Date: Thu, 7 Sep 2000 16:09:28 -0600 (MDT) From: Nate Williams <nate@yogotech.com> To: Bruce Evans <bde@zeta.org.au> Cc: Nate Williams <nate@yogotech.com>, Andrew Gallatin <gallatin@cs.duke.edu>, marcel@cup.hp.com, freebsd-emulation@FreeBSD.ORG Subject: Re: IBM JDK fails due to lack of SA_SIGINFO support Message-ID: <200009072209.QAA06661@nomad.yogotech.com> In-Reply-To: <Pine.BSF.4.21.0009080844380.30205-100000@besplex.bde.org> References: <200009071533.JAA05353@nomad.yogotech.com> <Pine.BSF.4.21.0009080844380.30205-100000@besplex.bde.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> > > > The change you're referring to isn't all that much different that what > > > > I'd been doing previously (setting MINSIGSTKSZ to 2048 in sys/signal.h). > > > > > > It is quite different. You change is not incorrect, but the committed > > > versions of Nate's patch results in memory beyond the end of the stack > > > being clobbered if the stack is actually use. > > > > Ouch. Why is that? I would have thought that the signalstack's were > > 'essentially' unallocated space, so by bumping up the size, we were > > giving the application more space to work with, not less. > > Alternative signal stacks are allocated by the application. And the size is allocated inside the application's space? I thought the size was allocated inside the kernel, hence the need for the system call. > Bumping the size in the kernel corresponds to using unallocated space > beyond the end of the space allocated by the application. So how would you propose fixing this? In Linux, the minimum size is << FreeBSD's minimum size. Either we decrease FreeBSD's minimum size or we abort the request, causing these applications to fail. Nate To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-emulation" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200009072209.QAA06661>