heads up: 2.4.18 bug causing bad value in siginfo.si_fd

It has been brought to my attention that there is a bug in the ia64 version of 2.4.18 which causes a bad value to be returned in the si_fd member of the siginfo structure. This bug occurs at least for siginfo values returned by sigtimedwait(), but I suspect the root cause is in the siginfo-copying routines so normal signal deliver probably also results in a bad si_fd value.

The bug is not present in 2.4.19 and 2.5.31.

	--david