The SPUs seem to become VERY irrealistic

Larry Lawrence larry at
Fri Oct 18 17:42:08 PDT 2002

Florian Hess (FLoH) wrote:

> Bill maltby - LFS Related schrieb:
>> On Fri, 18 Oct 2002, Florian Hess (FLoH) wrote:
>> You have a tough wait ahead of you. Here are some timings from when I ran
>> on a Cyrix Pr-166 with 64MB and 256MB swap. This was *before* static and
>> gcc-3.X. Using gcc-3.X added over 2 hours on an IBM6x86-II (Cyrix-II) 300
>> with 256MB PC-133 SDRAM and 256MB swap.
> Now at the legendary time of 01:40a, 10/19/2002, I stopped my LFS
> compilation per ^C because of a lack of patience. I'll even my LFS
> partition down and acquire a full Debian Woody distribution.
> May it be that I was a little bit absent of reality, compiling a lfs
> on a p150?
> However, it was worth trying! Congratulation to Gerard Beekmans for
> rising this project! I learned much about Linux.
> But, as a conclusion, something's confusing me: Gerard thought his SPU
> value to provide a relative base to appreciate compile times, and he
> tested this relation on many systems as he say. But it was not
> relative in my case: it exceeded 2,5h by far (6h until aborted).
There is no way for the SPU to be realistic with your conditions. I have one 
machine with a 486/66 and 16 MB.  You'll find the compile speed for 
gcc-3(something) posted in the archives, but it was in days.  Disk swapped 
memory is no match for the almost full memory compiles on today's machines.  
I tinker with a laptop with 8 MB, use debian on it, but to load dselect's  
database can take several hours versus seconds on a machine with 16 MB and 
its all related to memory swapping to disk.


Unsubscribe: send email to listar at
and put 'unsubscribe lfs-support' in the subject header of the message

More information about the lfs-support mailing list