FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Redhat > Fedora Development

 
 
LinkBack Thread Tools
 
Old 06-02-2008, 09:59 AM
Hans de Goede
 
Default Kernel headers changes in F10?

Bastien Nocera wrote:

Heya,

I'm having trouble compiling the GStreamer V4L2 plugin on F10, when it
work perfectly for the exact same source in F9.

F10:
http://koji.fedoraproject.org/koji/getfile?taskID=640403&name=build.log
F9:
http://koji.fedoraproject.org/koji/getfile?taskID=640384&name=build.log

Any changes somebody knows of?



There seem to be a number of changes in the 2.6.26 kernel headers causing
compile breakage (and/or in the new glibc), I've had to fix both of:


svgalib, really fixed

gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>, worked
around by no longer including <net/if.h>



So although I don't know anything about specific changes to the v4l headers,
there are definitely problems with the kernel / glibc headers in rawhide all
over the place.


Regards,

Hans

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 10:05 AM
Bastien Nocera
 
Default Kernel headers changes in F10?

Heya,

I'm having trouble compiling the GStreamer V4L2 plugin on F10, when it
work perfectly for the exact same source in F9.

F10:
http://koji.fedoraproject.org/koji/getfile?taskID=640403&name=build.log
F9:
http://koji.fedoraproject.org/koji/getfile?taskID=640384&name=build.log

Any changes somebody knows of?

Cheers

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 10:31 AM
David Woodhouse
 
Default Kernel headers changes in F10?

On Mon, 2008-06-02 at 11:05 +0100, Bastien Nocera wrote:
> Heya,
>
> I'm having trouble compiling the GStreamer V4L2 plugin on F10, when it
> work perfectly for the exact same source in F9.
>
> F10:
> http://koji.fedoraproject.org/koji/getfile?taskID=640403&name=build.log
> F9:
> http://koji.fedoraproject.org/koji/getfile?taskID=640384&name=build.log
>
> Any changes somebody knows of?

Rather than a link to a huge text file, it might have been useful to
include the failure mode, which is this:

v4l2_calls.c:268: error: 'V4L2_CID_HCENTER' undeclared (first use in this function)
v4l2_calls.c:268: error: (Each undeclared identifier is reported only once
v4l2_calls.c:268: error: for each function it appears in.)
v4l2_calls.c:269: error: 'V4L2_CID_VCENTER' undeclared (first use in this function)

A few moments with git-annotate showed that the missing ioctls were
removed in commit 26d507fcfef7f7d0cd2eec874a87169cc121c835 by Brandon
Philips. In their place, we have the following:

/* Deprecated, use V4L2_CID_PAN_RESET and V4L2_CID_TILT_RESET */
#define V4L2_CID_HCENTER_DEPRECATED (V4L2_CID_BASE+22)
#define V4L2_CID_VCENTER_DEPRECATED (V4L2_CID_BASE+23)

That seems like a rather dubious change to the user API -- shouldn't we
ensure that existing software continues to build, but maybe add a
compile-time or run-time warning for those using the deprecated ioctls?

Mauro, Brendan: unless the removal of those ioctls has been properly
documented in Documentation/feature-removal-schedule.txt for an
appropriate amount of time, please could you put them back as they were.
Changing the userspace API like this isn't acceptable.

--
dwmw2

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=26d507fc

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 10:33 AM
David Woodhouse
 
Default Kernel headers changes in F10?

On Mon, 2008-06-02 at 11:59 +0200, Hans de Goede wrote:
> There seem to be a number of changes in the 2.6.26 kernel headers causing
> compile breakage (and/or in the new glibc), I've had to fix both of:
>
> svgalib, really fixed
>
> gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>, worked
> around by no longer including <net/if.h>

I always like solutions which involve "include fewer kernel headers",
but we should probably investigate that in case there is some need for
someone to include both. John?

> So although I don't know anything about specific changes to the v4l headers,
> there are definitely problems with the kernel / glibc headers in rawhide all
> over the place.

Details or (preferably) bug numbers, please.

--
dwmw2

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 03:20 PM
David Woodhouse
 
Default Kernel headers changes in F10?

On Mon, 2008-06-02 at 11:55 -0300, Mauro Carvalho Chehab wrote:
> Hi David,
>
> On Mon, 02 Jun 2008 11:31:56 +0100
> David Woodhouse <dwmw2@infradead.org> wrote:
>
> > A few moments with git-annotate showed that the missing ioctls were
> > removed in commit 26d507fcfef7f7d0cd2eec874a87169cc121c835 by Brandon
> > Philips. In their place, we have the following:
> >
> > /* Deprecated, use V4L2_CID_PAN_RESET and V4L2_CID_TILT_RESET */
> > #define V4L2_CID_HCENTER_DEPRECATED (V4L2_CID_BASE+22)
> > #define V4L2_CID_VCENTER_DEPRECATED (V4L2_CID_BASE+23)
> >
> > That seems like a rather dubious change to the user API -- shouldn't we
> > ensure that existing software continues to build, but maybe add a
> > compile-time or run-time warning for those using the deprecated ioctls?
> >
> > Mauro, Brendan: unless the removal of those ioctls has been properly
> > documented in Documentation/feature-removal-schedule.txt for an
> > appropriate amount of time, please could you put them back as they were.
> > Changing the userspace API like this isn't acceptable.
>
> IMO, there's nothing to be added at feature-removal-schedule.txt. Those controls
> aren't used by any kernel drivers on a long time (I suspect that they were
> never used).
>
> The issue is that they are targeted to do exactly what V4L2_CID_PAN_RESET and
> V4L2_CID_TILT_RESET do. So, they are duplicated stuff.
>
> I was not aware that an userspace is using those symbols, but, if so, it is
> likely to be related to some binary-only or out-of-tree drivers [1].

It was part of the published API, and it was being used by the gstreamer
v4l2 plugin -- which now fails to build.

Yes, of course we'll fix the code in question, but just removing APIs
from the headers and causing source code to break without any prior
warning is not acceptable. That's precisely what
feature-removal-schedule.txt is _for_.

--
dwmw2

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 05:09 PM
"John W. Linville"
 
Default Kernel headers changes in F10?

On Mon, Jun 02, 2008 at 11:33:45AM +0100, David Woodhouse wrote:
> On Mon, 2008-06-02 at 11:59 +0200, Hans de Goede wrote:
> > There seem to be a number of changes in the 2.6.26 kernel headers causing
> > compile breakage (and/or in the new glibc), I've had to fix both of:
> >
> > svgalib, really fixed
> >
> > gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>, worked
> > around by no longer including <net/if.h>
>
> I always like solutions which involve "include fewer kernel headers",
> but we should probably investigate that in case there is some need for
> someone to include both. John?

Almost certainly because of this commit:

commit 2218228392080f0ca2fc2974604e79f57b12c436
Author: Kirill A. Shutemov <kirill@shutemov.name>
Date: Tue Apr 22 16:38:55 2008 +0300

Make linux/wireless.h be able to compile

Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

I may have let this slip by due to the "able to compile" bit --
should I not have merged it? I don't have a record or recollection
of what motivated the patch originally.

John
--
John W. Linville
linville@redhat.com

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 05:13 PM
Hans de Goede
 
Default Kernel headers changes in F10?

John W. Linville wrote:

On Mon, Jun 02, 2008 at 11:33:45AM +0100, David Woodhouse wrote:

On Mon, 2008-06-02 at 11:59 +0200, Hans de Goede wrote:
There seem to be a number of changes in the 2.6.26 kernel headers causing
compile breakage (and/or in the new glibc), I've had to fix both of:


svgalib, really fixed

gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>, worked
around by no longer including <net/if.h>

I always like solutions which involve "include fewer kernel headers",
but we should probably investigate that in case there is some need for
someone to include both. John?


Almost certainly because of this commit:

commit 2218228392080f0ca2fc2974604e79f57b12c436
Author: Kirill A. Shutemov <kirill@shutemov.name>
Date: Tue Apr 22 16:38:55 2008 +0300

Make linux/wireless.h be able to compile

Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

I may have let this slip by due to the "able to compile" bit --
should I not have merged it? I don't have a record or recollection
of what motivated the patch originally.



Erm,

I'm no git guru, can you tranlate this:
commit 2218228392080f0ca2fc2974604e79f57b12c436

Into an url showing the diff for me, then I can take try to take a guess at
what this is trying to fix, and how this might be done without conflicting with
net/if.h


Thanks,

Hans

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 05:28 PM
"John W. Linville"
 
Default Kernel headers changes in F10?

On Mon, Jun 02, 2008 at 07:13:27PM +0200, Hans de Goede wrote:
> John W. Linville wrote:
> >On Mon, Jun 02, 2008 at 11:33:45AM +0100, David Woodhouse wrote:
> >>On Mon, 2008-06-02 at 11:59 +0200, Hans de Goede wrote:
> >>>There seem to be a number of changes in the 2.6.26 kernel headers
> >>>causing compile breakage (and/or in the new glibc), I've had to fix both
> >>>of:
> >>>
> >>>svgalib, really fixed
> >>>
> >>>gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>,
> >>>worked around by no longer including <net/if.h>
> >>I always like solutions which involve "include fewer kernel headers",
> >>but we should probably investigate that in case there is some need for
> >>someone to include both. John?
> >
> >Almost certainly because of this commit:
> >
> >commit 2218228392080f0ca2fc2974604e79f57b12c436
> >Author: Kirill A. Shutemov <kirill@shutemov.name>
> >Date: Tue Apr 22 16:38:55 2008 +0300
> >
> > Make linux/wireless.h be able to compile
> >
> > Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
> > Signed-off-by: John W. Linville <linville@tuxdriver.com>
> >
> >I may have let this slip by due to the "able to compile" bit --
> >should I not have merged it? I don't have a record or recollection
> >of what motivated the patch originally.
> >
>
> Erm,
>
> I'm no git guru, can you tranlate this:
> commit 2218228392080f0ca2fc2974604e79f57b12c436
>
> Into an url showing the diff for me, then I can take try to take a guess at
> what this is trying to fix, and how this might be done without conflicting
> with net/if.h

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=2218228392080f0ca2fc2974604 e79f57b12c436

As I said, I have no record or memory of why this patch was needed.
It looks like it was a mistake for me to let it though in the first
place. My guess is that he wanted to include linux/wireless.h from
userland without including other kernel headers...?

I am happy to entertain arguments in favor of reverting (or keeping)
this patch.

John
--
John W. Linville
linville@redhat.com

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-02-2008, 05:45 PM
Hans de Goede
 
Default Kernel headers changes in F10?

John W. Linville wrote:

On Mon, Jun 02, 2008 at 07:13:27PM +0200, Hans de Goede wrote:

John W. Linville wrote:

On Mon, Jun 02, 2008 at 11:33:45AM +0100, David Woodhouse wrote:

On Mon, 2008-06-02 at 11:59 +0200, Hans de Goede wrote:
There seem to be a number of changes in the 2.6.26 kernel headers
causing compile breakage (and/or in the new glibc), I've had to fix both
of:


svgalib, really fixed

gkrellm-wifi, conflict between <net/if.h> and <linux/wireless.h>,
worked around by no longer including <net/if.h>

I always like solutions which involve "include fewer kernel headers",
but we should probably investigate that in case there is some need for
someone to include both. John?

Almost certainly because of this commit:

commit 2218228392080f0ca2fc2974604e79f57b12c436
Author: Kirill A. Shutemov <kirill@shutemov.name>
Date: Tue Apr 22 16:38:55 2008 +0300

Make linux/wireless.h be able to compile

Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

I may have let this slip by due to the "able to compile" bit --
should I not have merged it? I don't have a record or recollection
of what motivated the patch originally.


Erm,

I'm no git guru, can you tranlate this:
commit 2218228392080f0ca2fc2974604e79f57b12c436

Into an url showing the diff for me, then I can take try to take a guess at
what this is trying to fix, and how this might be done without conflicting
with net/if.h


http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=2218228392080f0ca2fc2974604 e79f57b12c436

As I said, I have no record or memory of why this patch was needed.
It looks like it was a mistake for me to let it though in the first
place. My guess is that he wanted to include linux/wireless.h from
userland without including other kernel headers...?



Looks like a good candidate for reverting. I see little arguments to keep this
patch in, it will probably break compilation of other users of linux/wireless.h
too, as those probably also already include <net/if.h> to get the necessary
stuff from there.


Regards,

Hans


--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 
Old 06-03-2008, 01:47 PM
"John W. Linville"
 
Default Kernel headers changes in F10?

On Mon, Jun 02, 2008 at 07:45:07PM +0200, Hans de Goede wrote:
> John W. Linville wrote:
> >On Mon, Jun 02, 2008 at 07:13:27PM +0200, Hans de Goede wrote:
> >>John W. Linville wrote:

> >>>Almost certainly because of this commit:
> >>>
> >>>commit 2218228392080f0ca2fc2974604e79f57b12c436
> >>>Author: Kirill A. Shutemov <kirill@shutemov.name>
> >>>Date: Tue Apr 22 16:38:55 2008 +0300
> >>>
> >>> Make linux/wireless.h be able to compile
> >>>
> >>> Signed-off-by: Kirill A. Shutemov <kirill@shutemov.name>
> >>> Signed-off-by: John W. Linville <linville@tuxdriver.com>
> >>>
> >>>I may have let this slip by due to the "able to compile" bit --
> >>>should I not have merged it? I don't have a record or recollection
> >>>of what motivated the patch originally.

> >As I said, I have no record or memory of why this patch was needed.
> >It looks like it was a mistake for me to let it though in the first
> >place. My guess is that he wanted to include linux/wireless.h from
> >userland without including other kernel headers...?
>
> Looks like a good candidate for reverting. I see little arguments to keep
> this patch in, it will probably break compilation of other users of
> linux/wireless.h too, as those probably also already include <net/if.h> to
> get the necessary stuff from there.

I have reverted that patch in rawhide, and I'm working with Kirill
for a more permanent solution.

Thanks,

John
--
John W. Linville
linville@redhat.com

--
fedora-devel-list mailing list
fedora-devel-list@redhat.com
https://www.redhat.com/mailman/listinfo/fedora-devel-list
 

Thread Tools




All times are GMT. The time now is 10:07 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org