Date:   Mon, 6 Mar 2017 10:52:00 +1100
From:   Stephen Rothwell <sfr@...b.auug.org.au>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: linux-next: stats (Was: Linux 4.11-rc1)

Hi Linus,

On Sun, 5 Mar 2017 13:41:04 -0800 Linus Torvalds <torvalds@...ux-foundation.org> wrote:
>
> It _does_ feel like there was more stuff that I was asked to pull than
> was in linux-next. That always happens, but seems to have happened
> more now than usually. Comparing to the linux-next tree at the time of
> the 4.10 release, almost 18% of the non-merge commits were not in
> Linux-next. That seems higher than usual, although I guess Stephen
> Rothwell has actual numbers from past merges.
> 
> Now, about a quarter of the patches that weren't in linux-next do end
> up having the same patch ID as something that was, so some of it was
> due to just rebasing. But still - we have about 13% of the merge
> window that wasn't in linux-next when 4.10 was released.
> 
> Looking at the sources of that, there's a few different classes:
> 
>  - fixes.
> 
>    This is obviously ok and inevitable. I don't expect everything to
> have been in linux-next, after all.
> 
>  - the statx() systen call thing.
> 
>    Yeah, I'll allow this one too, because quite frankly, the first
> version of that patch was posted over six years ago.
> 
>  - there's the quite noticeable <linux/sched.h> split-up series
> 
>    This one was posted and discussed before the merge window, and
> needed to be merged late (and even then caused some conflicts). So it
> had real reasons for late inclusion.
> 
>  - a couple of subsystems. drm, Infiniband, watchdog and btrfs stand out.

My stats:

As usual, the executive friendly graph is at
http://neuling.org/linux-next-size.html :-)

(No merge commits counted, next-20170220 was the first linux-next after
the merge window opened.)

Commits in v4.11-rc1 (relative to v4.10):          10960
Commits in next-20170220:                           9791
Commits with the same SHA1:                         9016
Commits with the same patch_id:                      479 (1)
Commits with the same subject line:                   68 (1)

(1) not counting those in the lines above.

So commits in -rc1 that were in next-20170220:      9563 87%

[
v4.10-rc1 like this:

Commits in v4.10-rc1 (relative to v4.9):           11455
Commits in next-20161212:                          10625
Commits with the same SHA1:                         9927
Commits with the same patch_id:                      437 (1)
Commits with the same subject line:                   25 (1)

(1) not counting those in the lines above.

So commits in -rc1 that were in next-20161212:     10389 90%

And v4.9-rc1 like this:

Commits in v4.9-rc1 (relative to v4.8):            14308
Commits in next-20161004:                          13539
Commits with the same SHA1:                        12716
Commits with the same patch_id:                      485 (1)
Commits with the same subject line:                   33 (1)

(1) not counting those in the lines above.

So commits in -rc1 that were in next-20161004:     13234 92%

So this time we have slightly more rebasing and overall more "extra" commits.
]

Some breakdown of the list of extra commits (relative to next-20170220)
in -rc1:

Top ten first word of commit summary:

    142 sched
    117 drm
     72 net
     66 ib
     63 watchdog
     61 btrfs
     60 scsi
     48 f2fs
     43 tools
     29 perf

Top ten authors:

    146 mingo@...nel.org
     46 bskeggs@...hat.com
     45 len.brown@...el.com
     39 bart.vanassche@...disk.com
     32 linux@...ck-us.net
     32 acourbot@...dia.com
     28 nborisov@...e.com
     26 idryomov@...il.com
     22 hch@....de
     21 anna.schumaker@...app.com

Top ten commiters:

    217 davem@...emloft.net
    161 mingo@...nel.org
     88 bskeggs@...hat.com
     79 dledford@...hat.com
     64 linux@...ck-us.net
     63 anna.schumaker@...app.com
     59 martin.petersen@...cle.com
     58 dsterba@...e.com
     57 axboe@...com
     52 idryomov@...il.com

There are also 229 commits in next-20170220 that didn't make it into
v4.11-rc1.

Top ten first word of commit summary:

     19 mm
     18 arm
     12 edac
      9 keys
      9 befs
      7 random
      6 target
      6 coresight
      6 bf609
      5 edac.txt

Top ten authors:

     19 mchehab@...nel.org
     17 akpm@...ux-foundation.org
     11 dhowells@...hat.com
     10 olof@...om.net
      9 luisbg@....samsung.com
      8 viro@...iv.linux.org.uk
      7 arnd@...db.de
      6 varun@...lsio.com
      5 sonic.zhang@...log.com
      4 realmz6@...il.com

Some of Andrew's patches are fixes for other patches in his tree (and
have been merged into those).

Top ten commiters:

     76 sfr@...b.auug.org.au
     20 mchehab@...nel.org
     16 steven@...ntu-virtualbox.(none)
     11 dhowells@...hat.com
     10 olof@...om.net
      9 luisbg@....samsung.com
      8 viro@...iv.linux.org.uk
      7 tytso@....edu
      7 bart.vanassche@...disk.com
      6 mathieu.poirier@...aro.org

Those commits by me are from the quilt series (mainly Andrew's mmotm
tree).  The steven@...ntu-virtualbox.(none) ones are form a very out of
date blackfin tree that has now been removed from linux-next.

-- 
Cheers,
Stephen Rothwell

Powered by blists - more mailing lists