rtl8192s_usb vs rtl8192u_usb device IDs

[Ben Hutchings <ben@xxxxxxxxxxxxxxx>]

Greg,

There seems to be some confusion between the two staging drivers
rtl8192s_usb (aka rtl8192su) and rtl8192u_usb (aka rtl8192u).

Guy Sheffer reported problems using rtl8192s_usb in 2.6.34:
> It is detected, but then returns: unknown rf chip, can't set channel map
> in function:rtl819x_set_channel_map() in dmesg

The USB device has device ID 0bda:8192.  This error appears to indicate
that this test in rtl8192SU_read_eeprom_info() failed:

	// To check autoload success or not.
	if (tmpU1b & CmdEEPROM_En)

In the vendor driver that came with this device,
<http://people.debian.org/~benh/rtl-wlan/linux_2.6.0003.0825.2009.tar.gz>, I see two separate lists of device IDs for 'RTL8192SU' and 'RTL8192U':

#ifdef RTL8192SU
	{USB_DEVICE(0x0bda, 0x8171)},
	{USB_DEVICE(0x0bda, 0x8172)},
	{USB_DEVICE(0x0bda, 0x8173)},
	{USB_DEVICE(0x0bda, 0x8174)},
	{USB_DEVICE(0x0bda, 0x8712)},
	{USB_DEVICE(0x0bda, 0x8713)},
	{USB_DEVICE(0x07aa, 0x0047)},
	{USB_DEVICE(0x07d1, 0x3303)},
	{USB_DEVICE(0x07d1, 0x3302)},
	{USB_DEVICE(0x07d1, 0x3300)},
	{USB_DEVICE(0x1740, 0x9603)},
	{USB_DEVICE(0x1740, 0x9605)},
	{USB_DEVICE(0x050d, 0x815F)},
	{USB_DEVICE(0x06f8, 0xe031)},
	{USB_DEVICE(0x7392, 0x7612)},
	{USB_DEVICE(0x0DF6, 0x0045)},
	{USB_DEVICE(0x0E66, 0x0015)},
	{USB_DEVICE(0x0E66, 0x0016)},
#else	
	{USB_DEVICE(0x0bda, 0x8192)},
	{USB_DEVICE(0x0bda, 0x8709)},
	{USB_DEVICE(0x07aa, 0x0043)},
	{USB_DEVICE(0x050d, 0x805E)},
	{USB_DEVICE(0x0df6, 0x0031)},
	{USB_DEVICE(0x1740, 0x9201)},
	{USB_DEVICE(0x2001, 0x3301)},
	{USB_DEVICE(0x5a57, 0x0290)},
	{USB_DEVICE(0x043E, 0x7A01)},
#endif	

In mainline, rtl8192su_usb claims a superset of the device IDs of
rtl8192u_usb, but does not seem to include the code to support the
'RTL8192U' devices.  For example, the vendor driver has different EEPROM
access code for the RTL8192U devices, which would explain the failure
Guy reported.

I think the device IDs listed in rtl8192u_usb should be removed from
rtl8192s_usb.

Ben.

-- 
Ben Hutchings
Once a job is fouled up, anything done to improve it makes it worse.

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/devel