strip: move driver to staging
Subsystems (2)
PATCH 1/4 strip: move driver to staging
Move the strip ("Starmode Radio IP") driver to drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.
This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
--- drivers/net/wireless/Kconfig | 23 - drivers/net/wireless/Makefile | 1 - drivers/net/wireless/strip.c | 2805 ---------------------------------------- drivers/staging/Kconfig | 2 + drivers/staging/Makefile | 1 + drivers/staging/strip/Kconfig | 22 + drivers/staging/strip/Makefile | 1 + drivers/staging/strip/strip.c | 2805 ++++++++++++++++++++++++++++++++++++++++ 8 files changed, 2831 insertions(+), 2829 deletions(-) delete mode 100644 drivers/net/wireless/strip.c create mode 100644 drivers/staging/strip/Kconfig create mode 100644 drivers/staging/strip/Makefile create mode 100644 drivers/staging/strip/strip.c
drivers/net/wireless/Kconfig +0 -23
@@ -24,29 +24,6 @@ menuconfig WLAN_PRE80211 This option does not affect the kernel build, it only lets you choose drivers. -config STRIP - tristate "STRIP (Metricom starmode radio IP)" - depends on INET && WLAN_PRE80211 - select WIRELESS_EXT - ---help--- - Say Y if you have a Metricom radio and intend to use Starmode Radio - IP. STRIP is a radio protocol developed for the MosquitoNet project - to send Internet traffic using Metricom radios. Metricom radios are - small, battery powered, 100kbit/sec packet radio transceivers, about - the size and weight of a cellular telephone. (You may also have heard - them called "Metricom modems" but we avoid the term "modem" because - it misleads many people into thinking that you can plug a Metricom - modem into a phone line and use it as a modem.) - - You can use STRIP on any Linux machine with a serial port, although - it is obviously most useful for people with laptop computers. If you - think you might get a Metricom radio in the future, there is no harm - in saying Y to STRIP now, except that it makes the kernel a bit - bigger. - - To compile this as a module, choose M here: the module will be - called strip. - config ARLAN tristate "Aironet Arlan 655 & IC2200 DS support" depends on ISA && !64BIT && WLAN_PRE80211
drivers/net/wireless/Makefile +0 -1
@@ -5,7 +5,6 @@ obj-$(CONFIG_IPW2100) += ipw2x00/ obj-$(CONFIG_IPW2200) += ipw2x00/ -obj-$(CONFIG_STRIP) += strip.o obj-$(CONFIG_ARLAN) += arlan.o arlan-objs := arlan-main.o arlan-proc.o
drivers/net/wireless/strip.c deleted +0 -2805
2,805 lines not on this page · raw message
drivers/staging/Kconfig +2 -0
@@ -127,5 +127,7 @@ source "drivers/staging/sep/Kconfig" source "drivers/staging/iio/Kconfig" +source "drivers/staging/strip/Kconfig" + endif # !STAGING_EXCLUDE_BUILD endif # STAGING
drivers/staging/Makefile +1 -0
@@ -45,3 +45,4 @@ obj-$(CONFIG_VME_BUS) += vme/ obj-$(CONFIG_RAR_REGISTER) += rar/ obj-$(CONFIG_DX_SEP) += sep/ obj-$(CONFIG_IIO) += iio/ +obj-$(CONFIG_STRIP) += strip/
drivers/staging/strip/Kconfig new file +22 -0
@@ -0,0 +1,22 @@ +config STRIP + tristate "STRIP (Metricom starmode radio IP)" + depends on INET + select WIRELESS_EXT + ---help--- + Say Y if you have a Metricom radio and intend to use Starmode Radio + IP. STRIP is a radio protocol developed for the MosquitoNet project + to send Internet traffic using Metricom radios. Metricom radios are + small, battery powered, 100kbit/sec packet radio transceivers, about + the size and weight of a cellular telephone. (You may also have heard + them called "Metricom modems" but we avoid the term "modem" because + it misleads many people into thinking that you can plug a Metricom + modem into a phone line and use it as a modem.) + + You can use STRIP on any Linux machine with a serial port, although + it is obviously most useful for people with laptop computers. If you + think you might get a Metricom radio in the future, there is no harm + in saying Y to STRIP now, except that it makes the kernel a bit + bigger. + + To compile this as a module, choose M here: the module will be + called strip.
drivers/staging/strip/Makefile new file +1 -0
@@ -0,0 +1 @@ +obj-$(CONFIG_STRIP) += strip.o
drivers/staging/strip/strip.c new file +2805 -0
125 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
1.6.2.5
Part of the diff is not shown here. View the raw message.
PATCH 2/4 arlan: move driver to staging
Move the arlan driver to drivers/staging. This is another pre-802.11 driver that has seen virtually no non-API-fixup activity in years, and for which no active hardware is likely to still exist. This driver represents unnecessary ongoing maintenance for no clear benefit.
This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
--- drivers/net/wireless/Kconfig | 16 - drivers/net/wireless/Makefile | 4 - drivers/net/wireless/arlan-main.c | 1887 ------------------------------------ drivers/net/wireless/arlan-proc.c | 1253 ------------------------ drivers/net/wireless/arlan.h | 539 ---------- drivers/staging/Kconfig | 2 + drivers/staging/Makefile | 1 + drivers/staging/arlan/Kconfig | 15 + drivers/staging/arlan/Makefile | 3 + drivers/staging/arlan/arlan-main.c | 1887 ++++++++++++++++++++++++++++++++++++ drivers/staging/arlan/arlan-proc.c | 1253 ++++++++++++++++++++++++ drivers/staging/arlan/arlan.h | 539 ++++++++++ 12 files changed, 3700 insertions(+), 3699 deletions(-) delete mode 100644 drivers/net/wireless/arlan-main.c delete mode 100644 drivers/net/wireless/arlan-proc.c delete mode 100644 drivers/net/wireless/arlan.h create mode 100644 drivers/staging/arlan/Kconfig create mode 100644 drivers/staging/arlan/Makefile create mode 100644 drivers/staging/arlan/arlan-main.c create mode 100644 drivers/staging/arlan/arlan-proc.c create mode 100644 drivers/staging/arlan/arlan.h
drivers/net/wireless/Kconfig +0 -16
@@ -24,22 +24,6 @@ menuconfig WLAN_PRE80211 This option does not affect the kernel build, it only lets you choose drivers. -config ARLAN - tristate "Aironet Arlan 655 & IC2200 DS support" - depends on ISA && !64BIT && WLAN_PRE80211 - select WIRELESS_EXT - ---help--- - Aironet makes Arlan, a class of wireless LAN adapters. These use the - www.Telxon.com chip, which is also used on several similar cards. - This driver is tested on the 655 and IC2200 series cards. Look at - <http://www.ylenurme.ee/~elmer/655/> for the latest information. - - The driver is built as two modules, arlan and arlan-proc. The latter - is the /proc interface and is not needed most of time. - - On some computers the card ends up in non-valid state after some - time. Use a ping-reset script to clear it. - config WAVELAN tristate "AT&T/Lucent old WaveLAN & DEC RoamAbout DS ISA support" depends on ISA && WLAN_PRE80211
drivers/net/wireless/Makefile +0 -4
@@ -5,10 +5,6 @@ obj-$(CONFIG_IPW2100) += ipw2x00/ obj-$(CONFIG_IPW2200) += ipw2x00/ -obj-$(CONFIG_ARLAN) += arlan.o - -arlan-objs := arlan-main.o arlan-proc.o - # Obsolete cards obj-$(CONFIG_WAVELAN) += wavelan.o obj-$(CONFIG_PCMCIA_NETWAVE) += netwave_cs.o
drivers/net/wireless/arlan-main.c deleted +0 -1887
1,887 lines not on this page · raw message
drivers/net/wireless/arlan-proc.c deleted +0 -1253
1,081 lines not on this page · raw message
drivers/net/wireless/arlan.h deleted +0 -539
539 lines not on this page · raw message
drivers/staging/arlan/Kconfig new file +15 -0
Lines of this file are not shown. View the raw message.
drivers/staging/arlan/Makefile new file +3 -0
Lines of this file are not shown. View the raw message.
drivers/staging/arlan/arlan-main.c new file +1887 -0
1,887 lines not on this page · raw message
drivers/staging/arlan/arlan-proc.c new file +1253 -0
1,253 lines not on this page · raw message
drivers/staging/arlan/arlan.h new file +539 -0
539 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
1.6.2.5
Part of the diff is not shown here. View the raw message.
A message that is not in the archive (sent off-list or not yet received): [email protected]
PATCH 4/4 netwave: move driver to staging
Move the netwave driver to drivers/staging. This is another pre-802.11 driver that has seen virtually no non-API-fixup activity in years, and for which no active hardware is likely to still exist. This driver represents unnecessary ongoing maintenance for no clear benefit.
This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
--- drivers/net/wireless/Kconfig | 65 +-- drivers/net/wireless/Makefile | 3 - drivers/net/wireless/netwave_cs.c | 1389 ---------------------------------- drivers/staging/Kconfig | 2 + drivers/staging/Makefile | 1 + drivers/staging/netwave/Kconfig | 10 + drivers/staging/netwave/Makefile | 1 + drivers/staging/netwave/netwave_cs.c | 1389 ++++++++++++++++++++++++++++++++++ 8 files changed, 1420 insertions(+), 1440 deletions(-) delete mode 100644 drivers/net/wireless/netwave_cs.c create mode 100644 drivers/staging/netwave/Kconfig create mode 100644 drivers/staging/netwave/Makefile create mode 100644 drivers/staging/netwave/netwave_cs.c
drivers/net/wireless/Kconfig +17 -48
@@ -7,47 +7,17 @@ menuconfig WLAN depends on !S390 default y ---help--- - This section contains all the pre 802.11 and 802.11 wireless - device drivers. For a complete list of drivers and documentation + This section contains all the 802.11 wireless LAN device + drivers. For a complete list of drivers and documentation on them refer to the wireless wiki: http://wireless.kernel.org/en/users/Drivers if WLAN -menuconfig WLAN_PRE80211 - bool "Wireless LAN (pre-802.11)" - depends on NETDEVICES - ---help--- - Say Y if you have any pre-802.11 wireless LAN hardware. - - This option does not affect the kernel build, it only - lets you choose drivers. - -config PCMCIA_NETWAVE - tristate "Xircom Netwave AirSurfer Pcmcia wireless support" - depends on PCMCIA && WLAN_PRE80211 - select WIRELESS_EXT - help - Say Y here if you intend to attach this type of PCMCIA (PC-card) - wireless Ethernet networking card to your computer. - - To compile this driver as a module, choose M here: the module will be - called netwave_cs. If unsure, say N. - - -menuconfig WLAN_80211 - bool "Wireless LAN (IEEE 802.11)" - depends on NETDEVICES - ---help--- - Say Y if you have any 802.11 wireless LAN hardware. - - This option does not affect the kernel build, it only - lets you choose drivers. - config PCMCIA_RAYCS tristate "Aviator/Raytheon 2.4GHz wireless support" - depends on PCMCIA && WLAN_80211 + depends on PCMCIA select WIRELESS_EXT ---help--- Say Y here if you intend to attach an Aviator/Raytheon PCMCIA
@@ -60,7 +30,6 @@ config PCMCIA_RAYCS config LIBERTAS tristate "Marvell 8xxx Libertas WLAN driver support" - depends on WLAN_80211 select WIRELESS_EXT select LIB80211 select FW_LOADER
@@ -100,7 +69,7 @@ config LIBERTAS_DEBUG config LIBERTAS_THINFIRM tristate "Marvell 8xxx Libertas WLAN driver support with thin firmware" - depends on WLAN_80211 && MAC80211 + depends on MAC80211 select FW_LOADER ---help--- A library for Marvell Libertas 8xxx devices using thinfirm.
@@ -113,7 +82,7 @@ config LIBERTAS_THINFIRM_USB config AIRO tristate "Cisco/Aironet 34X/35X/4500/4800 ISA and PCI cards" - depends on ISA_DMA_API && WLAN_80211 && (PCI || BROKEN) + depends on ISA_DMA_API && (PCI || BROKEN) select WIRELESS_EXT select CRYPTO ---help---
@@ -131,7 +100,7 @@ config AIRO config ATMEL tristate "Atmel at76c50x chipset 802.11b support" - depends on (PCI || PCMCIA) && WLAN_80211 + depends on (PCI || PCMCIA) select WIRELESS_EXT select FW_LOADER select CRC32
@@ -165,7 +134,7 @@ config PCMCIA_ATMEL config AT76C50X_USB tristate "Atmel at76c503/at76c505/at76c505a USB cards" - depends on MAC80211 && WLAN_80211 && USB + depends on MAC80211 && USB select FW_LOADER ---help--- Enable support for USB Wireless devices using Atmel at76c503,
@@ -173,7 +142,7 @@ config AT76C50X_USB config AIRO_CS tristate "Cisco/Aironet 34X/35X/4500/4800 PCMCIA cards" - depends on PCMCIA && (BROKEN || !M32R) && WLAN_80211 + depends on PCMCIA && (BROKEN || !M32R) select WIRELESS_EXT select CRYPTO select CRYPTO_AES
@@ -193,7 +162,7 @@ config AIRO_CS config PCMCIA_WL3501 tristate "Planet WL3501 PCMCIA cards" - depends on EXPERIMENTAL && PCMCIA && WLAN_80211 + depends on EXPERIMENTAL && PCMCIA select WIRELESS_EXT ---help--- A driver for WL3501 PCMCIA 802.11 wireless cards made by Planet.
@@ -202,7 +171,7 @@ config PCMCIA_WL3501 config PRISM54 tristate 'Intersil Prism GT/Duette/Indigo PCI/Cardbus (DEPRECATED)' - depends on PCI && EXPERIMENTAL && WLAN_80211 + depends on PCI && EXPERIMENTAL select WIRELESS_EXT select FW_LOADER ---help---
@@ -224,7 +193,7 @@ config PRISM54 config USB_ZD1201 tristate "USB ZD1201 based Wireless device support" - depends on USB && WLAN_80211 + depends on USB select WIRELESS_EXT select FW_LOADER ---help---
@@ -242,7 +211,7 @@ config USB_ZD1201 config USB_NET_RNDIS_WLAN tristate "Wireless RNDIS USB support" - depends on USB && WLAN_80211 && EXPERIMENTAL + depends on USB && EXPERIMENTAL depends on CFG80211 select USB_USBNET select USB_NET_CDCETHER
@@ -270,7 +239,7 @@ config USB_NET_RNDIS_WLAN config RTL8180 tristate "Realtek 8180/8185 PCI support" - depends on MAC80211 && PCI && WLAN_80211 && EXPERIMENTAL + depends on MAC80211 && PCI && EXPERIMENTAL select EEPROM_93CX6 ---help--- This is a driver for RTL8180 and RTL8185 based cards.
@@ -326,7 +295,7 @@ config RTL8180 config RTL8187 tristate "Realtek 8187 and 8187B USB support" - depends on MAC80211 && USB && WLAN_80211 + depends on MAC80211 && USB select EEPROM_93CX6 ---help--- This is a driver for RTL8187 and RTL8187B based cards.
@@ -355,7 +324,7 @@ config RTL8187_LEDS config ADM8211 tristate "ADMtek ADM8211 support" - depends on MAC80211 && PCI && WLAN_80211 && EXPERIMENTAL + depends on MAC80211 && PCI && EXPERIMENTAL select CRC32 select EEPROM_93CX6 ---help---
@@ -382,7 +351,7 @@ config ADM8211 config MAC80211_HWSIM tristate "Simulated radio testing tool for mac80211" - depends on MAC80211 && WLAN_80211 + depends on MAC80211 ---help--- This driver is a developer testing tool that can be used to test IEEE 802.11 networking stack (mac80211) functionality. This is not
@@ -395,7 +364,7 @@ config MAC80211_HWSIM config MWL8K tristate "Marvell 88W8xxx PCI/PCIe Wireless support" - depends on MAC80211 && PCI && WLAN_80211 && EXPERIMENTAL + depends on MAC80211 && PCI && EXPERIMENTAL ---help--- This driver supports Marvell TOPDOG 802.11 wireless cards.
drivers/net/wireless/Makefile +0 -3
@@ -5,9 +5,6 @@ obj-$(CONFIG_IPW2100) += ipw2x00/ obj-$(CONFIG_IPW2200) += ipw2x00/ -# Obsolete cards -obj-$(CONFIG_PCMCIA_NETWAVE) += netwave_cs.o - obj-$(CONFIG_HERMES) += orinoco/ obj-$(CONFIG_AIRO) += airo.o
drivers/net/wireless/netwave_cs.c deleted +0 -1389
1,389 lines not on this page · raw message
drivers/staging/Kconfig +2 -0
@@ -133,5 +133,7 @@ source "drivers/staging/arlan/Kconfig" source "drivers/staging/wavelan/Kconfig" +source "drivers/staging/netwave/Kconfig" + endif # !STAGING_EXCLUDE_BUILD endif # STAGING
drivers/staging/Makefile +1 -0
@@ -49,3 +49,4 @@ obj-$(CONFIG_STRIP) += strip/ obj-$(CONFIG_ARLAN) += arlan/ obj-$(CONFIG_WAVELAN) += wavelan/ obj-$(CONFIG_PCMCIA_WAVELAN) += wavelan/ +obj-$(CONFIG_PCMCIA_NETWAVE) += netwave/
drivers/staging/netwave/Kconfig new file +10 -0
@@ -0,0 +1,10 @@ +config PCMCIA_NETWAVE + tristate "Xircom Netwave AirSurfer Pcmcia wireless support" + depends on PCMCIA + select WIRELESS_EXT + help + Say Y here if you intend to attach this type of PCMCIA (PC-card) + wireless Ethernet networking card to your computer. + + To compile this driver as a module, choose M here: the module will be + called netwave_cs. If unsure, say N.
drivers/staging/netwave/Makefile new file +1 -0
@@ -0,0 +1 @@ +obj-$(CONFIG_PCMCIA_NETWAVE) += netwave_cs.o
drivers/staging/netwave/netwave_cs.c new file +1389 -0
1,389 lines not on this page · raw message
Signature · 1 line
1.6.2.5
Re: Staging: netwave: move driver to staging
Quoted 1 line · John W. Linville
-menuconfig
WLAN_80211
If you are removing this config option, what should we use to depend on the wireless stack? This is what the staging wireless drivers use, should they be changed to use something else?
thanks,
greg k-h
Re: Staging: netwave: move driver to staging
Quoted 6 lines · Greg KH
Quoted 1 line · John W. Linville
-menuconfig
WLAN_80211If you are removing this config option, what should we use to depend on the wireless stack? This is what the staging wireless drivers use, should they be changed to use something else?
Hmmm...probably WLAN will do. I should probably make that depend on NETDEVICES as well...
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Re: Staging: netwave: move driver to staging
Quoted 9 lines · John W. Linville
Quoted 6 lines · Greg KH
Quoted 1 line · John W. Linville
-menuconfig
WLAN_80211If you are removing this config option, what should we use to depend on the wireless stack? This is what the staging wireless drivers use, should they be changed to use something else?
Hmmm...probably WLAN will do. I should probably make that depend on NETDEVICES as well...
Ok, I'll go change the staging drivers now to depend on that so they don't "disappear" when this change goes through your tree.
thanks,
greg k-h
Quoted 7 lines · On Tue, 20 Oct 2009 14:17:07 +0900 John W. Linville wrote:
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
is drivers/staging/ a temporary home or a permanent home?
so no explanation of what it means to be in drivers/staging/ and no warning to users? either in Kconfig or in feature-removal-schedule.txt?
Quoted 15 lines
Signed-off-by John W. Linville <[email protected]>--- drivers/net/wireless/Kconfig | 23 - drivers/net/wireless/Makefile | 1 - drivers/net/wireless/strip.c | 2805 ---------------------------------------- drivers/staging/Kconfig | 2 + drivers/staging/Makefile | 1 + drivers/staging/strip/Kconfig | 22 + drivers/staging/strip/Makefile | 1 + drivers/staging/strip/strip.c | 2805 ++++++++++++++++++++++++++++++++++++++++ 8 files changed, 2831 insertions(+), 2829 deletions(-) delete mode 100644 drivers/net/wireless/strip.c create mode 100644 drivers/staging/strip/Kconfig create mode 100644 drivers/staging/strip/Makefile create mode 100644 drivers/staging/strip/strip.c
---
~Randy
Quoted 11 lines · Randy Dunlap
Quoted 7 lines · On Tue, 20 Oct 2009 14:17:07 +0900 John W. Linville wrote:
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
is
drivers/staging/a temporary home or a permanent home?
I would presume it to be "permanent until removed"...
Quoted 2 lines
so no explanation of what it means to be in
drivers/staging/and no warning to users? either inKconfigor in feature-removal-schedule.txt?
Moving it to staging _is_ the warning. As for the removal schedule, I imagined that Greg would "do the needful" before actually removing the driver.
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Quoted 11 lines · On Mon, 26 Oct 2009 12:46:45 -0400 John W. Linville wrote:
Quoted 10 lines · Randy Dunlap
Quoted 7 lines · On Tue, 20 Oct 2009 14:17:07 +0900 John W. Linville wrote:
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
Quoted 8 lines
Quoted 1 line
is
drivers/staging/a temporary home or a permanent home?I would presume it to be "permanent until removed"...
Quoted 2 lines
so no explanation of what it means to be in
drivers/staging/and no warning to users? either inKconfigor in feature-removal-schedule.txt?Moving it to staging _is_ the warning.
Nah. Maybe it is for developers, but not for users (IMO of course).
Quoted 3 lines
As for the removal schedule,
I imagined that Greg would "do the needful" before actually removing the driver.
He usually wants patches, but that's up to him.
---
~Randy
Quoted 22 lines · Randy Dunlap
Quoted 11 lines · On Mon, 26 Oct 2009 12:46:45 -0400 John W. Linville wrote:
Quoted 10 lines · Randy Dunlap
Quoted 7 lines · On Tue, 20 Oct 2009 14:17:07 +0900 John W. Linville wrote:
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This patch brought to you by the "hacking" session at the 2009 Kernel Summit in Tokyo, Japan...
Quoted 8 lines
Quoted 1 line
is
drivers/staging/a temporary home or a permanent home?I would presume it to be "permanent until removed"...
Quoted 2 lines
so no explanation of what it means to be in
drivers/staging/and no warning to users? either inKconfigor in feature-removal-schedule.txt?Moving it to staging _is_ the warning.
Quoted 7 lines
Nah. Maybe it is for developers, but not for users (IMO of course).
Quoted 3 lines
As for the removal schedule,
I imagined that Greg would "do the needful" before actually removing the driver.He usually wants patches, but that's up to him.
Nah, it's trivial for me to delete things on my own :)
thanks,
greg k-h
Quoted 4 lines · John W. Linville
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.
This seems like abuse of the staging process.
There's no TODO to say what needs to be fixed. You just don't want to maintain it. Because there's nothing to fix, noone has reason to patch it, and the (working, good enough) driver will just be removed.
It also marks driver as broken when it is not...
What about removing it in the regular way, that'sDocumentation/feature-removal.txt ?
Pavel
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Quoted 1 line
It also marks driver as broken when it is not...
Given the changes since anyone last used it I would assume it is.
Quoted 3 lines
What about removing it in the regular way, that's
Documentation/feature-removal.txt?
I'd go with that. And remove it ASAP along with about twenty other similar 'ex drivers'
Quoted 7 lines · Pavel Machek
Quoted 4 lines · John W. Linville
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This seems like abuse of the staging process.
My mailbox has been filling-up with discussions of using staging like this for the past few weeks. I'm reasonably certain those threads were on the public lists.
Quoted 3 lines
There's no TODO to say what needs to be fixed. You just don't want to maintain it. Because there's nothing to fix, noone has reason to patch it, and the (working, good enough) driver will just be removed.
"ongoing maintenance for no clear benefit" -- that is what is wrong with it.
Quoted 1 line
It also marks driver as broken when it is not...
I didn't mark it broken, I proposed moving it to staging. As for whether or not it actually is broken, how do you know?
The drivers in this thread are for pre-802.11 devices -- old ones (e.g. ISA) at that. If we actually have users that are willing to maintain them then maybe that is fine. But I don't see the benefit of maintaining these simply as extra targets for API change "bombing runs"...
Quoted 2 lines
What about removing it in the regular way, that's
Documentation/feature-removal.txt?
This has been discussed recently as an alternative for unmaintained drivers. I suspect this is better than adding a note to a file that no one reads...
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Quoted 13 lines · John W. Linville
Quoted 7 lines · Pavel Machek
Quoted 4 lines · John W. Linville
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.This seems like abuse of the staging process.
My mailbox has been filling-up with discussions of using staging like this for the past few weeks. I'm reasonably certain those threads were on the public lists.
Quoted 6 lines
Quoted 3 lines
There's no TODO to say what needs to be fixed. You just don't want to maintain it. Because there's nothing to fix, noone has reason to patch it, and the (working, good enough) driver will just be removed.
"ongoing maintenance for no clear benefit" -- that is what is wrong with it.
supporting existing hardware is no longer a 'clear benifit'?
is this driver broken (as in not working)?
other than the fact that you don't think many people have this hardware, is there anything wrong with this driver?
Quoted 11 lines
Quoted 1 line
It also marks driver as broken when it is not...
I didn't mark it broken, I proposed moving it to staging. As for whether or not it actually is broken, how do you know?
The drivers in this thread are for pre-802.11 devices -- old ones (e.g. ISA) at that. If we actually have users that are willing to maintain them then maybe that is fine. But I don't see the benefit of maintaining these simply as extra targets for API change "bombing runs"...
Quoted 6 lines
Quoted 2 lines
What about removing it in the regular way, that's
Documentation/feature-removal.txt?This has been discussed recently as an alternative for unmaintained drivers. I suspect this is better than adding a note to a file that no one reads...
this is exactly the type of thing I was afraid would start happening when I noticed the discussion. when I tried to bring it up I was told that this would only be used for things that were really broken or otherwise causing significant problems.
if this driver hasn't been changed other than for blanket API changes, how is it causing significant problems?
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
David Lang
Quoted 6 lines · On Mon, Oct 26, 2009 at 10:18:20AM -0700, [email protected] wrote:
Quoted 2 lines · John W. Linville
"ongoing maintenance for no clear benefit" -- that is what is wrong with it.
supporting existing hardware is no longer a 'clear benifit'?
Do you own or have access to such hardware? If the hardware is obsolete, unused, and mostly non-existant then what is the benefit to maintaining a driver that is rotting?
Quoted 1 line
is this driver broken (as in not working)?
I have no idea. I suspect that you don't know either.
Quoted 2 lines
other than the fact that you don't think many people have this hardware, is there anything wrong with this driver?
It sits there, bloating the target area for API maintenance and providing a likely broken example of how to write Linux drivers.
Quoted 2 lines
if this driver hasn't been changed other than for blanket API changes, how is it causing significant problems?
It is a distraction for no clear (or even murky) benefit, one more thing to keep compiling.
Quoted 3 lines
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
Not much different, possibly worse if they created a bigger distraction. But if they actually had the hardware we might at least be confident that the effort was in some way worthwhile...
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Quoted 3 lines · On Mon, Oct 26, 2009 at 10:18:20AM -0700, [email protected] wrote:
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
A person "claiming maintainership" would then be responsible for keeping the API up to date and ensuring that the driver worked. To do that, hardware would probably need to be present.
Do you have this kind of hardware and are willing to accept ownership of this driver?
thanks,
greg k-h
Quoted 8 lines · Greg KH
Quoted 3 lines · On Mon, Oct 26, 2009 at 10:18:20AM -0700, [email protected] wrote:
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
A person "claiming maintainership" would then be responsible for keeping the API up to date and ensuring that the driver worked. To do that, hardware would probably need to be present.
actually, I understood that the person changing the API was responsible for making the changes. when did this change?
Quoted 2 lines
Do you have this kind of hardware and are willing to accept ownership of this driver?
no, I do not have the hardware, but if there are no bugs reported against this driverit would seem that having a 'maintainer' who made absolutly no changes to the driver (just allowing API changes by others to be implemented) would be the same thing as having no maintainer, but in the first case you are willing to have the driver in the kernel, in the other you want to rip it out.
it used to be (not that long ago) that when people said that the reason they didn't push their driver upstream into the kernel because there wasn't that much demand for it, the response was that we wanted drivers for everything, no matter how small the user base. I remember seeing posts from core developers saying that we had drivers for hardware where there were only single digit quantities ever built.
now it appears that you have to have 'enough' users (an amount undefined) or a person to specificly take maintainership of the driver to keep it in.
prior to the kernel summit, the criteria for having something moved out of the kernel into staging was for fairly significant problems (with a sloppy edge of 'or an unreasonable maintinance burden')
I don't think anyone who read that would have thought that 'an unreasonable maintinance burden' could be "I don't want to change this driver when I change an API"
for old hardware the driver _should_ be static except for API changes. the hardware isn't changing, why should the driver.
David Lang
From: [email protected]
Date: Mon, 26 Oct 2009 21:17:55 -0700 (PDT)
Quoted 6 lines · Greg KH
Quoted 4 lines
Do you have this kind of hardware and are willing to accept ownership of
this driver?
Quoted 6 lines
no, I do not have the hardware, but if there are no bugs reported against this driverit would seem that having a 'maintainer' who made absolutly no changes to the driver (just allowing API changes by others to be implemented) would be the same thing as having no maintainer, but in the first case you are willing to have the driver in the kernel, in the other you want to rip it out.
Nobody is using this driver, and we're pretty damn confident of that.
Quoted 7 lines · [email protected] wrote:
Quoted 3 lines · Greg KH
A person "claiming maintainership" would then be responsible for keeping the API up to date and ensuring that the driver worked. To do that, hardware would probably need to be present.
actually, I understood that the person changing the API was responsible for making the changes. when did this change?
I'm not sure what the procedure with drivers/staging/ is, which is a middle ground between in-tree and out-of-tree. For in-tree code, last time I watched the procedure was thus:
- Those who change an infrastructure are responsible to /code/ the
necessary changes in all infrastructure using subsystems.
- Usually they are also responsible to submit those changes upstream;
if the change is done in a source-compatible manner during a transition period, then those API changes in dependent subsystems are sometimes submitted via the respective subsystem maintainers instead, but that's not typical.
- Those who introduce such changes usually cannot be expected to
runtime-test except with one or few subsystems for which they happen to have matching hardware. Thus, a small but real danger of regressions remains. Usually only the subsystem maintainer or users can detect and fix such regressions.
But remember, there are no formal rules, there are only "best practices". (And sometimes new procedures are tried out in order to find out whether they work or not, such as the 2.4+2.5 -> 2.6 switch of development models, the merge window, linux-next, drivers/staging/ as a way in for out-of-tree drivers, drivers/staging as a way out for broken or obsolete drivers...)
Signature · 3 lines
Stefan Richter -=====-==--= =-=- ==-== http://arcgraph.de/sr/
Quoted 7 lines · On Mon, Oct 26, 2009 at 09:17:55PM -0700, [email protected] wrote:
Quoted 5 lines · Greg KH
Quoted 3 lines · On Mon, Oct 26, 2009 at 10:18:20AM -0700, [email protected] wrote:
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
Quoted 6 lines
Quoted 3 lines
A person "claiming maintainership" would then be responsible for keeping the API up to date and ensuring that the driver worked. To do that, hardware would probably need to be present.
actually, I understood that the person changing the API was responsible for making the changes. when did this change?
It did not.
Quoted 10 lines
Quoted 2 lines
Do you have this kind of hardware and are willing to accept ownership of this driver?
no, I do not have the hardware, but if there are no bugs reported against this driverit would seem that having a 'maintainer' who made absolutly no changes to the driver (just allowing API changes by others to be implemented) would be the same thing as having no maintainer, but in the first case you are willing to have the driver in the kernel, in the other you want to rip it out.
Quoted 8 lines
it used to be (not that long ago) that when people said that the reason they didn't push their driver upstream into the kernel because there wasn't that much demand for it, the response was that we wanted drivers for everything, no matter how small the user base. I remember seeing posts from core developers saying that we had drivers for hardware where there were only single digit quantities ever built.
now it appears that you have to have 'enough' users (an amount undefined)
That amount would be 1. This driver does not have that, so it can be removed.
thanks,
greg k-h
Quoted 34 lines · Greg KH
Quoted 7 lines · On Mon, Oct 26, 2009 at 09:17:55PM -0700, [email protected] wrote:
Quoted 5 lines · Greg KH
Quoted 3 lines · On Mon, Oct 26, 2009 at 10:18:20AM -0700, [email protected] wrote:
if someone were to claim 'maintainership' and then do nothing other than complain if someone else were to change an API but not fix this in the process, how would this be different than the current situation?
Quoted 6 lines
Quoted 3 lines
A person "claiming maintainership" would then be responsible for keeping the API up to date and ensuring that the driver worked. To do that, hardware would probably need to be present.
actually, I understood that the person changing the API was responsible for making the changes. when did this change?
It did not.
Quoted 17 lines
Quoted 2 lines
Do you have this kind of hardware and are willing to accept ownership of this driver?
no, I do not have the hardware, but if there are no bugs reported against this driverit would seem that having a 'maintainer' who made absolutly no changes to the driver (just allowing API changes by others to be implemented) would be the same thing as having no maintainer, but in the first case you are willing to have the driver in the kernel, in the other you want to rip it out.
it used to be (not that long ago) that when people said that the reason they didn't push their driver upstream into the kernel because there wasn't that much demand for it, the response was that we wanted drivers for everything, no matter how small the user base. I remember seeing posts from core developers saying that we had drivers for hardware where there were only single digit quantities ever built.
Quoted 4 lines
Quoted 1 line
now it appears that you have to have 'enough' users (an amount undefined)
That amount would be 1. This driver does not have that, so it can be removed.
while I definantly agree that this driver is unlikly to have many users, unless it is known broken (which David Miller has stated that it is, but had not been mentioned previously in this thread), how can you know that the number of users has dropped below 1?
David Lang
Quoted 9 lines · On Tue, Oct 27, 2009 at 11:39:54AM -0700, [email protected] wrote:
Quoted 4 lines
Quoted 1 line
now it appears that you have to have 'enough' users (an amount undefined)
That amount would be 1. This driver does not have that, so it can be removed.
while I definantly agree that this driver is unlikly to have many users, unless it is known broken (which David Miller has stated that it is, but had not been mentioned previously in this thread), how can you know that the number of users has dropped below 1?
As the driver is known to be broken, I would think that the number of users has dropped below 1 :)
And again, if someone for this driver shows up, it is _trivial_ to add it back to the tree.
thanks,
greg k-h
Quoted 4 lines · On Tue, Oct 27, 2009 at 11:39:54AM -0700, [email protected] wrote:
while I definantly agree that this driver is unlikly to have many users, unless it is known broken (which David Miller has stated that it is, but had not been mentioned previously in this thread), how can you know that the number of users has dropped below 1?
You are free to disagree, but IMHO these devices are so unlikely to still be in use that the burden of proof of whether or not the driver works lies upon a forthcoming user. Further, such a user needs to be at least willing to test patches and probably needs to be in a position to analyze most "normal" problems that might materialize. IOW, such a user needs to be a de facto maintainer.
Blindly holding onto a driver not which not only has no known users but which is believed to have no users at all merely because we can't _prove_ it is broken is just a waste of resources.
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
O> while I definantly agree that this driver is unlikly to have many users,
Quoted 3 lines
unless it is known broken (which David Miller has stated that it is, but had not been mentioned previously in this thread), how can you know that the number of users has dropped below 1?
By the fact it never turns up in bugzilla, in kernel questions or in google - and then by deleting it and checking for screaming noises.
Do you have a better testing procedure ?
Quoted 7 lines · John W. Linville
Quoted 6 lines · Pavel Machek
Quoted 4 lines · John W. Linville
Move the strip ("Starmode Radio IP") driver to
drivers/staging. For several years this driver has only seen API "bombing-run" changes, and few people ever had the hardware. This driver represents unnecessary ongoing maintenance for no clear benefit.
Quoted 5 lines
Quoted 1 line
This seems like abuse of the staging process.
My mailbox has been filling-up with discussions of using staging like this for the past few weeks. I'm reasonably certain those threads were on the public lists.
Yes and I skimmed those.
Quoted 6 lines
Quoted 3 lines
There's no TODO to say what needs to be fixed. You just don't want to maintain it. Because there's nothing to fix, noone has reason to patch it, and the (working, good enough) driver will just be removed.
"ongoing maintenance for no clear benefit" -- that is what is wrong with it.
And how is moving to staging helping with that?
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement. So it looks like staging is wrong tool for the job.
Quoted 6 lines
Quoted 2 lines
What about removing it in the regular way, that's
Documentation/feature-removal.txt?This has been discussed recently as an alternative for unmaintained drivers. I suspect this is better than adding a note to a file that no one reads...
There are other alternatives, like printk() in module_init...
Pavel
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Quoted 5 lines
Quoted 3 lines
This has been discussed recently as an alternative for unmaintained drivers. I suspect this is better than adding a note to a file that no one reads...
There are other alternatives, like
printk()inmodule_init...
Put it in feature removal for tomrrow wait until midnight and then submit a deletion patch. Like anyone cares about the strip driver
Quoted 11 lines · Pavel Machek
Quoted 8 lines · John W. Linville
Quoted 3 lines · Pavel Machek
There's no TODO to say what needs to be fixed. You just don't want to maintain it. Because there's nothing to fix, noone has reason to patch it, and the (working, good enough) driver will just be removed.
"ongoing maintenance for no clear benefit" -- that is what is wrong with it.
And how is moving to staging helping with that?
Because no one is required to protect staging drivers from API-change breakage.
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Quoted 3 lines · Pavel Machek
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement. So it looks like staging is wrong tool for the job.
I will add such a TODO file.
It will say:
- A maintainer is needed to take ownership of this driver and they must be willing to maintain and test any future api changes needed.
thanks,
greg k-h
Quoted 6 lines · Greg KH
Quoted 2 lines · Pavel Machek
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement.
- A maintainer is needed to take ownership of this driver and they must be willing to maintain and test any future api changes needed.
There are a lot of files in the tree without assigned maintainers. (want a list? ;)
I think adding a requirement that files have maintainers before being added to not-staging is not a good thing.
It is likely that strip could have be deleted without anyone caring.
Quoted 8 lines · Joe Perches
Quoted 6 lines · Greg KH
Quoted 2 lines · Pavel Machek
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement.
- A maintainer is needed to take ownership of this driver and they must be willing to maintain and test any future api changes needed.
Quoted 7 lines
There are a lot of files in the tree without assigned maintainers. (want a list? ;)
I think adding a requirement that files have maintainers before being added to not-staging is not a good thing.
It is likely that strip could have be deleted without anyone caring.
Ok, then we can delete it, but I was trying to be nice here and provide a way for people to "rescue" it with a nice announcement and grace period.
Geesh, see what I get for being nice...
greg k-h
Quoted 8 lines · Joe Perches
Quoted 6 lines · Greg KH
Quoted 2 lines · Pavel Machek
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement.
- A maintainer is needed to take ownership of this driver and they must be willing to maintain and test any future api changes needed.
Quoted 5 lines
There are a lot of files in the tree without assigned maintainers. (want a list? ;)
I think adding a requirement that files have maintainers before being added to not-staging is not a good thing.
This is not a blanket policy. Please don't imply that it is.
Quoted 1 line
It is likely that strip could have be deleted without anyone caring.
Perhaps so...how would that be better? It would be less work for Greg, but he seems to be fine with this...
John
Signature · 2 lines
John W. Linville Someday the world will need a hero, and you [email protected] might be all we have. Be ready.
Quoted 15 lines · John W. Linville
Quoted 8 lines · Joe Perches
Quoted 6 lines · Greg KH
Quoted 2 lines · Pavel Machek
Anyway, for moving to staging, TODO file describing what needs to be fixed before it is brought back was listed as a requirement.
- A maintainer is needed to take ownership of this driver and they must be willing to maintain and test any future api changes needed.
Quoted 5 lines
There are a lot of files in the tree without assigned maintainers. (want a list? ;)
I think adding a requirement that files have maintainers before being added to not-staging is not a good thing.
Quoted 6 lines
This is not a blanket policy. Please don't imply that it is.
Quoted 1 line
It is likely that strip could have be deleted without anyone caring.
Perhaps so...how would that be better? It would be less work for Greg, but he seems to be fine with this...
Exactly, why are people who are not doing any work with these devices, suddenly worried about any additional work for me? As if people care about my patch workload suddenly :)
As everyone here has no problem with this specific driver, or the other ones posted, why is there even any discussion?
thanks,
greg k-h
(forgot the reply-all)
Quoted 4 lines · John W. Linville
Quoted 2 lines · Joe Perches
I think adding a requirement that files have maintainers before being added to not-staging is not a good thing.
This is not a blanket policy. Please don't imply that it is.
I believe it's an unnecessary precedent.
Quoted 4 lines
Quoted 1 line
It is likely that strip could have be deleted without anyone caring.
Perhaps so...how would that be better? It would be less work for Greg, but he seems to be fine with this...
It's fine by me if it moves to staging.
It's fine by me if it simply disappears as well.
It's fine by me if Greg works all he wants.