Performance comparison, ULE vs 4BSD and AMD64 vs i386

[Jem Matzan]

John Baldwin wrote:

>On Tuesday 24 February 2004 05:04 pm, Jem Matzan wrote:
>  
>
>>Brooks Davis wrote:
>>    
>>
>>>On Tue, Feb 24, 2004 at 04:37:44PM -0500, Jem Matzan wrote:
>>>      
>>>
>>>>How about AMD64 being slower than i386 on the same hardware? By
>>>>slower, I mean a buildworld -j4 took about 400 seconds longer in AMD64
>>>>mode.
>>>>        
>>>>
>>>You can't usefully compare compile times when you are compiling for
>>>a different instructions set.  The work involved is rairly the same
>>>so the results are meaning less.  If you could factor out the cost of
>>>building the native bootstrap tools since that isn't the same job on
>>>each machine, the speed of a cross buildworld would be an intresting
>>>test.  For comparing i386 and amd64, I'd probably build an alpha or
>>>sparc64 world so the target would be entierly different.
>>>
>>>-- Brooks
>>>      
>>>
>>I figured that the world would be the same for both AMD64 and i386. That
>>really sucks that all of this data and all of that time has been more or
>>less wasted on doing buildworld time benchmarks. As far as I know it
>>isn't possible to do a crossbuild (I've tried before, and I read on the
>>list several weeks ago that it won't work). Do you have any suggestions
>>for measuring compile times?
>>    
>>
>
>You can do a crossbuild easy, just do:
>
>make TARGET_ARCH=amd64 buildworld
>
>or subsitute whatever arch for amd64.
>
>  
>
Ah -- I wasn't considering all of the possibilities. It was the AMD64 
form i386 crossbuild that wasn't working for me (while actually trying 
to turn a working i386 system into an AMD64 build -- it was the 
installworld that failed and broke everything to hell. Now I remember). 
So instead I'll build a SPARC or ALPHA world on both AMD64 and i386. Of 
course this means another 18 hours of testing... unless I can build the 
i386 world with AMD64, in which case I only have half of everything to 
do again.

-Jem