strange "less" behaviour on big files

[Kai Mosebach]

Hi all,

i stumbled over a problem with less, when "lessing" big files.
i lessed a 1.0 gig file filled with zeros and less seems to slurp in as much as 
it can get at one time! this happens only on 5.2-RELEASE, a 4.9-RELEASE does 
fine.

my top says (on a 512 mb machine)

20041 root      76    0   382M   193M pfault   0:03  7.03%  7.03% less

furthermore i think the state "pfault" in top is pretty strange, and after a 
minute or so the process coredumps

backtrace is not very useful though ...

#0  0x2811dd4f in kill () from /lib/libc.so.5
(gdb) bt
#0  0x2811dd4f in kill () from /lib/libc.so.5
#1  0x281127f8 in raise () from /lib/libc.so.5
#2  0x2818af02 in abort () from /lib/libc.so.5
#3  0x2818967e in tcflow () from /lib/libc.so.5
#4  0x28189f1b in tcflow () from /lib/libc.so.5
#5  0x2818a2d6 in malloc () from /lib/libc.so.5
#6  0x28186ae1 in calloc () from /lib/libc.so.5
#7  0x0805255d in clear ()
#8  0x080535a4 in clear ()
#9  0x08053ab6 in clear ()
#10 0x08053dfb in clear ()
#11 0x08051366 in clear ()
#12 0x080523bf in clear ()
#13 0x0804dbc0 in clear ()
#14 0x0804dc27 in clear ()
#15 0x0804dfcf in clear ()
#16 0x080496b6 in free ()
#17 0x08049242 in free ()


Any ideas ?

Regards Kai

-- 
SapDB for FreeBSD --- see http://www.komadev.de/sapdb