Fixed a panic caused by over-optimizing npxdrop() in the non-FXSR case.

frstor can trap despite it being a control instruction, since it bogusly
checks for pending exceptions in the state that it is overwriting.
This used to be a non-problem because frstor was always paired with a
previous fnsave, and fnsave does an implicit fninit so any pending
exceptions only remain live in the saved state.  Now frstor is sometimes
paired with npxdrop() and we must do a little more than just forget
that the npx was used in npxdrop() to avoid a trap later.  This is a
non-problem in the FXSR case because fxrstor doesn't do the bogus check.

FXSR is part of SSE, and npxdrop() is only in FreeBSD-5.x, so this bug
only affected old machines running FreeBSD-5.x.

PR:		68058

sys/i386/isa/npx.c

@@ -872,6 +872,15 @@ npxdrop()
 {
 	struct thread *td;
 
+	/*
+	 * Discard pending exceptions in the !cpu_fxsr case so that unmasked
+	 * ones don't cause a panic on the next frstor.
+	 */
+#ifdef CPU_ENABLE_SSE
+	if (!cpu_fxsr)
+#endif
+		fnclex();
+
 	td = PCPU_GET(fpcurthread);
 	PCPU_SET(fpcurthread, NULL);
 	td->td_pcb->pcb_flags &= ~PCB_NPXINITDONE;