Re: Anyone tried Intel's compiler on Enigma?

[Guy Fraser <guy incentre net>]

Steve Snyder wrote:
> 
> I'm thinking of playing with Intel's v5.01 C/C++ compiler
> (http://developer.intel.com/software/products/compilers/c50/linux/noncom.htm).
> 
> I understand that it it not completely GCC-compatible so a lot of Linux
> software, especially the kernel, won't build with it.  Nevertheless, I'd
> like to check it out.  But not at the expense of my working Enigma box.
> 
> Intel says that the compile will work on RedHat v7.1, so presumably it will
> live happily on v7.2 as well.  All I want to know is, will the installation
> of Intel's compiler break my existing installation of GCC?
> 
> Thanks.

It looks good to me, so far!

Since I have just purchased a P4 and I doubt gcc has P4 optimizations,
it
looks to be encouraging. Since I have all the source packages I may be
able
to get a RH 7.2 Intel compile P4 set of packages done up {only god knows
how long it will take, and if it is possible.} Fortunately I have lots
of
experience programming with different compilers on different platforms.

The libraries are the biggest issue. If the intel compiler will work be
able
rebuild the normal libraries and work with the binary only libraries
made
with gnu's compilers, then we should be off to the races. With any good 
luck they will have good documentation.

I am encouraged with claimed compatibility and performance declarations.

I am told that the P4 relies on most optimizations to be done at compile 
time rather than at run time. This makes a lot of sense to me, this is
how
I used to have to write programs back when you optimized your code with
an
assembler. I lost about 5 years of programming {span of 386 to pentium 
launches.} at a job that only required a couple of piddly little fixed 
database viewers. The projects took about 4 days in total. While making
the viewers I developed a btree type of algorithm from scratch. Having 
never heard of such a beast I called it delta sort. I also made a simple
hashed index that was based on "pointers of pointers" from a Dr. Dobbs
issue.

Well maybe I'll take a couple of pokes at it this weekend.

Chow