Patch(2.4.6):serial unmaintained (bugfix pci timedia/sunix/exsys pci cards)

Hi,
this one-liner fixes a longstanding bug in serial for Timedia/Sunix/Exsys PCI cards !

The fix was sent to Ted several times since 02/2001 and uploaded to http://sourceforge.net/projects/serial/.
There was no reaction.

By listing a defunct MAINTAINER the progress of linux is stuck as patches go to /dev/null by
- the maintainer (who got lost for unknown, perhaps even valid reasons)
- Linus, as he waits for approval by the maintainer.

Linus, can you please include this patch?

Regards, Gunther

(Without this patch serial recognizes wrong number of serial ports!)

@@ -4193,7 +4193,7 @@
        for (i=0; timedia_data[i].num; i++) {
                ids = timedia_data[i].ids;
                for (j=0; ids[j]; j++) {
-                       if (pci_get_subvendor(dev) == ids[j]) {
+                       if (pci_get_subdevice(dev) == ids[j]) {
                                board->num_ports = timedia_data[i].num;
                                return 0;
                        }

This should already be fixed in 2.4.7-pre6, can you verify that it works for you?

Linus

Hi!

Am Donnerstag, 12. Juli 2001 18:31 schrieb Linus Torvalds:

There are a lot more bugs (and "send-in" patches to Ted) which fix a lot of problems even in the 2.4.7-x driver.
I know about a guy "Ian Abbott" (abbotti at mev.co.uk) which is active on the serial.sf.net bug-list and has done a lot good to the serial driver. Perhaps he should be heard regarding this.

Personally I found a (h/w) timinig problem in the inititalisation code which hangs the "sunix" chips (and freezes the complete box). But since I couldn't get my hands on a datasheet of these chips, I've until now only solved this problem through a hand full of delays after "critical" I/O operations (very ugly).

Salut, Jörg