MAINTAINERS: cleanups
Joe Perches
maintainercover MAINTAINERS: cleanups
A few more cleanups to MAINTAINERS
Joe Perches (7):
MAINTAINERS: IA64 - pair P:/M: entries properly
MAINTAINERS: Remove ivtv-user lists, add CX18 url
MAINTAINERS: QLGE 10Gb ETHERNET - pair P:/M: entries properly
MAINTAINERS: Use tabs in ACER ASPIRE ONE
MAINTAINERS: Remove L: [email protected]
MAINTAINERS: Move ARPD to CREDITS
MAINTAINERS: Update KERNEL JANITORS
CREDITS | 3 +++ MAINTAINERS | 33 ++++++++++----------------------- 2 files changed, 13 insertions(+), 23 deletions(-)
Joe Perches
maintainerPATCH 3/7 MAINTAINERS: QLGE 10Gb ETHERNET - pair P:/M: entries properly
--- MAINTAINERS | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
MAINTAINERS +1 -1
@@ -4810,8 +4810,8 @@ F: drivers/net/qla3xxx.* QLOGIC QLGE 10Gb ETHERNET DRIVER P: Ron Mercer -M: [email protected] M: [email protected] +M: [email protected] L: [email protected] S: Supported F: drivers/net/qlge/
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 1/7 MAINTAINERS: IA64 - pair P:/M: entries properly
In reply to Joe Perches--- MAINTAINERS | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
MAINTAINERS +1 -1
@@ -2851,8 +2851,8 @@ S: Maintained IA64 (Itanium) PLATFORM P: Tony Luck -P: Fenghua Yu M: [email protected] +P: Fenghua Yu M: [email protected] L: [email protected] W: http://www.ia64-linux.org/
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 4/7 MAINTAINERS: Use tabs in ACER ASPIRE ONE
In reply to Joe Perches--- MAINTAINERS | 10 +++++----- 1 files changed, 5 insertions(+), 5 deletions(-)
MAINTAINERS +5 -5
@@ -231,11 +231,11 @@ S: Maintained F: drivers/net/acenic* ACER ASPIRE ONE TEMPERATURE AND FAN DRIVER -P: Peter Feuerer -M: [email protected] -W: http://piie.net/?section=acerhdf -S: Maintained -F: drivers/platform/x86/acerhdf.c +P: Peter Feuerer +M: [email protected] +W: http://piie.net/?section=acerhdf +S: Maintained +F: drivers/platform/x86/acerhdf.c ACER WMI LAPTOP EXTRAS P: Carlos Corbacho
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 2/7 MAINTAINERS: Remove ivtv-user lists, add CX18 url
In reply to Joe Perches--- MAINTAINERS | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-)
MAINTAINERS +1 -2
@@ -1721,10 +1721,10 @@ M: [email protected] P: Andy Walls M: [email protected] L: [email protected] -L: [email protected] L: [email protected] T: git git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git W: http://linuxtv.org +W: http://www.ivtvdriver.org/index.php/Cx18 S: Maintained F: Documentation/video4linux/cx18.txt F: drivers/media/video/cx18/
@@ -3276,7 +3276,6 @@ IVTV VIDEO4LINUX DRIVER P: Hans Verkuil M: [email protected] L: [email protected] -L: [email protected] L: [email protected] T: git git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6.git W: http://www.ivtvdriver.org
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 5/7 MAINTAINERS: Remove L: [email protected]
In reply to Joe Perchesfrom sections that should not have them.
--- MAINTAINERS | 5 ----- 1 files changed, 0 insertions(+), 5 deletions(-)
MAINTAINERS +0 -5
@@ -2197,7 +2197,6 @@ F: drivers/scsi/lpfc/ ENE CB710 FLASH CARD READER DRIVER P: Michał Mirosław M: [email protected] -L: [email protected] S: Maintained F: drivers/misc/cb710/ F: drivers/mmc/host/cb710-mmc.*
@@ -3473,13 +3472,11 @@ P: Vegard Nossum M: [email protected] P Pekka Enberg M: [email protected] -L: [email protected] S: Maintained KMEMLEAK P: Catalin Marinas M: [email protected] -L: [email protected] S: Maintained F: Documentation/kmemleak.txt F: include/linux/kmemleak.h
@@ -4298,7 +4295,6 @@ F: drivers/video/omap/ OMAP MMC SUPPORT P: Jarkko Lavinen M: [email protected] -L: [email protected] L: [email protected] S: Maintained F: drivers/mmc/host/*omap*
@@ -4577,7 +4573,6 @@ P: Paul Mackerras M: [email protected] P: Ingo Molnar M: [email protected] -L: [email protected] S: Supported PERSONALITY HANDLING
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 6/7 MAINTAINERS: Move ARPD to CREDITS
In reply to Joe PerchesJonathan Layes is hard to find.
--- CREDITS | 3 +++ MAINTAINERS | 6 ------ 2 files changed, 3 insertions(+), 6 deletions(-)
CREDITS +3 -0
@@ -2006,6 +2006,9 @@ E: [email protected] D: Soundblaster driver fixes, ISAPnP quirk S: California, USA +N: Jonathan Layes +D: ARPD support + N: Tom Lees E: [email protected] W: http://www.lpsg.demon.co.uk/
MAINTAINERS +0 -6
@@ -951,12 +951,6 @@ W: http://www.arm.linux.org.uk/ S: Maintained F: arch/arm/vfp/ -ARPD SUPPORT -P: Jonathan Layes -L: [email protected] -S: Maintained -F: net/ipv4/arp.c - ASUS ACPI EXTRAS DRIVER P: Corentin Chary M: [email protected]
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Joe Perches
maintainerPATCH 7/7 MAINTAINERS: Update KERNEL JANITORS
In reply to Joe Perches--- MAINTAINERS | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-)
MAINTAINERS +2 -3
@@ -3366,10 +3366,9 @@ F: Makefile F: scripts/Makefile.* KERNEL JANITORS -P: Several L: [email protected] -W: http://www.kerneljanitors.org/ -S: Maintained +W: http://janitor.kernelnewbies.org/ +S: Odd fixes KERNEL NFSD, SUNRPC, AND LOCKD SERVERS P: J. Bruce Fields
Signature · 1 line
1.6.3.1.10.g659a0.dirty
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
In reply to Joe PerchesHi Joe,
Quoted 1 line · Joe Perches
from sections that should not have them.
What is that supposed to mean?
Quoted 10 lines
Signed-off-by Joe Perches <[email protected]>---
MAINTAINERS | 5 -----1 files changed, 0 insertions(+), 5 deletions(-)MAINTAINERS +0
-0@@ -3473,13 +3472,11 @@ P: Vegard Nossum
Quoted 5 lines
M: [email protected]
P Pekka Enberg
M: [email protected]
-L: [email protected]
S: Maintained
Why do you want to remove that? It's there to tell people that you're _supposed_ to cc LKML for kmemcheck patches in contrast to subsystems (e.g. networking) that want people to cc _their_ mailing list (keeping LKML is optional). Hmmm?
Pekka
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 1 line · Pekka Enberg
Hi Joe,
Hi Pekka.
Quoted 3 lines · Joe Perches
Quoted 1 line
from sections that should not have them.
What is that supposed to mean?
That there should be no sections that have LKML as a list entry other than "everything".
A earlier commit, b5472cddbe2c41fd434592ecf3c5b81a551d5bea, removed all the entries for LKML but "THE REST". These removed entries have been added since that commit.
Having just a few sections show LKML as an entry may lead people to think that LKML should not be used as a CC entry unless it is specifically listed.
Documentation/SubmittingPatches already states that LKML should be CC'd on all patches.
cheers, Joe
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Hi Joe,
Quoted 10 lines · Joe Perches
Quoted 3 lines · Joe Perches
Quoted 1 line
from sections that should not have them.
What is that supposed to mean?
That there should be no sections that have LKML as a list entry other than "everything".
A earlier commit,
b5472cddbe2c41fd434592ecf3c5b81a551d5bea, removed all the entries for LKML but "THE REST". These removed entries have been added since that commit.
OK, I don't agree, but as Andrew apparently thinks it's a good idea because he merged the previous patch, there's no reason not to merge the new one.
Quoted 2 lines · Joe Perches
Documentation/SubmittingPatchesalready states that LKML should be CC'd on all patches.
Maybe it does, but real world doesn't seem to agree with you here.
Pekka
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 5 lines · Pekka Enberg
Quoted 2 lines · Joe Perches
Documentation/SubmittingPatchesalready states that LKML should be CC'd on all patches.Maybe it does, but real world doesn't seem to agree with you here.
Well, if scripts/get_maintainer.pl gets used more often, it would.
cheers, Joe
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Hi Joe,
Quoted 2 lines · Joe Perches
Quoted 2 lines
Quoted 2 lines
Documentation/SubmittingPatchesalready states that LKML should be CC'd on all patches.
Quoted 1 line · Pekka Enberg
Quoted 1 line
Maybe it does, but real world doesn't seem to agree with you here.
Quoted 2 lines · Joe Perches
Well, if
scripts/get_maintainer.plgets used more often, it would.
And kernel hackers could fly if they had wings.
Look, all I care is that people who send patches to subsystems I maintain know what to CC. The get_maintainer.pl script is obviously really helpful here but the fact is, it's _not_ mandatory to CC LKML for all patches. So I worry that people will be confused and no list is CC'd because other subsystems that have a separate mailing list have that in MAINTAINERS.
But anyway, no need to argue about this, Andrew merged your first patch so the second patch makes sense as well.
Pekka
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 1 line · Pekka Enberg
And kernel hackers could fly if they had wings.
No need for wings -- enough thrust would suffice. ;)
Maciej
Re: [PATCH 7/7] MAINTAINERS: Update KERNEL JANITORS
In reply to Joe PerchesQuoted 12 lines · Joe Perches
Signed-off-by Joe Perches <[email protected]>--- MAINTAINERS | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-)MAINTAINERS +0
-0@@ -3366,10 +3366,9 @@ F: Makefile F: scripts/Makefile.*
MAINTAINERS F: Makefile · line 3374
Quoted 10 lines
MAINTAINERS +2
-3KERNEL JANITORS -P: Several L: [email protected] -W: http://www.kerneljanitors.org/ -S: Maintained +W: http://janitor.kernelnewbies.org/ +S: Odd fixes KERNEL NFSD, SUNRPC, AND LOCKD SERVERS P: J. Bruce Fields
Alexey could comment on that, but I agree with it.
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
In reply to Joe PerchesQuoted 2 lines · Joe Perches
Well, if
scripts/get_maintainer.plgets used more often, it would.
FWIW I've noticed that get_maintainer.pl has a tendency to pick up people it probably shouldn't when run on areas that have had few people committing to them since it tends to pick up people who've done generic cleanups but have no specific interest in that area of code. I think a heuristic based on filtering the git log based on the proportion of commits that were done by each committer would help with many of these cases.
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 7 lines · Mark Brown
FWIW I've noticed that get_maintainer.pl has a tendency to pick up people it probably shouldn't when run on areas that have had few people committing to them since it tends to pick up people who've done generic cleanups but have no specific interest in that area of code. I think a heuristic based on filtering the git log based on the proportion of commits that were done by each committer would help with many of these cases.
Hi Mark.
git log "-by:" lines are already sorted by number of commits before selecting maintainers. It doesn't matter if the line is acked-by:, Signed-off-by:, tested-by:, or brown-paper-bagged-by:, any "by:" signature is used.
get_maintainers does:
git log --since=${email_git_since} -- ${file} \
| grep -Ei "^[-_ a-z]+by:.*\@.*$" \
| cut -f2- -d":" | sort | uniq -c | sort -rnThere are a few existing options that could be used to minimize cleanup style committers.
--git => include recent git *-by: signers --git-min-signatures => number of signatures required (default: 1) --git-max-maintainers => maximum maintainers to add (default: 5) --git-since => git history to use (default: 1-year-ago)
Using --nogit lists just the entries in MAINTAINERS
Using --git-min-signatures=3 or so seems to minimize the generic cleanups committers.
Using --git-max-maintainers=3 or so, for files with more changes, seems also to minimize the generic cleanup committers
Using --git-since=6-months-ago reduces the history period
Using --stat on the commit log and weighting for things for lines changed would probably not be good because generic cleanups often change more of the code than real logic changes.
I suppose when the number of committers returned by the "git log | grep" is low, returning just the most frequent committers might be good.
Any suggestions on better heuristics?
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 11 lines · Joe Perches
Quoted 4 lines · Mark Brown
cleanups but have no specific interest in that area of code. I think a heuristic based on filtering the git log based on the proportion of commits that were done by each committer would help with many of these cases.
git log "-by:" lines are already sorted by number of commits before selecting maintainers. It doesn't matter if the line is acked-by:, Signed-off-by:, tested-by:, or brown-paper-bagged-by:, any "by:" signature is used.
Yup, I'm aware of that.
Quoted 5 lines
I suppose when the number of committers returned by the "git log | grep" is low, returning just the most frequent committers might be good.
Any suggestions on better heuristics?
That's pretty much what I was thinking of too - tweaking the values for the log queries based on the total number of hits so people only turn up if they did a certain proportion of the commits. It'd take some experimentation to work out what the values to use should be, I guess.
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 4 lines · Mark Brown
That's pretty much what I was thinking of too - tweaking the values for the log queries based on the total number of hits so people only turn up if they did a certain proportion of the commits. It'd take some experimentation to work out what the values to use should be, I guess.
Perhaps something like this:
scripts/get_maintainer.pl +17 -0
@@ -27,6 +27,7 @@ my $email_git = 1; my $email_git_penguin_chiefs = 0; my $email_git_min_signatures = 1; my $email_git_max_maintainers = 5; +my $email_git_min_percent = 0; my $email_git_since = "1-year-ago"; my $output_multiline = 1; my $output_separator = ", ";
@@ -65,6 +66,7 @@ if (!GetOptions( '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, 'm!' => \$email_maintainer, 'n!' => \$email_usename,
@@ -307,6 +309,7 @@ MAINTAINER field selection options: --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: 0) --git-since => git history to use (default: 1-year-ago) --m => include maintainer(s) if any --n => include name 'Full Name <addr\@domain.tld>'
@@ -497,6 +500,7 @@ sub recent_git_signoffs { my $output = ""; my $count = 0; my @lines = (); + my $total_sign_offs; if (which("git") eq "") { warn("$P: git not found. Add --nogit to options?\n");
@@ -520,6 +524,14 @@ sub recent_git_signoffs { $output =~ s/^\s*//gm; @lines = split("\n", $output); + + $total_sign_offs = 0; + foreach my $line (@lines) { + if ($line =~ m/([0-9]+)\s+(.*)/) { + $total_sign_offs += $1; + } + } + foreach my $line (@lines) { if ($line =~ m/([0-9]+)\s+(.*)/) { my $sign_offs = $1;
@@ -529,6 +541,11 @@ sub recent_git_signoffs { $count > $email_git_max_maintainers) { last; } + if ($email_git_min_percent > 0 && + (($sign_offs * 100 / $total_sign_offs) < $email_git_min_percent) + ) { + last; + } } else { die("$P: Unexpected git output: ${line}\n"); }
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 10 lines · Joe Perches
Quoted 4 lines · Mark Brown
That's pretty much what I was thinking of too - tweaking the values for the log queries based on the total number of hits so people only turn up if they did a certain proportion of the commits. It'd take some experimentation to work out what the values to use should be, I guess.
Perhaps something like this:
+ --git-min-percent => minimum percentage of commits required (default: 0)
That seems reasonably effective with 5% or so as the default. I do think the default should be non-zero if we're going to be encouraging people to use this as standard, though.
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 6 lines · Mark Brown
Quoted 5 lines · Joe Perches
Quoted 4 lines · Mark Brown
That's pretty much what I was thinking of too - tweaking the values for the log queries based on the total number of hits so people only turn up if they did a certain proportion of the commits. It'd take some experimentation to work out what the values to use should be, I guess.
Quoted 6 lines
Quoted 2 lines
Perhaps something like this:
+ --git-min-percent => minimum percentage of commits required (default: 0)That seems reasonably effective with 5% or so as the default. I do think the default should be non-zero if we're going to be encouraging people to use this as standard, though.
I'm not sure minimization of maintainers in the result is that desirable but I am for figuring out what works best.
I set it at 0 for now just to play with it.
What have you experimented with and how have the results changed?
If minimization is what's desired, what I've tried seems to work best at around 10.
Here's the little script I used.
#!/bin/bash
#
ShowMaintainers() {
file=$1
percent=$2
extraargs=$3
echo
echo "Maintainers for: $file @ $percent % $extraargs"
echo ./scripts/get_maintainer.pl $extraargs --git-min-percent=$percent -f $file
}
for dir in $(find drivers -type d) ; do
dir=$(echo $dir | sed -e 's/\.\///g') ShowMaintainers $dir 0 ShowMaintainers $dir 5 ShowMaintainers $dir 10 ShowMaintainers $dir 0 "--nogit"
done
Re: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 9 lines · Joe Perches
Quoted 3 lines · Mark Brown
That seems reasonably effective with 5% or so as the default. I do think the default should be non-zero if we're going to be encouraging people to use this as standard, though.
I'm not sure minimization of maintainers in
the result is that desirable but I am for
figuring out what works best.
It's not so much minimising maintainers per se that I'm concerned about as ensuring that people doing general cleanup work don't end up getting CCed on lots of random stuff, partly for their benefit and partly to make get_maintainers easier to use.
Quoted 2 lines
What have you experimented with and how have the
results changed?
A combination of patches I'm working on and source files in subsystems I maintain (basically, stuff where I know off the top of my head who should turn up). The noticable issues are things like Jean Delvare turning up for sound/soc/codecs/tlv320aic3x.c as a result of kernel wide API update work he was doing - by tweaking the percentage I was able to get the output to exclude people who I know have done non-specific cleanup work.
Joe Perches
maintainerRe: [PATCH 5/7] MAINTAINERS: Remove L: [email protected]
Quoted 5 lines · Mark Brown
Quoted 4 lines · Joe Perches
Quoted 3 lines · Mark Brown
That seems reasonably effective with 5% or so as the default. I do think the default should be non-zero if we're going to be encouraging people to use this as standard, though.
Andrew, do you have an opinion here?
You show up in these lists rather less often at 10% than 5.