Please describe the problem:
I tried to search in 
http://docs-pdf.sun.com/817-0689/817-0689.pdf
(that's Solaris 9 man pages, section 1, size ~5 MB, 1760 pages) -
very slow (and uses 100% CPU for the whole time)

Steps to reproduce:
- open document, wait until machine is idle
- type "/truss", start timer
- stop timer when "...% remaining to search" drops to 0%
- 520 sec

Actual results:


Expected results:


Does this happen every time?


Other information:
For comparison (on the same machine):
- Acrobat reader (7.0.9): searching for "truss" takes 41 sec 
  (measured from hitting Return in the "Find" entry until the 
  progress bar shows all pages have been scanned)
- xpdf (3.02): doesn't have an option to search for all occurences,
  but if I enter "truss" in the Find popup when at the beginning of the
  document, then click on Find repeatedly for every hit until the end
  of the document is reached, that's ~60 sec