zsh 3.0 pre6 VERY slow on nextstep 3.2

zsh 3.0 pre6 VERY slow on nextstep 3.2

[Timothy J. Luoma]

This newest version of the Z-shell is very slow.... How slow?  I  
can see each character as the cursor draws the PROMPT and as it  
redraws old history lines (up arrow).

Is anyone else noticing this on their OS?

Thanks
TjL

--------------------------------------------------------------------
Timothy J. Luoma <luomat@nerc.com>       http://www.nerc.com/~luomat
NeXT info via email: send message with SUBJECT: send-ascii info
NeXT Information:  http://www.nerc.com/~luomat/next
NeXT Printer Info: http://www.nerc.com/~luomat/next/printerinfo.html

Re: zsh 3.0 pre6 VERY slow on nextstep 3.2

[Bart Schaefer]

On Aug 4,  9:12pm, Timothy J. Luoma wrote:
} Subject: zsh 3.0 pre6 VERY slow on nextstep 3.2
}
} This newest version of the Z-shell is very slow.... How slow?  I  
} can see each character as the cursor draws the PROMPT and as it  
} redraws old history lines (up arrow).
} 
} Is anyone else noticing this on their OS?

I'm not, on Linux, but I've only been using pre6 since a few minutes ago.

Did you compile with debugging and zsh_mem and so forth?  To what previous
version are you comparing it (so we know what changes to re-examine)?

The only recent change that would have any conceivable effect on this is
that all references to the variable `columns' in display update code have
been replaced by the ternary expression `(columns < 1 ? 80 : columns)'.
But your compiler would have to be doing something really awful for that
to have any noticable effect.

-- 
Bart Schaefer                             Brass Lantern Enterprises
http://www.well.com/user/barts            http://www.nbn.com/people/lantern

New male in /home/schaefer:
>N  2 Justin William Schaefer  Sat May 11 03:43  53/4040  "Happy Birthday"

Re: zsh 3.0 pre6 VERY slow on nextstep 3.2

[Timothy J. Luoma]

"Bart Schaefer" on Sun, 4 Aug 1996 wrote:
> Did you compile with debugging and zsh_mem and so forth? 

No, just regular ./configure


> To what previous version are you comparing it (so we know what
> changes to re-examine)?

2.6-beta13 is what I've been using, but I've got 3.0-pre2 here too and that did not display this beahvior.

Would you need/like to see the compiling output?

Thanks for the help, I love zsh but it really is too slow to be usable...

TjL

Re: zsh 3.0 pre6 VERY slow on nextstep 3.2

[Charles Swiger]

On Mon,  5 Aug 96, "Timothy J. Luoma" wrote:
> Would you need/like to see the compiling output?
>
> Thanks for the help, I love zsh but it really is too slow to be usable...

I tried building under NEXTSTEP 3.3.  './configure' generates two core  
dumps; apparently the linker chokes trying to link POSIX stuff without  
'-posix', since building the test program which does  
tgetent((char*)0,"vt100") coredumps using either NeXT's cc or a gcc-2.7.2 I  
built.

Pressing on, I built zsh, but I didn't notice any obvious slowdowns.

-Chuck "yet-another-data-point" Swiger

      Charles Swiger | chuck@tertius.res.cmu.edu | standard disclaimer
      ---------------+---------------------------+--------------------
           I know you're an optimist if you think I'm a pessimist.

Re: zsh 3.0 pre6 VERY slow on nextstep 3.2

[Zoltan Hidvegi]

> On Mon,  5 Aug 96, "Timothy J. Luoma" wrote:
> > Would you need/like to see the compiling output?
> >
> > Thanks for the help, I love zsh but it really is too slow to be usable...
> 
> I tried building under NEXTSTEP 3.3.  './configure' generates two core  
> dumps; apparently the linker chokes trying to link POSIX stuff without  
> '-posix', since building the test program which does  
> tgetent((char*)0,"vt100") coredumps using either NeXT's cc or a gcc-2.7.2 I  
> built.

These coredumps are normal.  Configure tries call tgetent with a NULL
pointer which dumps core hence configure does not define the
TGETENT_ACCEPTS_NULL macro.

Zoltan