[parisc-linux] Further serial changes -- PLEASE TEST

[parisc-linux] Further serial changes -- PLEASE TEST

[Matthew Wilcox]

Richard reports the serial ports on his B180 are still swapped.  I
think I understand why and the following patch should fix it as well
as several other B/C/J `hundred' class machines.  It might also fix
the A180 (is this broken without the patch?  I haven't seen reports).

Please test this patch if you have one of the above machines.  This
mess might've been avoided if more people had tested the parisc_device
patch.

Index: drivers/gsc/serial.c
===================================================================
RCS file: /home/cvs/parisc/linux/drivers/gsc/serial.c,v
retrieving revision 1.20
diff -u -p -r1.20 serial.c
--- serial.c	2001/08/31 02:00:49	1.20
+++ serial.c	2001/09/12 22:32:40
@@ -97,23 +97,16 @@ static struct parisc_device_id serial_tb
  * only on those machines.  This needs to get fixed properly somehow.
  */
 static struct parisc_device_id serial1_tbl[] = {
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x019, 0x0008C }, /* B180L+ */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x022, 0x0008C }, /* B132L+ */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x03B, 0x0008C }, /* C1xx/C1xxL */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x03C, 0x0008C }, /* B132L */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x03D, 0x0008C }, /* B160L */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x03E, 0x0008C }, /* B132L+ */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x03F, 0x0008C }, /* B180L+ */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x040, 0x0008C }, /* B132L */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x041, 0x0008C }, /* B160L */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x046, 0x0008C }, /* Rocky2 120 */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x047, 0x0008C }, /* Rocky2 150 */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x048, 0x0008C }, /* Rocky2 120 */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x049, 0x0008C }, /* Rocky2 150 */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x04E, 0x0008C }, /* Kiji L2 132 */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x04F, 0x0008C }, /* Kiji L2 132 */
 	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x056, 0x0008C }, /* Raven+ */
-	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x05F, 0x00081 }, /* A-class 180 */
+	{ HPHW_FIO, HVERSION_REV_ANY_ID, 0x05F, 0x0008C }, /* A-class 180 */
 	{ 0 }
 };
 

-- 
Revolutions do not require corporate support.

Re: [parisc-linux] Further serial changes -- PLEASE TEST

[Richard Hirst]

On Wed, Sep 12, 2001 at 11:42:08PM +0100, Matthew Wilcox wrote:
> 
> Richard reports the serial ports on his B180 are still swapped.  I
> think I understand why and the following patch should fix it as well
> as several other B/C/J `hundred' class machines.  It might also fix
> the A180 (is this broken without the patch?  I haven't seen reports).

B180 is now ok.
C360 was ok anyway, and still is.


715/old orders its serial ports correctly, but as they are

6. Scorpio Sr. Core RS-232 (10) at 0xf0823000, versions 0x19, 0x0, 0x75
7. Scorpio Sr. Core RS-232 (10) at 0xf0822000, versions 0x19, 0x0, 0x75

I guess we are just lucky.  0xf0823000 is ttyS0, btw.


My 712 has
5. Gecko 100 Core RS-232 (10) at 0xf0105000, versions 0x1d, 0x0, 0x8c, 0x0, 0x0
13. Gecko Optional RS-232 (10) at 0xf0505000, versions 0x18, 0x0, 0x8c, 0x0, 0x0

It does put 0xf0105000 as ttyS0 as it should, but I guess there
should be an entry for 0x1d, 0x8c in serial1_tbl[] really?

Richard

Re: [parisc-linux] Further serial changes -- PLEASE TEST

[Matthew Wilcox]

On Thu, Sep 13, 2001 at 11:28:35AM +0100, Richard Hirst wrote:
> B180 is now ok.
> C360 was ok anyway, and still is.

Excellent!  Thanks for testing.

> 715/old orders its serial ports correctly, but as they are
> 
> 6. Scorpio Sr. Core RS-232 (10) at 0xf0823000, versions 0x19, 0x0, 0x75
> 7. Scorpio Sr. Core RS-232 (10) at 0xf0822000, versions 0x19, 0x0, 0x75
> 
> I guess we are just lucky.  0xf0823000 is ttyS0, btw.
> 
> 
> My 712 has
> 5. Gecko 100 Core RS-232 (10) at 0xf0105000, versions 0x1d, 0x0, 0x8c, 0x0, 0x0
> 13. Gecko Optional RS-232 (10) at 0xf0505000, versions 0x18, 0x0, 0x8c, 0x0, 0x0
> 
> It does put 0xf0105000 as ttyS0 as it should, but I guess there
> should be an entry for 0x1d, 0x8c in serial1_tbl[] really?

The way I look at this is that the serial ports really _ought_ to be
listed in the right order.  This is a table of exceptions to cope with
the few cases where that's not true.

-- 
Revolutions do not require corporate support.