Configuration Information [Automatically generated, do not change]:
Machine: i486
OS: linux-gnu
Compiler: gcc
Compilation CFLAGS: -DPROGRAM='bash' -DCONF_HOSTTYPE='i486'
-DCONF_OSTYPE='linux-gnu' -DCONF_MACHTYPE='i486-pc-linux-gnu'
-DCONF_VENDOR='pc' -DLOCALEDIR='/usr/share/locale' -DPACKAGE='bash' -DSHELL
-DHAVE_CONFIG_H -I. -I../bash -I../bash/include -I../bash/lib -g -O2
uname output: Linux johann 2.6.18-5-686 #1 SMP Wed Sep 26 17:54:59 UTC 2007
i686 GNU/Linux
Machine Type: i486-pc-linux-gnu
Bash Version: 3.1
Patch Level: 17
Release Status: release
(debian etch, bash 3.1dfsg-8)
Description:
aparently I hit a memory leak in the read builtin.
Repeat-By:
yes | bash -c 'ulimit -v 1000; while read a ; do :; done'
bash: xmalloc: ../bash/make_cmd.c:99: cannot allocate 2 bytes (0 bytes
allocated)
note that
yes | while read; do :; done
yes | while read a b ; do :; done
both run "forever".
this apears to be only triggered
when the number of input fields is larger
than the number of variables.
Fix:
I did not check too closely yet,
nor did I build a debug version or run it through gdb.
But from looking at the code,
I suspect that in builtins/read.def:647 ...
the allocated temporary space for t is never FREE'd.
| /* Check whether or not the number of fields is exactly the same as
the
| number of variables. */
| if (*input_string)
| {
| t1 = input_string;
| t = get_word_from_string (&input_string, ifs_chars, &e); <<<<====
| if (*input_string == 0)
| input_string = t;
| else
| input_string = strip_trailing_ifs_whitespace (t1, ifs_chars,
saw_escape);
| }
thanks for listening,
awaiting your feedback.
Lars Ellenberg
--
: Lars Ellenberg Tel +43-1-8178292-55 :
: LINBIT Information Technologies GmbH Fax +43-1-8178292-82 :
: Vivenotgasse 48, A-1120 Vienna/Europe http://www.linbit.com :
