Re: [PATCH 0 of 3] Add support for using gold instead of ld

[Bryan Hundven <bryanhundven at gmail dot com>]

On Mon, Dec 27, 2010 at 4:18 PM, Bryan Hundven <bryanhundven@gmail.com> wrote:
> On Mon, Dec 27, 2010 at 3:58 PM, Arnaud Lacombe <lacombar@gmail.com> wrote:
>> Hi,
>>
>> On Mon, Dec 27, 2010 at 6:33 PM, Yann E. MORIN
>> <yann.morin.1998@anciens.enib.fr> wrote:
>>> Hello All!
>>>
>>> This patchset is a tentative at using gold in place of ld with recent
>>> binutils (2.21 at least).
>>>
>>> Currently, glibc can not be built with gold (fails on the ld version
>>> check), and uClibc builds fine with a minor edit in the linker flags
>>> (which is already applied upstream).
>>>
>>> Basically, the patch series adds support for building gold or ld or
>>> both, and in the latter case, set which is the default. It also adds
>>> a (shell) wrapper that can override the default at runtime.
>>>
>>> Shortlog:
>>> Âbinutils/binutils: add BINUTILS_2_21_or_later blind option
>>> Âbinutils/binutils: add option to enable gold
>>> Âbinutils/binutils: add wrapper to gold and ld
>>>
>>> Diffstat:
>>> Âb/config/binutils/binutils.in      |  88  Â88   0   0 ++++++++++++++++++++++++++++++++
>>> Âb/scripts/build/binutils/binutils-ld.in | Â 11 Â Â11 Â Â 0 Â Â 0 ++++
>>> Âb/scripts/build/binutils/binutils.sh  Â|  37  Â36   1   0 +++++++++++++
>>> Â3 files changed, 135 insertions(+), 1 deletion(-)
>>>
>> That is incomplete You also need to pass --enable-gold to gcc's
>> configure to build the linker plugin.
>
> Right. Is this only when ld.gold is ld?
> Or just when gold is enabled at all.

Actually, to answer my own question, for gcc 4.5.x, the only option is
to --enable-gold (just like binutils 2.20x). for gcc 4.6.x the options
would be the same as binutils 2.21.

>> Â- Arnaud
>>
>> --
>> For unsubscribe information see http://sourceware.org/lists.html#faq
>>
>>
>
> Yann,
>
> I will test this stuff tonight (-0800) with the gcc --enable-gold option.
>
> -Bryan
>

-Bryan

--
For unsubscribe information see http://sourceware.org/lists.html#faq