Skip to contentSkip to conversation

epoll: use wrapper functions

PATCH Superseded Changli Gao 9 participants 45 messages May 6, 2010 – May 13, 2010 1 message not in the archive

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 epoll: use wrapper functions

use wrapper functions.

epoll should not touch flags in wait_queue_t. This patch introduces a new function add_wait_queue_head_exclusive_locked(), for the users, who use wait queue as a LIFO queue.

----

 fs/eventpoll.c       |    5 ++---
 include/linux/wait.h |   15 +++++++++++++--
 2 files changed, 15 insertions(+), 5 deletions(-)
fs/eventpoll.c +2 -3
index bd056a5..8137f6e 100644
@@ -1140,8 +1140,7 @@ retry:
 		 * ep_poll_callback() when events will become available.
 		 */
 		init_waitqueue_entry(&wait, current);
-		wait.flags |= WQ_FLAG_EXCLUSIVE;
-		__add_wait_queue(&ep->wq, &wait);
+		add_wait_queue_head_exclusive_locked(&ep->wq, &wait);
 
 		for (;;) {
 			/*
@@ -1161,7 +1160,7 @@ retry:
 			jtimeout = schedule_timeout(jtimeout);
 			spin_lock_irqsave(&ep->lock, flags);
 		}
-		__remove_wait_queue(&ep->wq, &wait);
+		remove_wait_queue_locked(&ep->wq, &wait);
 
 		set_current_state(TASK_RUNNING);
 	}
include/linux/wait.h +13 -2
index a48e16b..de2566d 100644
@@ -407,17 +407,28 @@ do {									\
  * Must be called with the spinlock in the wait_queue_head_t held.
  */
 static inline void add_wait_queue_exclusive_locked(wait_queue_head_t *q,
-						   wait_queue_t * wait)
+						   wait_queue_t *wait)
 {
 	wait->flags |= WQ_FLAG_EXCLUSIVE;
 	__add_wait_queue_tail(q,  wait);
 }
 
 /*
+ * Must be called with the spinlock in the wait_queue_head_t held, and
+ * q must be for exclusive wait only.
+ */
+static inline void add_wait_queue_head_exclusive_locked(wait_queue_head_t *q,
+							wait_queue_t *wait)
+{
+	wait->flags |= WQ_FLAG_EXCLUSIVE;
+	__add_wait_queue(q,  wait);
+}
+
+/*
  * Must be called with the spinlock in the wait_queue_head_t held.
  */
 static inline void remove_wait_queue_locked(wait_queue_head_t *q,
-					    wait_queue_t * wait)
+					    wait_queue_t *wait)
 {
 	__remove_wait_queue(q,  wait);
 }
#1 lore raw

indiscriminate get_maintainer.pl usage (was [PATCH] epoll: use wrapper functions)

(Cc list shortened)

Quoted 3 lines · Changli Gao
 fs/eventpoll.c       |    5 ++---
 include/linux/wait.h |   15 +++++++++++++--
 2 files changed, 15 insertions(+), 5 deletions(-)
$ git log fs/eventpoll.c include/linux/wait.h |grep stefanr
    Reported-by: Stefan Richter <[email protected]>
    Tested-by: Stefan Richter <[email protected]>

Please do not Cc me on subsystems that I do not develop nor maintain. My personal inbox is full enough and my todo-list growing. When I have time to look at patches to subsystems that I don't work with normally, I look into my inbox.lkml.

If you rely on scripts/get_maintainer.pl, then please use it more carefully or improve the script to produce more qualified Cc lists.

Thanks.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= --==-
http://arcgraph.de/sr/

A message that is not in the archive (sent off-list or not yet received): [email protected]

#3 lore raw

Re: indiscriminate get_maintainer.pl usage

Quoted 1 line

I should add --nogit option.

The script does get some useful information out of git (but obviously also some questionable).

--roles and --rolestats give away some more data...

$ scripts/get_maintainer.pl --rolestats -f fs/eventpoll.c
Alexander Viro <[email protected]> (maintainer:FILESYSTEMS (VFS...)
Andrew Morton <[email protected]> (commit_signer:2/6=33%)
"Eric W. Biederman" <[email protected]> (commit_signer:2/6=33%)
Davide Libenzi <[email protected]> (commit_signer:2/6=33%)
Roland Dreier <[email protected]> (commit_signer:1/6=17%)
Stefan Richter <[email protected]> (commit_signer:1/6=17%)
[email protected] (open list:FILESYSTEMS (VFS...)
[email protected] (open list)

...but still don't distinguish between signed-off-by or reviewed-by on one hand, and reported-by and tested-by on the other hand. Cc'ing a reporter or tester would only make sense if the patch changes something that directly affects the reporter's/ tester's setup, e.g. reworks a provisional fix.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= --==-
http://arcgraph.de/sr/
#4 lore raw

Re: indiscriminate get_maintainer.pl usage

Quoted 17 lines · Stefan Richter
Quoted 1 line

I should add --nogit option.

The script does get some useful information out of git (but obviously also some questionable).

--roles and --rolestats give away some more data...

$ scripts/get_maintainer.pl --rolestats -f fs/eventpoll.c
Alexander Viro <[email protected]> (maintainer:FILESYSTEMS (VFS...)
Andrew Morton <[email protected]> (commit_signer:2/6=33%)
"Eric W. Biederman" <[email protected]> (commit_signer:2/6=33%)
Davide Libenzi <[email protected]> (commit_signer:2/6=33%)
Roland Dreier <[email protected]> (commit_signer:1/6=17%)
Stefan Richter <[email protected]> (commit_signer:1/6=17%)
[email protected] (open list:FILESYSTEMS (VFS...)
[email protected] (open list)
Quoted 5 lines

...but still don't distinguish between signed-off-by or reviewed-by on one hand, and reported-by and tested-by on the other hand. Cc'ing a reporter or tester would only make sense if the patch changes something that directly affects the reporter's/ tester's setup, e.g. reworks a provisional fix.

I have always maintained that code w/out the need of explicit entry in the MAINTAINERS file. For isolated files people usually include the copyright owners and the ones listed in the comment header in general. No problem in being explicitly listed if this causes noise during bug reporting.

- Davide

maintainer
#5 lore raw

Re: indiscriminate get_maintainer.pl usage

Quoted 4 lines · Davide Libenzi
Quoted 3 lines · Stefan Richter
Quoted 1 line

I should add --nogit option.

The script does get some useful information out of git (but obviously also some questionable).

Stefan, I think that's the price you pay for fame. Try to do good, get more unsolicited email.

Quoted 17 lines
Quoted 11 lines

--roles and --rolestats give away some more data...

$ scripts/get_maintainer.pl --rolestats -f fs/eventpoll.c
Alexander Viro <[email protected]> (maintainer:FILESYSTEMS (VFS...)
Andrew Morton <[email protected]> (commit_signer:2/6=33%)
"Eric W. Biederman" <[email protected]> (commit_signer:2/6=33%)
Davide Libenzi <[email protected]> (commit_signer:2/6=33%)
Roland Dreier <[email protected]> (commit_signer:1/6=17%)
Stefan Richter <[email protected]> (commit_signer:1/6=17%)
[email protected] (open list:FILESYSTEMS (VFS...)
[email protected] (open list)
Quoted 5 lines

...but still don't distinguish between signed-off-by or reviewed-by on one hand, and reported-by and tested-by on the other hand. Cc'ing a reporter or tester would only make sense if the patch changes something that directly affects the reporter's/ tester's setup, e.g. reworks a provisional fix.

Quoted 5 lines

I have always maintained that code w/out the need of explicit entry in the MAINTAINERS file. For isolated files people usually include the copyright owners and the ones listed in the comment header in general. No problem in being explicitly listed if this causes noise during bug reporting.

There is also the --git-blame option for patches, but blame is also only a semi-useful tool, as it could show cleanup patches on individual as authors.

Another option to get_maintainers is --file-emails which looks inside a file for email addresses.

Right now, the default for # of git commits signed to be added to the CC list is 1. It's possible to change the default.

--git-min-signatures => number of signatures required (default: 1) --git-max-maintainers => maximum maintainers to add (default: 5) --git-min-percent => minimum percentage of commits required (default: 5)

Maybe some fine-tuning of the defaults could be useful.

Suggestions?

#6 lore raw

Re: indiscriminate get_maintainer.pl usage

Quoted 10 lines · Joe Perches

There is also the --git-blame option for patches, but blame is also only a semi-useful tool, as it could show cleanup patches on individual as authors.

Another option to get_maintainers is --file-emails which looks inside a file for email addresses.

Right now, the default for # of git commits signed to be added to the CC list is 1. It's possible to change the default.

Quoted 7 lines

--git-min-signatures => number of signatures required (default: 1) --git-max-maintainers => maximum maintainers to add (default: 5) --git-min-percent => minimum percentage of commits required (default: 5)

Maybe some fine-tuning of the defaults could be useful.

Suggestions?

(not knowing the get_maintainers internals ...)
Maybe if you weigh each contributor by the amount (in terms of diffstat modulo) of changes, you can get more accurate results.
Part of the weight might be related to the time distance (relative to current time) since the changeset.

- Davide

maintainer
#7 lore raw

Re: indiscriminate get_maintainer.pl usage

Quoted 10 lines · Davide Libenzi
Quoted 7 lines

Right now, the default for # of git commits signed to be
added to the CC list is 1. It's possible to change the default.
--git-min-signatures => number of signatures required (default: 1) --git-max-maintainers => maximum maintainers to add (default: 5) --git-min-percent => minimum percentage of commits required (default: 5)
Maybe some fine-tuning of the defaults could be useful. Suggestions?

(not knowing the get_maintainers internals ...)
Maybe if you weigh each contributor by the amount (in terms of diffstat modulo) of changes, you can get more accurate results.

That might tend to overweight cleanup against actual logic patches, especially as the period for commits tracked is defaulted to 1 year.

Quoted 2 lines

Part of the weight might be related to the time distance (relative to current time) since the changeset.

Which is kind of what activity alone does.

I think get_maintainer with --git (the default):

Without a named MAINTAINER:

For relatively active files: works moderately well.

For relatively inactive files: might work less well.

It tends to hide the original developer and show people that submit cleanup or fix patches.

Maybe the --git portion could be reduce the number of commit signers added when files have named MAINTAINERS.

Without named MAINTAINERS, maybe the file should be inspected for author info and those persons added.

If the original commit is after Linus' first wholesale commit of 2.6.12, maybe the first git author should be added.

Dunno. I think any of those might be slight improvements, but get_maintainers isn't really that much of an issue as-is. I think it more important to get the correct MAINTAINERS entries and file patterns.

#8 lore raw

PATCH 1/1 get_maintainer.pl: ignore non-maintainer tags

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch. Thus we restrict the candidates to be the one's listed with Reviewed-By, Signed-Off-By and Acked-By tags.

The Acked-By: is questionable also, but as people listed with this tag tend to be active linux gatekeepers, false positives are tolerable.

---
 scripts/get_maintainer.pl |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
scripts/get_maintainer.pl +2 -2
index 6f97a13..ad38ed3 100755
@@ -496,7 +496,7 @@ version: $V
 
 MAINTAINER field selection options:
   --email => print email address(es) if any
-    --git => include recent git \*-by: signers
+    --git => include recent git {Reviewed|Signed-Off|Acked}-by: signers
     --git-chief-penguins => include ${penguin_chiefs}
     --git-min-signatures => number of signatures required (default: 1)
     --git-max-maintainers => maximum maintainers to add (default: 5)
@@ -964,7 +964,7 @@ sub vcs_find_signers {
 
     $commits = grep(/$pattern/, @lines);	# of commits
 
-    @lines = grep(/^[-_ 	a-z]+by:.*\@.*$/i, @lines);
+    @lines = grep(/(signed-off|reviewed|acked)-by:.*\@.*$/i, @lines);
     if (!$email_git_penguin_chiefs) {
 	@lines = grep(!/${penguin_chiefs}/i, @lines);
     }
Signature · 1 line
1.7.0.4
maintainer
#9 lore raw

Re: [PATCH] get_maintainer.pl: ignore non-maintainer tags

Quoted 3 lines · On Fri, 2010-05-07 at 08:34 +0200, [email protected] wrote:

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch.

I think this is a questionable assumption.

People that test or otherwise sign a patch are also good candidates to review new patches.

cheers, Joe

#10 lore raw

Re: [PATCH] get_maintainer.pl: ignore non-maintainer tags

Quoted 11 lines · Joe Perches
Quoted 3 lines · On Fri, 2010-05-07 at 08:34 +0200, [email protected] wrote:

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch.

I think this is a questionable assumption.

People that test or otherwise sign a patch are also good candidates to review new patches.

cheers, Joe

Our views differ then.

It is unreasonable to assume, that someone who is listed via
Tested-By: in a random patch is able to review another patch to the same file. More likely will he be annoyed, because he see's a patch, dig's in and then thinks: "wtf? why were i cc'ed on this. I don't know this code. I don't have time for this!"

Thats just broken.

cheers,
Flo

#11 lore raw

Re: [PATCH] get_maintainer.pl: ignore non-maintainer tags

Quoted 14 lines · Florian Mickler
Quoted 10 lines · Joe Perches
Quoted 3 lines · On Fri, 2010-05-07 at 08:34 +0200, [email protected] wrote:

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch.

I think this is a questionable assumption.

People that test or otherwise sign a patch are also good candidates to review new patches.

cheers, Joe

Our views differ then.

Quoted 7 lines

It is unreasonable to assume, that someone who is listed via
Tested-By: in a random patch is able to review another patch to the same file. More likely will he be annoyed, because he see's a patch, dig's in and then thinks: "wtf? why were i cc'ed on this. I don't know this code. I don't have time for this!"

Thats just broken.

Indeed. Three points:

- Reported-by: and Tested-by: signed commits were most likely about something radically different from what the new patch submission is about.

- get_maintainer.pl is supposed to list maintainer addresses. Reporters and testers are not maintainers.

- If a get_maintainer.pl user expects to get not only maintainer addresses but also addresses of potential reviewers, then he shall use the mailinglist addresses for that.
People who /want/ to review patches subscribe to that mailinglist or register themselves in MAINTAINERS.

Pushing patches to people of whom it is unknown whether they want or can review the patches may actually be rude. If it can be avoided that get_maintainer.pl lists addresses of people who are likely of that category, then please teach it to.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= --===
http://arcgraph.de/sr/
#12 lore raw

PATCH 1/1 get_maintainer.pl: optionally ignore non-maintainer tags

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch. Thus we restrict the candidates to be the one's listed with Reviewed-By, Signed-Off-By and Acked-By tags.

Signed-Off-By:'s because that are people who are responsible for the code in question.

Reviewed-By:'s because people responsible for the code in question obviously thought that the review-feedback for this changeset by that person was valuable.

The Acked-By: is questionable, but as people listed with this tag tend to be active linux gatekeepers, false positives are tolerable.

---
 scripts/get_maintainer.pl |   21 ++++++++++++++++++++-
 1 files changed, 20 insertions(+), 1 deletions(-)
scripts/get_maintainer.pl +20 -1
index 6f97a13..8e974bb 100755
@@ -25,6 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
+my $email_git_maintainers = 0;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
@@ -50,6 +51,15 @@ my $help = 0;
 
 my $exit = 0;
 
+#
+# tags for people who are either
+# 	a) responsible for the code in question, or
+# 	b) familiar enough with it to give relevant feedback
+my @maintainer_tags = ();
+push(@maintainer_tags,"Signed-Off");
+push(@maintainer_tags,"Reviewed");
+push(@maintainer_tags,"Acked");
+
 my @penguin_chief = ();
 push(@penguin_chief,"Linus Torvalds:torvalds\@linux-foundation.org");
 #Andrew wants in on most everything - 2009/01/14
@@ -100,6 +110,7 @@ my %VCS_cmds_hg = (
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
+		'git-maintainers!' => \$email_git_maintainers,
 		'git-blame!' => \$email_git_blame,
 		'git-chief-penguins!' => \$email_git_penguin_chiefs,
 		'git-min-signatures=i' => \$email_git_min_signatures,
@@ -497,6 +508,7 @@ version: $V
 MAINTAINER field selection options:
   --email => print email address(es) if any
     --git => include recent git \*-by: signers
+    --git-maintainers => only use 'Signed-Off-By:', 'Reviewed-By:' and 'Acked-By:' signers
     --git-chief-penguins => include ${penguin_chiefs}
     --git-min-signatures => number of signatures required (default: 1)
     --git-max-maintainers => maximum maintainers to add (default: 5)
@@ -957,6 +969,13 @@ sub vcs_find_signers {
     my ($cmd) = @_;
     my @lines = ();
     my $commits;
+    my $tagPattern;
+    if ($email_git_maintainers) {
+	    $tagPattern = join("|",@maintainer_tags);
+    } else {
+	    $tagPattern = "[^ \t]+";
+    }
+
 
     @lines = &{$VCS_cmds{"execute_cmd"}}($cmd);
 
@@ -964,7 +983,7 @@ sub vcs_find_signers {
 
     $commits = grep(/$pattern/, @lines);	# of commits
 
-    @lines = grep(/^[-_ 	a-z]+by:.*\@.*$/i, @lines);
+    @lines = grep(/^[ \t]*($tagPattern)-by:.*\@.*$/i, @lines);
     if (!$email_git_penguin_chiefs) {
 	@lines = grep(!/${penguin_chiefs}/i, @lines);
     }
Signature · 1 line
1.7.0.4
#13 lore raw

RFC PATCH 1/1 get_maintainer.pl: only list maintainers by default

If scm meta-information is used, only list people that signed-off or acked or reviewed the code in question.

after all it's called get_maintainer.pl and not
get_random_people_to_look_at_the_code.pl

---
 scripts/get_maintainer.pl |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
scripts/get_maintainer.pl +1 -1
index 8e974bb..2c53476 100755
@@ -25,7 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
-my $email_git_maintainers = 0;
+my $email_git_maintainers = 1;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
Signature · 1 line
1.7.0.4
maintainer
#14 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 5 lines · On Sat, 2010-05-08 at 23:39 +0200, [email protected] wrote:

If scm meta-information is used, only list people that signed-off or acked or reviewed the code in question.

after all it's called get_maintainer.pl and not
get_random_people_to_look_at_the_code.pl

Anyone that actually signs a patch is not "random".

#15 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 8 lines · Joe Perches
Quoted 5 lines · On Sat, 2010-05-08 at 23:39 +0200, [email protected] wrote:

If scm meta-information is used, only list people that signed-off or acked or reviewed the code in question.

after all it's called get_maintainer.pl and not
get_random_people_to_look_at_the_code.pl

Anyone that actually signs a patch is not "random".

Reported-by and tested-by signatures in the git history commit /are/ in fact random WRT getting a list of handlers/ reviewers of a new patch. It is more likely that reporters and testers cannot or/and do not want to do anything about such a patch. Such a mailing is called UBE.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
maintainer
#16 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 3 lines · Stefan Richter

Joe Perches wrote:
Reported-by and tested-by signatures in the git history commit /are/ in fact random WRT getting a list of handlers/ reviewers of a new patch.

Supporting stats please.

$ git log --since=1-year-ago | grep -Pi "(reported|tested)-by:.*@" \
  | cut -f2- -d":" | sort | uniq -c | sort -rn | head -10
     77  Ingo Molnar <[email protected]>
     47  Wolfram Sang <[email protected]>
     42  Stephen Rothwell <[email protected]>
     35  Dan Carpenter <[email protected]>
     24  Randy Dunlap <[email protected]>
     23  Larry Finger <[email protected]>
     20  Brian Cavagnolo <[email protected]>
     17  Frans Pop <[email protected]>
     17  Christian Kujau <[email protected]>
     16  Alexander Beregalov <[email protected]>

Look like mostly patch authors and maintainers to me.

Brian Cavagnolo isn't a patch author, but he's an active tester.

Quoted 2 lines

It is more likely that reporters and testers cannot or/and do not want to do anything about such a patch. Such a mailing is called UBE.

I doubt that's true of even half of the reporters and testers.

I also doubt such patch emails fit any reasonable definition of UBE.

cheers, Joe

#17 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 6 lines · Joe Perches
Quoted 3 lines · Stefan Richter

Joe Perches wrote:
Reported-by and tested-by signatures in the git history commit /are/ in fact random WRT getting a list of handlers/ reviewers of a new patch.

Supporting stats please.

...

Quoted 3 lines

Look like mostly patch authors and maintainers to me.

Brian Cavagnolo isn't a patch author, but he's an active tester.

So you looked up signatures across the entire tree, found some people who appear more often than an average reporter/tester, and one of these is not a classic developer/maintainer.

But how does this indicate a need to include these addresses into get_maintainer.pl output? In no way at all.

To get your patch reviewed, Cc the relevant mailinglists. To get your patch into the submission pipeline, Cc the maintainer.

Quoted 7 lines
Quoted 2 lines

It is more likely that reporters and testers cannot or/and do not want to do anything about such a patch. Such a mailing is called UBE.

I doubt that's true of even half of the reporters and testers.

I also doubt such patch emails fit any reasonable definition of UBE.

The U, B, and E all describe such mail exactly.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
maintainer
#18 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 2 lines · Stefan Richter

But how does this indicate a need to include these addresses into get_maintainer.pl output? In no way at all.

Supporting stats please.

I gave you some that seem to differ from your position.

I'd send you Linus' original email from 2007, but you'd already commented on that, so for now, unless you come up with some stats that support your view, I'll stand by mine.

cheers, Joe

#19 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 8 lines · Joe Perches
Quoted 2 lines · Stefan Richter

But how does this indicate a need to include these addresses into get_maintainer.pl output? In no way at all.

Supporting stats please.

I gave you some that seem to differ from your
position.

What do want with statistics here? Look at semantics, not at quantities. Look at what this thread started.

You apparently overlooked what I wrote earlier, so I repeat it to be sure:

Quoted 1 line
Quoted 1 line
Quoted 1 line
$ git log fs/eventpoll.c include/linux/wait.h |grep stefanr
    Reported-by: Stefan Richter <[email protected]>
    Tested-by: Stefan Richter <[email protected]>

Please do not Cc me on subsystems that I do not develop nor maintain. My personal inbox is full enough and my todo-list growing. When I have time to look at patches to subsystems that I don't work with normally, I look into my inbox.lkml.

If you rely on scripts/get_maintainer.pl, then please use it more carefully or improve the script to produce more qualified Cc lists.

Thanks.
<<<

Quoted 1 line
Quoted 1 line
Quoted 1 line

Cc'ing a
reporter or tester would only make sense if the patch changes something that directly affects the reporter's/ tester's setup, e.g. reworks a provisional fix.
<<<
[which is highly unlikely, and in those special cases get_maintainer.pl is not where we take that info from]

Quoted 1 line
Quoted 1 line
Quoted 1 line

- Reported-by: and Tested-by: signed commits were most likely about something radically different from what the new patch submission is about.
<<<

But wait, you want stats, you get stats.

  $ git log drivers/firewire/ drivers/ieee1394/ |
    grep -e "Tested-by: " -e "Reported-by: "
lists 32 people of whom
  - 0 need to be/ should be included in the address list of a patch
    submission,
  - 2 or 3 were active in firewire subsystem development (but not
    anymore) but were perfectly fine to reach via MAINTAINERS' L: entry
  - 3 or 2 are still somewhat involved in firewire subsystem
    development, but only in special application domains, and they are
    reached via MAINTAINERS' L: entry

You on the other had found e.g. Ingo among testers signatures. Could you show a single source file of which a patch should be Cc'd to Ingo and get_maintainer.pl finds him only among testers or reporters but no among the MAINTAINERS M: entries nor among the signed-off-bys? (Or any of the other maintainers that you found.)

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
#20 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

Quoted 2 lines · Stefan Richter
Quoted 1 line · Joe Perches

Supporting stats please.

...

Quoted 7 lines
  $ git log drivers/firewire/ drivers/ieee1394/ |
    grep -e "Tested-by: " -e "Reported-by: "
lists 32 people of whom
  - 0 need to be/ should be included in the address list of a patch
    submission,
  - 2 or 3 were active in firewire subsystem development (but not
    anymore) but were perfectly fine to reach via MAINTAINERS' L: entry

PS: And when they were still active, they also clearly showed up as patch authors.

Quoted 3 lines

- 3 or 2 are still somewhat involved in firewire subsystem development, but only in special application domains, and they are reached via MAINTAINERS' L: entry

Think of mailinglists as a pull model to reach potential reviewers, whereas personal mail constitute a push model which must be used with a lot of care --- i.e. when you as a submitter _know_ that these people are going to want to be directly informed about the patch submission.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
#21 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

In reply to Joe Perches
Quoted 5 lines · On Sat, 08 May 2010 17:57:13 -0700 Joe Perches <[email protected]> wrote:
Quoted 2 lines · Stefan Richter

But how does this indicate a need to include these addresses into get_maintainer.pl output? In no way at all.

Supporting stats please.

#number of one-off tested-by/reported-by appearences
git log --since=1-year-ago | grep -Pi "(reported|tested)-by:.*@" | cut -f2- -d":" | sort | uniq -c | awk '{print $1}' | grep '^1$' | wc -l
822

cheers,
Flo

p.s.: please double check the cmdline, but i think this is correct.

#22 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

In reply to Joe Perches
Quoted 4 lines · Joe Perches

I'd send you Linus' original email from 2007, but you'd already commented on that, so for now, unless you come up with some stats that support your view, I'll stand by mine.

I don't remember the particular mail of his that you refer to. But I suspect he recommended to /look/ at the history, not exactly to /blindly grep/ through the history and Cc all addresses that got caught.

If the thread starter had looked at the history of fs/eventpoll.c, he wouldn't have Cc'd me. A 5 seconds glance at gitk fs/eventpoll.c would have sufficed.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
#23 lore raw

Re: [PATCH] [RFC] get_maintainer.pl: only list maintainers by default

In reply to Joe Perches
Quoted 9 lines · On Sat, 08 May 2010 15:44:34 -0700 Joe Perches <[email protected]> wrote:
Quoted 5 lines · On Sat, 2010-05-08 at 23:39 +0200, [email protected] wrote:

If scm meta-information is used, only list people that signed-off or acked or reviewed the code in question.

after all it's called get_maintainer.pl and not
get_random_people_to_look_at_the_code.pl

Anyone that actually signs a patch is not "random".

I actually think that tested-by and reviewed-by is not "signing", because it is not the person who does the "tag insertion".

it was one of the signed-off-by's who put the reviewed- and testet-tags in there. presumably this is done in agreement with the mentioned person.

And for facts, please check this out:

$ git log --since=1-year-ago | grep -Pi "(reported|tested)-by:.*@"  \
 | cut -f2- -d":" | sort | uniq | wc -l
1280

so i would say, these numbers (combied with your top ten which accounts for only 318 signatures ) suggest that those mentioned with these tag's are one-off bug reporters and not maintainers.

cheers,
Flo

maintainer
#24 lore raw

Re: [PATCH] get_maintainer.pl: optionally ignore non-maintainer tags

In reply to [email protected]
Quoted 4 lines · On Sat, 2010-05-08 at 23:32 +0200, [email protected] wrote:

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch. Thus we restrict the candidates to be the one's listed with Reviewed-By, Signed-Off-By and Acked-By tags.

[]

scripts/get_maintainer.pl sub vcs_find_signers { · line 976

Quoted 5 lines
scripts/get_maintainer.pl +5 -0
+    my $tagPattern;
+    if ($email_git_maintainers) {
+	    $tagPattern = join("|",@maintainer_tags);
+    } else {
+	    $tagPattern = "[^ \t]+";

I think the else case should be "[a-z _-]+" as some non-standard signatures have embedded spaces.

#25 lore raw

PATCH v2 1/1 get_maintainer.pl: optionally ignore non-maintainer tags

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch. Thus we restrict the candidates to be the one's listed with Reviewed-By, Signed-Off-By and Acked-By tags.

Signed-Off-By:'s because that are people who are responsible for the code in question.

Reviewed-By:'s because people responsible for the code in question obviously thought that the review-feedback for this changeset by that person was valuable.

The Acked-By: is questionable, but as people listed with this tag tend to be active linux gatekeepers, false positives are tolerable.

---

change from v1:fixed the tagPattern as per suggestion by Joe.

 scripts/get_maintainer.pl |   21 ++++++++++++++++++++-
 1 files changed, 20 insertions(+), 1 deletions(-)
scripts/get_maintainer.pl +20 -1
index 6f97a13..8125b51 100755
@@ -25,6 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
+my $email_git_maintainers = 0;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
@@ -50,6 +51,15 @@ my $help = 0;
 
 my $exit = 0;
 
+#
+# tags for people who are either
+# 	a) responsible for the code in question, or
+# 	b) familiar enough with it to give relevant feedback
+my @maintainer_tags = ();
+push(@maintainer_tags,"Signed-Off");
+push(@maintainer_tags,"Reviewed");
+push(@maintainer_tags,"Acked");
+
 my @penguin_chief = ();
 push(@penguin_chief,"Linus Torvalds:torvalds\@linux-foundation.org");
 #Andrew wants in on most everything - 2009/01/14
@@ -100,6 +110,7 @@ my %VCS_cmds_hg = (
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
+		'git-maintainers!' => \$email_git_maintainers,
 		'git-blame!' => \$email_git_blame,
 		'git-chief-penguins!' => \$email_git_penguin_chiefs,
 		'git-min-signatures=i' => \$email_git_min_signatures,
@@ -497,6 +508,7 @@ version: $V
 MAINTAINER field selection options:
   --email => print email address(es) if any
     --git => include recent git \*-by: signers
+    --git-maintainers => only use 'Signed-Off-By:', 'Reviewed-By:' and 'Acked-By:' signers
     --git-chief-penguins => include ${penguin_chiefs}
     --git-min-signatures => number of signatures required (default: 1)
     --git-max-maintainers => maximum maintainers to add (default: 5)
@@ -957,6 +969,13 @@ sub vcs_find_signers {
     my ($cmd) = @_;
     my @lines = ();
     my $commits;
+    my $tagPattern;
+    if ($email_git_maintainers) {
+	    $tagPattern = join("|",@maintainer_tags);
+    } else {
+	    $tagPattern = "[a-z _-]+";
+    }
+
 
     @lines = &{$VCS_cmds{"execute_cmd"}}($cmd);
 
@@ -964,7 +983,7 @@ sub vcs_find_signers {
 
     $commits = grep(/$pattern/, @lines);	# of commits
 
-    @lines = grep(/^[-_ 	a-z]+by:.*\@.*$/i, @lines);
+    @lines = grep(/^[ \t]*($tagPattern)-by:.*\@.*$/i, @lines);
     if (!$email_git_penguin_chiefs) {
 	@lines = grep(!/${penguin_chiefs}/i, @lines);
     }
Signature · 1 line
1.7.0.4
#26 lore raw

Re: [PATCH v2] get_maintainer.pl: optionally ignore non-maintainer tags

Quoted 8 lines · [email protected] wrote:

Using --git to determine who to send a patch to, it is not reasonable to include people that only reported an issue or tested a patch. Thus we restrict the candidates to be the one's listed with Reviewed-By, Signed-Off-By and Acked-By tags.

Signed-Off-By:'s because that are people who are responsible for the code in question.

Quoted 6 lines

Reviewed-By:'s because people responsible for the code in question obviously thought that the review-feedback for this changeset by that person was valuable.

The Acked-By: is questionable, but as people listed with this tag tend to be active linux gatekeepers, false positives are tolerable.

This and "get_maintainer.pl: only list maintainers by default" in
conjunction: Reviewed-by: Stefan Richter <[email protected]> But *don't* record this in the git history because then somebody might think I want to be Cc'd on get_maintainer.pl patches.

Signature · 3 lines
Stefan Richter
-=====-==-=- -=-= -=--=
http://arcgraph.de/sr/
maintainer
#27 lore raw

PATCH 1/2 scripts/get_maintainer.pl: optionally ignore non-maintainer signatures

In reply to [email protected]

When using --git to determine who to send a patch to, get_maintainers will currently include all signatures. This can include signers that simply report an issue or test a patch. Signers that use this tag are not necessarily good candidates to review new patches.

This patch allows get_maintainers to optionally restrict output to only signatures that use Signed-off-by:, Acked-by:, or Reviewed-by:.

Signed-off-by: is included because those are people who are responsible for the code.

Acked-by: is questionable, but as signers that use this tag tend to be active linux gatekeepers, false positives are tolerable.

Reviewed-by: is included because signers responsible for the code thought that the review feedback for a changeset by that signer was valuable.

This patch has been modified from Florian's original submission to change the supported signature types to the canonical forms and use slightly different spacing. A couple of spacing issues were also corrected in the original source. The command line argument was also renamed.

---
 scripts/get_maintainer.pl |   35 ++++++++++++++++++++++++++---------
 1 files changed, 26 insertions(+), 9 deletions(-)
scripts/get_maintainer.pl +26 -9
index 6f97a13..f66018d 100755
@@ -25,6 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
+my $email_git_all_signature_types = 1;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
@@ -51,9 +52,9 @@ my $help = 0;
 my $exit = 0;
 
 my @penguin_chief = ();
-push(@penguin_chief,"Linus Torvalds:torvalds\@linux-foundation.org");
+push(@penguin_chief, "Linus Torvalds:torvalds\@linux-foundation.org");
 #Andrew wants in on most everything - 2009/01/14
-#push(@penguin_chief,"Andrew Morton:akpm\@linux-foundation.org");
+#push(@penguin_chief, "Andrew Morton:akpm\@linux-foundation.org");
 
 my @penguin_chief_names = ();
 foreach my $chief (@penguin_chief) {
@@ -63,7 +64,16 @@ foreach my $chief (@penguin_chief) {
 	push(@penguin_chief_names, $chief_name);
     }
 }
-my $penguin_chiefs = "\(" . join("|",@penguin_chief_names) . "\)";
+my $penguin_chiefs = "\(" . join("|", @penguin_chief_names) . "\)";
+
+# Signature types of people who are either
+# 	a) responsible for the code in question, or
+# 	b) familiar enough with it to give relevant feedback
+my @signature_tags = ();
+push(@signature_tags, "Signed-off-by:");
+push(@signature_tags, "Reviewed-by:");
+push(@signature_tags, "Acked-by:");
+my $signaturePattern = "\(" . join("|", @signature_tags) . "\)";
 
 # rfc822 email address - preloaded methods go here.
 my $rfc822_lwsp = "(?:(?:\\r\\n)?[ \\t])";
@@ -100,6 +110,7 @@ my %VCS_cmds_hg = (
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
+		'git-all-signature-types!' => \$email_git_all_signature_types,
 		'git-blame!' => \$email_git_blame,
 		'git-chief-penguins!' => \$email_git_penguin_chiefs,
 		'git-min-signatures=i' => \$email_git_min_signatures,
@@ -180,6 +191,10 @@ if (!top_of_kernel_tree($lk_path)) {
 	. "a linux kernel source tree.\n";
 }
 
+if ($email_git_all_signature_types) {
+    $signaturePattern = "(.+?)[Bb][Yy]:";
+}
+
 ## Read MAINTAINERS for type/value pairs
 
 my @typevalue = ();
@@ -497,13 +512,15 @@ version: $V
 MAINTAINER field selection options:
   --email => print email address(es) if any
     --git => include recent git \*-by: signers
+    --git-all-signature-types => include signers regardless of signature type
+        or use only ${signaturePattern} signers (default: $email_git_all_signature_types)
     --git-chief-penguins => include ${penguin_chiefs}
-    --git-min-signatures => number of signatures required (default: 1)
-    --git-max-maintainers => maximum maintainers to add (default: 5)
-    --git-min-percent => minimum percentage of commits required (default: 5)
+    --git-min-signatures => number of signatures required (default: $email_git_min_signatures)
+    --git-max-maintainers => maximum maintainers to add (default: $email_git_max_maintainers)
+    --git-min-percent => minimum percentage of commits required (default: $email_git_min_percent)
     --git-blame => use git blame to find modified commits for patch or file
-    --git-since => git history to use (default: 1-year-ago)
-    --hg-since => hg history to use (default: -365)
+    --git-since => git history to use (default: $email_git_since)
+    --hg-since => hg history to use (default: $email_hg_since)
     --m => include maintainer(s) if any
     --n => include name 'Full Name <addr\@domain.tld>'
     --l => include list(s) if any
@@ -964,7 +981,7 @@ sub vcs_find_signers {
 
     $commits = grep(/$pattern/, @lines);	# of commits
 
-    @lines = grep(/^[-_ 	a-z]+by:.*\@.*$/i, @lines);
+    @lines = grep(/^[ \t]*${signaturePattern}.*\@.*$/, @lines);
     if (!$email_git_penguin_chiefs) {
 	@lines = grep(!/${penguin_chiefs}/i, @lines);
     }
Signature · 1 line
1.7.0.3.311.g6a6955
#28 lore raw

Re: [PATCH 1/2] scripts/get_maintainer.pl: optionally ignore non-maintainer signatures

Quoted 8 lines · On Sun, 9 May 2010 21:56:02 -0700 Joe Perches <[email protected]> wrote:

This patch has been modified from Florian's original submission to change the supported signature types to the canonical forms and use slightly different spacing. A couple of spacing issues were also corrected in the original source. The command line argument was also renamed.

I'm ok with your changes. Looks good.

Quoted 96 lines
---
 scripts/get_maintainer.pl |   35 ++++++++++++++++++++++++++---------
 1 files changed, 26 insertions(+), 9 deletions(-)
scripts/get_maintainer.pl +26 -9
index 6f97a13..f66018d 100755
@@ -25,6 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
+my $email_git_all_signature_types = 1;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
@@ -51,9 +52,9 @@ my $help = 0;
 my $exit = 0;
 
 my @penguin_chief = ();
-push(@penguin_chief,"Linus Torvalds:torvalds\@linux-foundation.org");
+push(@penguin_chief, "Linus Torvalds:torvalds\@linux-foundation.org");
 #Andrew wants in on most everything - 2009/01/14
-#push(@penguin_chief,"Andrew Morton:akpm\@linux-foundation.org");
+#push(@penguin_chief, "Andrew Morton:akpm\@linux-foundation.org");
 
 my @penguin_chief_names = ();
 foreach my $chief (@penguin_chief) {
@@ -63,7 +64,16 @@ foreach my $chief (@penguin_chief) {
 	push(@penguin_chief_names, $chief_name);
     }
 }
-my $penguin_chiefs = "\(" . join("|",@penguin_chief_names) . "\)";
+my $penguin_chiefs = "\(" . join("|", @penguin_chief_names) . "\)";
+
+# Signature types of people who are either
+# 	a) responsible for the code in question, or
+# 	b) familiar enough with it to give relevant feedback
+my @signature_tags = ();
+push(@signature_tags, "Signed-off-by:");
+push(@signature_tags, "Reviewed-by:");
+push(@signature_tags, "Acked-by:");
+my $signaturePattern = "\(" . join("|", @signature_tags) . "\)";
 
 # rfc822 email address - preloaded methods go here.
 my $rfc822_lwsp = "(?:(?:\\r\\n)?[ \\t])";
@@ -100,6 +110,7 @@ my %VCS_cmds_hg = (
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
+		'git-all-signature-types!' => \$email_git_all_signature_types,
 		'git-blame!' => \$email_git_blame,
 		'git-chief-penguins!' => \$email_git_penguin_chiefs,
 		'git-min-signatures=i' => \$email_git_min_signatures,
@@ -180,6 +191,10 @@ if (!top_of_kernel_tree($lk_path)) {
 	. "a linux kernel source tree.\n";
 }
 
+if ($email_git_all_signature_types) {
+    $signaturePattern = "(.+?)[Bb][Yy]:";
+}
+
 ## Read MAINTAINERS for type/value pairs
 
 my @typevalue = ();
@@ -497,13 +512,15 @@ version: $V
 MAINTAINER field selection options:
   --email => print email address(es) if any
     --git => include recent git \*-by: signers
+    --git-all-signature-types => include signers regardless of signature type
+        or use only ${signaturePattern} signers (default: $email_git_all_signature_types)
     --git-chief-penguins => include ${penguin_chiefs}
-    --git-min-signatures => number of signatures required (default: 1)
-    --git-max-maintainers => maximum maintainers to add (default: 5)
-    --git-min-percent => minimum percentage of commits required (default: 5)
+    --git-min-signatures => number of signatures required (default: $email_git_min_signatures)
+    --git-max-maintainers => maximum maintainers to add (default: $email_git_max_maintainers)
+    --git-min-percent => minimum percentage of commits required (default: $email_git_min_percent)
     --git-blame => use git blame to find modified commits for patch or file
-    --git-since => git history to use (default: 1-year-ago)
-    --hg-since => hg history to use (default: -365)
+    --git-since => git history to use (default: $email_git_since)
+    --hg-since => hg history to use (default: $email_hg_since)
     --m => include maintainer(s) if any
     --n => include name 'Full Name <addr\@domain.tld>'
     --l => include list(s) if any
@@ -964,7 +981,7 @@ sub vcs_find_signers {
 
     $commits = grep(/$pattern/, @lines);	# of commits
 
-    @lines = grep(/^[-_ 	a-z]+by:.*\@.*$/i, @lines);
+    @lines = grep(/^[ \t]*${signaturePattern}.*\@.*$/, @lines);
     if (!$email_git_penguin_chiefs) {
 	@lines = grep(!/${penguin_chiefs}/i, @lines);
     }
#29 lore raw

PATCH 1/1 scripts/get_maintainer.pl: default to not include unspecified tags

In reply to Joe Perches

This changes the default of the option --git-all-signature-types to be disabled by default.

The effect being, that only certain (currently Signed-Off-By:,
Acked-By: and Reviewed-By:) tags are used to get adresses of potential maintainers.

Motivated is this change by the desire to not 'spam' people unnecessary:
A Tested-By or a Reported-By is not ment as a hint that those people want to/are able to review patches to the code in question.

In a quest to find resilient statistics for this i came up with this:

I produced a list of all the tag-signers not already covered
with a signed-off/acked/reviewed tag somewhere in the last year of git history.

Those were 650 addresses of "assumed non-developers".

And to check if those "assumed non-developers" are professional testers/reporters worth cc'ing, i then counted their total appearences in the git log:

469 were mentioned only once.
123 were mentioned twice.
38 three times
8 four times
5 six times
5 five times
1 eight times
1 fourteen times

I believe this supports my thesis, that the ''non-maintainer-tags'' are not actively useful for patch-review. (except probably the guy mentioned fourteen times...)

But of course one could also find arguments to poke holes in this statistics, for example does this statistic not include code-locality:
A tested-by on a patch that touches some specific piece of code can be more worth than a signed-off in another part of the tree.

But... let's play it safe and let's err on the "safe" side meaning to not spam those people when in doubt. We already have the signed-off's and Maintainers file. So this should be ok. And if need be, the maintainers can always forward the patch.

[i probably could make a diploma thesis out of this changelog :)]

---
 scripts/get_maintainer.pl |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
scripts/get_maintainer.pl +1 -1
index b82ac95..b228198 100755
@@ -25,7 +25,7 @@ my $email_list = 1;
 my $email_subscriber_list = 0;
 my $email_git_penguin_chiefs = 0;
 my $email_git = 1;
-my $email_git_all_signature_types = 1;
+my $email_git_all_signature_types = 0;
 my $email_git_blame = 0;
 my $email_git_min_signatures = 1;
 my $email_git_max_maintainers = 5;
Signature · 1 line
1.7.0.4
Acked-by on v1 1/1
#30 lore raw

Re: [PATCH] scripts/get_maintainer.pl: default to not include unspecified tags

Quoted 40 lines · On Tue, May 11, 2010 at 08:36:36AM +0200, [email protected] wrote:

This changes the default of the option --git-all-signature-types to be disabled by default.

The effect being, that only certain (currently Signed-Off-By:,
Acked-By: and Reviewed-By:) tags are used to get adresses of potential maintainers.

Motivated is this change by the desire to not 'spam' people unnecessary:
A Tested-By or a Reported-By is not ment as a hint that those people want to/are able to review patches to the code in question.

In a quest to find resilient statistics for this i came up with this:

I produced a list of all the tag-signers not already covered
with a signed-off/acked/reviewed tag somewhere in the last year of git history.

Those were 650 addresses of "assumed non-developers".

And to check if those "assumed non-developers" are professional testers/reporters worth cc'ing, i then counted their total appearences in the git log:

469 were mentioned only once.
123 were mentioned twice.
38 three times
8 four times
5 six times
5 five times
1 eight times
1 fourteen times

I believe this supports my thesis, that the ''non-maintainer-tags'' are not actively useful for patch-review. (except probably the guy mentioned fourteen times...)

But of course one could also find arguments to poke holes in this statistics, for example does this statistic not include code-locality:
A tested-by on a patch that touches some specific piece of code can be more worth than a signed-off in another part of the tree.

Quoted 6 lines

But... let's play it safe and let's err on the "safe" side meaning to not spam those people when in doubt. We already have the signed-off's and Maintainers file. So this should be ok. And if need be, the maintainers can always forward the patch.

[i probably could make a diploma thesis out of this changelog :)]

:D

Quoted 2 lines

I second this reasoning:

Signature · 2 lines
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
maintainer
#31 lore raw

Re: [PATCH] scripts/get_maintainer.pl: default to not include unspecified tags

In reply to [email protected]
Quoted 3 lines · On Tue, 2010-05-11 at 08:36 +0200, [email protected] wrote:

This changes the default of the option --git-all-signature-types to be disabled by default.
[i probably could make a diploma thesis out of this changelog :)]

Then of course you would spell check it before submission. ;)

Luckily for you, most here don't seem to care much about that.

Quoted 1 line

I-can-be-persuaded-by-logic-and-statistics-to-Sign-off-by: Joe Perches <[email protected]>

#32 lore raw

Re: [PATCH] scripts/get_maintainer.pl: default to not include unspecified tags

Quoted 7 lines · On Tue, 11 May 2010 08:59:55 -0700 Joe Perches <[email protected]> wrote:
Quoted 3 lines · On Tue, 2010-05-11 at 08:36 +0200, [email protected] wrote:

This changes the default of the option --git-all-signature-types to be disabled by default.
[i probably could make a diploma thesis out of this changelog :)]

Then of course you would spell check it before submission. ;)

Come on! There were some difficult words in there! Give me some credit... :)

Cheers,
Flo

p.s.:
And of course i don't wanna miss out on the obligatory "lalala sorry, i'm not a native speaker lalala".

maintainer
#33 lore raw

Re: [PATCH] scripts/get_maintainer.pl: default to not include unspecified tags

Quoted 2 lines · Florian Mickler

Come on! There were some difficult words in there! Give me some credit... :)

If I had to rely on my prehistoric 2 semesters of German to communicate in restaurants, I'd go hungry but not thirsty.

cheers, Joe

maintainer
#34 lore raw

cover scripts/get_maintainer.pl: trivial improvements

In reply to [email protected]

A desire exists to reduce the number of people that are listed by get_maintainers that may not actually be interested in receiving patches. Non-standard signature types may not indicate a desire to receive patches. Add the ability to only use selected signature types.

Joe Perches (2):
scripts/get_maintainer.pl: optionally ignore non-maintainer signatures scripts/get_maintainer.pl: add .get_maintainer.conf default options file

 scripts/get_maintainer.pl |   66 ++++++++++++++++++++++++++++++++++++++-------
 1 files changed, 56 insertions(+), 10 deletions(-)
maintainer
#35 lore raw

PATCH 2/2 scripts/get_maintainer.pl: add .get_maintainer.conf default options file

In reply to [email protected]

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf can contain any valid command-line argument.

File contents are prepended to any additional command line arguments.

Multiple lines may be used, blank lines ignored, # is a comment.

Updated scripts/get_maintainer.pl version to 0.24

---
 scripts/get_maintainer.pl |   31 ++++++++++++++++++++++++++++++-
 1 files changed, 30 insertions(+), 1 deletions(-)
scripts/get_maintainer.pl +30 -1
index f66018d..b82ac95 100755
@@ -13,7 +13,7 @@
 use strict;
 
 my $P = $0;
-my $V = '0.23';
+my $V = '0.24';
 
 use Getopt::Long qw(:config no_auto_abbrev);
 
@@ -107,6 +107,30 @@ my %VCS_cmds_hg = (
     "blame_commit_pattern" => "^([0-9a-f]+):"
 );
 
+if (-f "${lk_path}.get_maintainer.conf") {
+    my @conf_args;
+    open(my $conffile, '<', "${lk_path}.get_maintainer.conf")
+	or warn "$P: Can't open .get_maintainer.conf: $!\n";
+    while (<$conffile>) {
+	my $line = $_;
+
+	$line =~ s/\s*\n?$//g;
+	$line =~ s/^\s*//g;
+	$line =~ s/\s+/ /g;
+
+	next if ($line =~ m/^\s*#/);
+	next if ($line =~ m/^\s*$/);
+
+	my @words = split(" ", $line);
+	foreach my $word (@words) {
+	    last if ($word =~ m/^#/);
+	    push (@conf_args, $word);
+	}
+    }
+    close($conffile);
+    unshift(@ARGV, @conf_args) if @conf_args;
+}
+
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
@@ -573,6 +597,11 @@ Notes:
           --git-min-signatures, --git-max-maintainers, --git-min-percent, and
           --git-blame
       Use --hg-since not --git-since to control date selection
+  File ".get_maintainer.conf", if it exists in the linux kernel source root
+      directory, can change whatever get_maintainer defaults are desired.
+      Entries in this file can be any command line argument.
+      This file is prepended to any additional command line arguments.
+      Multiple lines and # comments are allowed.
 EOT
 }
 
Signature · 1 line
1.7.0.3.311.g6a6955
#36 lore raw

Re: [PATCH 2/2] scripts/get_maintainer.pl: add .get_maintainer.conf default options file

Quoted 8 lines · Joe Perches

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf can contain any valid command-line argument.

File contents are prepended to any additional command line arguments.

Multiple lines may be used, blank lines ignored, # is a comment.

Do we really have to do this? This looks odd to me.

If the user of get_maintainer.pl uses some long command line arguments, to save his input, he should use bash aliases, not configure files. I never see configure files used like this.

Also, it doesn't worthy a configure file for such a script like get_maintainer.pl.

#37 lore raw

Re: [PATCH 2/2] scripts/get_maintainer.pl: add .get_maintainer.conf default options file

Quoted 11 lines · On Wed, 12 May 2010 14:30:42 +0800 Américo Wang <[email protected]>…
Quoted 8 lines · Joe Perches

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf can contain any valid command-line argument.

File contents are prepended to any additional command line arguments.

Multiple lines may be used, blank lines ignored, # is a comment.

Do we really have to do this? This looks odd to me.

it's a little hackish, but the code impact is small. if you prefer standard config-file syntax (i.e. "bla=" ) check the patch below.

joe, what do you think?

Quoted 3 lines

If the user of get_maintainer.pl uses some long command line arguments, to save his input, he should use bash aliases, not configure files. I never see configure files used like this.

bash aliases don't work (at least here) with "git send-email --cc-cmd="

Quoted 3 lines

Also, it doesn't worthy a configure file for such a script like get_maintainer.pl.

hm.. the only other way to use it via git send-email would be a wrapper script?

cheers,
Flo

Quoted 1 line

From 4f81b09c346075a062c868c59b724442c382b690 Mon Sep 17 00:00:00 2001

From: Florian Mickler <[email protected]>
Date: Wed, 12 May 2010 10:57:24 +0200
Subject: [PATCH] get_maintainer.pl: change config file to use key=value pairs.

the solution to prepend the contents of the config file to the arguments is unexpected for most people.

so we change it to be key = value syntax.

this approach uses a hash to store references to the config variables as this makes it easier to implement default/override semantics for the config file and cmdline.

---
 scripts/get_maintainer.pl |  169 ++++++++++++++++++++++++++++-----------------
 1 files changed, 106 insertions(+), 63 deletions(-)
scripts/get_maintainer.pl +106 -63
index b228198..e8aad5c 100755
@@ -48,6 +48,54 @@ my $from_filename = 0;
 my $pattern_depth = 0;
 my $version = 0;
 my $help = 0;
+my %prefs = (
+		'email' => \$email,
+		'git' => \$email_git,
+		'git-all-signature-types' => \$email_git_all_signature_types,
+		'git-blame' => \$email_git_blame,
+		'git-chief-penguins' => \$email_git_penguin_chiefs,
+		'git-min-signatures' => \$email_git_min_signatures,
+		'git-max-maintainers' => \$email_git_max_maintainers,
+		'git-min-percent' => \$email_git_min_percent,
+		'git-since' => \$email_git_since,
+		'hg-since' => \$email_hg_since,
+		'remove-duplicates' => \$email_remove_duplicates,
+		'maintainer' => \$email_maintainer,
+		'names' => \$email_usename,
+		'list' => \$email_list,
+		'subscribers' => \$email_subscriber_list,
+		'multiline' => \$output_multiline,
+		'roles' => \$output_roles,
+		'rolestats' => \$output_rolestats,
+		'separator' => \$output_separator,
+		'subsystem' => \$subsystem,
+		'status' => \$status,
+		'scm' => \$scm,
+		'web' => \$web,
+		'pattern-depth' => \$pattern_depth,
+		'keywords' => \$keywords,
+		'sections' => \$sections,
+		'file-emails' => \$file_emails,
+		'file' => \$from_filename,
+		'version' => \$version,
+		'help' => \$help,
+);
+
+my $conffile =  "${lk_path}.get_maintainer.conf";
+if (-f $conffile) {
+    open(my $conffile, '<', "${lk_path}.get_maintainer.conf")
+	or warn "$P: Can't open .get_maintainer.conf: $!\n";
+    while (<$conffile>) {
+	chomp;                  # no newline
+	s/#.*//;                # no comments
+	s/^\s+//;               # no leading white
+	s/\s+$//;               # no trailing white
+	next unless length;     # anything left?
+	my ($key, $val) = split(/\s*=\s*/, $_, 2);
+	${$prefs{$key}} = $val;
+    }
+    close($conffile);
+}
 
 my $exit = 0;
 
@@ -107,61 +155,37 @@ my %VCS_cmds_hg = (
     "blame_commit_pattern" => "^([0-9a-f]+):"
 );
 
-if (-f "${lk_path}.get_maintainer.conf") {
-    my @conf_args;
-    open(my $conffile, '<', "${lk_path}.get_maintainer.conf")
-	or warn "$P: Can't open .get_maintainer.conf: $!\n";
-    while (<$conffile>) {
-	my $line = $_;
-
-	$line =~ s/\s*\n?$//g;
-	$line =~ s/^\s*//g;
-	$line =~ s/\s+/ /g;
-
-	next if ($line =~ m/^\s*#/);
-	next if ($line =~ m/^\s*$/);
-
-	my @words = split(" ", $line);
-	foreach my $word (@words) {
-	    last if ($word =~ m/^#/);
-	    push (@conf_args, $word);
-	}
-    }
-    close($conffile);
-    unshift(@ARGV, @conf_args) if @conf_args;
-}
-
-if (!GetOptions(
-		'email!' => \$email,
-		'git!' => \$email_git,
-		'git-all-signature-types!' => \$email_git_all_signature_types,
-		'git-blame!' => \$email_git_blame,
-		'git-chief-penguins!' => \$email_git_penguin_chiefs,
-		'git-min-signatures=i' => \$email_git_min_signatures,
-		'git-max-maintainers=i' => \$email_git_max_maintainers,
-		'git-min-percent=i' => \$email_git_min_percent,
-		'git-since=s' => \$email_git_since,
-		'hg-since=s' => \$email_hg_since,
-		'remove-duplicates!' => \$email_remove_duplicates,
-		'm!' => \$email_maintainer,
-		'n!' => \$email_usename,
-		'l!' => \$email_list,
-		's!' => \$email_subscriber_list,
-		'multiline!' => \$output_multiline,
-		'roles!' => \$output_roles,
-		'rolestats!' => \$output_rolestats,
-		'separator=s' => \$output_separator,
-		'subsystem!' => \$subsystem,
-		'status!' => \$status,
-		'scm!' => \$scm,
-		'web!' => \$web,
-		'pattern-depth=i' => \$pattern_depth,
-		'k|keywords!' => \$keywords,
-		'sections!' => \$sections,
-		'fe|file-emails!' => \$file_emails,
-		'f|file' => \$from_filename,
-		'v|version' => \$version,
-		'h|help|usage' => \$help,
+if (!GetOptions( \%prefs,
+		'email!',
+		'git!',
+		'git-all-signature-types!',
+		'git-blame!',
+		'git-chief-penguins!',
+		'git-min-signatures=i',
+		'git-max-maintainers=i',
+		'git-min-percent=i',
+		'git-since=s',
+		'hg-since=s',
+		'remove-duplicates!',
+		'maintainer|m!',
+		'names|n!',
+		'list|l!',
+		'subscribers|s!',
+		'multiline!',
+		'roles!',
+		'rolestats!',
+		'separator=s',
+		'subsystem!',
+		'status!',
+		'scm!',
+		'web!',
+		'pattern-depth=i',
+		'keywords|k!',
+		'sections!',
+		'file-emails|fe!',
+		'file|f',
+		'version|v',
+		'help|h|usage',
 		)) {
     die "$P: invalid argument - use --help if necessary\n";
 }
@@ -545,10 +569,10 @@ MAINTAINER field selection options:
     --git-blame => use git blame to find modified commits for patch or file
     --git-since => git history to use (default: $email_git_since)
     --hg-since => hg history to use (default: $email_hg_since)
-    --m => include maintainer(s) if any
-    --n => include name 'Full Name <addr\@domain.tld>'
-    --l => include list(s) if any
-    --s => include subscriber only list(s) if any
+    --maintainer | --m  => include maintainer(s) if any
+    --names | --n => include name 'Full Name <addr\@domain.tld>'
+    --list | --l  => include list(s) if any
+    --subscribers | --s => include subscriber only list(s) if any
     --remove-duplicates => minimize duplicate email names/addresses
     --roles => show roles (status:subsystem, git-signer, list, etc...)
     --rolestats => show roles and statistics (commits/total_commits, %)
@@ -582,26 +606,45 @@ Notes:
           no individual file within the directory or subdirectory
           is matched.
       Used with "--git-blame", does not iterate all files in directory
+  
   Using "--git-blame" is slow and may add old committers and authors
       that are no longer active maintainers to the output.
+  
   Using "--roles" or "--rolestats" with git send-email --cc-cmd or any
       other automated tools that expect only ["name"] <email address>
       may not work because of additional output after <email address>.
+  
   Using "--rolestats" and "--git-blame" shows the #/total=% commits,
       not the percentage of the entire file authored.  # of commits is
       not a good measure of amount of code authored.  1 major commit may
       contain a thousand lines, 5 trivial commits may modify a single line.
+  
   If git is not installed, but mercurial (hg) is installed and an .hg
       repository exists, the following options apply to mercurial:
           --git,
           --git-min-signatures, --git-max-maintainers, --git-min-percent, and
           --git-blame
       Use --hg-since not --git-since to control date selection
-  File ".get_maintainer.conf", if it exists in the linux kernel source root
+
+  The file ".get_maintainer.conf", if it exists in the linux kernel source root
       directory, can change whatever get_maintainer defaults are desired.
-      Entries in this file can be any command line argument.
-      This file is prepended to any additional command line arguments.
-      Multiple lines and # comments are allowed.
+      Entries in this file can be any command line argument without the
+      preceding "--" followed by an equal sign ('=') and a 0 for disabled
+      or a 1 for enabled. 
+      This makes it easy to use it with non-default options via 
+      'git send-email --cc-cmd'.
+
+      An example file would look like this:
+      -------------------------------------------------------------------------
+      
+      ### comments and blank lines are allowed. ###
+      git-all-signature-types=1
+      lists=0 # don't include mailinglists
+      names=0 # no names
+
+      -------------------------------------------------------------------------
+
+  
 EOT
 }
 
Signature · 1 line
1.7.0.4
#38 lore raw

Re: [PATCH 2/2] scripts/get_maintainer.pl: add .get_maintainer.conf default options file

Quoted 9 lines · Florian Mickler
Quoted 6 lines · On Wed, 12 May 2010 14:30:42 +0800 Américo Wang <[email protected]>…
Quoted 5 lines · Joe Perches

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf can contain any valid command-line argument.

Quoted 9 lines
Quoted 5 lines
Quoted 3 lines

File contents are prepended to any additional command line arguments.

Multiple lines may be used, blank lines ignored, # is a comment.

Do we really have to do this? This looks odd to me.

it's a little hackish, but the code impact is small. if you prefer standard config-file syntax (i.e. "bla=" ) check the patch below.

Better somwhat, but...

Quoted 8 lines

joe, what do you think?

Quoted 3 lines

If the user of get_maintainer.pl uses some long command line arguments, to save his input, he should use bash aliases, not configure files. I never see configure files used like this.

bash aliases don't work (at least here) with "git send-email --cc-cmd="

...

Quoted 7 lines
Quoted 3 lines

Also, it doesn't worthy a configure file for such a script like get_maintainer.pl.

hm.. the only other way to use it via git send-email would be a wrapper script?

I still think we should avoid config files as possible as we can.

Thanks.

maintainer
#39 lore raw

Re: [PATCH 2/2] scripts/get_maintainer.pl: add .get_maintainer.conf default options file

Quoted 3 lines · Américo Wang
Quoted 2 lines

hm.. the only other way to use it via git send-email would be a wrapper script?

I still think we should avoid config files as possible as we can.

I had a request for a config capability awhile ago.

It wasn't then or is now a big deal to me.

If it's picked up by Andrew at all, I prefer the simpler style I posted as the code is smaller and the arguments are just the same as the command line so all of aliases work in the .conf file as well.

I do understand the desire for a Windows style .ini though. If it's an .ini I think it should use:

[sections]
key=value

Maybe that only if checkpatch or some other script uses a config as well.

maintainer
#40 lore raw

PATCH 1/1 scripts/get_maintainer.pl: add .get_maintainer.conf default options file

In reply to [email protected]

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf entries can be any valid argument.

Multiple lines may be used, blank lines ignored, # is a comment

---
 scripts/get_maintainer.pl |   24 ++++++++++++++++++++++++
 1 files changed, 24 insertions(+), 0 deletions(-)
scripts/get_maintainer.pl +24 -0
index 6f97a13..0c4c7e0 100755
@@ -97,6 +97,30 @@ my %VCS_cmds_hg = (
     "blame_commit_pattern" => "^([0-9a-f]+):"
 );
 
+if (-f ${lk_path}.get_maintainer.conf) {
+    my @conf_args;
+    open(my $conffile, '<', "${lk_path}.get_maintainer.conf")
+	or warn "$P: Can't open .get_maintainer.conf: $!\n";
+    while (<$conffile>) {
+	my $line = $_;
+
+	$line =~ s/\s*\n?$//g;
+	$line =~ s/^\s*//g;
+	$line =~ s/\s+/ /g;
+
+	next if ($line =~ m/^\s*#/);
+	next if ($line =~ m/^\s*$/);
+
+	my @words = split(" ", $line);
+	foreach my $word (@words) {
+	    last if ($word =~ m/^#/);
+	    push (@conf_args, $word);
+	}
+    }
+    close($conffile);
+    unshift(@ARGV, @conf_args) if @conf_args;
+}
+
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
Reviewed-by on v1 1/1
#41 lore raw

Re: [PATCH] scripts/get_maintainer.pl: add .get_maintainer.conf default options file

Quoted 8 lines · On Sat, 08 May 2010 15:26:49 -0700 Joe Perches <[email protected]> wrote:

Allow the use of a .get_maintainer.conf file to control the default options applied when scripts/get_maintainer.pl is run.

.get_maintainer.conf entries can be any valid argument.

Multiple lines may be used, blank lines ignored, # is a comment

probably worth noting, that an entry is of the form

"--cmdlinearg"

as that's unexpected for a config file.

scripts/get_maintainer.pl my %VCS_cmds_hg = ( · line 100

Quoted 13 lines
---
 scripts/get_maintainer.pl |   24 ++++++++++++++++++++++++
 1 files changed, 24 insertions(+), 0 deletions(-)
scripts/get_maintainer.pl +1 -0
index 6f97a13..0c4c7e0 100755
@@ -97,6 +97,30 @@ my %VCS_cmds_hg = (
     "blame_commit_pattern" => "^([0-9a-f]+):"
 );
 
+if (-f ${lk_path}.get_maintainer.conf) {

missing "" around the argument to -f ...

i wouldn't put the trailing slash in the variables but add it when composing path's. but it seems this is the convention for get_maintainer.pl...

Quoted 12 lines
scripts/get_maintainer.pl +12 -0
+    my @conf_args;
+    open(my $conffile, '<', "${lk_path}.get_maintainer.conf")
+	or warn "$P: Can't open .get_maintainer.conf: $!\n";
+    while (<$conffile>) {
+	my $line = $_;
+
+	$line =~ s/\s*\n?$//g;
+	$line =~ s/^\s*//g;
+	$line =~ s/\s+/ /g;
+
+	next if ($line =~ m/^\s*#/);
+	next if ($line =~ m/^\s*$/);

scripts/get_maintainer.pl my %VCS_cmds_hg = ( · line 122

Quoted 10 lines
scripts/get_maintainer.pl +10 -0
+
+	my @words = split(" ", $line);
+	foreach my $word (@words) {
+	    last if ($word =~ m/^#/);
+	    push (@conf_args, $word);
+	}
+    }
+    close($conffile);
+    unshift(@ARGV, @conf_args) if @conf_args;
+}

this works.

scripts/get_maintainer.pl my %VCS_cmds_hg = ( · line 126

Quoted 5 lines
scripts/get_maintainer.pl +1 -0
+
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
#42 lore raw

Re: indiscriminate get_maintainer.pl usage

In reply to Stefan Richter
Quoted 9 lines
$ scripts/get_maintainer.pl --rolestats -f fs/eventpoll.c
Alexander Viro <[email protected]> (maintainer:FILESYSTEMS (VFS...)
Andrew Morton <[email protected]> (commit_signer:2/6=33%)
"Eric W. Biederman" <[email protected]> (commit_signer:2/6=33%)
Davide Libenzi <[email protected]> (commit_signer:2/6=33%)
Roland Dreier <[email protected]> (commit_signer:1/6=17%)
Stefan Richter <[email protected]> (commit_signer:1/6=17%)
[email protected] (open list:FILESYSTEMS (VFS...)
[email protected] (open list)

And my appearing here is a bit silly as well, since I have made one commit to fs/eventpoll.c that changed one line. A bit different in scope from Davide who designed and wrote the whole thing...

OTOH it's not clear how smart we want get_maintainer.pl to try to be.

- R.

Signature · 2 lines
Roland Dreier <[email protected]> || For corporate legal information go to:
http://www.cisco.com/web/about/doing_business/legal/cri/index.html
#43 lore raw
In reply to Changli Gao
Quoted 5 lines · Changli Gao

use wrapper functions.

epoll should not touch flags in wait_queue_t. This patch introduces a new function add_wait_queue_head_exclusive_locked(), for the users, who use wait queue as a LIFO queue.

Since we already have __add_wait_queue(), __add_wait_queue_tail() and __remove_wait_queue() (which all means "locked"), and while I agree in having the exclusive-add wrapped into a function, I much better prefer a:

static inline void __add_wait_queue_excl(wait_queue_head_t *head,
wait_queue_t *new)

{
	new->flags |= WQ_FLAG_EXCLUSIVE;
	__add_wait_queue(head, new);
}

The patch you posted introduces a different naming, which leaves all the other __*() untouched, and wraps the already one-liner __remove_wait_queue() with yet another one-liner.

Quoted 64 lines

----

 fs/eventpoll.c       |    5 ++---
 include/linux/wait.h |   15 +++++++++++++--
 2 files changed, 15 insertions(+), 5 deletions(-)
fs/eventpoll.c +2 -3
index bd056a5..8137f6e 100644
@@ -1140,8 +1140,7 @@ retry:
 		 * ep_poll_callback() when events will become available.
 		 */
 		init_waitqueue_entry(&wait, current);
-		wait.flags |= WQ_FLAG_EXCLUSIVE;
-		__add_wait_queue(&ep->wq, &wait);
+		add_wait_queue_head_exclusive_locked(&ep->wq, &wait);
 
 		for (;;) {
 			/*
@@ -1161,7 +1160,7 @@ retry:
 			jtimeout = schedule_timeout(jtimeout);
 			spin_lock_irqsave(&ep->lock, flags);
 		}
-		__remove_wait_queue(&ep->wq, &wait);
+		remove_wait_queue_locked(&ep->wq, &wait);
 
 		set_current_state(TASK_RUNNING);
 	}
include/linux/wait.h +13 -3
index a48e16b..de2566d 100644
@@ -407,17 +407,28 @@ do {									\
  * Must be called with the spinlock in the wait_queue_head_t held.
  */
 static inline void add_wait_queue_exclusive_locked(wait_queue_head_t *q,
-						   wait_queue_t * wait)
+						   wait_queue_t *wait)
 {
 	wait->flags |= WQ_FLAG_EXCLUSIVE;
 	__add_wait_queue_tail(q,  wait);
 }
 
 /*
+ * Must be called with the spinlock in the wait_queue_head_t held, and
+ * q must be for exclusive wait only.
+ */
+static inline void add_wait_queue_head_exclusive_locked(wait_queue_head_t *q,
+							wait_queue_t *wait)
+{
+	wait->flags |= WQ_FLAG_EXCLUSIVE;
+	__add_wait_queue(q,  wait);
+}
+
+/*
  * Must be called with the spinlock in the wait_queue_head_t held.
  */
 static inline void remove_wait_queue_locked(wait_queue_head_t *q,
-					    wait_queue_t * wait)
+					    wait_queue_t *wait)
 {
 	__remove_wait_queue(q,  wait);
 }
--
Quoted 5 lines

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

- Davide

maintainer
#44 lore raw
Quoted 6 lines · Davide Libenzi

Since we already have __add_wait_queue(), __add_wait_queue_tail() and __remove_wait_queue() (which all means "locked"), and while I agree in having the exclusive-add wrapped into a function, I much better prefer a:

static inline void __add_wait_queue_excl(wait_queue_head_t *head,
wait_queue_t *new)

Quoted 8 lines
{
        new->flags |= WQ_FLAG_EXCLUSIVE;
        __add_wait_queue(head, new);
}

The patch you posted introduces a different naming, which leaves all the other __*() untouched, and wraps the already one-liner __remove_wait_queue() with yet another one-liner.

I concur, I always get confused by the _locked postfix (and its more typing). Also, it goes against the lock data not code paradigm.

#45 lore raw
Quoted 12 lines · Peter Zijlstra
Quoted 11 lines · Davide Libenzi

Since we already have __add_wait_queue(), __add_wait_queue_tail() and __remove_wait_queue() (which all means "locked"), and while I agree in having the exclusive-add wrapped into a function, I much better prefer a:

static inline void __add_wait_queue_excl(wait_queue_head_t *head, wait_queue_t *new) {
        new->flags |= WQ_FLAG_EXCLUSIVE; __add_wait_queue(head, new);
}

Quoted 8 lines
Quoted 3 lines

The patch you posted introduces a different naming, which leaves all the other __*() untouched, and wraps the already one-liner __remove_wait_queue() with yet another one-liner.

I concur, I always get confused by the _locked postfix (and its more typing). Also, it goes against the lock data not code paradigm.

I greped all the code, and found that add_wait_queue_head_exclusive_locked() and remove_wait_queue_locked() aren't used. It seems that no users like these APIs. So I will remove these two APIs, and add __add_wait_queue_excl() instead. Thanks.

Signature · 2 lines
Regards,
Changli Gao([email protected])