Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Gary Vaughan
2008/10/28 Simon Josefsson <[EMAIL PROTECTED]>: > I looked into this more, and it seems v2.4.x and older called strerror > inside cdk_strerror. However, the cdk_strerror function is itself > unused, so you can safely remove its definition from lib/opencdk/main.c. > I have made this change on the v

Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Simon Josefsson
Eric Blake <[EMAIL PROTECTED]> writes: >> I found this while linking against gnutls, which uses the string >> module, which then uses '#define strerror rpl_strerror' in its lgpl >> library when strerror is not found. That, in turn, leads to an >> undefined rpl_strerror symbol in libgnutls.so. > >

Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Bruno Haible
Gary V. Vaughan wrote: > The easy fix would be to change both instances of 'strerror(-2)' > to 'strerror(1)' in strerror.m4 The 'strerror(-2)' is there on purpose: We don't want a strerror function that crashes. The comment in strerror.m4 says it: if test $gl_cv_func_working_strerror = no; th

Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Gary V. Vaughan on 10/27/2008 11:49 PM: > The current test code inside strerror.m4 says: > > AC_LANG_PROGRAM( > [[#include > ]], > [[return !*strerror (-2);]]) > > Which is fine for Solaris 9 and 10 w

Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Simon Josefsson
I looked into this more, and it seems v2.4.x and older called strerror inside cdk_strerror. However, the cdk_strerror function is itself unused, so you can safely remove its definition from lib/opencdk/main.c. I have made this change on the v2.4.x branch as well, in case we ever make another relea

Re: strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-28 Thread Simon Josefsson
"Gary V. Vaughan" <[EMAIL PROTECTED]> writes: > [[Simon, I'm Cc:ing you for the gnutls bug at the bottom]] Thanks for looking into this. > Even with the strerror detection fixed, gnutls will still create > an unusable libgnutls.so whenever it is built on a machine that > doesn't provide a strerr

strerror.m4 test fails incorrectly on solaris 8 and earlier [and bonus gnutls bug]

2008-10-27 Thread Gary V. Vaughan
[[Simon, I'm Cc:ing you for the gnutls bug at the bottom]] The current test code inside strerror.m4 says: AC_LANG_PROGRAM( [[#include ]], [[return !*strerror (-2);]]) Which is fine for Solaris 9 and 10 which return "Unknown Error", but causes a SIGSEGV on ol