Re: [Buildroot] [PATCH 1/1] Enable quadmath on aarch64

Re: [Buildroot] [PATCH 1/1] Enable quadmath on aarch64

[Thomas Petazzoni]

On Sat, 21 Aug 2021 21:44:01 +0200
Bensuperpc <bensuperpc@gmail.com> wrote:

> Enable quadmath on aarch64
> 
> Signed-off-by: Bensuperpc <bensuperpc@gmail.com>

We do need a real name to accept contributions.

> ---
>  toolchain/Config.in | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/toolchain/Config.in b/toolchain/Config.in
> index 1d0469a13e..b4cd6d511f 100644
> --- a/toolchain/Config.in
> +++ b/toolchain/Config.in
> @@ -794,6 +794,6 @@ config BR2_TOOLCHAIN_HAS_ATOMIC
>  # - When available, libquadmath requires wchar support.
>  config BR2_TOOLCHAIN_HAS_LIBQUADMATH
>  	bool
> -	default y if BR2_i386 || BR2_x86_64
> +	default y if BR2_i386 || BR2_x86_64 || BR2_aarch64

This option is not about "enabling" something, but about saying whether
libquadmath is supported/available on a given CPU architecture. Since
when is libquadmath available/supported by gcc?

What was the motivation behind this change?

Thanks!

Thomas
-- 
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot@busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Re: [Buildroot] [PATCH 1/1] Enable quadmath on aarch64

[Bensuperpc]

[-- Attachment #1.1: Type: text/plain, Size: 1685 bytes --]

Some headers need quadmath to work (Ex: boost::multiprecision::float128),

I am getting following compile error on aarch64 without
*BR2_TOOLCHAIN_HAS_LIBQUADMATH=y *:
*boost/multiprecision/float128.hpp:40:10: fatal error: 'quadmath.h' file
not found*
when I use float128 in boost.

Config file:
https://github.com/dockcross/dockcross/blob/master/linux-arm64-full/buildroot.config

Bensuperpc,


Le sam. 21 août 2021 à 23:27, Thomas Petazzoni <thomas.petazzoni@bootlin.com>
a écrit :

> On Sat, 21 Aug 2021 21:44:01 +0200
> Bensuperpc <bensuperpc@gmail.com> wrote:
>
> > Enable quadmath on aarch64
> >
> > Signed-off-by: Bensuperpc <bensuperpc@gmail.com>
>
> We do need a real name to accept contributions.
>
> > ---
> >  toolchain/Config.in | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/toolchain/Config.in b/toolchain/Config.in
> > index 1d0469a13e..b4cd6d511f 100644
> > --- a/toolchain/Config.in
> > +++ b/toolchain/Config.in
> > @@ -794,6 +794,6 @@ config BR2_TOOLCHAIN_HAS_ATOMIC
> >  # - When available, libquadmath requires wchar support.
> >  config BR2_TOOLCHAIN_HAS_LIBQUADMATH
> >       bool
> > -     default y if BR2_i386 || BR2_x86_64
> > +     default y if BR2_i386 || BR2_x86_64 || BR2_aarch64
>
> This option is not about "enabling" something, but about saying whether
> libquadmath is supported/available on a given CPU architecture. Since
> when is libquadmath available/supported by gcc?
>
> What was the motivation behind this change?
>
> Thanks!
>
> Thomas
> --
> Thomas Petazzoni, co-owner and CEO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
>

[-- Attachment #1.2: Type: text/html, Size: 2834 bytes --]

[-- Attachment #2: Type: text/plain, Size: 145 bytes --]

_______________________________________________
buildroot mailing list
buildroot@busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Re: [Buildroot] [PATCH 1/1] Enable quadmath on aarch64

[Giulio Benetti]

[-- Attachment #1.1: Type: text/plain, Size: 2259 bytes --]

Hi Bensuperpc,

you have to use your real name. Then you have to use plain text, no html at all in Mailing List.
This commit log seems written directly in email client, instead you should write it during git commit.

Also, you need to send v2 patch, so pass -v2 to git format-patch and indicate V1->V2 changes(basically you’re written the commit log).

Best regards
Giulio Benetti

Inviato da iPhone

> Il giorno 22 ago 2021, alle ore 06:27, Bensuperpc <bensuperpc@gmail.com> ha scritto:
> 
> 
> Some headers need quadmath to work (Ex: boost::multiprecision::float128), 
> 
> I am getting following compile error on aarch64 without BR2_TOOLCHAIN_HAS_LIBQUADMATH=y :
> boost/multiprecision/float128.hpp:40:10: fatal error: 'quadmath.h' file not found
> when I use float128 in boost.
> 
> Config file: https://github.com/dockcross/dockcross/blob/master/linux-arm64-full/buildroot.config
> 
> Bensuperpc,
> 
> 
>> Le sam. 21 août 2021 à 23:27, Thomas Petazzoni <thomas.petazzoni@bootlin.com> a écrit :
>> On Sat, 21 Aug 2021 21:44:01 +0200
>> Bensuperpc <bensuperpc@gmail.com> wrote:
>> 
>> > Enable quadmath on aarch64
>> > 
>> > Signed-off-by: Bensuperpc <bensuperpc@gmail.com>
>> 
>> We do need a real name to accept contributions.
>> 
>> > ---
>> >  toolchain/Config.in | 2 +-
>> >  1 file changed, 1 insertion(+), 1 deletion(-)
>> > 
>> > diff --git a/toolchain/Config.in b/toolchain/Config.in
>> > index 1d0469a13e..b4cd6d511f 100644
>> > --- a/toolchain/Config.in
>> > +++ b/toolchain/Config.in
>> > @@ -794,6 +794,6 @@ config BR2_TOOLCHAIN_HAS_ATOMIC
>> >  # - When available, libquadmath requires wchar support.
>> >  config BR2_TOOLCHAIN_HAS_LIBQUADMATH
>> >       bool
>> > -     default y if BR2_i386 || BR2_x86_64
>> > +     default y if BR2_i386 || BR2_x86_64 || BR2_aarch64
>> 
>> This option is not about "enabling" something, but about saying whether
>> libquadmath is supported/available on a given CPU architecture. Since
>> when is libquadmath available/supported by gcc?
>> 
>> What was the motivation behind this change?
>> 
>> Thanks!
>> 
>> Thomas
>> -- 
>> Thomas Petazzoni, co-owner and CEO, Bootlin
>> Embedded Linux and Kernel engineering
>> https://bootlin.com

[-- Attachment #1.2: Type: text/html, Size: 3765 bytes --]

[-- Attachment #2: Type: text/plain, Size: 145 bytes --]

_______________________________________________
buildroot mailing list
buildroot@busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot

Re: [Buildroot] [PATCH 1/1] Enable quadmath on aarch64

[Arnout Vandecappelle]

On 21/08/2021 21:56, Bensuperpc wrote:
> Enable quadmath on aarch64
> 
> Signed-off-by: Bensuperpc <bensuperpc@gmail.com>

 Since no signoff with a real name was provided, and Thomas's question hasn't
been answered (I checked the current master GCC source and couldn't find a
quadmath implementation for arm64), I marked this patch as Rejected.

 Regards,
 Arnout

> ---
>  toolchain/Config.in | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/toolchain/Config.in b/toolchain/Config.in
> index 1d0469a13e..b4cd6d511f 100644
> --- a/toolchain/Config.in
> +++ b/toolchain/Config.in
> @@ -794,6 +794,6 @@ config BR2_TOOLCHAIN_HAS_ATOMIC
>  # - When available, libquadmath requires wchar support.
>  config BR2_TOOLCHAIN_HAS_LIBQUADMATH
>  	bool
> -	default y if BR2_i386 || BR2_x86_64
> +	default y if BR2_i386 || BR2_x86_64 || BR2_aarch64
>  
>  endmenu
> 
_______________________________________________
buildroot mailing list
buildroot@lists.buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

[Buildroot] [PATCH 1/1] Enable quadmath on aarch64

[Bensuperpc]

Enable quadmath on aarch64

Signed-off-by: Bensuperpc <bensuperpc@gmail.com>
---
 toolchain/Config.in | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/toolchain/Config.in b/toolchain/Config.in
index 1d0469a13e..b4cd6d511f 100644
--- a/toolchain/Config.in
+++ b/toolchain/Config.in
@@ -794,6 +794,6 @@ config BR2_TOOLCHAIN_HAS_ATOMIC
 # - When available, libquadmath requires wchar support.
 config BR2_TOOLCHAIN_HAS_LIBQUADMATH
 	bool
-	default y if BR2_i386 || BR2_x86_64
+	default y if BR2_i386 || BR2_x86_64 || BR2_aarch64
 
 endmenu
-- 
2.32.0

_______________________________________________
buildroot mailing list
buildroot@busybox.net
http://lists.busybox.net/mailman/listinfo/buildroot