|
Thank you CP staffers and regulars for creating a friendly and helpful climate here.
I just got done working really hard on ostensibly cross platform project but i couldn't compile on windows because my windows machine is down, and then @Randor came along with a shiny MSVC compiler and some C++ expertise and put in serious work helping me shore up my code.
And that's the sort of thing I have actually come to count on here, and it occurred to me that I do just now, and so that is the inspiration for this message.
Y'all are great. This site is great. There's magic here.
It's why you aren't getting rid of me any time soon. 
Real programmers use butterflies
|
|
|
|
|
I expected your new job to markedly reduce your presence on this site, but it seems I underestimated your tenacity!
|
|
|
|
|
You aren't getting rid of me that easily.
In seriousness though, I love to create, and so working inspires more creation. That and really, I only work part time, but in spurts, so I absences shall be abrupt and short lived.
You're stuck with me. 
Real programmers use butterflies
|
|
|
|
|
honey the codewitch wrote: I only work part time, but in spurt
In the fullness of time, and speaking of such things is very timely at this time, I realize a one-time mistake I made. Unfortunately, such is hindsight, that one can only pass on learned wisdom and hope it will help others.
Take my advice: next time be born rich.
OK - what's done is done and you really cannot blame yourself - yet, it does lead one to realize such mistakes, so early in life, can have long term consequences.
42
| Ravings en masse^ |
|---|
| "The difference between genius and stupidity is that genius has its limits." - Albert Einstein | | "If you are searching for perfection in others, then you seek disappointment. If you seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|
I mean, I probably would have chosen to not be schizo if I had been able to make those kinds of choices. Money has never really been anything other than a temporary obstacle for me. I don't really care about money, or I should say, it's not money that motivates me. The only time i care about it is if i don't have any and i happen to need some. I've restructured my life such that this rarely happens.
Real programmers use butterflies
|
|
|
|
|
Nothing about your on-line persona qua diva, your productivity, your obvious intellect, suggests schizophrenia.
The crucible of "not fitting in," the forge and hammered anvil of capricious social rejection in the often vicious rip-tides of adolescence, the presence of unusual abilities ... all these, in some unfathomable way, remind me of:
"I speak in a poem of the ancient food of heroes: humiliation, unhappiness, discord. Those things are given to us to transform, so that we may make from the miserable circumstances of our lives things that are eternal, or aspire to be so." Jorge Luis Borges May you have a blessed year.
"Your task is not to seek for love, but merely to seek and find all the barriers within yourself that you have built against it." Jalaluddin Rumi
«One day it will have to be officially admitted that what we have christened reality is an even greater illusion than the world of dreams.» Salvador Dali
|
|
|
|
|
I'm not schizophrenic. I'm schizoaffective. We share some symptoms.
There are plenty of disorders under the cluster a "schizo" umbrella, some more benign, some absolutely devastating.
Have a happy new year.
Real programmers use butterflies
|
|
|
|
|
Quote: "Never imagine yourself not to be otherwise than what it might appear to others that what you were or might have been was not otherwise than what you had been would have appeared to them to be otherwise."
Duchess to Alice, in "Alice In Wonderland," by Lewis Carroll.
«One day it will have to be officially admitted that what we have christened reality is an even greater illusion than the world of dreams.» Salvador Dali
|
|
|
|
|
If this exists I want it. If it doesn't exist I might build it:
I want something that allows me to highlight a routine or chunk of code in C++ and disassemble, showing me only that section.
It's probably a lot trickier than it sounds to implement, if it's even possible.
But if such an animal exists, particularly as a VS code plugin, sign me up.
Real programmers use butterflies
|
|
|
|
|
When I did a lot of C coding (at least before Windows NT), I handled such things per-emptively with one of my favorite parts of C, to which I mean, the asm { } directive. It ended when directly accessing the bios and O/S, in general, became problematic and broke so many wonderful programs (they weren't called "apps" yet).
You can do it! And inline assembly is somehow remarkably satisfying.
| Ravings en masse^ |
|---|
| "The difference between genius and stupidity is that genius has its limits." - Albert Einstein | | "If you are searching for perfection in others, then you seek disappointment. If you seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|
Are you kidding me?
I'm still miffed that I haven't figured out a better way to do this (gcc on x86/64):
if(someTemporaryCheckConditionForDebugging) asm("int $3");
Real programmers use butterflies
|
|
|
|
|
You're making me think. How to do this without excessive pain?
OK - how about a flank attack? Temporarily hide the source of the debug code and when an error is thrown (force one?) then when asked to debug then VS will open up what it has, which is the asm code? It used to open up at the right place - so a cleverly placed somewhat fatal error should bring your to where you need to be.
| Ravings en masse^ |
|---|
| "The difference between genius and stupidity is that genius has its limits." - Albert Einstein | | "If you are searching for perfection in others, then you seek disappointment. If you seek perfection in yourself, then you will find failure." - Balboos HaGadol Mar 2010 |
|
|
|
|
|
With Visual Studio I can set conditional breakpoints. Although even there, I fall back on this technique, because variants are usable for it everywhere - even on command line builds, even with GCC, and even in C#!
Real programmers use butterflies
|
|
|
|
|
If it is .net code then you can use the free tool that I use for C#:
dotPeek: Free .NET Decompiler & Assembly Browser by JetBrains[^]
I don't know if any other tool that is free to use that does the same thing for non .net code.
This tool reads in a .dll and shows you all the members, methods, classes, etc. in tree form (and searchable). I have used it to port over VB.net code to C#, that was only in a .dll (missing source code). It helps a lot.
... of course, providing the assembly code is was not obfuscated first. 
|
|
|
|
|
Yeah, I use VS's builtins, reflector or ildasm for that. This is unmanaged code though so you don't have type information in a happy metadata file to work with, and the C++ compiler may take that neat little struct you made and totally deconstruct it into its component parts and scatter it across your binary.
So things get ... complicated.
Real programmers use butterflies
|
|
|
|
|
I suppose that you are curious as to how that multiply affected your code.
movsd xmm0, QWORD PTR __real@4024000000000000
; 457 : st.real+=(cp*(pow(10, (st.flags.fracCount * -1))));
neg eax <------------That's the sign reversal
xorps xmm1, xmm1
|
|
|
|
|
I assumed that, but it's nice to verify. One thing I miss about Visual Studio is getting disassembly out of your code was pretty easy. No wrangling with much.
What's interesting to me is what registers it's using. looks like SIMD instructions.
Real programmers use butterflies
|
|
|
|
|
Actually, it's not that hard. The hardest part is determining how far to go. What I mean is I have a little library that will disassemble X86 code up to the 32-bit, Pentium level. It does not do X86-64 stuff. It began life as a 486 disassembler and I added more to it for the Pentium instruction set and tweaked it a bit more. Here's the prototype of the entry point :
int DisAsm( unsigned char *buffer, int length, FILE *fpout ); As I mentioned, I'm not sure how to determine the length in compiled code. I used this when I wrote a C-like scripting language and then converted its java-style, interpreted byte code into Pentium machine code and this let me see the emitted code so, of course, I knew the length. You could probably set up a table of your functions addresses and use some pointer arithmetic to determine the bounds of each function's code. You could also use this along with a PE file reader and make a disassembler for executable files if you wanted to. FWIW, I have compared its output with listings output from VS C++ and they agreed.
Anyway, I can share this with you if you would like it. I found it years ago and it has a GNU license. Here's a comment from it : Quote: The code was originally snaffled from the GNU C++ debugger, as ported
to DOS by DJ Delorie and Kent Williams (williams@herky.cs.uiowa.edu).
Extensively modified by Robin Hilliard in Jan and May 1992.
"They have a consciousness, they have a life, they have a soul! Damn you! Let the rabbits wear glasses! Save our brothers! Can I get an amen?"
|
|
|
|
|
Cool. I probably don't need it right now but that is cool code. 😎
Real programmers use butterflies
|
|
|
|
|
Something that would enhance a stack trace log (on an exception, for example) would be including each stack frame and the contents of registers so that the values of locals could be looked at during debugging.
|
|
|
|
|
Cool for making core dumps useful maybe 
Real programmers use butterflies
|
|
|
|
|
Maybe is right, because it can be hard to tell where the compiler squirreled the locals away!
|
|
|
|
|
Running into that right now, looking at my .S file for a project compiled with all optimizations enabled, and the asm code is all very postmodern.
What is a type, really? Definitely my compiler is eschewing grand narratives about my data and getting very creative about how it interprets it. I'm having a hard time keeping up.
Real programmers use butterflies
|
|
|
|
|
When we went from a stack machine (running on a CPU developed in-house during the 1980s) to the Motorola 68K series, it created lots of work for our compiler and tools groups. 
|
|
|
|
|
Writing that machine code generator is one of the funnest things I have ever done. It was done purely for performance reasons. We were trying to sell a targeted application framework and we got lots of questions about the performance of the script engine. As it turns out, it was more than adequate and get even more so as machines got faster but people didn't buy that so I figured out a high performance option. Once I got it done it was fascinating to evaluate its performance. I found that it was neck and neck with Visual C++'s unoptimized code and I was very, very happy with that. The interesting thing was comparing the performance with VC++ code that was optimized. That really showed me the power of optimization because in some cases I saw a 50% improvement. In the stuff I was most interested in, which was numerical computations, it was more like 20% but that is significant. Back then, the fastest hardware was just getting into the 200MHz range so optimization was very important. Now, we are in the multi-gigahertz range and CPUs barely register any usage at all when running that application, which is still being actively deployed.
"They have a consciousness, they have a life, they have a soul! Damn you! Let the rabbits wear glasses! Save our brothers! Can I get an amen?"
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
