Orinoco driver update
A message that is not in the archive (sent off-list or not yet received): 20020927025227.GC1898@zax
David,
This doesn't compile at all... stuff like "static void orinoco_reset[...]" in orinoco.c being called from orinoco_pci.c could never hope to build...
Linus applied it... would you mind sending an update patch please?
Jeff
Quoted 7 lines · Jeff Garzik
David,
This doesn't compile at all... stuff like "static void orinoco_reset[...]" in
orinoco.cbeing called fromorinoco_pci.ccould never hope to build...Linus applied it... would you mind sending an update patch please?
Crud. Looks like my patch making script was borken, so orinoco_pci.c wasn't updated properly. The patch below should fix that, and adds some other minor updates (driver version 0.13a) as well.
drivers/net/wireless/airport.c +2 -2
@@ -1,4 +1,4 @@ -/* airport.c 0.13 +/* airport.c 0.13a * * A driver for "Hermes" chipset based Apple Airport wireless * card.
@@ -264,7 +264,7 @@kfree (dev); }/* airport_detach */ -static char version[]__initdata ="airport.c 0.13 (Benjamin Herrenschmidt <[email protected]>)" ; +static char version[]__initdata ="airport.c 0.13a (Benjamin Herrenschmidt <[email protected]>)" ;MODULE_AUTHOR ("Benjamin Herrenschmidt <[email protected]>" );MODULE_DESCRIPTION ("Driver for the Apple Airport wireless card." );MODULE_LICENSE ("Dual MPL/GPL" );
drivers/net/wireless/hermes.h +12 -0
@@ -251,6 +251,18 @@struct hermes_scan_apinfo aps[35 ];/* Scan result */ } __attribute__ ((packed)); +#define HERMES_LINKSTATUS_NOT_CONNECTED (0x0000) +#define HERMES_LINKSTATUS_CONNECTED (0x0001) +#define HERMES_LINKSTATUS_DISCONNECTED (0x0002) +#define HERMES_LINKSTATUS_AP_CHANGE (0x0003) +#define HERMES_LINKSTATUS_AP_OUT_OF_RANGE (0x0004) +#define HERMES_LINKSTATUS_AP_IN_RANGE (0x0005) +#define HERMES_LINKSTATUS_ASSOC_FAILED (0x0006) + +struct hermes_linkstatus { +u16 linkstatus;/* Link status */ +} __attribute__ ((packed)); +// #define HERMES_DEBUG_BUFFER 1 #define HERMES_DEBUG_BUFSIZE 4096struct hermes_debug_entry {
drivers/net/wireless/orinoco.c +69 -13
136 lines not on this page · raw message
drivers/net/wireless/orinoco.h +5 -0
@@ -7,6 +7,11 @@#ifndef _ORINOCO_H#define _ORINOCO_H +#include <linux/types.h> +#include <linux/spinlock.h> +#include <linux/netdevice.h> +#include <linux/wireless.h> +#include <linux/tqueue.h> #include "hermes.h" /* To enable debug messages */
drivers/net/wireless/orinoco_cs.c +2 -2
@@ -1,4 +1,4 @@ -/* orinoco_cs.c 0.13 - (formerly known as dldwd_cs.c) +/* orinoco_cs.c 0.13a - (formerly known as dldwd_cs.c) * * A driver for "Hermes" chipset based PCMCIA wireless adaptors, such * as the Lucent WavelanIEEE/Orinoco cards and their OEM (Cabletron/
@@ -693,7 +693,7 @@/* Can't be declared "const" or the whole __initdata section will * become const */ -static char version[]__initdata ="orinoco_cs.c 0.13 (David Gibson <[email protected]> and others)" ; +static char version[]__initdata ="orinoco_cs.c 0.13a (David Gibson <[email protected]> and others)" ;static int __init init_orinoco_cs (void )
drivers/net/wireless/orinoco_pci.c +24 -64
180 lines not on this page · raw message
drivers/net/wireless/orinoco_plx.c +2 -2
@@ -1,4 +1,4 @@ -/* orinoco_plx.c 0.13 +/* orinoco_plx.c 0.13a * * Driver for Prism II devices which would usually be driven by orinoco_cs, * but are connected to the PCI bus by a PLX9052.
@@ -341,7 +341,7 @@ .resume = 0, }; -static char version[]__initdata ="orinoco_plx.c 0.13 (Daniel Barlow <[email protected]>, David Gibson <[email protected]>)" ; +static char version[]__initdata ="orinoco_plx.c 0.13a (Daniel Barlow <[email protected]>, David Gibson <[email protected]>)" ;MODULE_AUTHOR ("Daniel Barlow <[email protected]>" );MODULE_DESCRIPTION ("Driver for wireless LAN cards using the PLX9052 PCI bridge" );#ifdef MODULE_LICENSE
Signature · 4 lines
David Gibson | For every complex problem there is a [email protected] | solution which is simple, neat and | wrong. http://www.ozlabs.org/people/dgibson
David,
Linus applied 0.13a and that fixes things, thanks.
Two quick comments:
* you need pci_set_drvdata(pdev,NULL) after pci_disable_disable in your pci_driver::remove hook
* I think it would look better to remove the struct pci_driver ->suspend and ->resume hook references, if they are NULL (0)...
Quoted 7 lines · Jeff Garzik
David,
Linus applied 0.13a and that fixes things, thanks.
Two quick comments:
* you needpci_set_drvdata(pdev,NULL)afterpci_disable_disablein yourpci_driver::removehook
Ok, I've added that before kfree()ing the net_device structure in both orinoco_pci.c and orinoco_plx.c
Quoted 2 lines
* I think it would look better to remove the struct
pci_driver->suspend and ->resume hook references, if they are NULL (0)...
Hmm... I'd kind of prefer to leave them there, to remind me that the suspend/resume hooks need to be implemented.
Signature · 4 lines
David Gibson | For every complex problem there is a [email protected] | solution which is simple, neat and | wrong. http://www.ozlabs.org/people/dgibson
Quoted 6 lines · David Gibson
Quoted 1 line · Jeff Garzik
* I think it would look better to remove the struct
pci_driver->suspend and ->resume hook references, if they are NULL (0)...Hmm... I'd kind of prefer to leave them there, to remind me that the suspend/resume hooks need to be implemented.
heh, with that argument I'm certainly convinced :)