[prev in list] [next in list] [prev in thread] [next in thread] 

List:       koffice-devel
Subject:    Re: Portable Code (am I dreaming? :))
From:       Clarence Dang <CTRL_CD () bigpond ! com>
Date:       2002-01-24 11:22:36
[Download RAW message or body]

On Wed, 23 Jan 2002 13:39, shaheed wrote:
> Clarence,
> 3. Don't worry about the direction of bit shifts for the reason Nicolas
> gave, and don't worry about the performance of bit shifts since the
> compiler will almost certainly put the original value in a register. 

It's not the performance of bit shifts that I'm worried about but it's the overhead \
of function calls. e.g. if I had:

class SOME_IMPORTANT_STRUCTURE
{
public:
	short a;
	char b;
	long c;
}

Currently, I can just make one function call (fread) and read in the whole structure.
But if I'm going to make it portable, I would have to make 3 function calls (to read \
in a, then b, then c). Imagine a structure with 15 variables :(
n number of function calls (even inline ones) cannot beat 1 function call in terms of \
speed (n > 1). But I guess this is unavoidable anyway.

btw, If I write "fread (buffer, 4, 1, filepointer)", will this always read in a long \
int "correctly" or will I get endian issues? And also, what is the difference \
between:

fread (buffer, 4, 1, filepointer);
fread (buffer, 2, 2, filepointer);
fread (buffer, 1, 4, filepointer);

Are they all the same or can they give different output?
 
> The
> current msword code does some of this, and a typical routine looks like
> this:
> 
> unsigned MsWordGenerated::read(const U8 *in, __UNAL SHD *out, unsigned
> count) {
> U32 shifterU32;
> U16 shifterU16;
> U8 shifterU8;
> U8 *ptr;
> unsigned bytes = 0;
> 
> ptr = (U8 *)out;
> shifterU32 = shifterU16 = shifterU8 = 0;
> for (unsigned i = 0; i < count; i++)
> {
> bytes += read(in + bytes, &shifterU16);
> out->icoFore = shifterU16;
> shifterU16 >>= 5;
> out->icoBack = shifterU16;
> shifterU16 >>= 5;
> out->ipat = shifterU16;
> shifterU16 >>= 6;
> out++;
> }
> return bytes;
> } // SHD
> 
> Note that the _UNAL will disappear as soon as I get round to eliminating a
> few sizeof()s used by the current code. In the example, I have an
> overloaded read() function for the different primitive types.
> 
Can all the packing/alignment business be avoided if I get rid of sizeof()s
and instead have #define <structure name>_SIZEOF everywhere instead?

> Finally, if you can, I heartily recommend a code generator...that way, if
> you get the implementation wrong, you can at least change it.
> 
A code generator?  Do you mean generating data structures from a data format \
specification automatically? Or do you mean something else?  I mean a code generator \
sounds like a great way to reduce work :)

Thanks!
Clarence

_______________________________________________
koffice-devel mailing list
koffice-devel@mail.kde.org
http://mail.kde.org/mailman/listinfo/koffice-devel


[prev in list] [next in list] [prev in thread] [next in thread]