From: Cesar Rabak <csrabak@yahoo.com.br>
Subject: Re: Translating an embedded C algorithm
Date: Thu, 18 Jan 2007 21:27:37 -0200
Date: 2007-01-18T21:27:37-02:00 [thread overview]
Message-ID: <eoovl1$6ps$1@aioe.org> (raw)
In-Reply-To:
Jean-Pierre Rosen escreveu:
> Talulah a �crit :
>>> There is a least one documented case of Ada producing smaller code than
>>> hand-optimized assembler.
>>
>> I love these statistics. There's a Java vendor who reckons the byte
>> code runs faster than C as well. You can prove anything if y0ou have
>> choice over the tools that you use to produce the results. This "hand
>> coded assembler" could mean anything - it could mean taking Ada
>> compiled code and ADDING instructions to it!
> This was not the case.
> See http://www.acm.org/sigada/education/pages/lawlis.html for the full
> story
>
>> Thinking logically, if the assembler coder was any good, he can always
>> produce code of equal size to compiled code, and should always produce
>> tighter code.
> The reference above has a good explanation why this is not the case.
>
This paper is the one it should never shown for people are not already
knowledgeable of Ada:
1) It is old;
2) It hides labour under reporting the effort to write the Ada version:
in section 5. "This 700 lines of Ada development had taken three weeks
of labor, approximately 150 hours."
It took also (section 2.) "...experts with various backgrounds in
widely diverse organizations--the chip manufacturer, an Ada vendor, an
independent consultant, a government laboratory, and the Air Force's
graduate school, the Air Force Institute of Technology.", plus I
understand the authors of the paper itself.
3) It puts away costs that are not insignificant (US$1000 in 1996
dollars) *per system*;
4) A close reading of section 4. "The Proof Backfired" shows that for
the Ada version arrive at the same performance of the assembly code, one
engineer from the compiler vendor had to perform a series of non trivial
experiments, write a hand crafted version of Ada code to unroll loops
*after* analysis of the assembly code and binary. So, not Ada technology
for itself, after of all.
Again, I remember: the report is old, and for present state of art chips
with multiple datapaths, positively its more likely than a hand made
assembly code will not match a HLL compiler generated binary.
So, I think that it is the case Talulah points out.
next prev parent reply other threads:[~2007-01-18 23:27 UTC|newest]
Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-01-15 14:36 Translating an embedded C algorithm Talulah
2007-01-16 0:06 ` Jeffrey Carter
2007-01-16 1:10 ` Marc A. Criley
2007-01-16 2:21 ` Cesar Rabak
2007-01-16 10:40 ` Markus E Leypold
2007-01-16 14:48 ` Larry Kilgallen
2007-01-16 17:32 ` Jeffrey Carter
2007-01-16 18:04 ` Cesar Rabak
2007-01-17 0:09 ` Jeffrey Carter
2007-01-17 1:07 ` Cesar Rabak
2007-01-16 4:37 ` Alexander E. Kopilovich
2007-01-16 13:37 ` Cesar Rabak
2007-01-16 23:47 ` Simon Wright
2007-01-17 1:02 ` Cesar Rabak
2007-01-16 3:50 ` Vo, Anh (US SSA)
2007-01-16 12:15 ` Niklas Holsti
2007-01-16 23:50 ` Simon Wright
2007-01-18 16:17 ` Niklas Holsti
2007-01-18 16:41 ` Ludovic Brenta
2007-01-18 20:02 ` Niklas Holsti
2007-01-18 22:25 ` Cesar Rabak
2007-01-19 8:32 ` Niklas Holsti
2007-01-19 19:15 ` Cesar Rabak
2007-01-19 20:49 ` Simon Wright
2007-01-18 16:55 ` Robert A Duff
2007-01-18 18:54 ` Jeffrey Carter
2007-01-19 0:45 ` Robert A Duff
2007-01-18 21:25 ` Niklas Holsti
2007-01-19 0:50 ` Robert A Duff
2007-01-19 4:43 ` Jeffrey Carter
2007-01-18 18:43 ` Jeffrey Carter
2007-01-18 20:19 ` Niklas Holsti
2007-01-18 20:30 ` Niklas Holsti
2007-01-18 23:34 ` Cesar Rabak
2007-01-19 8:57 ` Niklas Holsti
2007-01-19 2:11 ` Steve Whalen
2007-01-19 10:27 ` Niklas Holsti
2007-01-16 13:32 ` Cesar Rabak
2007-01-16 14:47 ` Gautier
2007-01-16 15:15 ` Cesar Rabak
2007-01-16 15:16 ` Jean-Pierre Rosen
2007-01-16 16:12 ` Ludovic Brenta
2007-01-16 17:10 ` Georg Bauhaus
2007-01-16 22:32 ` Ludovic Brenta
2007-01-17 20:22 ` Georg Bauhaus
2007-01-18 9:23 ` Ludovic Brenta
2007-01-16 17:12 ` Cesar Rabak
2007-01-16 17:20 ` Frank J. Lhota
2007-01-16 18:09 ` Cesar Rabak
2007-01-16 17:36 ` Dmitry A. Kazakov
2007-01-16 18:08 ` Cesar Rabak
2007-01-16 18:48 ` Dmitry A. Kazakov
2007-01-16 20:03 ` Cesar Rabak
2007-01-18 19:33 ` Björn Persson
2007-01-18 22:32 ` Cesar Rabak
2007-01-19 20:26 ` Björn Persson
2007-01-19 23:25 ` Cesar Rabak
2007-01-19 7:15 ` Maciej Sobczak
2007-01-19 20:27 ` Björn Persson
2007-01-19 20:34 ` Robert A Duff
2007-01-17 13:48 ` Maciej Sobczak
2007-01-17 23:32 ` Translating an embedded C algorithm -- OT Cesar Rabak
2007-01-18 8:56 ` Talulah
2007-01-18 22:05 ` Cesar Rabak
2007-01-18 9:03 ` Maciej Sobczak
2007-01-18 10:22 ` Alex R. Mosteo
2007-01-18 18:34 ` Jeffrey Carter
2007-01-18 22:26 ` Cesar Rabak
2007-01-19 4:45 ` Jeffrey Carter
2007-01-18 22:18 ` Cesar Rabak
2007-01-19 20:53 ` Simon Wright
2007-01-16 15:55 ` Translating an embedded C algorithm Cesar Rabak
2007-01-17 3:00 ` Vo, Anh (US SSA)
2007-01-17 10:48 ` Cesar Rabak
2007-01-17 11:44 ` Niklas Holsti
2007-01-17 13:31 ` Talulah
2007-01-17 19:20 ` Jeffrey Carter
2007-01-18 14:19 ` Talulah
2007-01-18 15:28 ` Jean-Pierre Rosen
2007-01-18 23:27 ` Cesar Rabak [this message]
2007-01-18 18:51 ` Jeffrey Carter
2007-01-18 22:30 ` Cesar Rabak
2007-01-19 4:48 ` Jeffrey Carter
2007-01-19 19:13 ` Cesar Rabak
2007-01-20 20:56 ` Jeffrey Carter
2007-01-19 2:21 ` Alexander E. Kopilovich
2007-01-19 3:25 ` Larry Kilgallen
2007-01-20 0:46 ` Alexander E. Kopilovich
2007-01-20 13:03 ` Larry Kilgallen
2007-01-20 16:54 ` Alexander E. Kopilovich
2007-01-20 23:53 ` Larry Kilgallen
2007-01-20 21:02 ` Jeffrey Carter
2007-01-25 21:59 ` Markus E Leypold
2007-01-26 4:06 ` Larry Kilgallen
2007-01-26 11:26 ` Markus E Leypold
2007-01-26 12:25 ` Cesar Rabak
2007-01-19 4:52 ` Jeffrey Carter
2007-01-19 10:13 ` Warner BRUNS
2007-01-19 14:54 ` Robert A Duff
2007-01-19 4:08 ` Steve
2007-01-19 20:41 ` Simon Wright
-- strict thread matches above, loose matches on Subject: below --
2007-01-17 7:07 AW: " Grein, Christoph (Fa. ESG)
2007-01-17 10:26 ` Ludovic Brenta
2007-01-17 16:44 ` Markus E Leypold
2007-01-18 8:49 ` Ludovic Brenta
2007-01-19 9:33 ` Stephen Leake
2007-01-19 19:23 ` Cesar Rabak
2007-01-19 20:27 ` Robert A Duff
2007-01-20 9:54 ` Dmitry A. Kazakov
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox