usbnet.c: remove an unused function

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

The patch below removes an unused function from drivers/usb/net/usbnet.c

diffstat output:

 drivers/usb/net/usbnet.c |    6 ------
 1 files changed, 6 deletions(-)

- --- linux-2.6.10-rc1-mm1-full/drivers/usb/net/usbnet.c.old 2004-10-28 23:32:50.000000000 +0200
+++ linux-2.6.10-rc1-mm1-full/drivers/usb/net/usbnet.c 2004-10-28 23:33:23.000000000 +0200
@@ -2127,12 +2127,6 @@
}

static inline int
- -pl_clear_QuickLink_features (struct usbnet *dev, int val)
- -{
- - return pl_vendor_req (dev, 1, (u8) val, 0);
- -}
- -
- -static inline int
pl_set_QuickLink_features (struct usbnet *dev, int val) {
return pl_vendor_req (dev, 3, (u8) val, 0);

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFBgX/HmfzqmE8StAARAvVCAJ9wqcniAa5kgAtve4KqGdgI+zXx4ACeLPet /dF6osevC+GlXF3HSbiFWVM=
=nlSX
-----END PGP SIGNATURE-----

[ this time without the problems due to a digital signature... ]

The patch below removes an unused function from drivers/usb/net/usbnet.c

diffstat output:

 drivers/usb/net/usbnet.c |    6 ------
 1 files changed, 6 deletions(-)

@@ -2127,12 +2127,6 @@
 }
 
 static inline int
-pl_clear_QuickLink_features (struct usbnet *dev, int val)
-{
-	return pl_vendor_req (dev, 1, (u8) val, 0);
-}
-
-static inline int
 pl_set_QuickLink_features (struct usbnet *dev, int val)
 {
 	return pl_vendor_req (dev, 3, (u8) val, 0);

Actually in this case I'd rather leave the function there; the documentation on this chip is hard to find, and this function will be needed when someone finally spends the time to fix some of the init/reset handshake issues for these PL2301/2302 chips. The current init code cloned some very ancient code from about the 2.2.10 (!) or so kernel.

Speaking of which ... anyone motivated to solve that can probably find answers in a file transfer package I saw a while back (but lost!) for file transferring files to/from Linux, especialy for Playstation 2 boxes. That used a very different handshaking scheme... possibly one that's incompatible with Linux 2.2, but evidently one that worked reliably and noticed connect change events. (Neither is true of the current PL2301/2302 code!)

- Dave

p.s. Last I looked, GCC ignored unused inlines; no code
generated, no warnings. Did that change?

23:32:50.000000000 +0200

23:33:23.000000000 +0200

OK.

It didn't change.

But there are three different possible reactions on my patches:
1. ACK, kill this dead code
2. ups, I really wanted to use this function
3. please keep, code using this function will/might follow in the future

Case 1 is the most common case (and this simply removes some dead code).

I had until now two times case 2 (which means the code is now better).

You are the first person for case 3.

cu
Adrian

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

And presumably there will also be at least a few case 4:

4. no response, treated as an ACK.

:)