See this comment for a full explanation: https://bugzilla.redhat.com/show_bug.cgi?id=553834#c43
Reported upstream as: https://bugs.g10code.com/gnupg/issue1197
*ping* Is anyone working on this?
Created attachment 447791 [details] libgcrypt-invalidate-thread-cbs.patch (untested) Here is an untested patch to demonstrate what I think might be one approach to the problem. When GCRYCTL_SET_THREAD_CBS is called and we already have an interface installed, a copy is made of this new interface. If it is called several times by different libraries, we have several different interfaces to choose from. When a library is dlclosed, a new API command GCRYCTL_INVALIDATE_THREAD_CBS should be called with the same function pointers as originally given to GCRYCTL_SET_THREAD_CBS. If the invalidated interface is the one in use, this has the effect of switching to one (and initialising) of the other available interfaces. Is this a viable approach?
Tim, the best thing would be to discuss it with the libgcrypt upstream. I do not have much influence on it.
Created attachment 447988 [details] libgcrypt-invalidate-thread-cbs.patch (more work needed by upstream) Here is as far as I can go with this patch. The problem I've come up against is that libgcrypt doesn't have any way of simply re-initialising the mutexes it uses. Basically, it has no concept of how to behave if the mutexes created by the calling library become invalid. I don't know who/how to contact upstream so hopefully I can leave this with you. In the mean time I think I'll need to get libcups to serialize all calls to gnutls. :-(((
The gcrypt ath_ stuff is designed to cope with the long-since-obsolete user-space thread libraries like GNU pth and IMO should simply be ripped out and hard-coded to use POSIX (or whatever the platforms supports). At minimum we could make it DTRT by default and ignore GCRYCTL_SET_THREAD_CBS completely.
Joe: absolutely agree.
Created attachment 448055 [details] libgcrypt-use-pthread.patch (untested) Perhaps something along these lines could be applied to the Fedora package as a work-around?
This message is a reminder that Fedora 13 is nearing its end of life. Approximately 30 (thirty) days from now Fedora will stop maintaining and issuing updates for Fedora 13. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as WONTFIX if it remains open with a Fedora 'version' of '13'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version prior to Fedora 13's end of life. Bug Reporter: Thank you for reporting this issue and we are sorry that we may not be able to fix it before Fedora 13 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora please change the 'version' of this bug to the applicable version. If you are unable to change the version, please add a comment here and someone will do it for you. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete. The process we are following is described here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping
Looks like this is fixed in gnutls-2.12.x?