reviewing Ubuntu kernel patches

reviewing Ubuntu kernel patches

[Randy Dunlap]

Hi,

A few of us at Oracle have been reviewing the Ubuntu kernel git tree
to see what "extra" patches they are using that might make sense in
the mainline kernel.

I'll be posting these for comments/consideration etc. in no particular
order (over a period of days, not all at one time).

Thanks,
~Randy

Re: reviewing Ubuntu kernel patches

[Ben Collins]

On Wed, 2006-06-14 at 16:22 -0700, Randy Dunlap wrote:
> Hi,
> 
> A few of us at Oracle have been reviewing the Ubuntu kernel git tree
> to see what "extra" patches they are using that might make sense in
> the mainline kernel.
> 
> I'll be posting these for comments/consideration etc. in no particular
> order (over a period of days, not all at one time).

Excellent. I had that slated for two weeks from now, but if someone else
is doing it, I'm happy to just review :)

If you have any questions on the rhyme or reason of a patch, let me
know.

Thanks

-- 
Ubuntu     - http://www.ubuntu.com/
Debian     - http://www.debian.org/
Linux 1394 - http://www.linux1394.org/
SwissDisk  - http://www.swissdisk.com/

Re: reviewing Ubuntu kernel patches

[Arjan van de Ven]

On Wed, 2006-06-14 at 16:22 -0700, Randy Dunlap wrote:
> Hi,
> 
> A few of us at Oracle have been reviewing the Ubuntu kernel git tree
> to see what "extra" patches they are using that might make sense in
> the mainline kernel.
> 
> I'll be posting these for comments/consideration etc. in no particular
> order (over a period of days, not all at one time).

while I highly applaud this effort this does put the interesting
question of "who puts his signed-off-by on these patches".....

Re: reviewing Ubuntu kernel patches

[Randy Dunlap]

Arjan van de Ven wrote:
> On Wed, 2006-06-14 at 16:22 -0700, Randy Dunlap wrote:
>> Hi,
>>
>> A few of us at Oracle have been reviewing the Ubuntu kernel git tree
>> to see what "extra" patches they are using that might make sense in
>> the mainline kernel.
>>
>> I'll be posting these for comments/consideration etc. in no particular
>> order (over a period of days, not all at one time).
> 
> while I highly applaud this effort this does put the interesting
> question of "who puts his signed-off-by on these patches".....

Certainly a good question IMO.  If Andrew or Linus knows whether
I need to add my Signed-off-by, I'll be glad to listen.
(That's not a general call for opinions.)

~Randy

Re: reviewing Ubuntu kernel patches

[Theodore Tso]

On Thu, Jun 15, 2006 at 10:39:49PM -0700, Randy Dunlap wrote:
> Certainly a good question IMO.  If Andrew or Linus knows whether
> I need to add my Signed-off-by, I'll be glad to listen.
> (That's not a general call for opinions.)

If you're submitting the patch, then surely you need to add your
Signed-off-by:, since you're asserting either (b) or (c):

        (b) The contribution is based upon previous work that, to the best
            of my knowledge, is covered under an appropriate open source
            license and I have the right under that license to submit that
            work with modifications, whether created in whole or in part
            by me, under the same open source license (unless I am
            permitted to submit under a different license), as indicated
            in the file; or

        (c) The contribution was provided directly to me by some other
            person who certified (a), (b) or (c) and I have not modified
            it.

We've gotten into the habit of assuming the Signed-off-by: also has
the meaning of "I vouch for it from technical point of view", but
really, that's presumably true since otherwise you wouldn't have
e-mailed it to Andrew or Linus.  The original meaning of the
Signed-off-by: is in the Developer's Certificate of Origin
statement...

						- Ted

Re: reviewing Ubuntu kernel patches

[Adrian Bunk]

On Fri, Jun 16, 2006 at 10:03:34AM -0400, Theodore Tso wrote:
> On Thu, Jun 15, 2006 at 10:39:49PM -0700, Randy Dunlap wrote:
> > Certainly a good question IMO.  If Andrew or Linus knows whether
> > I need to add my Signed-off-by, I'll be glad to listen.
> > (That's not a general call for opinions.)
> 
> If you're submitting the patch, then surely you need to add your
> Signed-off-by:, since you're asserting either (b) or (c):
> 
>         (b) The contribution is based upon previous work that, to the best
>             of my knowledge, is covered under an appropriate open source
>             license and I have the right under that license to submit that
>             work with modifications, whether created in whole or in part
>             by me, under the same open source license (unless I am
>             permitted to submit under a different license), as indicated
>             in the file; or
> 
>         (c) The contribution was provided directly to me by some other
>             person who certified (a), (b) or (c) and I have not modified
>             it.
> 
> We've gotten into the habit of assuming the Signed-off-by: also has
> the meaning of "I vouch for it from technical point of view", but
> really, that's presumably true since otherwise you wouldn't have
> e-mailed it to Andrew or Linus.  The original meaning of the
> Signed-off-by: is in the Developer's Certificate of Origin
> statement...

Re-reading this text, I'm getting some doubts whether the way Andrew 
handles this when merging patches is correct:

When Andrew merges some small fix into an existing patch in -mm
(e.g. "make a function static" or "fix a compile error"), he adds the 
Signed-off-by line of the small fix to the Signed-off-by lines of the 
patch.

But the submitter of the fix does not necessarly have checked the 
legal or technical status of the rest of the patch.

As an example, my only contribution to commit 
6b3934ef52712ece50605dfc72e55d00c580831a was making signal_cachep 
static, and I do refuse any legal or technical responsibility for 
the rest of the patch (this shouldn't imply the rest of the patch was 
bad - it's simply not me who can is responsible for it).

> 						- Ted

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: reviewing Ubuntu kernel patches

[Andrew Morton]

On Wed, 21 Jun 2006 19:38:41 +0200
Adrian Bunk <bunk@stusta.de> wrote:

> When Andrew merges some small fix into an existing patch in -mm
> (e.g. "make a function static" or "fix a compile error"), he adds the 
> Signed-off-by line of the small fix to the Signed-off-by lines of the 
> patch.
> 
> But the submitter of the fix does not necessarly have checked the 
> legal or technical status of the rest of the patch.
> 
> As an example, my only contribution to commit 
> 6b3934ef52712ece50605dfc72e55d00c580831a was making signal_cachep 
> static, and I do refuse any legal or technical responsibility for 
> the rest of the patch (this shouldn't imply the rest of the patch was 
> bad - it's simply not me who can is responsible for it).

Yes, nowadays I give the folded-in patch a mini-changelog, as in

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.17/2.6.17-mm1/broken-out/slab-kmalloc-kzalloc-comments-cleanup-and-fix.patch