On Jan 5, 2007, at 12:45 PM, Brian Buhrow wrote:

> 	Hello. In looking further at the wm(4) driver, I find I wonder about
> the correctness of the enumerated list of chip types.  Specifically,  
> I'm
> assuming this is an ordered list, where the first item is number 0,  
> and the
> index increments for each chip type in the enumerated list.  I'll  
> quote the
> enumerated list below, but my question is, is this order correct?
> Specifically, does the 82541 series of chips really come after the  
> 82546
> series?  If I'm reading this code corectly, then places where the  
> chip type
> is tested for some capability, like for all chips greater than the  
> 82543, for
> example, then the 82541 chip would test to be more capable than the  
> 82543,
> which seems counter intuitive to me.
> 	Could someone comment on this who  knows more than I about this
> driver?

Yes, this is essentially correct.  Also, sometimes it's not  
necessarily "more capable" but rather "differently capable" or  
"introduced later".


> -thanks
> -Brian
>
>
> /*	$NetBSD: if_wm.c,v 1.100.2.5 2006/07/07 06:24:40 tron Exp $	*/
> typedef enum {
> 	WM_T_unknown		= 0,
> 	WM_T_82542_2_0,			/* i82542 2.0 (really old) */
> 	WM_T_82542_2_1,			/* i82542 2.1+ (old) */
> 	WM_T_82543,			/* i82543 */
> 	WM_T_82544,			/* i82544 */
> 	WM_T_82540,			/* i82540 */
> 	WM_T_82545,			/* i82545 */
> 	WM_T_82545_3,			/* i82545 3.0+ */
> 	WM_T_82546,			/* i82546 */
> 	WM_T_82546_3,			/* i82546 3.0+ */
> 	WM_T_82541,			/* i82541 */
> 	WM_T_82541_2,			/* i82541 2.0+ */
> 	WM_T_82547,			/* i82547 */
> 	WM_T_82547_2,			/* i82547 2.0+ */
> } wm_chip_type;

-- thorpej