This is the mail archive of the libc-hacker@sourceware.cygnus.com mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: libc/836: There appears to be a bug in the sigmask handling in Linuxthreads.



William,

please note that I've CC'ed the libc developer list - perhaps somebody 
with more experience in debugging linuxthreads problems on alpha might 
help.

Richard, do you have any idea - it really looks as if this is an alpha 
related bug (if you don't have the thread, I can easily send it).


>>>>> William T Jones writes:

William> At 03:34 AM 11/1/98 -0500, you wrote:
>> 
>> `Andreas Jaeger' changed the state to `suspended'.
>> 
>> 
>> State-Changed-From-To: open-suspended
>> State-Changed-By: jaeger
>> State-Changed-When: Sun Nov  1 03:33:23 1998
>> State-Changed-Why:
>> I'm suspending (suspending is different from closing a report, it
>> means that nobody is looking into it now but somebody might do it
>> later) this report now since I don't have enough time to check this in
>> detail and I prefer self-contained tests programs that don't need
>> addional libraries/programs..
>> 
>> The bug you've noticed might be caused by any of the following:
>> - a compiler/linker/... bug
>> - a bug in glibc
>> - a bug in mysql
>> - a bug in the kernel
>> 
>> Matthias Urlichs reported that mysql works for him (AFAIK on Intel)
>> running Linux 2.0.35 and 2.1.125 without any problems at all with
>> glibc2.  So it might be - as you stated already - a bug unique to
>> alpha.
>> 
>> If you come up with a small, self-contained, test program which fails
>> (e.g. causes locks) on your system (but shouldn't fail according to
>> the documentation), please send it to us and somebody will look into
>> it.
>> 
>> Andreas

William> So much for my enthusiasm and recent promoting for the "open" UNIX
William> community.  I am convinced this is an alpha related bug, as many (other
William> than Matthias Urlichs) are successfully running mySQL on Intel.  I have
William> complied with suggestions from Michael Widenius a mySQL developer, and will
William> now have less of an argument against using commercial "enterprise solution"
William> database servers.  I have offered my services which, with a little
William> direction, I feel would have been of use.  I do not have glibc source to
William> work with, nor am I a kernel developer.  As such I guess I am totally
William> dependent on others.  Oh well...what do you do.

Sorry, glibc development is done by volunteers and I don't have the
time now to come up with a test case to reproduce the bug you're
noticing - and that's all we need at the moment, just one or two small
programs to reproduce this.

Andreas
-- 
 Andreas Jaeger   aj@arthur.rhein-neckar.de    jaeger@informatik.uni-kl.de
  for pgp-key finger ajaeger@aixd1.rhrk.uni-kl.de


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]