2007-05-10 01:39:11

by Jeff Garzik

[permalink] [raw]
Subject: [git patch] move USB net drivers to drivers/net


This was ACK'd by Greg, as you see in the sign-offs. See the commit
below for rationale.

USB is now treated like other buses, for network drivers:
* USB network driver patches should go to me and netdev
* Just like in PCI or PCMCIA land, bus-specific code may
originate from the USB side of the house (Greg's tree).



Please pull from 'usb-move' branch of
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git usb-move

to receive the following update:


commit 5b2fc499917e5897a13add780e181b4cef197072
Author: Jeff Garzik <[email protected]>
Date: Wed May 9 21:31:55 2007 -0400

Move USB network drivers to drivers/net/usb.

It is preferable to group drivers by usage (net, scsi, ATA, ...) than
by bus. When reviewing drivers, the [PCI|USB|PCMCIA|...] maintainer
is probably less qualified on networking issues than a networking
maintainer. Also, from a practical standpoint, chips often
appear on multiple buses, which is why we do not put drivers into
drivers/pci/net.

Signed-off-by: Greg Kroah-Hartman <[email protected]>
Signed-off-by: Jeff Garzik <[email protected]>


drivers/net/Kconfig | 2 ++
drivers/net/Makefile | 8 ++++++++
drivers/{usb/net => net/usb}/Kconfig | 0
drivers/{usb/net => net/usb}/Makefile | 0
drivers/{usb/net => net/usb}/asix.c | 0
drivers/{usb/net => net/usb}/catc.c | 0
drivers/{usb/net => net/usb}/cdc_ether.c | 0
drivers/{usb/net => net/usb}/cdc_subset.c | 0
drivers/{usb/net => net/usb}/dm9601.c | 0
drivers/{usb/net => net/usb}/gl620a.c | 0
drivers/{usb/net => net/usb}/kaweth.c | 0
drivers/{usb/net => net/usb}/kawethfw.h | 0
drivers/{usb/net => net/usb}/mcs7830.c | 0
drivers/{usb/net => net/usb}/net1080.c | 0
drivers/{usb/net => net/usb}/pegasus.c | 0
drivers/{usb/net => net/usb}/pegasus.h | 0
drivers/{usb/net => net/usb}/plusb.c | 0
drivers/{usb/net => net/usb}/rndis_host.c | 0
drivers/{usb/net => net/usb}/rtl8150.c | 0
drivers/{usb/net => net/usb}/usbnet.c | 0
drivers/{usb/net => net/usb}/usbnet.h | 0
drivers/{usb/net => net/usb}/zaurus.c | 0
drivers/usb/Kconfig | 2 --
drivers/usb/Makefile | 7 -------
24 files changed, 10 insertions(+), 9 deletions(-)
rename drivers/{usb/net => net/usb}/Kconfig (100%)
rename drivers/{usb/net => net/usb}/Makefile (100%)
rename drivers/{usb/net => net/usb}/asix.c (100%)
rename drivers/{usb/net => net/usb}/catc.c (100%)
rename drivers/{usb/net => net/usb}/cdc_ether.c (100%)
rename drivers/{usb/net => net/usb}/cdc_subset.c (100%)
rename drivers/{usb/net => net/usb}/dm9601.c (100%)
rename drivers/{usb/net => net/usb}/gl620a.c (100%)
rename drivers/{usb/net => net/usb}/kaweth.c (100%)
rename drivers/{usb/net => net/usb}/kawethfw.h (100%)
rename drivers/{usb/net => net/usb}/mcs7830.c (100%)
rename drivers/{usb/net => net/usb}/net1080.c (100%)
rename drivers/{usb/net => net/usb}/pegasus.c (100%)
rename drivers/{usb/net => net/usb}/pegasus.h (100%)
rename drivers/{usb/net => net/usb}/plusb.c (100%)
rename drivers/{usb/net => net/usb}/rndis_host.c (100%)
rename drivers/{usb/net => net/usb}/rtl8150.c (100%)
rename drivers/{usb/net => net/usb}/usbnet.c (100%)
rename drivers/{usb/net => net/usb}/usbnet.h (100%)
rename drivers/{usb/net => net/usb}/zaurus.c (100%)


diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index b86ccd2..d9842d8 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2499,6 +2499,8 @@ source "drivers/net/tokenring/Kconfig"

source "drivers/net/wireless/Kconfig"

+source "drivers/net/usb/Kconfig"
+
source "drivers/net/pcmcia/Kconfig"

source "drivers/net/wan/Kconfig"
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index 59c0459..c5d8423 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -206,6 +206,14 @@ obj-$(CONFIG_TR) += tokenring/
obj-$(CONFIG_WAN) += wan/
obj-$(CONFIG_ARCNET) += arcnet/
obj-$(CONFIG_NET_PCMCIA) += pcmcia/
+
+obj-$(CONFIG_USB_CATC) += usb/
+obj-$(CONFIG_USB_KAWETH) += usb/
+obj-$(CONFIG_USB_PEGASUS) += usb/
+obj-$(CONFIG_USB_RTL8150) += usb/
+obj-$(CONFIG_USB_USBNET) += usb/
+obj-$(CONFIG_USB_ZD1201) += usb/
+
obj-y += wireless/
obj-$(CONFIG_NET_TULIP) += tulip/
obj-$(CONFIG_HAMRADIO) += hamradio/

[SNIP ADD PATCHES AS OBVIOUS]

diff --git a/drivers/usb/Kconfig b/drivers/usb/Kconfig
index b847bbc..278a22c 100644
--- a/drivers/usb/Kconfig
+++ b/drivers/usb/Kconfig
@@ -87,8 +87,6 @@ source "drivers/usb/storage/Kconfig"

source "drivers/usb/image/Kconfig"

-source "drivers/usb/net/Kconfig"
-
source "drivers/usb/mon/Kconfig"

comment "USB port drivers"
diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index 0ef090b..72464b5 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -23,13 +23,6 @@ obj-$(CONFIG_USB_PRINTER) += class/
obj-$(CONFIG_USB_STORAGE) += storage/
obj-$(CONFIG_USB) += storage/

-obj-$(CONFIG_USB_CATC) += net/
-obj-$(CONFIG_USB_KAWETH) += net/
-obj-$(CONFIG_USB_PEGASUS) += net/
-obj-$(CONFIG_USB_RTL8150) += net/
-obj-$(CONFIG_USB_USBNET) += net/
-obj-$(CONFIG_USB_ZD1201) += net/
-
obj-$(CONFIG_USB_MDC800) += image/
obj-$(CONFIG_USB_MICROTEK) += image/

[SNIP REMOVE PATCHES AS OBVIOUS]


2007-05-10 12:13:09

by Indan Zupancic

[permalink] [raw]
Subject: Re: [git patch] move USB net drivers to drivers/net

Hello,

On Thu, May 10, 2007 03:38, Jeff Garzik wrote:
>
> This was ACK'd by Greg, as you see in the sign-offs. See the commit
> below for rationale.
>
> USB is now treated like other buses, for network drivers:
> * USB network driver patches should go to me and netdev
> * Just like in PCI or PCMCIA land, bus-specific code may
> originate from the USB side of the house (Greg's tree).
>
>
>
> Please pull from 'usb-move' branch of
> master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git usb-move
>
> to receive the following update:
>
>
> commit 5b2fc499917e5897a13add780e181b4cef197072
> Author: Jeff Garzik <[email protected]>
> Date: Wed May 9 21:31:55 2007 -0400
>
> Move USB network drivers to drivers/net/usb.
>
> It is preferable to group drivers by usage (net, scsi, ATA, ...) than
> by bus. When reviewing drivers, the [PCI|USB|PCMCIA|...] maintainer
> is probably less qualified on networking issues than a networking
> maintainer. Also, from a practical standpoint, chips often
> appear on multiple buses, which is why we do not put drivers into
> drivers/pci/net.

What about drivers/usb/atm/?

Greetings,

Indan


2007-05-10 13:53:34

by Duncan Sands

[permalink] [raw]
Subject: Re: [git patch] move USB net drivers to drivers/net

On Thursday 10 May 2007 14:12:47 Indan Zupancic wrote:
> Hello,
>
> On Thu, May 10, 2007 03:38, Jeff Garzik wrote:
> >
> > This was ACK'd by Greg, as you see in the sign-offs. See the commit
> > below for rationale.
> >
> > USB is now treated like other buses, for network drivers:
> > * USB network driver patches should go to me and netdev
> > * Just like in PCI or PCMCIA land, bus-specific code may
> > originate from the USB side of the house (Greg's tree).
> >
> >
> >
> > Please pull from 'usb-move' branch of
> > master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git usb-move
> >
> > to receive the following update:
> >
> >
> > commit 5b2fc499917e5897a13add780e181b4cef197072
> > Author: Jeff Garzik <[email protected]>
> > Date: Wed May 9 21:31:55 2007 -0400
> >
> > Move USB network drivers to drivers/net/usb.
> >
> > It is preferable to group drivers by usage (net, scsi, ATA, ...) than
> > by bus. When reviewing drivers, the [PCI|USB|PCMCIA|...] maintainer
> > is probably less qualified on networking issues than a networking
> > maintainer. Also, from a practical standpoint, chips often
> > appear on multiple buses, which is why we do not put drivers into
> > drivers/pci/net.
>
> What about drivers/usb/atm/?

I don't mind if it moves to drivers/atm/, though I don't see the point of it.

Best wishes,

Duncan.

2007-05-10 16:16:53

by Jan Engelhardt

[permalink] [raw]
Subject: Re: [another git patch] move USB net drivers to drivers/net

Hi Jeff,


On May 9 2007 21:38, Jeff Garzik wrote:
>diff --git a/drivers/net/Makefile b/drivers/net/Makefile
>index 59c0459..c5d8423 100644
>--- a/drivers/net/Makefile
>+++ b/drivers/net/Makefile
>@@ -206,6 +206,14 @@ obj-$(CONFIG_TR) += tokenring/
> obj-$(CONFIG_WAN) += wan/
> obj-$(CONFIG_ARCNET) += arcnet/
> obj-$(CONFIG_NET_PCMCIA) += pcmcia/
>+
>+obj-$(CONFIG_USB_CATC) += usb/
>+obj-$(CONFIG_USB_KAWETH) += usb/
>+obj-$(CONFIG_USB_PEGASUS) += usb/
>+obj-$(CONFIG_USB_RTL8150) += usb/
>+obj-$(CONFIG_USB_USBNET) += usb/
>+obj-$(CONFIG_USB_ZD1201) += usb/
>+

This looks, well, a bit ugly. Since we seem to be going in the direction
of using "menuconfig"s (Kconfig language object) anyway, I propose the
patch below.

diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
new file mode 100644
index 0000000..3de564b
--- /dev/null
+++ b/drivers/net/usb/Kconfig
@@ -0,0 +1,338 @@
+#
+# USB Network devices configuration
+#
+comment "Networking support is needed for USB Network Adapter support"
+ depends on USB && !NET

This comment is superfluous. If !NET, then NETDEVICES is not even available.
Since you just moved/renamed it, you are not to blame. Though, I remove
it in below's patch (applies on top of
commit 5b2fc499917e5897a13add780e181b4cef197072).


Signed-off-by: Jan Engelhardt <[email protected]>

diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index c5d8423..5f6112d 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -207,12 +207,7 @@ obj-$(CONFIG_WAN) += wan/
obj-$(CONFIG_ARCNET) += arcnet/
obj-$(CONFIG_NET_PCMCIA) += pcmcia/

-obj-$(CONFIG_USB_CATC) += usb/
-obj-$(CONFIG_USB_KAWETH) += usb/
-obj-$(CONFIG_USB_PEGASUS) += usb/
-obj-$(CONFIG_USB_RTL8150) += usb/
-obj-$(CONFIG_USB_USBNET) += usb/
-obj-$(CONFIG_USB_ZD1201) += usb/
+obj-$(CONFIG_NETDEV_USB) += usb/

obj-y += wireless/
obj-$(CONFIG_NET_TULIP) += tulip/
diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
index 3de564b..2458f99 100644
--- a/drivers/net/usb/Kconfig
+++ b/drivers/net/usb/Kconfig
@@ -1,11 +1,12 @@
#
# USB Network devices configuration
#
-comment "Networking support is needed for USB Network Adapter support"
- depends on USB && !NET
-
-menu "USB Network Adapters"
+menuconfig NETDEVICES_USB
+ bool "USB Network Adapters"
depends on USB && NET
+ default y
+
+if NETDEVICES_USB

config USB_CATC
tristate "USB CATC NetMate-based Ethernet device support (EXPERIMENTAL)"
@@ -334,5 +335,4 @@ config USB_NET_ZAURUS
really need this non-conformant variant of CDC Ethernet (or in
some cases CDC MDLM) protocol, not "g_ether".

-
-endmenu
+endif # NETDEVICES_USB
#<EOF>

Jan
--

2007-08-17 10:46:16

by Jan Engelhardt

[permalink] [raw]
Subject: Re: [another git patch] move USB net drivers to drivers/net

Hi,


On May 10 2007 18:12, Jan Engelhardt wrote:
>Subject: Re: [another git patch] move USB net drivers to drivers/net
>
>Hi Jeff,
>
>
>On May 9 2007 21:38, Jeff Garzik wrote:
>>diff --git a/drivers/net/Makefile b/drivers/net/Makefile
>>index 59c0459..c5d8423 100644
>>--- a/drivers/net/Makefile
>>+++ b/drivers/net/Makefile
>>@@ -206,6 +206,14 @@ obj-$(CONFIG_TR) += tokenring/
>> obj-$(CONFIG_WAN) += wan/
>> obj-$(CONFIG_ARCNET) += arcnet/
>> obj-$(CONFIG_NET_PCMCIA) += pcmcia/
>>+
>>+obj-$(CONFIG_USB_CATC) += usb/
>>+obj-$(CONFIG_USB_KAWETH) += usb/
>>+obj-$(CONFIG_USB_PEGASUS) += usb/
>>+obj-$(CONFIG_USB_RTL8150) += usb/
>>+obj-$(CONFIG_USB_USBNET) += usb/
>>+obj-$(CONFIG_USB_ZD1201) += usb/
>>+
>
>This looks, well, a bit ugly. Since we seem to be going in the direction
>of using "menuconfig"s (Kconfig language object) anyway, I propose the
>patch below.


Please consider, thank you.



>
>diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
>new file mode 100644
>index 0000000..3de564b
>--- /dev/null
>+++ b/drivers/net/usb/Kconfig
>@@ -0,0 +1,338 @@
>+#
>+# USB Network devices configuration
>+#
>+comment "Networking support is needed for USB Network Adapter support"
>+ depends on USB && !NET
>
>This comment is superfluous. If !NET, then NETDEVICES is not even available.
>Since you just moved/renamed it, you are not to blame. Though, I remove
>it in below's patch (applies on top of
>commit 5b2fc499917e5897a13add780e181b4cef197072).
>
>
>Signed-off-by: Jan Engelhardt <[email protected]>
>
>diff --git a/drivers/net/Makefile b/drivers/net/Makefile
>index c5d8423..5f6112d 100644
>--- a/drivers/net/Makefile
>+++ b/drivers/net/Makefile
>@@ -207,12 +207,7 @@ obj-$(CONFIG_WAN) += wan/
> obj-$(CONFIG_ARCNET) += arcnet/
> obj-$(CONFIG_NET_PCMCIA) += pcmcia/
>
>-obj-$(CONFIG_USB_CATC) += usb/
>-obj-$(CONFIG_USB_KAWETH) += usb/
>-obj-$(CONFIG_USB_PEGASUS) += usb/
>-obj-$(CONFIG_USB_RTL8150) += usb/
>-obj-$(CONFIG_USB_USBNET) += usb/
>-obj-$(CONFIG_USB_ZD1201) += usb/
>+obj-$(CONFIG_NETDEV_USB) += usb/
>
> obj-y += wireless/
> obj-$(CONFIG_NET_TULIP) += tulip/
>diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
>index 3de564b..2458f99 100644
>--- a/drivers/net/usb/Kconfig
>+++ b/drivers/net/usb/Kconfig
>@@ -1,11 +1,12 @@
> #
> # USB Network devices configuration
> #
>-comment "Networking support is needed for USB Network Adapter support"
>- depends on USB && !NET
>-
>-menu "USB Network Adapters"
>+menuconfig NETDEVICES_USB
>+ bool "USB Network Adapters"
> depends on USB && NET
>+ default y
>+
>+if NETDEVICES_USB
>
> config USB_CATC
> tristate "USB CATC NetMate-based Ethernet device support (EXPERIMENTAL)"
>@@ -334,5 +335,4 @@ config USB_NET_ZAURUS
> really need this non-conformant variant of CDC Ethernet (or in
> some cases CDC MDLM) protocol, not "g_ether".
>
>-
>-endmenu
>+endif # NETDEVICES_USB
>#<EOF>
>
> Jan
>--

Jan
--