Daniel Jacobowitz wrote:

Or should we provide a C-level fix? Thoughts?


I'm not sure, but if you dig up the message where I posted the patch
(probably in the archives of libc-alpha at: http://sources.redhat.com/ml/libc-alpha)
then there may be an application workaround.

Are you talking about this patch? http://sources.redhat.com/ml/libc-alpha/2002-10/msg00510.html


Yes, that's the one.  If you can arrange for the rest of it to be
cleared in perl/mod_perl, it becomes a non-issue.

Similar to what was done for the cloned perl interpreters I suppose. Though I see that Doug worked around only the current_saltbits entry. He didn't touch current_salt[0-1]. So let's see if this patch works first.


David, can you please test with this patch?

Index: src/modules/perl/mod_perl.c
===================================================================
RCS file: /home/cvs/modperl-2.0/src/modules/perl/mod_perl.c,v
retrieving revision 1.156
diff -u -r1.156 mod_perl.c
--- src/modules/perl/mod_perl.c 8 Mar 2003 09:49:26 -0000       1.156
+++ src/modules/perl/mod_perl.c 17 Mar 2003 00:50:45 -0000
@@ -172,6 +172,11 @@
     endav = PL_endav;
     PL_endav = Nullav;

+#if defined(USE_REENTRANT_API) && defined(HAS_CRYPT_R) && defined(__GLIBC__)
+    /* workaround 5.8.0 bug */
+    PL_reentrant_buffer->_crypt_struct.current_saltbits = 0;
+#endif
+
     perl_run(perl);

PL_endav = endav;




__________________________________________________________________ Stas Bekman JAm_pH ------> Just Another mod_perl Hacker http://stason.org/ mod_perl Guide ---> http://perl.apache.org mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com http://modperlbook.org http://apache.org http://ticketmaster.com


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to