Re: Slow linking for ARM

[Bryan Hundven <bryanhundven at gmail dot com>]

Bill, Titus, all,

On Fri, Dec 24, 2010 at 6:53 AM, Bill Pringlemeir <bpringle@sympatico.ca> wrote:
> On 22 Dec 2010, titus@v9g.de wrote:
>> Am 22.12.2010 um 21:09 schrieb Bryan Hundven:
>
>>>> As a test apart from my company's own software, I used the http
>>>> server example of asio today, and got a factor of 2 between ARM
>>>> and other archs (though linking time is quite short in total:
>>>> 0.21s user for PPC/X86 and 0.45 s user for ARM on a 2GHz Core 2
>>>> Duo Mac using binutils-2.20).
>
>>> So your building apache2? lighttpd? custom httpd server? ÂJust
>>> curious so I can repo the problem, and eventually start to provide
>>> help.
>
>> Sorry, that was a bit short. ÂI meant boost's asio lib
>> (www.boost.org). Included are small samples e.g. for a http
>> server. That was something I had lying around for testing. ÂBut
>> maybe also other applications do the trick.
>
> It sounds like titus's primary apps are C++, perhaps with heavy
> template usage. ÂThe 'gold' linker has far higher performance for C++
> applications with large name mangling, etc.
>
> Ian Lance Taylor, golds creator, measures it as 5x faster. ÂFor
> pathological cases, it may be even more.
>
> Bryan and other ARM users may not notice the link times unless they
> are linking C++ templates. ÂIf 'gold' is not building for the ARM,
> this might explain everything as 'ld' can be 'gold' with the right
> configuration options. ÂI think that 'whole program optimization'
> (-whopr?) might require 'gold'.

That makes sense. The usage of my toolchains is mostly systems
software, and very few c++ applications.

> Fwiw,
> Bill Pringlemeir.
>
> Some links on gold,
> Âhttp://lwn.net/Articles/274859/
> Âhttp://www.airs.com/blog/archives/78
>
> --
> For unsubscribe information see http://sourceware.org/lists.html#faq
>
>



-- 
Bryan Hundven
bryanhundven@gmail.com

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