Error in example code for EQEP in SPRC892

The exmple code for the eqep_pos_speed example has the following code:

//**** Low-speed computation using QEP capture counter ****//
	if(EQep1Regs.QEPSTS.bit.UPEVNT==1)                 // Unit position event
	{
		if(EQep1Regs.QEPSTS.bit.COEF==0)               // No Capture overflow
			temp1=(unsigned long)EQep1Regs.QCPRDLAT;   // temp1 = t2-t1
		else							               // Capture overflow, saturate the result
			temp1=0xFFFF;

		p->Speed_pr = _IQdiv(p->SpeedScaler,temp1);    // p->Speed_pr = p->SpeedScaler/temp1
		Tmp1=p->Speed_pr;

		if (Tmp1>_IQ(1))
	 		p->Speed_pr = _IQ(1);
		else
	 		p->Speed_pr = Tmp1;

	    // Convert p->Speed_pr to RPM
		if (p->DirectionQep==0)                                 // Reverse direction = negative
			p->SpeedRpm_pr = -_IQmpy(p->BaseRpm,p->Speed_pr); 	// Q0 = Q0*GLOBAL_Q => _IQXmpy(), X = GLOBAL_Q
		else                                                    // Forward direction = positive
			p->SpeedRpm_pr = _IQmpy(p->BaseRpm,p->Speed_pr); 	// Q0 = Q0*GLOBAL_Q => _IQXmpy(), X = GLOBAL_Q


		EQep1Regs.QEPSTS.all=0x88;					// Clear Unit position event flag
													// Clear overflow error flag

The last line is incorrect.  The line should be

EQep1Regs.QEPSTS.bit.COEFF=1;

This clears the capture overflow flag which is a read/write sticky bit, cleared by writing a 1.  

The 0x80 bit is the unit position event flag (UPEVNT) bit which is read only.  

Writing to this bit caused erroneous reading to be pulled from the period latch register QCPRDLAT  in the sample code ( at least on the 28035 chip I'm using)

 These reading were always 65535, with an occasional valid result.

Not sure where else to post, to avoid other programmers running in to the same issue.