CONFIG_EXPERT was designed to toggle visibility of config options.
However, we used it in several different places to change default values
or default selections. This may produce unwanted side-effects for users
who enable EXPERT mode to toggle a single config option.
Regards
David
David Herrmann (7):
HID: Do not switch default values in EXPERT mode
Media: Do not change default values in EXPERT mode
rfkill: Do not change default values in EXPERT mode
Kconfig: Do not switch default values in EXPERT mode
usb: remove unneeded default value
pcmcia: Do not change selected options in EXPERT mode
drm: do not change selected options based on EXPERT mode
drivers/gpu/drm/Kconfig | 1 -
drivers/gpu/drm/nouveau/Kconfig | 1 -
drivers/hid/Kconfig | 30 +++++++++++++++---------------
drivers/media/common/tuners/Kconfig | 1 -
drivers/media/dvb/frontends/Kconfig | 1 -
drivers/media/video/Kconfig | 2 +-
drivers/pcmcia/Kconfig | 1 -
drivers/usb/core/Kconfig | 1 -
init/Kconfig | 2 +-
net/rfkill/Kconfig | 2 +-
usr/Kconfig | 8 ++++----
11 files changed, 22 insertions(+), 28 deletions(-)
--
1.7.10.2
Expert mode was designed to toggle visibility of options and not to
change their behavior. If a user wants to tweak a single value and
enables EXPERT mode to do this, he will have several unexpected
side effects when other subsystems toggle their default values based
on EXPERT mode. This makes EXPERT pretty useless.
Cc: Jiri Kosina <[email protected]>
Signed-off-by: David Herrmann <[email protected]>
---
drivers/hid/Kconfig | 30 +++++++++++++++---------------
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/drivers/hid/Kconfig b/drivers/hid/Kconfig
index ffddcba..ce21ea1 100644
--- a/drivers/hid/Kconfig
+++ b/drivers/hid/Kconfig
@@ -63,7 +63,7 @@ menu "Special HID drivers"
config HID_A4TECH
tristate "A4 tech mice" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for A4 tech X5 and WOP-35 / Trust 450L mice.
@@ -84,7 +84,7 @@ config HID_ACRUX_FF
config HID_APPLE
tristate "Apple {i,Power,Mac}Books" if EXPERT
depends on (USB_HID || BT_HIDP)
- default !EXPERT
+ default y
---help---
Support for some Apple devices which less or more break
HID specification.
@@ -95,21 +95,21 @@ config HID_APPLE
config HID_BELKIN
tristate "Belkin Flip KVM and Wireless keyboard" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Belkin Flip KVM and Wireless keyboard.
config HID_CHERRY
tristate "Cherry Cymotion keyboard" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Cherry Cymotion keyboard.
config HID_CHICONY
tristate "Chicony Tactical pad" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Chicony Tactical pad.
@@ -131,7 +131,7 @@ config HID_PRODIKEYS
config HID_CYPRESS
tristate "Cypress mouse and barcode readers" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for cypress mouse and barcode readers.
@@ -173,7 +173,7 @@ config HID_ELECOM
config HID_EZKEY
tristate "Ezkey BTC 8193 keyboard" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Ezkey BTC 8193 keyboard.
@@ -236,7 +236,7 @@ config HID_TWINHAN
config HID_KENSINGTON
tristate "Kensington Slimblade Trackball" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Kensington Slimblade Trackball.
@@ -249,7 +249,7 @@ config HID_LCPOWER
config HID_LOGITECH
tristate "Logitech devices" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Logitech devices that are not fully compliant with HID standard.
@@ -325,14 +325,14 @@ config HID_MAGICMOUSE
config HID_MICROSOFT
tristate "Microsoft non-fully HID-compliant devices" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Microsoft devices that are not fully compliant with HID standard.
config HID_MONTEREY
tristate "Monterey Genius KB29E keyboard" if EXPERT
depends on USB_HID
- default !EXPERT
+ default y
---help---
Support for Monterey Genius KB29E.
@@ -438,7 +438,7 @@ config HID_PICOLCD
config HID_PICOLCD_FB
bool "Framebuffer support" if EXPERT
- default !EXPERT
+ default y
depends on HID_PICOLCD
depends on HID_PICOLCD=FB || FB=y
select FB_DEFERRED_IO
@@ -452,7 +452,7 @@ config HID_PICOLCD_FB
config HID_PICOLCD_BACKLIGHT
bool "Backlight control" if EXPERT
- default !EXPERT
+ default y
depends on HID_PICOLCD
depends on HID_PICOLCD=BACKLIGHT_CLASS_DEVICE || BACKLIGHT_CLASS_DEVICE=y
---help---
@@ -461,7 +461,7 @@ config HID_PICOLCD_BACKLIGHT
config HID_PICOLCD_LCD
bool "Contrast control" if EXPERT
- default !EXPERT
+ default y
depends on HID_PICOLCD
depends on HID_PICOLCD=LCD_CLASS_DEVICE || LCD_CLASS_DEVICE=y
---help---
@@ -469,7 +469,7 @@ config HID_PICOLCD_LCD
config HID_PICOLCD_LEDS
bool "GPO via leds class" if EXPERT
- default !EXPERT
+ default y
depends on HID_PICOLCD
depends on HID_PICOLCD=LEDS_CLASS || LEDS_CLASS=y
---help---
--
1.7.10.2
If no default value is specified, then 'n' is used so the default value
used here is not needed. Furthermore, we should never change default
values depending on EXPERT mode. EXPERT mode should only make options
visible, not change them.
Cc: Greg Kroah-Hartman <[email protected]>
Signed-off-by: David Herrmann <[email protected]>
---
drivers/usb/core/Kconfig | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/usb/core/Kconfig b/drivers/usb/core/Kconfig
index 18d02e3..04aec50 100644
--- a/drivers/usb/core/Kconfig
+++ b/drivers/usb/core/Kconfig
@@ -125,7 +125,6 @@ config USB_OTG_WHITELIST
bool "Rely on OTG Targeted Peripherals List"
depends on USB_OTG || EXPERT
default y if USB_OTG
- default n if EXPERT
help
If you say Y here, the "otg_whitelist.h" file will be used as a
product whitelist, so USB peripherals not listed there will be
--
1.7.10.2
EXPERT mode is designed to toggle visibility of config options. It should
not have any other effect on config values. Imagine if a user wants to
tweak a single value that is protected by EXPERT mode. If they enable
EXPERT mode to change that single option, they might get side effects from
other subsystems which they didn't expect. This makes EXPERT mode totally
useless. Hence, simply let the user always select the framebuffer-console
manually.
Furthermore, with kmscon or wayland, users might want to disable CONFIG_VT
which is currently automatically selected by DRM_KMS_HELPER through
FRAMEBUFFER_CONSOLE if !EXPERT.
Cc: David Airlie <[email protected]>
Signed-off-by: David Herrmann <[email protected]>
---
drivers/gpu/drm/Kconfig | 1 -
drivers/gpu/drm/nouveau/Kconfig | 1 -
2 files changed, 2 deletions(-)
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index e354bc0..ded991c 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -28,7 +28,6 @@ config DRM_KMS_HELPER
tristate
depends on DRM
select FB
- select FRAMEBUFFER_CONSOLE if !EXPERT
select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE
help
FB and CRTC helpers for KMS drivers.
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index 97a8126..d7e2536 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -8,7 +8,6 @@ config DRM_NOUVEAU
select FB_CFB_COPYAREA
select FB_CFB_IMAGEBLIT
select FB
- select FRAMEBUFFER_CONSOLE if !EXPERT
select FB_BACKLIGHT if DRM_NOUVEAU_BACKLIGHT
select ACPI_VIDEO if ACPI && X86 && BACKLIGHT_CLASS_DEVICE && VIDEO_OUTPUT_CONTROL && INPUT
select ACPI_WMI if ACPI
--
1.7.10.2
Expert mode was designed to toggle visibility of options and not to
change their behavior. If a user wants to tweak a single value and
enables EXPERT mode to do this, he will have several unexpected
side effects when other subsystems toggle their default values based
on EXPERT mode. This makes EXPERT pretty useless.
Cc: Mauro Carvalho Chehab <[email protected]>
Signed-off-by: David Herrmann <[email protected]>
---
drivers/media/common/tuners/Kconfig | 1 -
drivers/media/dvb/frontends/Kconfig | 1 -
drivers/media/video/Kconfig | 2 +-
3 files changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/media/common/tuners/Kconfig b/drivers/media/common/tuners/Kconfig
index 4a6d5ce..0bb2956 100644
--- a/drivers/media/common/tuners/Kconfig
+++ b/drivers/media/common/tuners/Kconfig
@@ -35,7 +35,6 @@ config MEDIA_TUNER
config MEDIA_TUNER_CUSTOMISE
bool "Customize analog and hybrid tuner modules to build"
depends on MEDIA_TUNER
- default y if EXPERT
help
This allows the user to deselect tuner drivers unnecessary
for their hardware from the build. Use this option with care
diff --git a/drivers/media/dvb/frontends/Kconfig b/drivers/media/dvb/frontends/Kconfig
index 2124670..1b0c854 100644
--- a/drivers/media/dvb/frontends/Kconfig
+++ b/drivers/media/dvb/frontends/Kconfig
@@ -1,7 +1,6 @@
config DVB_FE_CUSTOMISE
bool "Customise the frontend modules to build"
depends on DVB_CORE
- default y if EXPERT
help
This allows the user to select/deselect frontend drivers for their
hardware from the build.
diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig
index ce1e7ba..843d317 100644
--- a/drivers/media/video/Kconfig
+++ b/drivers/media/video/Kconfig
@@ -100,7 +100,7 @@ config VIDEO_FIXED_MINOR_RANGES
config VIDEO_HELPER_CHIPS_AUTO
bool "Autoselect pertinent encoders/decoders and other helper chips"
- default y if !EXPERT
+ default y
---help---
Most video cards may require additional modules to encode or
decode audio/video standards. This option will autoselect
--
1.7.10.2
EXPERT mode is supposed to toggle the visibility of config options, it
should never modify them, though. If users enable EXPERT mode to change a
single value, they might get unwanted side-effects with our current
behavior.
Signed-off-by: David Herrmann <[email protected]>
---
drivers/pcmcia/Kconfig | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/pcmcia/Kconfig b/drivers/pcmcia/Kconfig
index bba3ab2..ef0dbf9 100644
--- a/drivers/pcmcia/Kconfig
+++ b/drivers/pcmcia/Kconfig
@@ -69,7 +69,6 @@ comment "PC-card bridges"
config YENTA
tristate "CardBus yenta-compatible bridge support"
depends on PCI
- select CARDBUS if !EXPERT
select PCCARD_NONSTATIC if PCMCIA != n
---help---
This option enables support for CardBus host bridges. Virtually
--
1.7.10.2
We should avoid changing default values when EXPERT mode is enabled. Often
users enable EXPERT mode to change a single value. If other subsystems
change their defaults in EXPERT mode, the user might get a totally
different result as expected.
Signed-off-by: David Herrmann <[email protected]>
---
init/Kconfig | 2 +-
usr/Kconfig | 8 ++++----
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/init/Kconfig b/init/Kconfig
index 6cfd71d..f16400d 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -801,7 +801,7 @@ config CHECKPOINT_RESTORE
menuconfig NAMESPACES
bool "Namespaces support" if EXPERT
- default !EXPERT
+ default y
help
Provides the way to make tasks work with different objects using
the same id. For example same IPC id may refer to different objects
diff --git a/usr/Kconfig b/usr/Kconfig
index 65b845b..065a0e3 100644
--- a/usr/Kconfig
+++ b/usr/Kconfig
@@ -56,7 +56,7 @@ config RD_GZIP
config RD_BZIP2
bool "Support initial ramdisks compressed using bzip2" if EXPERT
- default !EXPERT
+ default y
depends on BLK_DEV_INITRD
select DECOMPRESS_BZIP2
help
@@ -65,7 +65,7 @@ config RD_BZIP2
config RD_LZMA
bool "Support initial ramdisks compressed using LZMA" if EXPERT
- default !EXPERT
+ default y
depends on BLK_DEV_INITRD
select DECOMPRESS_LZMA
help
@@ -74,7 +74,7 @@ config RD_LZMA
config RD_XZ
bool "Support initial ramdisks compressed using XZ" if EXPERT
- default !EXPERT
+ default y
depends on BLK_DEV_INITRD
select DECOMPRESS_XZ
help
@@ -83,7 +83,7 @@ config RD_XZ
config RD_LZO
bool "Support initial ramdisks compressed using LZO" if EXPERT
- default !EXPERT
+ default y
depends on BLK_DEV_INITRD
select DECOMPRESS_LZO
help
--
1.7.10.2
Expert mode was designed to toggle visibility of options and not to
change their behavior. If a user wants to tweak a single value and
enables EXPERT mode to do this, he will have several unexpected
side effects when other subsystems toggle their default values based
on EXPERT mode. This makes EXPERT pretty useless.
Cc: Johannes Berg <[email protected]>
Signed-off-by: David Herrmann <[email protected]>
---
net/rfkill/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/rfkill/Kconfig b/net/rfkill/Kconfig
index 78efe89..3d96f42 100644
--- a/net/rfkill/Kconfig
+++ b/net/rfkill/Kconfig
@@ -21,7 +21,7 @@ config RFKILL_INPUT
bool "RF switch input support" if EXPERT
depends on RFKILL
depends on INPUT = y || RFKILL = INPUT
- default y if !EXPERT
+ default y
config RFKILL_REGULATOR
tristate "Generic rfkill regulator driver"
--
1.7.10.2
> EXPERT mode is designed to toggle visibility of config options. It should
> not have any other effect on config values. Imagine if a user wants to
> tweak a single value that is protected by EXPERT mode. If they enable
> EXPERT mode to change that single option, they might get side effects from
> other subsystems which they didn't expect. This makes EXPERT mode totally
> useless. Hence, simply let the user always select the framebuffer-console
> manually.
> Furthermore, with kmscon or wayland, users might want to disable CONFIG_VT
> which is currently automatically selected by DRM_KMS_HELPER through
> FRAMEBUFFER_CONSOLE if !EXPERT.
NAK,
not going to happen, the whole point is that we get FRAMEBUFFER_CONSOLE
selected, the number of bugs filed where people build kernels with kms
and forget the console is why we did this, it would be stupid to undo it.
The !EXPERT thing was added just for the kmscon/wayland use case, where
some expert was configuring the kernel.
Dave.
On Sun, 13 May 2012, David Herrmann wrote:
> Expert mode was designed to toggle visibility of options and not to
> change their behavior. If a user wants to tweak a single value and
> enables EXPERT mode to do this, he will have several unexpected
> side effects when other subsystems toggle their default values based
> on EXPERT mode. This makes EXPERT pretty useless.
>
> Cc: Jiri Kosina <[email protected]>
> Signed-off-by: David Herrmann <[email protected]>
Please see
d04b431e3d769fbbf26c4f4072002375c8cc4ed9
b36299bcc0feae4c1bcff6a1561b8beb635e9c80
for some reasoning why this is done the way it's done.
Thanks,
--
Jiri Kosina
SUSE Labs