Re: Instruction Tracing

Liste des GroupesRevenir à c arch 
Sujet : Re: Instruction Tracing
De : cr88192 (at) *nospam* gmail.com (BGB)
Groupes : comp.arch
Date : 10. Aug 2024, 23:49:42
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <v98qq8$ukip$1@dont-email.me>
References : 1 2
User-Agent : Mozilla Thunderbird
On 8/10/2024 2:41 PM, John Dallman wrote:
In article <2024Aug10.121802@mips.complang.tuwien.ac.at>,
anton@mips.complang.tuwien.ac.at (Anton Ertl) wrote:
 
IIRC IA-64 has no FP division.
 You recall correctly. It has an "approximation to reciprocal" instruction,
which gives you about 8 bits of precision, and then requires the compiler
to generate Newton-Raphson sequences. Intel's manual, 2010 edition, says
this is advantageous because users can generate only the precision they
need. Writing Itanium assembler for customised precision? Not many people
would have wanted to do that in 2001, let alone 2010.
 In, I think, 1996, my employers had visitors from Intel trying to
persuade us to adopt their C/C++ compiler for IA-32. They had been able
to speed up one of our competitors' code by a factor of two, and hoped to
do the same for us.
 They failed. We already had that factor of two, which was "ordinary
compiler optimisation." That competitor had some rather odd coding
standards at the time, which meant most compilers failed if asked to
optimise their code. Someone from Intel had stayed at their site for most
of a year, reporting the bugs and getting them fixed until Intel's
compiler could optimise the code.
 While visiting us, Intel asked what may have been a significant question
about the mixture of floating-point arithmetic instructions we used. We
didn't have precise figures, but were sure that we used at least as many
square roots as divides. IA-64 does square roots like divides, with a
starter approximation and Newton-Raphson sequences. Slowly, because the
N-R instructions all depend on the previous instruction, and can't be run
in parallel.
 
FWIW:
In my case it is similar (if not using the FDIV) instruction, where there are approximations for divide / reciprocal / square root.
Meanwhile, saw a video recently where someone had ported Doom to a 233 MHz PowerPC (running Windows NT4) machine and, its performance was not good...
Not obvious is what combination of factors conspired to cause Doom to apparently run at single-digit framerates.
Video mentioned that it was drawing using GDI calls, but this by itself wouldn't explain the level of slowness seen in the video.
Like, presumably, this would require around 90% + of the clock cycles going into overhead, which seems a bit much.
Reference:
https://www.youtube.com/watch?v=LAkSJ-HqKw8

John

Date Sujet#  Auteur
10 Aug 24 * Instruction Tracing31Lawrence D'Oliveiro
10 Aug 24 +* Re: Instruction Tracing29Anton Ertl
10 Aug 24 i+- Re: Instruction Tracing1MitchAlsup1
10 Aug 24 i+* Re: Instruction Tracing8John Dallman
10 Aug 24 ii+- Re: Instruction Tracing1MitchAlsup1
11 Aug 24 ii`* Re: Instruction Tracing6BGB
11 Aug 24 ii +* Re: Instruction Tracing4Lawrence D'Oliveiro
11 Aug 24 ii i`* Re: Instruction Tracing3BGB
11 Aug 24 ii i `* Re: Instruction Tracing2George Neuner
11 Aug 24 ii i  `- Re: Instruction Tracing1BGB
12 Aug 24 ii `- Re: Instruction Tracing1Michael S
10 Aug 24 i+* Re: Instruction Tracing3BGB
11 Aug 24 ii`* Re: Instruction Tracing2MitchAlsup1
11 Aug 24 ii `- Re: Instruction Tracing1BGB
11 Aug 24 i`* Re: Instruction Tracing16John Levine
11 Aug 24 i +* Re: Instruction Tracing3OrangeFish
11 Aug 24 i i`* Re: Instruction Tracing2John Levine
12 Aug 24 i i `- Re: Instruction Tracing1Lynn Wheeler
11 Aug 24 i `* Re: Instruction Tracing12Anton Ertl
11 Aug 24 i  +* Re: Instruction Tracing2MitchAlsup1
12 Aug 24 i  i`- Re: Instruction Tracing1Lawrence D'Oliveiro
12 Aug 24 i  `* Re: Instruction Tracing9Lawrence D'Oliveiro
12 Aug 24 i   +* Re: Instruction Tracing2Terje Mathisen
12 Aug 24 i   i`- Re: Instruction Tracing1Anton Ertl
12 Aug 24 i   `* Re: Instruction Tracing6Anton Ertl
12 Aug 24 i    `* Re: Instruction Tracing5Lawrence D'Oliveiro
12 Aug 24 i     `* Re: Instruction Tracing4Michael S
12 Aug 24 i      `* Re: Instruction Tracing3Lawrence D'Oliveiro
12 Aug 24 i       `* Re: Instruction Tracing2Michael S
12 Aug 24 i        `- Re: Instruction Tracing1MitchAlsup1
10 Aug 24 `- Re: Instruction Tracing1MitchAlsup1

Haut de la page

Les messages affichés proviennent d'usenet.

NewsPortal