[cfe-commits] r88819 - in /cfe/trunk: lib/Driver/Tools.cpp test/Driver/clang-translation.c

[Carl-Daniel Hailfinger]

On 22.11.2009 18:22, Daniel Dunbar wrote:
> On Sun, Nov 22, 2009 at 7:26 AM, Chris Lattner <clattner at apple.com> wrote:
>   
>> On Nov 14, 2009, at 2:04 PM, Daniel Dunbar wrote:
>>     
>>> Author: ddunbar
>>> New Revision: 88819
>>>
>>> Add clang -mcpu=native support, patch by Roman Divacky, varioustweaks by me.
>>> - We still need support for detecting the target features, since the name
>>>   doesn't actually do a good job of decribing what the CPU supports (for LLVM).
>>>       
>> Very cool.  Should this be the default on linux and BSDs?
>>     
>
> I personally think -march=native is not the right default, but I see
> the arguments for it. I guess the argument for making it the default
> on Linux/BSDs is that many users build their own software? What do
> modern GCC's do? IIRC, you brought this up before and I don't think I
> was the only opponent.
>   

Given that I often have to deal with non-developer people who screwed up
one machine and need to compile a few recovery apps on another machine,
then transfer them via pen drive, my personal preference is to have
-mcpu not use anything that may cause problems or extreme slowdowns on
the screwed up machine.
This is purely from a user perspective though, feel free to ignore it.

Regards,
Carl-Daniel

-- 
Developer quote of the month: 
"We are juggling too many chainsaws and flaming arrows and tigers."