Re: Fixing Re: Is libtool being maintained at all?

[Dalibor Topic]

Hi Albert,

Albert Chin wrote:
> On Mon, Jul 14, 2003 at 08:25:06PM +0200, Dalibor Topic wrote:
>
> > The proper way to fix it could be to do a binary search over multiples 
> > of 2 instead of the linear search emplyed now. Use a resonable minamal 
> > value like 2**13, which leaves only a few options between 2**13 and 
> > 2**17, which is the current upper limit for the test. Starting with 
> > 2**15, you only need three tests to figure out the maximal command line 
> > length, instead of 17. How does that sound?
>
>
> How do you start out with 2^13 printable characters?

I haven't really programmed m4 yet, so I can't say for sure. The 
existing test apparently does two things:
* spawn some kind of a new shell
* concatenate test string with itself

So for example just doing a loop that concatenates the test string with 
itself, thereby doubling its size on each iteration, until the length is 
2**13 would be simple. You don't really need to measure test string 
length, just terminating the loop when the counter becomes 14 should be 
enough. This would save 13 shell spawnings.

For an even better effect, just insert an 8K test string into 
libtool.m4. It's almost 200k already, I doubt another 8k would hurt 
much. That saves another 13 string concatenations and tests. It's also 
much simpler than the first method, where one has to be careful with 
loop limits.

If noone volunteers to write it, and I end up doing it, I'll take the 
second method, since we need all the speed we can get on slow platforms. 
 If you have a better idea, or comments to my proposals, I'd like to 
hear them.

cheers,
dalibor topic