php.net |  support |  documentation |  report a bug |  advanced search |  search howto |  statistics |  random bug |  login

go to bug id or search bugs for

Bug #47824 64-bit build make check segmentation fault
Submitted: 2009-03-29 03:31 UTC Modified: 2009-04-08 01:00 UTC
Votes:1
Avg. Score:5.0 ± 0.0
Reproduced:1 of 1 (100.0%)
Same Version:0 (0.0%)
Same OS:0 (0.0%)
From: jr at e-integration dot net Assigned:
Status: No Feedback Package: Reproducible crash
PHP Version: 5.2.9 OS: Solaris 10 SPARC u6
Private report: No CVE-ID: None
View Developer Edit
Welcome! If you don't have a Git account, you can't do anything here.
If you reported this bug, you can edit this bug over here.
(description)
Block user comment
Status: Assign to:
Package:
Bug Type:
Summary:
From: jr at e-integration dot net
New email:
PHP Version: OS:

 

 [2009-03-29 03:31 UTC] jr at e-integration dot net 
Description:
------------
# make

The make completes without error.

# ./sapi/cli/php --version
PHP 5.2.9 (cli) (built: Mar 26 2009 23:43:41)
Copyright (c) 1997-2009 The PHP Group
Zend Engine v2.2.0, Copyright (c) 1998-2009 Zend Technologies

# make test

Build complete.
Don't forget to run 'make test'.

Segmentation Fault - core dumped
make: [test] Error 139 (ignored)

Build environment:  Solaris 10 SPARC u6; 64-bit used gcc 4.2.4 and GNU make 3.81

Expected result:
----------------
make test creates two core files if you have coreadm set like so:

# coreadm
     global core file pattern:
     global core file content: default
       init core file pattern: /var/core/init_core.%n.%z.%u.%f.%p
       init core file content: default
            global core dumps: disabled
       per-process core dumps: enabled
      global setid core dumps: disabled
 per-process setid core dumps: enabled
     global core dump logging: enabled

# ls -lt /var/core/
total 106272
-rw------- 1 root root 27865628 Mar 27 10:38 init_core.halo.global.0.php.7926
-rw------- 1 root root 26489372 Mar 27 10:38 init_core.halo.global.0.php.7924
-rw------- 1 root root 27865628 Mar 27 10:30 init_core.halo.global.0.php.7757
-rw------- 1 root root 26489372 Mar 27 10:30 init_core.halo.global.0.php.7751

Patches

Pull Requests

History

AllCommentsChangesGit/SVN commitsRelated reports
 [2009-03-29 03:34 UTC] jr at e-integration dot net 
Used the following build tools:

gcc 4.2.4
gmake 3.81
re2c 0.13.5
bison 2.4.1
 [2009-03-31 07:09 UTC] jani@php.net 
Can you provide the backtrace of the crash..?
 [2009-04-02 03:50 UTC] Michael dot Zehmeister at dier dot tas dot gov dot au 
I am experiencing the exact same issue as the original submitter on a Sparc 64-bit sun4u system.

I recompiled with --enable-debug to obtain a backtrace to submit but having done so make test no longer seg faults.  This is reproducible: without debug it seg faults at the start of make test, with debug enabled and all other options identical it doesn't.  Given that I can't get a backtrace what alternative information can I provide?

---

Build details:

GNU gcc version 4.3.3 (GCC)
GNU Make 3.81

./configure \
--enable-debug \
--prefix=/opt/php \
--with-config-file-path=/opt/php/etc \
--enable-inline-optimization \
--with-gd \
--with-jpeg-dir=/opt/libjpeg \
--with-png-dir=/opt/libpng \
--enable-bcmath \
--with-openssl=/opt/openssl \
--with-zlib=/opt/zlib \
--enable-wddx \
--enable-ftp \
--disable-cgi \
--with-curl=/opt/curl \
--with-gettext \
--enable-sockets \
--enable-pcntl \
--with-pspell=shared,/opt/aspell \
--enable-mbstring=all \
--enable-shmop \
--with-pear \
--with-ldap=/opt/openldap \
--enable-exif \
--with-freetype-dir=/opt/freetype \
--with-pdo-oci \
--with-oci8=/opt/oracle/product/10.2.0/client \
 --with-apxs2=/opt/httpd/bin/apxs \
 --enable-gd-native-ttf \
> ../doc/config.out 2>&1
 [2009-04-02 23:30 UTC] Michael dot Zehmeister at dier dot tas dot gov dot au 
I no longer think this is a PHP bug.  I believe the problem is actually with the build of gcc.

Configuring php without --enable-debug but setting CFLAGS to -g doesn't segfault (indicating the problem is with gcc).  So I tried again after setting CFLAGS to -O1 and make test runs fine.

As such I think the bug is probably in the gcc optimisation process.  I'll investigate this further, but if this is the case then it's not relevant to PHP.

If the original poster wants to try 'export CFLAGS=-O1' to see if he encounters the same behaviour then this bug can probably be closed.

Cheers.
 [2009-04-08 01:00 UTC] php-bugs at lists dot php dot net 
No feedback was provided for this bug for over a week, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".
 
PHP Copyright © 2001-2025 The PHP Group
All rights reserved. 		Last updated: Mon Nov 03 16:00:02 2025 UTC