"]
+   is_late -> is_next[label="yes"]
+
+   is_next -> next_2[label="no"]
+   is_next -> next_fixes_0[label="yes"]
 }
-- 
2.36.0

[PATCH] drm-misc: Another check in the bugfix flow

[Daniel Vetter]

Even in feature freeze it can happen that the broken patch is only in
drm-misc-next, and then applying to drm-misc-next-fixes will
rightfully stumble over dim's Fixes: validation since the sha1 wont be
an ancestor.

Fix that by adding another check.

Cc: Marek Vasut 
Signed-off-by: Daniel Vetter 
---
 drm-misc-commit-flow.dot | 8 +++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drm-misc-commit-flow.dot b/drm-misc-commit-flow.dot
index 04afab6f4074..c2740b76b5b9 100644
--- a/drm-misc-commit-flow.dot
+++ b/drm-misc-commit-flow.dot
@@ -1,6 +1,7 @@
 digraph {
next_0[shape=box, style=rounded, color=blue, label="drm-misc-next"];
next_1[shape=box, style=rounded, color=blue, label="drm-misc-next"];
+   next_2[shape=box, style=rounded, color=blue, label="drm-misc-next"];
fixes_0[shape=box, style=rounded, color=blue, label="drm-misc-fixes"];
next_fixes_0[shape=box, style=rounded, color=blue,
 label="drm-misc-next-fixes"];
@@ -10,6 +11,8 @@ digraph {
  label="Is the bug in the current rc?"];
is_late[shape=diamond, color=red,
label="Is drm in feature freeze?\n(occurs after -rc6)"];
+   is_next[shape=diamond, color=red,
+   label="Is the bug in drm-next\nand not only in drm-misc-next?"];
 
is_fix -> next_0[label="no"];
is_fix -> in_origin[label="yes"];
@@ -18,5 +21,8 @@ digraph {
in_origin -> is_late[label="no"];
 
is_late -> next_1[label="no"]
-   is_late -> next_fixes_0[label="yes"]
+   is_late -> is_next[label="yes"]
+
+   is_next -> next_2[label="no"]
+   is_next -> next_fixes_0[label="yes"]
 }
-- 
2.36.0