[Zlib-devel] [PATCH] deflate.c: identify slide_Pos() for later optimization

Jan Seiffert kaffeemonster at googlemail.com
Tue Jul 24 09:24:46 EDT 2012


John Reiser schrieb:
> On 07/24/2012 02:26 AM, Jan Seiffert wrote:
> 
>> https://github.com/kaffeemonster/zlib/commit/aa30d206d93755c1f5c287b78dd7d165fb0998a8
>    etc.
> 
> Thank you for the code!  Where are the discussion threads about such optimizations?
> 

This specific code or SIMD optimization or optimization in general?
For this specific code there was no discussion, i announced it and crickets:
http://mail.madler.net/pipermail/zlib-devel_madler.net/2011-May/002564.html

>> And no, most compiler do not get it that this is a subtraction with saturation.
> 
> I suggested "a very good compiler".  I have worked with a large handful of
> compilers of different origins, and only one has been good enough [a compiler
> for APL.]  On the other hand, subtraction with saturation was a primitive [library]
> routine in Fortran II (1958): MIN0, DMIN0, etc.; and I have used those.  So this
> looks like a case of lost or discarded history.
> 

Ahhhh, you are talking about compiler in general.
I was talking about C compiler. They can't get this because of C type promotion
rules in this construct (or they have to cut through a lot of red tape).

Greetings
	Jan

-- 
Anyone can build a fast processor. The trick is to build a fast system.
(Seymour Cray)




More information about the Zlib-devel mailing list