Casper.Dik at Sun.COM wrote:
> >/home/test001/ksh93/on_build002_m1/test1/proto/root_i386/lib/libcmd.so.1
> >/lib/libcmd.so.1
> >-rwxr-xr-x   1 test001  users     186060 Apr 26 22:24
> >/home/test001/ksh93/on_build002_m1/test1/proto/root_i386/lib/libcmd.so.1
> >-rwxr-xr-x   1 root     bin        20840 Jan 23  2005 /lib/libcmd.so.1
> 
> And size?
> 
> size /lib/libcmd.so
> 2190 + 404 + 4 = 2598
> 
> "Size is what matters".
> 
> Not so much the "text" as well as the "data" as the data is unshared.

Currently it looks like this:
-- snip --
% size
/home/test001/ksh93/on_build002_m1/test1/proto/root_i386/lib/libcmd.so.1
/lib/libcmd.so.1
/home/test001/ksh93/on_build002_m1/test1/proto/root_i386/lib/libcmd.so.1:
170999 + 3556 + 10248 = 184803
/lib/libcmd.so.1: 8996 + 505 + 31 = 9532
-- snip --
around ~1.5kb can be saved here via allocating some buffers from the
stack (I have a patch for that in the queue which uses C99 features,
giving both a performace boost via avoiding the heap allocator and (in
other places) reduce the need for static buffers).
If there is still someone complaining about the increase in size then I
propose to recompile libnsl/libsocket/libc with -xstrconst which will
more than compensate the increase in libcmd... :-) 

[snip]
> >Thanks to that list libcmd.so.1 is more or less guranteed to be loaded
> >and locked in memory in a multiuser system
> 
> So what's the data/bss.

See above...

> >I agree, however we're talking about three new libraries plus some new
> >functions in libcmd. It is really not that bad as Mozilla with all it's
> >modules and ever-changing API, also the size is much smaller (42MB
> >statically linked Mozilla Seamonkey vs. ~~1.8MB statically linked ksh93
> >(debug; the stripped dynamically linked version of ksh93 is just
> >9k(!!))).
> 
> Sure; but as the man said "adding a few cycles here cannot be
> measured (except when you do it a 1000 times)".

The last time I tested this ksh93q was - for larger scipts - much faster
(and ran fewer syscalls) than Solaris /usr/bin/ksh... the question is
whether the impact of the merged libcmd to other parts of the system is
really that bad...

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) roland.mainz at nrubsig.org
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 7950090
 (;O/ \/ \O;)

Reply via email to