imacat wrote:
Dear Stas Bekman,

Hi. This is imacat from Taiwan.
Thank you for the patch imacat from Taiwan - would please be so kind to
try the attached version?
On Mon, 01 Sep 2008 17:24:00 -0700
Stas Bekman wrote:
imacat wrote:
Ok, so 0.06 had the same problem. That's what I needed to check.
Now, I can't reproduce the problem. So if someone else who can reproduce
it could look at the issue, that would be great. I think the difference
is x64, but it could be something else.
I have digged into your code for that "panic: free from wrong pool
during global destruction". It seems to be caused by calling setenv()
at line 75 in your C.xs, which threw out an error in in Perl
5.10. Further investigation turns out that, according to:

Perl write to the memory directly when implementing putenv(), which
might cause problem. To avoid that and use the safe (but potentially
leaky) system putenv() you may set PL_use_safe_putenv = 1.

The above does work for my Perl 5.8.8 and 5.10.0 on x86_64. I do
not know if there will be any side effect on other platforms. I made a
simple patch below, in the hope that it helps. Please tell me if you
need any more information, or if I could be of any help. Thank you.

Hash: SHA1

diff -u -r Env-C-0.07.orig/C.xs Env-C-0.07/C.xs
- --- Env-C-0.07.orig/C.xs 2008-09-01 06:26:13.000000000 +0800
+++ Env-C-0.07/C.xs 2008-09-03 00:40:50.000000000 +0800
@@ -72,6 +72,7 @@
RETVAL = -1;
+ PL_use_safe_putenv = 1;
RETVAL = setenv(key, val, override);

Version: GnuPG v1.4.9 (GNU/Linux)
Stas Bekman

Search Discussions

Discussion Posts


Follow ups

Related Discussions

Discussion Navigation
viewthread | post
posts ‹ prev | 4 of 6 | next ›
Discussion Overview
groupcpan-testers-discuss @
postedSep 2, '08 at 12:09a
activeSep 4, '08 at 3:44p

2 users in discussion

Stas Bekman: 3 posts Imacat: 3 posts



site design / logo © 2018 Grokbase