#328396 pthread_rwlock_wrlock() hangs even after all locks should be unlocked!

Package:
libc6
Source:
glibc
Description:
GNU C Library: Shared libraries
Submitter:
Paul Taylor
Date:
2018-12-11 17:48:02 UTC
Severity:
important
#328396#5
Date:
2005-09-15 03:16:20 UTC
From:
To:
where the rwlock variable should be free of all locks,
it still hangs on a pthread_rwlock_wrlock(), but
recursively/cascades just find on a
pthread_rwlock_rdlock()s. What can I do to fix this?
#328396#10
Date:
2005-09-15 06:23:30 UTC
From:
To:
reassign 328396 glibc
tags 328396 + moreinfo

thanks

Paul Taylor <progmanpaul@yahoo.com> writes:

Send us a test case, and tell us what version on which architecture
you are using.
#328396#17
Date:
2005-09-15 06:23:30 UTC
From:
To:
reassign 328396 glibc
tags 328396 + moreinfo

thanks

Paul Taylor <progmanpaul@yahoo.com> writes:

Send us a test case, and tell us what version on which architecture
you are using.
#328396#22
Date:
2005-09-15 17:05:06 UTC
From:
To:
http://dls.palacecommunity.com/open/mansion/mansion.tar.gz

Has all my source code (free software) server to
replace The Palace (thepalace.com) Linux server. You
can imagine how big of a drag trying to support 700
connections or more on a single threaded server could
be lol. So I'm trying to make a multi-threaded version
of the server as it should've been done the first time
(in my eyes of course). This server is far from
completetion, but getting there...

If you check the logs/mansion.log file after a run
down, it runs just as my previous message says,
recursive read threads, after everything is unlocked,
it still hangs on a new write lock. I use functions
like pluginInfo() and apiInfo() to recursively check
to see if for instance, an 'ID' is already in use. But
it's when I try to register a second API under the
pointer/chain system, that it hangs upon the
pthread_rwlock_wrlock().

I'm using the latest x86 (686) Intel 1.8 GHz P4
(before HyperThreading) Debian, 2.6.10 kernel, but my
distro is MEPIS 3.3.1-1 (forked off of Debian) -
(mepis.org). My 'libstdc++6' is at (4.0.1-7), don't
know what package version 'glibc' is up to...

If you have any other questions, I'll be happy to
assist! Thanks for the reply and have a nice day!

--- Falk Hueffner <falk@debian.org> wrote:
#328396#27
Date:
2005-09-16 20:57:36 UTC
From:
To:
I forgot to mention along with that mansion.tar.gz
package, instead of a normal Makefile I use a bash
script to compile with g++.

'./mf mansion' is what you'd type to compile.

Sorry if I caused any inconvience(s)...

--- Falk Hueffner <falk@debian.org> wrote:
#328396#32
Date:
2005-09-16 21:38:58 UTC
From:
To:
Report:

If I comment out the PTHREAD_LOCK_UNLOCK, #define for
pthread_rwlock_unlock(), then the problem dissappears
for that thread only. I know that's not what's suppose
to happen but just reporting what I've found through
observation.

--- Falk Hueffner <falk@debian.org> wrote:
#328396#39
Date:
2018-12-11 14:39:46 UTC
From:
To:
Dear Google User,

 Your email has emerged winner on the ongoing Google End of year promotion 2018.

  Read the attached letter for more information on how to receive your funds.

 Sundar Pichai
Chief Executive Officer of Google LLC