This is the mail archive of the libc-alpha@sources.redhat.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: Re-use of user-defined stacks for threads

To: drepper at cygnus dot com
Subject: Re: Re-use of user-defined stacks for threads
From: Wolfram Gloger <Wolfram dot Gloger at dent dot med dot uni-muenchen dot de>
Date: Wed, 29 Nov 2000 12:36:48 +0100 ("MET)
CC: libc-alpha at sources dot redhat dot com
References: <20001125113341.29693.qmail@md.dent.med.uni-muenchen.de> <m3n1ejst6b.fsf@otr.mynet.cygnus.com>

Hello,

> I agree with the analysis but the patch is not OK.  It creates a race.
> If the manager is restarting the joining thread before the latter
> actually suspended itself we have a problem.

That is correct, however aren't there already similar (minuscule)
races in the code, eg. in __pthread_create_2_1()'s use of suspend()?

> Probably the joining thread should deallocate the stack somehow.

That would be possible, but complicate things quite a bit, because
currently the manager only handles these things.

> Don't know yet what the solution should be.

I'll think about it some more.

Regards,
Wolfram.

References:
- Re-use of user-defined stacks for threads
  - From: Wolfram Gloger
- Re: Re-use of user-defined stacks for threads
  - From: Ulrich Drepper

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