[RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Khem Raj]

Hi

I would like to propose to move gcc 4.4.3 recipes to gcc 4.4.4, 
there are no additional patches that we need and 4.4.4 is superset
of bugfixes. I want to avoid adding additional recipes if we could.
4.4.3 is not so widely used in distributions unlike 4.4.2

I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
to keep using 4.4.3

Please cast your concerns

Thanks
-Khem

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Martin Jansa]

On Mon, May 10, 2010 at 04:27:20PM -0700, Khem Raj wrote:
> Hi
> 
> I would like to propose to move gcc 4.4.3 recipes to gcc 4.4.4, 
> there are no additional patches that we need and 4.4.4 is superset
> of bugfixes. I want to avoid adding additional recipes if we could.
> 4.4.3 is not so widely used in distributions unlike 4.4.2
> 
> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> to keep using 4.4.3

SHR is using 4.4.3 and I don't mind moving to 4.4.4. Thanks for adding
it.

> Please cast your concerns

Any news about 4.5.0 producing segfaulting binaries on armv[45]t?

-- 
uin:136542059                jid:Martin.Jansa@gmail.com
Jansa Martin                 sip:jamasip@voip.wengo.fr 
JaMa                         

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Khem Raj]

On (11/05/10 06:37), Martin Jansa wrote:
> On Mon, May 10, 2010 at 04:27:20PM -0700, Khem Raj wrote:
> > Hi
> > 
> > I would like to propose to move gcc 4.4.3 recipes to gcc 4.4.4, 
> > there are no additional patches that we need and 4.4.4 is superset
> > of bugfixes. I want to avoid adding additional recipes if we could.
> > 4.4.3 is not so widely used in distributions unlike 4.4.2
> > 
> > I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> > to keep using 4.4.3
> 
> SHR is using 4.4.3 and I don't mind moving to 4.4.4. Thanks for adding
> it.

OK

> 
> > Please cast your concerns
> 
> Any news about 4.5.0 producing segfaulting binaries on armv[45]t?

I have been able to reproduce it using 2.6.33 on osk5912 but no time to
debug so far :(

Thx
-Khem

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Roman I Khimov]

В сообщении от Вторник 11 мая 2010 03:27:20 автор Khem Raj написал:
> I would like to propose to move gcc 4.4.3 recipes to gcc 4.4.4,
> there are no additional patches that we need and 4.4.4 is superset
> of bugfixes. I want to avoid adding additional recipes if we could.
> 4.4.3 is not so widely used in distributions unlike 4.4.2
> 
> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> to keep using 4.4.3

Using 4.4.3 here, think it'd be nice to move to 4.4.4.

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Martin Jansa]

On Tue, May 11, 2010 at 10:39 AM, Roman I Khimov <khimov@altell.ru> wrote:
> В сообщении от Вторник 11 мая 2010 03:27:20 автор Khem Raj написал:
>> 4.4.3 is not so widely used in distributions unlike 4.4.2

Hi,

My guess: just because it's not preferred by sane-toolchain.inc?

>> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
>> to keep using 4.4.3
>
> Using 4.4.3 here, think it'd be nice to move to 4.4.4.

what about updating
conf/distro/include/sane-toolchain.inc
to 4.4.4 if we don't find any new issue with it?

I haven't found any new issue with 4.4.3 (using it since it was
added), rebuilding images with 4.4.4 now.

Cheers,

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Michael 'Mickey' Lauer]

Am Mittwoch, den 12.05.2010, 11:39 +0200 schrieb Martin Jansa:
> On Tue, May 11, 2010 at 10:39 AM, Roman I Khimov <khimov@altell.ru> wrote:
> > В сообщении от Вторник 11 мая 2010 03:27:20 автор Khem Raj написал:
> >> 4.4.3 is not so widely used in distributions unlike 4.4.2
> 
> Hi,
> 
> My guess: just because it's not preferred by sane-toolchain.inc?
> 
> >> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> >> to keep using 4.4.3
> >
> > Using 4.4.3 here, think it'd be nice to move to 4.4.4.
> 
> what about updating
> conf/distro/include/sane-toolchain.inc
> to 4.4.4 if we don't find any new issue with it?

Sounds good.

-- 
:M:

Re: [RFC] Move gcc 4.4.3 recipes in favour of 4.4.4

[Roman I Khimov]

В сообщении от Среда 12 мая 2010 13:39:02 автор Martin Jansa написал:
> what about updating
> conf/distro/include/sane-toolchain.inc
> to 4.4.4 if we don't find any new issue with it?

No objections.
 
> I haven't found any new issue with 4.4.3 (using it since it was
> added), rebuilding images with 4.4.4 now.

Just built mine, works like charm. Thanks, Khem!

Re: [RFC] Move sane-toolchain defaults for gcc binutils and eglibc

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 1214 bytes --]

On (12/05/10 11:44), Michael 'Mickey' Lauer wrote:
> Am Mittwoch, den 12.05.2010, 11:39 +0200 schrieb Martin Jansa:
> > On Tue, May 11, 2010 at 10:39 AM, Roman I Khimov <khimov@altell.ru> wrote:
> > > В сообщении от Вторник 11 мая 2010 03:27:20 автор Khem Raj написал:
> > >> 4.4.3 is not so widely used in distributions unlike 4.4.2
> > 
> > Hi,
> > 
> > My guess: just because it's not preferred by sane-toolchain.inc?
> > 
> > >> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> > >> to keep using 4.4.3
> > >
> > > Using 4.4.3 here, think it'd be nice to move to 4.4.4.
> > 
> > what about updating
> > conf/distro/include/sane-toolchain.inc
> > to 4.4.4 if we don't find any new issue with it?
> 
> Sounds good.

Here is a patch to move gcc to 4.4.4, eglibc to 2.11 and binutils to
2.20.1 releases as defaults. I have tested this combination on ARM
and x86 without any regressions that I dont have on existing combination. 

sane-toolchain.in is used by kaeilos-2010, shr, micro and minimal DISTROs
if the maintainers are ok with this then please ack this patch

Thanks
-Khem

Signed-off-by: Khem Raj <raj.khem@gmail.com>

[-- Attachment #2: diff --]
[-- Type: text/plain, Size: 1043 bytes --]

diff --git a/conf/distro/include/sane-toolchain.inc b/conf/distro/include/sane-toolchain.inc
index daa2def..7f249b0 100644
--- a/conf/distro/include/sane-toolchain.inc
+++ b/conf/distro/include/sane-toolchain.inc
@@ -5,14 +5,14 @@
 #############################################################################
 
 # Base line versions, good for most things
-PREFERRED_GCC_VERSION                ?= "4.4.2"
-PREFERRED_BINUTILS                   ?= "2.20"
+PREFERRED_GCC_VERSION                ?= "4.4.4"
+PREFERRED_BINUTILS                   ?= "2.20.1"
 
 # Prefer glibc 2.6 and uclibc 0.9.30, these have had the most testing.
 PREFERRED_VERSION_glibc             ?= "2.10.1"
 PREFERRED_VERSION_glibc-initial     ?= "2.10.1"
-PREFERRED_VERSION_eglibc             ?= "2.10"
-PREFERRED_VERSION_eglibc-initial     ?= "2.10"
+PREFERRED_VERSION_eglibc             ?= "2.11"
+PREFERRED_VERSION_eglibc-initial     ?= "2.11"
 PREFERRED_VERSION_uclibc            ?= "0.9.30.1"
 PREFERRED_VERSION_uclibc-initial    ?= "0.9.30.1"
 

Re: [RFC] Move sane-toolchain defaults for gcc binutils and eglibc

[Martin Jansa]

On Wed, May 12, 2010 at 10:49:02AM -0700, Khem Raj wrote:
> On (12/05/10 11:44), Michael 'Mickey' Lauer wrote:
> > Am Mittwoch, den 12.05.2010, 11:39 +0200 schrieb Martin Jansa:
> > > On Tue, May 11, 2010 at 10:39 AM, Roman I Khimov <khimov@altell.ru> wrote:
> > > > В сообщении от Вторник 11 мая 2010 03:27:20 автор Khem Raj написал:
> > > >> 4.4.3 is not so widely used in distributions unlike 4.4.2
> > > 
> > > Hi,
> > > 
> > > My guess: just because it's not preferred by sane-toolchain.inc?
> > > 
> > > >> I can also just add 4.4.4 and leave 4.4.3 as it is if someone wants
> > > >> to keep using 4.4.3
> > > >
> > > > Using 4.4.3 here, think it'd be nice to move to 4.4.4.
> > > 
> > > what about updating
> > > conf/distro/include/sane-toolchain.inc
> > > to 4.4.4 if we don't find any new issue with it?
> > 
> > Sounds good.
> 
> Here is a patch to move gcc to 4.4.4, eglibc to 2.11 and binutils to
> 2.20.1 releases as defaults. I have tested this combination on ARM
> and x86 without any regressions that I dont have on existing combination. 
> 
> sane-toolchain.in is used by kaeilos-2010, shr, micro and minimal DISTROs
> if the maintainers are ok with this then please ack this patch
> 
> Thanks
> -Khem
> 
> Signed-off-by: Khem Raj <raj.khem@gmail.com>

We (SHR) are using this combination already without problems, so ACK.

Acked-by: Martin Jansa <Martin.Jansa@gmail.com>

> diff --git a/conf/distro/include/sane-toolchain.inc b/conf/distro/include/sane-toolchain.inc
> index daa2def..7f249b0 100644
> --- a/conf/distro/include/sane-toolchain.inc
> +++ b/conf/distro/include/sane-toolchain.inc
> @@ -5,14 +5,14 @@
>  #############################################################################
>  
>  # Base line versions, good for most things
> -PREFERRED_GCC_VERSION                ?= "4.4.2"
> -PREFERRED_BINUTILS                   ?= "2.20"
> +PREFERRED_GCC_VERSION                ?= "4.4.4"
> +PREFERRED_BINUTILS                   ?= "2.20.1"
>  
>  # Prefer glibc 2.6 and uclibc 0.9.30, these have had the most testing.
>  PREFERRED_VERSION_glibc             ?= "2.10.1"
>  PREFERRED_VERSION_glibc-initial     ?= "2.10.1"
> -PREFERRED_VERSION_eglibc             ?= "2.10"
> -PREFERRED_VERSION_eglibc-initial     ?= "2.10"
> +PREFERRED_VERSION_eglibc             ?= "2.11"
> +PREFERRED_VERSION_eglibc-initial     ?= "2.11"
>  PREFERRED_VERSION_uclibc            ?= "0.9.30.1"
>  PREFERRED_VERSION_uclibc-initial    ?= "0.9.30.1"
>  

> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel


-- 
uin:136542059                jid:Martin.Jansa@gmail.com
Jansa Martin                 sip:jamasip@voip.wengo.fr 
JaMa