https://bugs.exim.org/show_bug.cgi?id=2344
Bug ID: 2344
Summary: Sporadic failure in test 0 from RunTest
Product: PCRE
Version: 10.32 (PCRE2)
Hardware: All
OS: NetBSD
Status: NEW
Severity: bug
Priority: medium
Component: Code
Assignee: ph10@???
Reporter: carenas@???
CC: pcre-dev@???
the call :
checkspecial '-S 1 -t 10 testSinput'
will fail sometimes in NetBSD in the call to setrlimit (src/pcre2test.c:8539)
with an errno of EINVAL and as documented in their man page[1] (apparently in
conformance with SUS)
the rationale being that the stack size used is already over 1MB and therefore
the request to set it to less of what is currently being used is invalid.
the commit message doesn't explain the rationale behind this test, and it seems
to be only relevant on deciding if -bigstack is to be honoured, so it might be
as well that the "solution" is to use instead a bigger value that would be
unlikely to fail?, why is 1 used currently?
[1]
http://netbsd.gw.com/cgi-bin/man-cgi?setrlimit+2+NetBSD-current
--
You are receiving this mail because:
You are on the CC list for the bug.
