On Jun 16 15:28, Eric Blake wrote: > Haojun Bao <baohaojun <at> gmail.com> writes: > > > type the command: > > cat|cat <Enter> > > > > type ^Z, type > > bg <Enter> > > I can also reproduce it under cygwin 1.7, and further clarified that it is > the > second cat that is failing: > > $ sleep 100 | cat > ^Z > $ bg > cat: -: Bad address > > I'm not quite sure how to strace this, since you can't use ^Z to suspend an > active strace. But my guess is that there is a bug in cygwin1.dll pipe > handling code, such that it dereferences bad memory when resuming execution > of > a read from a pipe after a SIGSTOP; cat is just reporting the EFAULT from a > failed read() call.
I have a patch for this in the loop. Actually it is supposed to fix the "Bad address" problem in git(*), but the underlying problem is apparently the same. Corinna (*) http://cygwin.com/ml/cygwin/2009-06/msg00546.html -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
