[Buildroot] [PATCH 1/1] Toolchain - Add option to enable Ada language.

[Thomas Petazzoni]

Dear Patrick Noffke,

Sorry for the long response time. However, unless I missed it, you
never resent your patch with git send-email. Can you do this? Also, see
some comments below.

On Tue, 07 Apr 2015 14:10:28 -0500, Patrick Noffke wrote:
> From d429cdacd2584e561c754bc7d7cf0eb78d7c03f0 Mon Sep 17 00:00:00 2001
> From: Patrick Noffke <patrick at noffke.me>
> Date: Tue, 7 Apr 2015 09:04:45 -0500
> Subject: [PATCH 1/1] Toolchain - Add option to enable Ada language.
> 
> Signed-off-by: Patrick Noffke <patrick at noffke.me>
> ---
>  package/gcc/Config.in.host         | 10 ++++++++++
>  package/gcc/gcc-final/gcc-final.mk |  9 +++++++++
>  toolchain/toolchain-common.in      |  3 +++
>  3 files changed, 22 insertions(+)
> 
> diff --git a/package/gcc/Config.in.host b/package/gcc/Config.in.host
> index cf96c73..be55b4f 100644
> --- a/package/gcc/Config.in.host
> +++ b/package/gcc/Config.in.host
> @@ -108,6 +108,16 @@ config BR2_TOOLCHAIN_BUILDROOT_CXX
>  	  C++ language and you want C++ libraries to be installed on
>  	  your target system.
>  
> +config BR2_TOOLCHAIN_BUILDROOT_ADA
> +	bool "Enable Ada support"
> +	select BR2_INSTALL_LIBADA

Technically speaking, this hidden boolean will only become necessary
when some packages will need Ada support. We've recently added Fortran
support, without a similar boolean. I guess we'll have to add it only
when some packages will start using Fortran support.

> +	depends on !(BR2_GCC_VERSION_4_2_2_AVR32_2_1_5 && \

This gcc version has since then been remove.

> +		     BR2_TOOLCHAIN_BUILDROOT_LOCALE)

So a toolchain with locale support cannot have Ada support? This seems
weird, can you add a comment that explains why?

> +	help
> +	  Enable this option if you want your toolchain to support the
> +	  Ada language and you want Ada libraries to be installed on
> +	  your target system.
> +
>  config BR2_GCC_ENABLE_TLS
>  	bool "Enable compiler tls support" if BR2_TOOLCHAIN_BUILDROOT_UCLIBC
>  	default y
> diff --git a/package/gcc/gcc-final/gcc-final.mk b/package/gcc/gcc-final/gcc-final.mk
> index d76eb31..7defb97 100644
> --- a/package/gcc/gcc-final/gcc-final.mk
> +++ b/package/gcc/gcc-final/gcc-final.mk
> @@ -44,6 +44,7 @@ endef
>  # Languages supported by the cross-compiler
>  GCC_FINAL_CROSS_LANGUAGES-y = c
>  GCC_FINAL_CROSS_LANGUAGES-$(BR2_INSTALL_LIBSTDCPP) += c++
> +GCC_FINAL_CROSS_LANGUAGES-$(BR2_INSTALL_LIBADA) += ada
>  GCC_FINAL_CROSS_LANGUAGES = $(subst $(space),$(comma),$(GCC_FINAL_CROSS_LANGUAGES-y))
>  
>  HOST_GCC_FINAL_CONF_OPTS = \
> @@ -65,6 +66,10 @@ else
>  HOST_GCC_FINAL_CONF_OPTS += --disable-libgomp
>  endif
>  
> +ifeq ($(BR2_INSTALL_LIBADA),y)
> +HOST_GCC_FINAL_CONF_OPTS += --enable-libada
> +endif
> +
>  # End with user-provided options, so that they can override previously
>  # defined options.
>  HOST_GCC_FINAL_CONF_OPTS += \
> @@ -125,6 +130,10 @@ ifeq ($(BR2_INSTALL_LIBSTDCPP),y)
>  HOST_GCC_FINAL_USR_LIBS += libstdc++
>  endif
>  
> +ifeq ($(BR2_INSTALL_LIBADA),y)
> +#HOST_GCC_FINAL_USR_LIBS += libgnat
> +endif

This is commented out, why? If you need some library on the target to
run Ada programs, you indeed need to add something to
HOST_GCC_FINAL_USR_LIBS.

Also, if I understood correctly, building the Ada support in gcc
requires having an already working Ada compiler on your machine. Is
this correct? If it is, then support/dependencies/dependencies.sh need
to be adjusted to also check for an Ada compiler when Ada support is
selected in the toolchain.

Could you work on these issues and submit an updated version of your
patch? In the mean time, I'll mark your patch as "Changes Requested" in
our patch tracking system.

Thanks a lot!

Thomas
-- 
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com