Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 9 Aug 2020 20:06:22 -0400
From: Rich Felker <dalias@...c.org>
To: musl@...ts.openwall.com
Subject: Re: Revisiting sigaltstack and implementation-internal signals

On Sun, Aug 09, 2020 at 09:54:31AM +0200, Markus Wichmann wrote:
> On Sat, Aug 08, 2020 at 08:39:58PM -0400, Rich Felker wrote:
> > on it (possibly not even any signal handlers installed), and (2)
> > whether we should care about breaking code that swaps off of and back
> > onto the alternate signal stack with swapcontext.
> 
> Would anything bad happen in that case? I thought, when a signal handler
> with SA_ONSTACK is invoked, the altstack is marked with SS_ONSTACK and
> will not be reset until the signal handler returns. If the handler does
> not return, and does not call sigaltstack(), then the SS_ONSTACK remains
> set, and therefore further signals with SA_ONSTACK will be delivered on
> the current stack. Otherwise, if a signal were to arrive while the
> altstack is in use, it would overwrite the old stack.
> 
> I cannot find a source code for swapcontext, but to my knowledge it
> merely combines setjmp() and longjmp(), right? (setjmp() for the current
> context and longjmp() for the other one). So no call to sigaltstack().

My understanding is that SA_ONSTACK is just reported by the kernel if
the current stack pointer is inside the alternate stack. If the
application has moved off that stack and a signal arrives, it has
nowhere to know "where in the alternate stack it was" or that the
alternate stack was even already in use, and clobbers it from the
beginning if a new signal arrives that is to execute on the alternate
stack.

If you think this understanding is incorrect, we should research/test.

Rich

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.