Can we get warnings from the x86 userspace headers fixed?

Can we get warnings from the x86 userspace headers fixed?

[Sam Ravnborg]

Hi Ingo.

We have recently introduced a lot of new warnings for our
userspace headers.

Try to do a "make headers_check" and enjoy...

Are there any chance you could ask one of your helpers to
start to look into the x86 specific ones?

If we get x86 clean this would create more incentive to
make the generic clean too.

I had originally planned to attack the generic headers
(include/linux/*) but I'm faced by reality after my
vacation and has almost no spare time for the time being.

I already fixed the sparc headers but that is not 
visible compared to x86 (the fixes was btw easy).

I can try to help by doing a few reviews in the
beginning but I need to fix kbuild stuff first where
I have a few serious issues pending. And lacks time...

	Sam

Re: Can we get warnings from the x86 userspace headers fixed?

[Harvey Harrison]

On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
> Hi Ingo.
> 
> We have recently introduced a lot of new warnings for our
> userspace headers.
> 
> Try to do a "make headers_check" and enjoy...
> 
> Are there any chance you could ask one of your helpers to
> start to look into the x86 specific ones?
> 
> If we get x86 clean this would create more incentive to
> make the generic clean too.
> 
> I had originally planned to attack the generic headers
> (include/linux/*) but I'm faced by reality after my
> vacation and has almost no spare time for the time being.
> 
> I already fixed the sparc headers but that is not 
> visible compared to x86 (the fixes was btw easy).
> 
> I can try to help by doing a few reviews in the
> beginning but I need to fix kbuild stuff first where
> I have a few serious issues pending. And lacks time...
> 

I'll have a go at some of them this afternoon unless someone beats me to it.

Harvey

Re: Can we get warnings from the x86 userspace headers fixed?

[Sam Ravnborg]

On Mon, Jan 12, 2009 at 01:31:03PM -0800, Harvey Harrison wrote:
> On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
> > Hi Ingo.
> > 
> > We have recently introduced a lot of new warnings for our
> > userspace headers.
> > 
> > Try to do a "make headers_check" and enjoy...
> > 
> > Are there any chance you could ask one of your helpers to
> > start to look into the x86 specific ones?
> > 
> > If we get x86 clean this would create more incentive to
> > make the generic clean too.
> > 
> > I had originally planned to attack the generic headers
> > (include/linux/*) but I'm faced by reality after my
> > vacation and has almost no spare time for the time being.
> > 
> > I already fixed the sparc headers but that is not 
> > visible compared to x86 (the fixes was btw easy).
> > 
> > I can try to help by doing a few reviews in the
> > beginning but I need to fix kbuild stuff first where
> > I have a few serious issues pending. And lacks time...
> > 
> 
> I'll have a go at some of them this afternoon unless someone beats me to it.

Thanks Harvey!

	Sam

Re: Can we get warnings from the x86 userspace headers fixed?

[Ingo Molnar]

* Harvey Harrison <harvey.harrison@gmail.com> wrote:

> On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
> > Hi Ingo.
> > 
> > We have recently introduced a lot of new warnings for our
> > userspace headers.
> > 
> > Try to do a "make headers_check" and enjoy...
> > 
> > Are there any chance you could ask one of your helpers to
> > start to look into the x86 specific ones?
> > 
> > If we get x86 clean this would create more incentive to
> > make the generic clean too.
> > 
> > I had originally planned to attack the generic headers
> > (include/linux/*) but I'm faced by reality after my
> > vacation and has almost no spare time for the time being.
> > 
> > I already fixed the sparc headers but that is not 
> > visible compared to x86 (the fixes was btw easy).
> > 
> > I can try to help by doing a few reviews in the
> > beginning but I need to fix kbuild stuff first where
> > I have a few serious issues pending. And lacks time...
> > 
> 
> I'll have a go at some of them this afternoon unless someone beats me to 
> it.

thanks Harvey! Please work against tip/master if you do that, there's a 
few header file changes in the queue already.

	Ingo

Re: Can we get warnings from the x86 userspace headers fixed?

[Cyrill Gorcunov]

[Ingo Molnar - Mon, Jan 12, 2009 at 10:37:19PM +0100]
| 
| * Harvey Harrison <harvey.harrison@gmail.com> wrote:
| 
| > On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
| > > Hi Ingo.
| > > 
| > > We have recently introduced a lot of new warnings for our
| > > userspace headers.
| > > 
| > > Try to do a "make headers_check" and enjoy...
| > > 
| > > Are there any chance you could ask one of your helpers to
| > > start to look into the x86 specific ones?
| > > 
| > > If we get x86 clean this would create more incentive to
| > > make the generic clean too.
| > > 
| > > I had originally planned to attack the generic headers
| > > (include/linux/*) but I'm faced by reality after my
| > > vacation and has almost no spare time for the time being.
| > > 
| > > I already fixed the sparc headers but that is not 
| > > visible compared to x86 (the fixes was btw easy).
| > > 
| > > I can try to help by doing a few reviews in the
| > > beginning but I need to fix kbuild stuff first where
| > > I have a few serious issues pending. And lacks time...
| > > 
| > 
| > I'll have a go at some of them this afternoon unless someone beats me to 
| > it.
| 
| thanks Harvey! Please work against tip/master if you do that, there's a 
| few header file changes in the queue already.
| 
| 	Ingo
| 

hmm... what I've got:

---
cyrill@lenovo linux-2.6.git $ make headers_check
  CHK     include/linux/version.h
make[1]: `scripts/unifdef' is up to date.
cyrill@lenovo linux-2.6.git $ 
---

I don't know what to think...

		- Cyrill -

Re: Can we get warnings from the x86 userspace headers fixed?

[Cyrill Gorcunov]

[Cyrill Gorcunov - Tue, Jan 13, 2009 at 12:44:20AM +0300]
| [Ingo Molnar - Mon, Jan 12, 2009 at 10:37:19PM +0100]
| | 
| | * Harvey Harrison <harvey.harrison@gmail.com> wrote:
| | 
| | > On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
| | > > Hi Ingo.
| | > > 
| | > > We have recently introduced a lot of new warnings for our
| | > > userspace headers.
| | > > 
| | > > Try to do a "make headers_check" and enjoy...
| | > > 
| | > > Are there any chance you could ask one of your helpers to
| | > > start to look into the x86 specific ones?
| | > > 
| | > > If we get x86 clean this would create more incentive to
| | > > make the generic clean too.
| | > > 
| | > > I had originally planned to attack the generic headers
| | > > (include/linux/*) but I'm faced by reality after my
| | > > vacation and has almost no spare time for the time being.
| | > > 
| | > > I already fixed the sparc headers but that is not 
| | > > visible compared to x86 (the fixes was btw easy).
| | > > 
| | > > I can try to help by doing a few reviews in the
| | > > beginning but I need to fix kbuild stuff first where
| | > > I have a few serious issues pending. And lacks time...
| | > > 
| | > 
| | > I'll have a go at some of them this afternoon unless someone beats me to 
| | > it.
| | 
| | thanks Harvey! Please work against tip/master if you do that, there's a 
| | few header file changes in the queue already.
| | 
| | 	Ingo
| | 
| 
| hmm... what I've got:
| 
| ---
| cyrill@lenovo linux-2.6.git $ make headers_check
|   CHK     include/linux/version.h
| make[1]: `scripts/unifdef' is up to date.
| cyrill@lenovo linux-2.6.git $ 
| ---
| 
| I don't know what to think...
| 
| 		- Cyrill -

Just checked out last -tip. Ugh, damn, so many warnings... :(

		- Cyrill -

Re: Can we get warnings from the x86 userspace headers fixed?

[Jaswinder Singh Rajput]

On Tue, Jan 13, 2009 at 3:07 AM, Ingo Molnar <mingo@elte.hu> wrote:
>
> * Harvey Harrison <harvey.harrison@gmail.com> wrote:
>
>> On Mon, 2009-01-12 at 22:28 +0100, Sam Ravnborg wrote:
>> > Hi Ingo.
>> >
>> > We have recently introduced a lot of new warnings for our
>> > userspace headers.
>> >
>> > Try to do a "make headers_check" and enjoy...
>> >
>> > Are there any chance you could ask one of your helpers to
>> > start to look into the x86 specific ones?
>> >
>> > If we get x86 clean this would create more incentive to
>> > make the generic clean too.
>> >
>> > I had originally planned to attack the generic headers
>> > (include/linux/*) but I'm faced by reality after my
>> > vacation and has almost no spare time for the time being.
>> >
>> > I already fixed the sparc headers but that is not
>> > visible compared to x86 (the fixes was btw easy).
>> >
>> > I can try to help by doing a few reviews in the
>> > beginning but I need to fix kbuild stuff first where
>> > I have a few serious issues pending. And lacks time...
>> >
>>
>> I'll have a go at some of them this afternoon unless someone beats me to
>> it.
>
> thanks Harvey! Please work against tip/master if you do that, there's a
> few header file changes in the queue already.
>

Some patches are also available in:

http://git.kernel.org/?p=linux/kernel/git/jaswinder/linux-2.6-tiptop.git;a=summary

Thanks
--
JSR

Re: Can we get warnings from the x86 userspace headers fixed?

[Cyrill Gorcunov]

[Sam Ravnborg - Mon, Jan 12, 2009 at 10:28:55PM +0100]
| Hi Ingo.
| 
| We have recently introduced a lot of new warnings for our
| userspace headers.
| 
| Try to do a "make headers_check" and enjoy...
| 
| Are there any chance you could ask one of your helpers to
| start to look into the x86 specific ones?
| 
| If we get x86 clean this would create more incentive to
| make the generic clean too.
| 
| I had originally planned to attack the generic headers
| (include/linux/*) but I'm faced by reality after my
| vacation and has almost no spare time for the time being.
| 
| I already fixed the sparc headers but that is not 
| visible compared to x86 (the fixes was btw easy).
| 
| I can try to help by doing a few reviews in the
| beginning but I need to fix kbuild stuff first where
| I have a few serious issues pending. And lacks time...
| 
| 	Sam
| 

Sam, I hope to find some spare time for this
tomorrow (it's night here). At least in one
moment headers_check assume that CONFIG is exported
into user space even being bound by pure C comments.

---
/home/cyrill/projects/kernel/linux-2.6.git/usr/include/asm/e820.h:13: leaks
CONFIG_NODES to userspace where it is not valid
---

Just a note.

		- Cyrill -