www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Signaling NaNs Rise Again

reply Walter Bright <newshound1 digitalmars.com> writes:
Inspired by Don Clugston's recent compiler patch.

http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/
Mar 26 2009
next sibling parent reply Georg Wrede <georg.wrede iki.fi> writes:
Walter Bright wrote:
 Inspired by Don Clugston's recent compiler patch.
 
 http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/ 
Yess! (And I loved the war story at the beginning!)
Mar 27 2009
parent reply Walter Bright <newshound1 digitalmars.com> writes:
Georg Wrede wrote:
 Walter Bright wrote:
 Inspired by Don Clugston's recent compiler patch.

 http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/ 
Yess! (And I loved the war story at the beginning!)
It's interesting because it's right before computers moved in and replaced everything. If I'd stayed another year, I'd have bought (with my own funds) a desktop computer and put it on my desk. I'm sure that within 2 years Boeing would have automatically supplied me with one.
Mar 27 2009
parent reply Paul D. Anderson <paul.d.removethis.anderson comcast.andthis.net> writes:
Walter Bright Wrote:

 Georg Wrede wrote:
 Walter Bright wrote:
 Inspired by Don Clugston's recent compiler patch.

 http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/ 
Yess! (And I loved the war story at the beginning!)
It's interesting because it's right before computers moved in and replaced everything. If I'd stayed another year, I'd have bought (with my own funds) a desktop computer and put it on my desk. I'm sure that within 2 years Boeing would have automatically supplied me with one.
Hmmm... I'm trying to recall when it was that Boeing started "automatically" supplying desktop PCs. I know they didn't do that when I started here (1985). I know they first started supplying the office assistants (they were "secretaries" back then) with Wang word processors not too long after I started. I seem to recall my boss getting a PC at some point. I can't recall exactly when, but it had to be five years or more after that. I was hired at Boeing as a software engineer (my degree is in mechanical engineering) so I always had access to a computer, but it was always either a computer for the system we were designing or it was test support equipment. But a desktop computer for general use didn't occur for a long time. Paul
Mar 27 2009
parent Walter Bright <newshound1 digitalmars.com> writes:
Paul D. Anderson wrote:
 Walter Bright Wrote:
 It's interesting because it's right before computers moved in and 
 replaced everything. If I'd stayed another year, I'd have bought
 (with my own funds) a desktop computer and put it on my desk. I'm
 sure that within 2 years Boeing would have automatically supplied
 me with one.
Hmmm... I'm trying to recall when it was that Boeing started "automatically" supplying desktop PCs. I know they didn't do that when I started here (1985). I know they first started supplying the office assistants (they were "secretaries" back then) with Wang word processors not too long after I started. I seem to recall my boss getting a PC at some point. I can't recall exactly when, but it had to be five years or more after that. I was hired at Boeing as a software engineer (my degree is in mechanical engineering) so I always had access to a computer, but it was always either a computer for the system we were designing or it was test support equipment. But a desktop computer for general use didn't occur for a long time.
When I was there the Wangs were in a separate, locked room. The operators had to go through a Wang training course, and then they "Wanged" full time. I picked up the Wang manual and started using one in about 5 minutes, which really annoyed the manager of the Wang pool. (The managers of these fiefdoms worked hard at creating a mythology around their priestesses of the Wangs. In reality, it wasn't harder than any J-random text editor.) I had blustered my way into the Wang pool because I was expected to handwrite the whole document and then hand it over to a Wang priestess to Wang it in. No way was I going to waste time doing that.
Mar 27 2009
prev sibling next sibling parent bearophile <bearophileHUGS lycos.com> writes:
Walter Bright Wrote:

 Inspired by Don Clugston's recent compiler patch.
I have appreciated the story at the beginning and I always like the work of Don. Every one that tries to both speed up things and reduce the probability of bugs has my appreciation. I have to try write some programs with a compiler with such sNaNs, and then I'll be able to tell if I like this feature.
I tried turning on -Wuninitialized to compile dmd under Linux using g++, and
got pages of spurious warnings about possible use of uninitialized variables,
every one of which was wrong.<
on dotnet. My limited experience in it shows me that most of the uninitialized variables warnings it gives are for the better. So for me the situation to me doesn't look as bad as you say. Bye, bearophile
Mar 27 2009
prev sibling parent reply dennis luehring <dl.soluz gmx.net> writes:
Walter Bright schrieb:
 Inspired by Don Clugston's recent compiler patch.
 
 http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/
what about Witold Baryluk "Converting FPU exceptions to D" comment on the codetalk article - seems to be another good extension for the fpu exception party http://smp.if.uj.edu.pl/~baryluk/d/onpd/onp/ddoc/floatexp.html http://smp.if.uj.edu.pl/~baryluk/d/onpd/onp/arithmetic/interval/floatexp.d
Mar 29 2009
parent Don <nospam nospam.com> writes:
dennis luehring wrote:
 Walter Bright schrieb:
 Inspired by Don Clugston's recent compiler patch.

 http://www.reddit.com/r/programming/comments/87vqv/signaling_nans_rise_again/ 
what about Witold Baryluk "Converting FPU exceptions to D" comment on the codetalk article - seems to be another good extension for the fpu exception party http://smp.if.uj.edu.pl/~baryluk/d/onpd/onp/ddoc/floatexp.html http://smp.if.uj.edu.pl/~baryluk/d/onpd/onp/arithmetic/interval/floatexp.d
Definitely. The signalling NaN change is only the compiler side of things. We need library support as well. BTW, is Witold contactable? There's a lot of his code that I'd love to see in a standard library.
Mar 30 2009