[PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Jonathan Cameron]

Fix compile bug for arm with recent kernel headers
../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No such file or directory

Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
---
Update to previous posting under 
[PATCH] glibc 2.9: Remove unnecessary asm/page.h include, not present in recent headers

This patch is already included by the 2.6.1 recipe (thanks Khem)
Prevents the following error with recent kernel headers
../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No such file or directory

Fixes problem with 2.6.29 headers against both glibc 2.7 and 2.9

Related bug reports (with equivalent solutions)
http://www.mail-archive.com/debian-glibc@lists.debian.org/msg38393.html
http://sourceware.org/ml/crossgcc/2008-05/msg00014.html
http://sourceware.org/ml/libc-ports/2008-04/msg00005.html

Bug no longer exists at head of glibc ports tree. May effect other versions.

 recipes/glibc/glibc_2.7.bb |    3 ++-
 recipes/glibc/glibc_2.9.bb |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/recipes/glibc/glibc_2.7.bb b/recipes/glibc/glibc_2.7.bb
index f152756..dcea69c 100644
--- a/recipes/glibc/glibc_2.7.bb
+++ b/recipes/glibc/glibc_2.7.bb
@@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
 PACKAGES_DYNAMIC = "libc6*"
 RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
 
-PR = "${INC_PR}.0"
+PR = "${INC_PR}.1"
 
 # the -isystem in bitbake.conf screws up glibc do_stage
 BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
@@ -68,6 +68,7 @@ SRC_URI = "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
            file://etc/ld.so.conf \
            file://generate-supported.mk \
            file://march-i686.patch;patch=1;pnum=0 \
+	   file://glibc-arm-no-asm-page.patch;patch=1 \
            "
 
 
diff --git a/recipes/glibc/glibc_2.9.bb b/recipes/glibc/glibc_2.9.bb
index 6e323aa..9849c0a 100644
--- a/recipes/glibc/glibc_2.9.bb
+++ b/recipes/glibc/glibc_2.9.bb
@@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
 PACKAGES_DYNAMIC = "libc6*"
 RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
 
-PR = "${INC_PR}.0"
+PR = "${INC_PR}.1"
 
 # the -isystem in bitbake.conf screws up glibc do_stage
 BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
@@ -65,6 +65,7 @@ SRC_URI = "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
            file://generate-supported.mk \
            file://march-i686.patch;patch=1;pnum=0 \
 	   file://tls_i486.patch;patch=1 \
+	   file://glibc-arm-no-asm-page.patch;patch=1 \
            "
 
 
-- 
1.6.3.3

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Holger Hans Peter Freyther]

On Monday 17 August 2009 15:53:00 you wrote:
> Fix compile bug for arm with recent kernel headers
> ../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No
> such file or directory
>
> Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>

Signed-off-by: Holger Hans Peter Freyther <zecke@selfish.org>

One note worthy comment. Normally we only bump the PR when the packaged 
content is changing. In one way we had a build error so the packaging didn't 
change at all, in another way we do have existing glibc packages...

To be honest I wonder why you are triggering this area, and why other builds 
don't do that... it is  a bit frightening.



> ---
> Update to previous posting under
> [PATCH] glibc 2.9: Remove unnecessary asm/page.h include, not present in
> recent headers
>
> This patch is already included by the 2.6.1 recipe (thanks Khem)
> Prevents the following error with recent kernel headers
> ../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No
> such file or directory
>
> Fixes problem with 2.6.29 headers against both glibc 2.7 and 2.9
>
> Related bug reports (with equivalent solutions)
> http://www.mail-archive.com/debian-glibc@lists.debian.org/msg38393.html
> http://sourceware.org/ml/crossgcc/2008-05/msg00014.html
> http://sourceware.org/ml/libc-ports/2008-04/msg00005.html
>
> Bug no longer exists at head of glibc ports tree. May effect other
> versions.
>
>  recipes/glibc/glibc_2.7.bb |    3 ++-
>  recipes/glibc/glibc_2.9.bb |    3 ++-
>  2 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/recipes/glibc/glibc_2.7.bb b/recipes/glibc/glibc_2.7.bb
> index f152756..dcea69c 100644
> --- a/recipes/glibc/glibc_2.7.bb
> +++ b/recipes/glibc/glibc_2.7.bb
> @@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
>  PACKAGES_DYNAMIC = "libc6*"
>  RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
>
> -PR = "${INC_PR}.0"
> +PR = "${INC_PR}.1"
>
>  # the -isystem in bitbake.conf screws up glibc do_stage
>  BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
> @@ -68,6 +68,7 @@ SRC_URI =
> "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
> file://etc/ld.so.conf \
>             file://generate-supported.mk \
>             file://march-i686.patch;patch=1;pnum=0 \
> +	   file://glibc-arm-no-asm-page.patch;patch=1 \
>             "
>
>
> diff --git a/recipes/glibc/glibc_2.9.bb b/recipes/glibc/glibc_2.9.bb
> index 6e323aa..9849c0a 100644
> --- a/recipes/glibc/glibc_2.9.bb
> +++ b/recipes/glibc/glibc_2.9.bb
> @@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
>  PACKAGES_DYNAMIC = "libc6*"
>  RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
>
> -PR = "${INC_PR}.0"
> +PR = "${INC_PR}.1"
>
>  # the -isystem in bitbake.conf screws up glibc do_stage
>  BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
> @@ -65,6 +65,7 @@ SRC_URI =
> "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
> file://generate-supported.mk \
>             file://march-i686.patch;patch=1;pnum=0 \
>  	   file://tls_i486.patch;patch=1 \
> +	   file://glibc-arm-no-asm-page.patch;patch=1 \
>             "

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Phil Blundell]

On Mon, 2009-08-17 at 16:06 +0200, Holger Hans Peter Freyther wrote:
> To be honest I wonder why you are triggering this area, and why other builds 
> don't do that... it is  a bit frightening.

I expect it's nothing more sinister than different versions of
linux-libc-headers.  <asm/page.h> was there in 2.6.23 but, it seems,
gone by 2.6.29.  So, probably no cause for alarm.

p.

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Phil Blundell]

On Mon, 2009-08-17 at 16:01 +0100, Phil Blundell wrote:
> On Mon, 2009-08-17 at 16:06 +0200, Holger Hans Peter Freyther wrote:
> > To be honest I wonder why you are triggering this area, and why other builds 
> > don't do that... it is  a bit frightening.
> 
> I expect it's nothing more sinister than different versions of
> linux-libc-headers.  <asm/page.h> was there in 2.6.23 but, it seems,
> gone by 2.6.29.  So, probably no cause for alarm.

Oh, and, as for bumping PR: in this case I would suggest not doing it.
The output binaries should be identical with and without that header,
and for folks doing online upgrades it is a bit of a pain to have glibc
rebuild unnecessarily.

p.

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Otavio Salvador]

Hello Phil

On Mon, Aug 17, 2009 at 12:02 PM, Phil Blundell<pb@reciva.com> wrote:
> Oh, and, as for bumping PR: in this case I would suggest not doing it.
> The output binaries should be identical with and without that header,
> and for folks doing online upgrades it is a bit of a pain to have glibc
> rebuild unnecessarily.

I disagree and think it ought to be bumped so it gets deployed and
tested as soon as possible.

People that cares about online updates are supposed to be using stable
to reduce it.

-- 
Otavio Salvador                  O.S. Systems
E-mail: otavio@ossystems.com.br  http://www.ossystems.com.br
Mobile: +55 53 9981-7854         http://projetos.ossystems.com.br

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Jonathan Cameron]

Holger Hans Peter Freyther wrote:
> On Monday 17 August 2009 15:53:00 you wrote:
>> Fix compile bug for arm with recent kernel headers
>> ../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No
>> such file or directory
>>
>> Signed-off-by: Jonathan Cameron <jic23@cam.ac.uk>
> 
> Signed-off-by: Holger Hans Peter Freyther <zecke@selfish.org>
> 
> One note worthy comment. Normally we only bump the PR when the packaged 
> content is changing. In one way we had a build error so the packaging didn't 
> change at all, in another way we do have existing glibc packages...
Cool, hadn't realized that. 
> 
> To be honest I wonder why you are triggering this area, and why other builds 
> don't do that... it is  a bit frightening.
My guess is the majority of distros haven't bumped up either one of the kernel
header or glibc. Either that or I'm just unlucky in picking combinations!

Quick browse of those in the repository:  (headers will cause trouble form 2.6.25
onwards, I think).

angstrom-2008.1.conf headers = 2.6.23
chinook-compat.conf headers = 2.6.20
jlime2-2009.1 headers headers = 2.6.18
jlime-donkey.conf  glibc 2.6.1 (the fixed version) headers 2.6.23 (so don't need it)
jlime-henchman.conf headers = 2.6.23
jlime-mongo.conf headers = 2.6.18
kaeilos.conf glibc = 2.6.1 (fixed) headers =2.6.23 (fine anyway)
mamona.conf headers 2.6.22
nylon.conf headers 2.6.18
openprotium headers 2.6.11.1
oplinux.conf headers 2.6.18

Normally people don't care about kernel header versions, but I need some netlink stuff
that only just went in.

Thanks,

Jonathan







> 
> 
> 
>> ---
>> Update to previous posting under
>> [PATCH] glibc 2.9: Remove unnecessary asm/page.h include, not present in
>> recent headers
>>
>> This patch is already included by the 2.6.1 recipe (thanks Khem)
>> Prevents the following error with recent kernel headers
>> ../ports/sysdeps/unix/sysv/linux/arm/ioperm.c:48:22: error: asm/page.h: No
>> such file or directory
>>
>> Fixes problem with 2.6.29 headers against both glibc 2.7 and 2.9
>>
>> Related bug reports (with equivalent solutions)
>> http://www.mail-archive.com/debian-glibc@lists.debian.org/msg38393.html
>> http://sourceware.org/ml/crossgcc/2008-05/msg00014.html
>> http://sourceware.org/ml/libc-ports/2008-04/msg00005.html
>>
>> Bug no longer exists at head of glibc ports tree. May effect other
>> versions.
>>
>>  recipes/glibc/glibc_2.7.bb |    3 ++-
>>  recipes/glibc/glibc_2.9.bb |    3 ++-
>>  2 files changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/recipes/glibc/glibc_2.7.bb b/recipes/glibc/glibc_2.7.bb
>> index f152756..dcea69c 100644
>> --- a/recipes/glibc/glibc_2.7.bb
>> +++ b/recipes/glibc/glibc_2.7.bb
>> @@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
>>  PACKAGES_DYNAMIC = "libc6*"
>>  RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
>>
>> -PR = "${INC_PR}.0"
>> +PR = "${INC_PR}.1"
>>
>>  # the -isystem in bitbake.conf screws up glibc do_stage
>>  BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
>> @@ -68,6 +68,7 @@ SRC_URI =
>> "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
>> file://etc/ld.so.conf \
>>             file://generate-supported.mk \
>>             file://march-i686.patch;patch=1;pnum=0 \
>> +	   file://glibc-arm-no-asm-page.patch;patch=1 \
>>             "
>>
>>
>> diff --git a/recipes/glibc/glibc_2.9.bb b/recipes/glibc/glibc_2.9.bb
>> index 6e323aa..9849c0a 100644
>> --- a/recipes/glibc/glibc_2.9.bb
>> +++ b/recipes/glibc/glibc_2.9.bb
>> @@ -5,7 +5,7 @@ ARM_INSTRUCTION_SET = "arm"
>>  PACKAGES_DYNAMIC = "libc6*"
>>  RPROVIDES_${PN}-dev = "libc6-dev virtual-libc-dev"
>>
>> -PR = "${INC_PR}.0"
>> +PR = "${INC_PR}.1"
>>
>>  # the -isystem in bitbake.conf screws up glibc do_stage
>>  BUILD_CPPFLAGS = "-I${STAGING_INCDIR_NATIVE}"
>> @@ -65,6 +65,7 @@ SRC_URI =
>> "ftp://ftp.gnu.org/pub/gnu/glibc/glibc-${PV}.tar.bz2 \
>> file://generate-supported.mk \
>>             file://march-i686.patch;patch=1;pnum=0 \
>>  	   file://tls_i486.patch;patch=1 \
>> +	   file://glibc-arm-no-asm-page.patch;patch=1 \
>>             "
> 
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
> 

Re: [PATCH] glibc 2.7 and 2.9 (arm): Remove unnecessary asm/page.h include

[Jonathan Cameron]

Otavio Salvador wrote:
> Hello Phil
> 
> On Mon, Aug 17, 2009 at 12:02 PM, Phil Blundell<pb@reciva.com> wrote:
>> Oh, and, as for bumping PR: in this case I would suggest not doing it.
>> The output binaries should be identical with and without that header,
>> and for folks doing online upgrades it is a bit of a pain to have glibc
>> rebuild unnecessarily.
> 
> I disagree and think it ought to be bumped so it gets deployed and
> tested as soon as possible.
> 
> People that cares about online updates are supposed to be using stable
> to reduce it.
> 
Happy to go with either option, but keen to get this merged soon.

Can we come to a consensus on whether to bump the PR?

If we do want to, can someone push this to the repository.

Thanks,

Jonathan