[pcre-dev] [Bug 1048] RunGrepTest 62 and 63 fails on s390x a…

Top Page
Delete this message
Author: Philip Hazel
Date:  
To: pcre-dev
Old-Topics: [pcre-dev] [Bug 1048] New: RunGrepTest 62 and 63 fails on s390x
Subject: [pcre-dev] [Bug 1048] RunGrepTest 62 and 63 fails on s390x and ppc64
------- You are receiving this mail because: -------
You are on the CC list for the bug.

http://bugs.exim.org/show_bug.cgi?id=1048




--- Comment #2 from Philip Hazel <ph10@???> 2010-12-17 09:21:05 ---
On Fri, 17 Dec 2010, Dan Horák wrote:

> Same situation is in RHEL-6 on ppc64, the only difference is that the default
> stack size is 10240 kB.


These tests are supposed to be checking the setting of limits so that
running out of stack does not happen; instead there is an error return
from pcre_exec(). So the size of the stack (as long as it is not
ridiculously small) should make no difference.

The expected result of test 62 is an error message "pcregrep:
pcre_exec() gave error -8...".

You'll have to help me with this, as I have no access to any
architecture other than i386. First thing is to check that the
--match-limit and --recursion-limit options are setting the correct
values. This can be checked by inserting a print() just before line 2547
in pcregrep.c. It should be straightforward to run the test on its own.
For example, test 62 is this:

./pcregrep --match-limit=1000 -M 'This is a file(.|\R)*file.' \
./testdata/grepinput

If the values of the match_limit and match_limit_recursion variables
are not set correctly, we can investigate why; if they are, the next
test is to see if they are being correctly passed to pcre_exec().

Philip


--
Configure bugmail: http://bugs.exim.org/userprefs.cgi?tab=email