Bug in compiler

The interesting thing with printf - and you can see examples of it if you search this forum - is that the Keil tools are good at analyzing how you use it.

printf() can require linking of more or less code. In this case, you didn't actually perform any "formatting", so you got a minimalistic overhead.

You think so?

Integer formatting:

#include <stdio.h>


void main(void)
{
   int d=1;

   printf("Hello world %d\n",d);

   while(1);
}

000003H   00035EH   00035CH   BYTE   UNIT     CODE           ?PR?PRINTF?PRINTF

No formatting:

#include <stdio.h>


void main(void)
{
   float f=1.0;

   printf("Hello world\n");

   while(1);
}


00051BH   000989H   00046FH   BYTE   UNIT     CODE           ?PR?PRINTF?PRINTF

Floating point formatting:

#include <stdio.h>


void main(void)
{
   float f=1.0;

   printf("Hello world %f\n",f);


   while(1);
}

00051BH   000989H   00046FH   BYTE   UNIT     CODE           ?PR?PRINTF?PRINTF

Doesn't seem to be much analysis going on there.

"Doesn't seem to be much analysis going on there."

There is obviously sufficient analysis going on to allow it to omit a load of stuff if the program uses no floating point.

Presumably, if the program contains floating point and uses printf, the compiler cannot safely assume that there aren't any dynamically-created floating-point formats - so it just has to include them, just in case...

"There is obviously sufficient analysis going on to allow it to omit a load of stuff if the program uses no floating point."

From previous experience of Microsoft compiler technologies, I would say that it is not so much analyzing that the stuff can be omitted but rather that they should not be included.

Specifically - When floating point is required in a module, it makes reference(s) to any external functions that are required for the compiler chosen task.

At link time these external references would normally pull in the libraries that satisfy the external references.

So - It is not normally the formatting string of the printf that is interpreted by the compiler to determine whether the libraries that are included, but rather the fact that a floating point access is somewhere in the project.

Remember people, C is even more limited!

Just curious, but how is C more limited than C++?

That's like saying that assembly is more limited than C. Or, that the Russian language is more limited than German.

Jon

There is obviously sufficient analysis going on to allow it to omit a load of stuff if the program uses no floating point.

Yes. But that doesn't require, or imply, any analysis of the printf() calls themselves.

Presumably, if the program contains floating point and uses printf, the compiler cannot safely assume that there aren't any dynamically-created floating-point formats - so it just has to include them, just in case.

Indeed.

I want to use C++ because I know C++.

Sure, but be honest and say "I have no idea of anything else, if C++ is right for this job or not does not matter"

NOW DO ANSWER have you ever done ANY small embeeded before, we all know that the answer is 'no' but it would be nice to hear it from you.

Erik

In this instance, dismissing C++ out of academic principal I would say is narrow minded.
NOBODY has "dismissed C++ out of academic principal" many have "dismissed C++ out of knowledge of the architecture of the '51"

Erik

The problem you are going to run into if you use the Ceibo + Keil
and you reply

void main(void)
{
   printf("Hello world\n");

   while(1);
}

where does the Ceibo C++ come in, in the above mr smokied sardine

Erik

The problem you are going to run into if you use the Ceibo + Keil
and you reply

void main(void)
{ printf("Hello world\n");

while(1);
}

where does the Ceibo C++ come in, in the above mr smokied sardine

Oh dear, here we go again.

I quoted two sentences as follows:

As I have said before: The problem you are going to run into if you use the Ceibo + Keil combination is that the evaluation version of Keil C51 only allows a code size of 4 kB. Any complex string formatting function (this includes printf and cout) will easily need upwards of 10 kB code space.

You snipped this down to half a sentence:

The problem you are going to run into if you use the Ceibo + Keil

Thereby removing all the content I was actually replying to.

You then quoted part of my response, this time snipping out all the text that gave meaning to the code snippet.

Go back and read my post again. If you still do not understand it let me know and I will explain it to you in whatever level of detail is necessary.

Go back and read my post again. If you still do not understand it let me know and I will explain it to you in whatever level of detail is necessary.

I do not give a hoot about your post, the subject of this thread is NOT that printf works in C, which we all know, but C++ on the '51.

Erik

FYI, at the start of the thread I was (as I admitted) not aware of the nuances.

The fact that you still call the difference between two separate programming languages a "nuance" is rather strong evidence that you're still just as unaware.

Erik,

I think in this case you really have missed Jack's point. He was showing the m51 file to indicate the using printf() in Keil didn't increase the code by 10kB, but rather by 1kBish which was an order-of-magnitude less "badness" than the original statement. It had nothing to do with C++, but was just continuing illustration of what he sees as bad in the responses on this forum -- namely that sometimes the estimates people provide from experience can be exaggerated and overly-emphatic.

Also, just as a point of reference for your future usage: The name he's chosen (Jack Sprat) doesn't have anything to do with sardines. It's from an old nursery-rhyme (though I don't know it's origin) that begins like this: "Jack Sprat could eat no fat. His wife could eat no lean."

-Jay Daniel

Also, just as a point of reference for your future usage: The name he's chosen (Jack Sprat) doesn't have anything to do with sardines.

I know that, but sardines are canned :)

Erik

"[Jack Sprat] was showing the m51 file to indicate the using printf() in Keil didn't increase the code by 10kB, but rather by 1kBish"

But did it show that?

It showed that the size of ?PR?PRINTF?PRINTF is 1K-ish, but it doesn't consider what other stuff may also get pulled-in as a result of having printf that wouldn't other wise have been included.

I haven't had the time for a detailed look at the map file, but the summary line from building a simple "Hello, world" example as shown indicates that the total CODE space usage is on the order of 2K...

It probably also varies with the Memory Model chosen...

So: still not 10K, but it certainly does show that a simple printf can easily use up virtually the whole CODE size limitation of the Evaluation version...!

"Just curious, but how is C more limited than C++?"

C++ is (basically) a superset of C

Embedded C++ is (basically) a subset of C++

And Embedded C++ is (basically) a subset of C++

So, expressed in terms of facilities:

C < Embedded C++ < C++

I said it is more limited purely because it is considered to be lower level, less abstract etc

So it could be extended to the expression:

Assembler < C < Embedded C++ < C++

The extrapolation to "Russian is more limited than German" is clearly wrong.

But everyone knows that Esperanto is superior to all others ;)