tree-wide: remove "select FW_LOADER" uses
Subsystems (29)
RFC 1/1 tree-wide: remove "select FW_LOADER" uses
There are some some recursive dependency issues one can run into when a kconfig option is selected by a driver, and then kconfig has to figure out whether or not some dependencies are valid for that selected option. For instance in terms of simplifying the logic issue a best as possible on my part and its implications:
If a driver selects FW_LOADER but also has a few other selects, kconfig will not allow another driver which also selects FW_LOADER to negate dependency on what that original driver's extra dependencies were.
This was recenlty discovered while working on extensions to the firmware_class module with some code being developed [0] to make firmware_class APIs extensible.
A possible fix to this is to change a "select FW_LOADER" to "depends FW_LOADER". If we're going to do that we would first need to evaluate if the recursive dependency issue is real or not, and if it is not we should do a tree wide change from all "select FW_LOADER" to "depend FW_LOADER".
Upon review with Paul, the recursive dependency issue is actually real mostly because of the complexity of the amount of symbols we handle and how we ended up deciding in logic for "select" and "depends on" to work [1].
Paul did some git archaeological analysis of previous trends on gramatically similar reported recursive issues and found that the generic solution to such problems were to convert "select FOO" to "depends on FOO". This was based on a study of 20 commits with similar reported recursive issue reported since v2.6.36 [2].
Paul also found commit d9b19199e489 ("always enable FW_LOADER unless EMBEDDED=y") which explains that we can then get rid of all "select FW_LOADER" [...] and as per Paul it seems the plan to get rid of those never materialized [2].
Paul also insists that *there is no reason* to think that current logic of the kconfig tools, as it is currently implemented, is flawed [2].
The above history is just *one* set of reasons to remove all kernel "select FW_LOADER" entries. Another simple reason is that since FW_LOADER can only be disabled by an EXPERT and it defaults to y its now a superfluous select. We will ensure things build for you if using EXPERT but if you have selected EXPERT you should know what run time requirements you need.
This goes build tested with an allyesconfig and allmodconfig configuration and then disabling FW_LOADER on each build target. This was tested on linux-next tag next-20150518.
Should this require rebasing you can regenerate the target patches with:
for i in $(git grep "select FW_LOADER$" | \
awk -F":" '{print $1}' | sort | uniq); do \
perl -i -ne 'print if !/select FW_LOADER$/' $i; done[0] https://lkml.org/lkml/2015/5/7/814
[1] https://lkml.org/lkml/2015/5/8/748
[2] https://lkml.org/lkml/2015/5/18/835
---
I realize tree-wide changes are not encourated like this, but I am doing this as an RFC first to save a few electrons to spare sending out 21 separate e-mails and separate reviews and discussions on one topic alone, should there be any. If this seems reasonable please send me your respective Acked-by's, and I can split up into 21 patches by subsystem:
network
media
sound
scsi
drm
inputdell_rbu
crypto
bluetooth
leds
atm
staging
usb
tty
uwb
misc
pcmcia
remoteproc
arm mach-ixp4xx
cris
x86 microcode
If you wish me to just merge a bunch of these together let me know as well.
arch/arm/mach-ixp4xx/Kconfig | 1 - arch/cris/arch-v32/drivers/Kconfig | 1 - arch/x86/Kconfig | 3 --- drivers/atm/Kconfig | 2 -- drivers/bluetooth/Kconfig | 7 ------- drivers/crypto/qat/Kconfig | 1 - drivers/firmware/Kconfig | 1 - drivers/gpu/drm/Kconfig | 3 --- drivers/gpu/drm/nouveau/Kconfig | 1 - drivers/input/touchscreen/Kconfig | 1 - drivers/leds/Kconfig | 1 - drivers/media/dvb-frontends/Kconfig | 1 - drivers/media/pci/saa7164/Kconfig | 1 - drivers/media/platform/exynos4-is/Kconfig | 1 - drivers/media/radio/Kconfig | 1 - drivers/media/usb/as102/Kconfig | 1 - drivers/misc/Kconfig | 1 - drivers/misc/ti-st/Kconfig | 1 - drivers/net/ethernet/broadcom/Kconfig | 2 -- drivers/net/ethernet/chelsio/Kconfig | 2 -- drivers/net/ethernet/myricom/Kconfig | 1 - drivers/net/ethernet/qlogic/Kconfig | 2 -- drivers/net/ethernet/realtek/Kconfig | 1 - drivers/net/ethernet/toshiba/Kconfig | 1 - drivers/net/irda/Kconfig | 1 - drivers/net/wimax/i2400m/Kconfig | 1 - drivers/net/wireless/Kconfig | 6 ------ drivers/net/wireless/ath/ar5523/Kconfig | 1 - drivers/net/wireless/ath/carl9170/Kconfig | 1 - drivers/net/wireless/b43/Kconfig | 1 - drivers/net/wireless/b43legacy/Kconfig | 1 - drivers/net/wireless/brcm80211/Kconfig | 4 ---- drivers/net/wireless/ipw2x00/Kconfig | 2 -- drivers/net/wireless/iwlegacy/Kconfig | 1 - drivers/net/wireless/iwlwifi/Kconfig | 1 - drivers/net/wireless/libertas/Kconfig | 1 - drivers/net/wireless/mwifiex/Kconfig | 3 --- drivers/net/wireless/orinoco/Kconfig | 2 -- drivers/net/wireless/p54/Kconfig | 1 - drivers/net/wireless/rt2x00/Kconfig | 1 - drivers/net/wireless/rtlwifi/Kconfig | 1 - drivers/net/wireless/ti/wl1251/Kconfig | 1 - drivers/net/wireless/ti/wlcore/Kconfig | 1 - drivers/net/wireless/zd1211rw/Kconfig | 1 - drivers/pcmcia/Kconfig | 1 - drivers/remoteproc/Kconfig | 1 - drivers/scsi/Kconfig | 1 - drivers/scsi/aic94xx/Kconfig | 1 - drivers/scsi/csiostor/Kconfig | 1 - drivers/scsi/mvsas/Kconfig | 1 - drivers/scsi/qla2xxx/Kconfig | 1 - drivers/staging/rtl8192e/rtl8192e/Kconfig | 1 - drivers/staging/rtl8712/Kconfig | 1 - drivers/staging/vt6656/Kconfig | 1 - drivers/tty/Kconfig | 3 --- drivers/tty/serial/Kconfig | 2 -- drivers/usb/atm/Kconfig | 3 --- drivers/usb/misc/Kconfig | 1 - drivers/uwb/Kconfig | 1 - sound/drivers/Kconfig | 1 - sound/isa/Kconfig | 5 ----- sound/pci/Kconfig | 21 --------------------- sound/pci/hda/Kconfig | 2 -- sound/soc/sh/Kconfig | 1 - sound/usb/Kconfig | 1 - 65 files changed, 121 deletions(-)
arch/arm/mach-ixp4xx/Kconfig +0 -1
@@ -230,7 +230,6 @@ config IXP4XX_QMGR config IXP4XX_NPE tristate "IXP4xx Network Processor Engine support" - select FW_LOADER help This driver supports IXP4xx built-in network coprocessors and is automatically selected by Ethernet and HSS drivers.
arch/cris/arch-v32/drivers/Kconfig +0 -1
@@ -331,7 +331,6 @@ config PCI config ETRAX_IOP_FW_LOAD tristate "IO-processor hotplug firmware loading support" depends on ETRAX_ARCH_V32 - select FW_LOADER help Enables IO-processor hotplug firmware loading support.
arch/x86/Kconfig +0 -3
@@ -1099,7 +1099,6 @@ config X86_REBOOTFIXUPS config MICROCODE tristate "CPU microcode loading support" depends on CPU_SUP_AMD || CPU_SUP_INTEL - select FW_LOADER ---help--- If you say Y here, you will be able to update the microcode on
@@ -1119,7 +1118,6 @@ config MICROCODE_INTEL bool "Intel microcode loading support" depends on MICROCODE default MICROCODE - select FW_LOADER ---help--- This options enables microcode patch loading support for Intel processors.
@@ -1131,7 +1129,6 @@ config MICROCODE_INTEL config MICROCODE_AMD bool "AMD microcode loading support" depends on MICROCODE - select FW_LOADER ---help--- If you select this option, microcode patch loading support for AMD processors will be enabled.
drivers/atm/Kconfig +0 -2
@@ -328,7 +328,6 @@ config ATM_IA_DEBUG config ATM_FORE200E tristate "FORE Systems 200E-series" depends on (PCI || SBUS) - select FW_LOADER ---help--- This is a driver for the FORE Systems 200E-series ATM adapter cards. It simultaneously supports PCA-200E and SBA-200E models
@@ -394,7 +393,6 @@ config ATM_HE_USE_SUNI config ATM_SOLOS tristate "Solos ADSL2+ PCI Multiport card driver" depends on PCI - select FW_LOADER help Support for the Solos multiport ADSL2+ card.
drivers/bluetooth/Kconfig +0 -7
@@ -7,11 +7,9 @@ config BT_INTEL config BT_BCM tristate - select FW_LOADER config BT_RTL tristate - select FW_LOADER config BT_HCIBTUSB tristate "HCI USB driver"
@@ -153,7 +151,6 @@ config BT_HCIUART_BCM config BT_HCIBCM203X tristate "HCI BCM203x USB driver" depends on USB - select FW_LOADER help Bluetooth HCI BCM203x USB driver. This driver provides the firmware loading mechanism for the Broadcom
@@ -176,7 +173,6 @@ config BT_HCIBPA10X config BT_HCIBFUSB tristate "HCI BlueFRITZ! USB driver" depends on USB - select FW_LOADER help Bluetooth HCI BlueFRITZ! USB driver. This driver provides support for Bluetooth USB devices with AVM
@@ -202,7 +198,6 @@ config BT_HCIDTL1 config BT_HCIBT3C tristate "HCI BT3C (PC Card) driver" depends on PCMCIA - select FW_LOADER help Bluetooth HCI BT3C (PC Card) driver. This driver provides support for Bluetooth PCMCIA devices with
@@ -265,7 +260,6 @@ config BT_MRVL config BT_MRVL_SDIO tristate "Marvell BT-over-SDIO driver" depends on BT_MRVL && MMC - select FW_LOADER select WANT_DEV_COREDUMP help The driver for Marvell Bluetooth chipsets with SDIO interface.
@@ -280,7 +274,6 @@ config BT_MRVL_SDIO config BT_ATH3K tristate "Atheros firmware download driver" depends on BT_HCIBTUSB - select FW_LOADER help Bluetooth firmware download driver. This driver loads the firmware into the Atheros Bluetooth
drivers/crypto/qat/Kconfig +0 -1
@@ -8,7 +8,6 @@ config CRYPTO_DEV_QAT select CRYPTO_SHA1 select CRYPTO_SHA256 select CRYPTO_SHA512 - select FW_LOADER config CRYPTO_DEV_QAT_DH895xCC tristate "Support for Intel(R) DH895xCC"
drivers/firmware/Kconfig +0 -1
@@ -63,7 +63,6 @@ config EFI_PCDP config DELL_RBU tristate "BIOS update support for DELL systems via sysfs" depends on X86 - select FW_LOADER select FW_LOADER_USER_HELPER help Say m if you want to have the option of updating the BIOS for your
drivers/gpu/drm/Kconfig +0 -3
@@ -91,7 +91,6 @@ config DRM_TDFX config DRM_R128 tristate "ATI Rage 128" depends on DRM && PCI - select FW_LOADER help Choose this option if you have an ATI Rage 128 graphics card. If M is selected, the module will be called r128. AGP support for
@@ -103,7 +102,6 @@ config DRM_RADEON select FB_CFB_FILLRECT select FB_CFB_COPYAREA select FB_CFB_IMAGEBLIT - select FW_LOADER select DRM_KMS_HELPER select DRM_KMS_FB_HELPER select DRM_TTM
@@ -136,7 +134,6 @@ source "drivers/gpu/drm/i915/Kconfig" config DRM_MGA tristate "Matrox g200/g400" depends on DRM && PCI - select FW_LOADER help Choose this option if you have a Matrox G200, G400 or G450 graphics card. If M is selected, the module will be called mga. AGP
drivers/gpu/drm/nouveau/Kconfig +0 -1
@@ -1,7 +1,6 @@ config DRM_NOUVEAU tristate "Nouveau (NVIDIA) cards" depends on DRM && PCI - select FW_LOADER select DRM_KMS_HELPER select DRM_KMS_FB_HELPER select DRM_TTM
drivers/input/touchscreen/Kconfig +0 -1
@@ -105,7 +105,6 @@ config TOUCHSCREEN_AR1021_I2C config TOUCHSCREEN_ATMEL_MXT tristate "Atmel mXT I2C Touchscreen" depends on I2C - select FW_LOADER help Say Y here if you have Atmel mXT series I2C touchscreen, such as AT42QT602240/ATMXT224, connected to your system.
drivers/leds/Kconfig +0 -1
@@ -221,7 +221,6 @@ config LEDS_LP3944 config LEDS_LP55XX_COMMON tristate "Common Driver for TI/National LP5521/5523/55231/5562/8501" depends on LEDS_LP5521 || LEDS_LP5523 || LEDS_LP5562 || LEDS_LP8501 - select FW_LOADER help This option supports common operations for LP5521/5523/55231/5562/8501 devices.
drivers/media/dvb-frontends/Kconfig +0 -1
@@ -763,7 +763,6 @@ config DVB_LGS8GL5 config DVB_LGS8GXX tristate "Legend Silicon LGS8913/LGS8GL5/LGS8GXX DMB-TH demodulator" depends on DVB_CORE && I2C - select FW_LOADER default m if !MEDIA_SUBDRV_AUTOSELECT help A DMB-TH tuner module. Say Y when you want to support this frontend.
drivers/media/pci/saa7164/Kconfig +0 -1
@@ -2,7 +2,6 @@ config VIDEO_SAA7164 tristate "NXP SAA7164 support" depends on DVB_CORE && VIDEO_DEV && PCI && I2C select I2C_ALGOBIT - select FW_LOADER select VIDEO_TUNER select VIDEO_TVEEPROM select VIDEOBUF_DVB
drivers/media/platform/exynos4-is/Kconfig +0 -1
@@ -60,7 +60,6 @@ config VIDEO_EXYNOS4_FIMC_IS depends on HAS_DMA select VIDEOBUF2_DMA_CONTIG depends on OF - select FW_LOADER help This is a V4L2 driver for Samsung EXYNOS4x12 SoC series FIMC-IS (Imaging Subsystem).
drivers/media/radio/Kconfig +0 -1
@@ -214,7 +214,6 @@ config RADIO_WL1273 depends on I2C && VIDEO_V4L2 select MFD_CORE select MFD_WL1273_CORE - select FW_LOADER ---help--- Choose Y here if you have this FM radio chip.
drivers/media/usb/as102/Kconfig +0 -1
@@ -1,7 +1,6 @@ config DVB_AS102 tristate "Abilis AS102 DVB receiver" depends on DVB_CORE && USB && I2C && INPUT - select FW_LOADER help Choose Y or M here if you have a device containing an AS102
drivers/misc/Kconfig +0 -1
@@ -489,7 +489,6 @@ config USB_SWITCH_FSA9480 config LATTICE_ECP3_CONFIG tristate "Lattice ECP3 FPGA bitstream configuration via SPI" depends on SPI && SYSFS - select FW_LOADER default n help This option enables support for bitstream configuration (programming
drivers/misc/ti-st/Kconfig +0 -1
@@ -6,7 +6,6 @@ menu "Texas Instruments shared transport line discipline" config TI_ST tristate "Shared transport core driver" depends on NET && GPIOLIB && TTY - select FW_LOADER help This enables the shared transport core driver for TI BT / FM and GPS combo chips. This enables protocol drivers
drivers/net/ethernet/broadcom/Kconfig +0 -2
@@ -74,7 +74,6 @@ config BNX2 tristate "QLogic bnx2 support" depends on PCI select CRC32 - select FW_LOADER ---help--- This driver supports QLogic bnx2 gigabit Ethernet cards.
@@ -122,7 +121,6 @@ config BNX2X tristate "Broadcom NetXtremeII 10Gb support" depends on PCI select PTP_1588_CLOCK - select FW_LOADER select ZLIB_INFLATE select LIBCRC32C select MDIO
drivers/net/ethernet/chelsio/Kconfig +0 -2
@@ -49,7 +49,6 @@ config CHELSIO_T1_1G config CHELSIO_T3 tristate "Chelsio Communications T3 10Gb Ethernet support" depends on PCI && INET - select FW_LOADER select MDIO ---help--- This driver supports Chelsio T3-based gigabit and 10Gb Ethernet
@@ -69,7 +68,6 @@ config CHELSIO_T3 config CHELSIO_T4 tristate "Chelsio Communications T4/T5 Ethernet support" depends on PCI && (IPV6 || IPV6=n) - select FW_LOADER select MDIO ---help--- This driver supports Chelsio T4 and T5 based gigabit, 10Gb Ethernet
drivers/net/ethernet/myricom/Kconfig +0 -1
@@ -21,7 +21,6 @@ if NET_VENDOR_MYRI config MYRI10GE tristate "Myricom Myri-10G Ethernet support" depends on PCI && INET - select FW_LOADER select CRC32 ---help--- This driver supports Myricom Myri-10G Dual Protocol interface in
drivers/net/ethernet/qlogic/Kconfig +0 -2
@@ -30,7 +30,6 @@ config QLA3XXX config QLCNIC tristate "QLOGIC QLCNIC 1/10Gb Converged Ethernet NIC Support" depends on PCI - select FW_LOADER ---help--- This driver supports QLogic QLE8240 and QLE8242 Converged Ethernet devices.
@@ -89,7 +88,6 @@ config QLGE config NETXEN_NIC tristate "NetXen Multi port (1/10) Gigabit Ethernet NIC" depends on PCI - select FW_LOADER ---help--- This enables the support for NetXen's Gigabit Ethernet card.
drivers/net/ethernet/realtek/Kconfig +0 -1
@@ -103,7 +103,6 @@ config 8139_OLD_RX_RESET config R8169 tristate "Realtek 8169 gigabit ethernet support" depends on PCI - select FW_LOADER select CRC32 select MII ---help---
drivers/net/ethernet/toshiba/Kconfig +0 -1
@@ -43,7 +43,6 @@ config GELIC_WIRELESS config SPIDER_NET tristate "Spider Gigabit Ethernet driver" depends on PCI && PPC_IBM_CELL_BLADE - select FW_LOADER select SUNGEM_PHY ---help--- This driver supports the Gigabit Ethernet chips present on the
drivers/net/irda/Kconfig +0 -1
@@ -243,7 +243,6 @@ comment "FIR device drivers" config USB_IRDA tristate "IrDA USB dongles" depends on IRDA && USB - select FW_LOADER ---help--- Say Y here if you want to build support for the USB IrDA FIR Dongle device driver. To compile it as a module, choose M here: the module
drivers/net/wimax/i2400m/Kconfig +0 -1
@@ -2,7 +2,6 @@ config WIMAX_I2400M tristate depends on WIMAX - select FW_LOADER comment "Enable USB support to see WiMAX USB drivers" depends on USB = n
drivers/net/wireless/Kconfig +0 -6
@@ -35,7 +35,6 @@ config PCMCIA_RAYCS config LIBERTAS_THINFIRM tristate "Marvell 8xxx Libertas WLAN driver support with thin firmware" depends on MAC80211 - select FW_LOADER ---help--- A library for Marvell Libertas 8xxx devices using thinfirm.
@@ -76,7 +75,6 @@ config ATMEL depends on CFG80211 && (PCI || PCMCIA) select WIRELESS_EXT select WEXT_PRIV - select FW_LOADER select CRC32 ---help--- A driver 802.11b wireless cards based on the Atmel fast-vnet
@@ -100,7 +98,6 @@ config PCMCIA_ATMEL tristate "Atmel at76c502/at76c504 PCMCIA cards" depends on ATMEL && PCMCIA select WIRELESS_EXT - select FW_LOADER select CRC32 ---help--- Enable support for PCMCIA cards containing the
@@ -109,7 +106,6 @@ config PCMCIA_ATMEL config AT76C50X_USB tristate "Atmel at76c503/at76c505/at76c505a USB cards" depends on MAC80211 && USB - select FW_LOADER ---help--- Enable support for USB Wireless devices using Atmel at76c503, at76c505 or at76c505a chips.
@@ -152,7 +148,6 @@ config PRISM54 select WIRELESS_EXT select WEXT_SPY select WEXT_PRIV - select FW_LOADER ---help--- This enables support for FullMAC PCI/Cardbus prism54 devices. This driver is now deprecated in favor for the SoftMAC driver, p54pci.
@@ -171,7 +166,6 @@ config USB_ZD1201 depends on CFG80211 && USB select WIRELESS_EXT select WEXT_PRIV - select FW_LOADER ---help--- Say Y if you want to use wireless LAN adapters based on the ZyDAS ZD1201 chip.
drivers/net/wireless/ath/ar5523/Kconfig +0 -1
@@ -2,7 +2,6 @@ config AR5523 tristate "Atheros AR5523 wireless driver support" depends on MAC80211 && USB select ATH_COMMON - select FW_LOADER ---help--- This module add support for AR5523 based USB dongles such as D-Link DWL-G132, Netgear WPN111 and many more.
drivers/net/wireless/ath/carl9170/Kconfig +0 -1
@@ -2,7 +2,6 @@ config CARL9170 tristate "Linux Community AR9170 802.11n USB support" depends on USB && MAC80211 select ATH_COMMON - select FW_LOADER select CRC32 help This is another driver for the Atheros "otus" 802.11n USB devices.
drivers/net/wireless/b43/Kconfig +0 -1
@@ -3,7 +3,6 @@ config B43 depends on (BCMA_POSSIBLE || SSB_POSSIBLE) && MAC80211 && HAS_DMA select BCMA if B43_BCMA select SSB if B43_SSB - select FW_LOADER ---help--- b43 is a driver for the Broadcom 43xx series wireless devices.
drivers/net/wireless/b43legacy/Kconfig +0 -1
@@ -2,7 +2,6 @@ config B43LEGACY tristate "Broadcom 43xx-legacy wireless support (mac80211 stack)" depends on SSB_POSSIBLE && MAC80211 && HAS_DMA select SSB - select FW_LOADER ---help--- b43legacy is a driver for 802.11b devices from Broadcom (BCM4301 and BCM4303) and early model 802.11g chips (BCM4306 Ver. 2) used in the
drivers/net/wireless/brcm80211/Kconfig +0 -4
@@ -9,7 +9,6 @@ config BRCMSMAC select NEW_LEDS if BCMA_DRIVER_GPIO select LEDS_CLASS if BCMA_DRIVER_GPIO select BRCMUTIL - select FW_LOADER select CORDIC ---help--- This module adds support for PCIe wireless adapters based on Broadcom
@@ -38,7 +37,6 @@ config BRCMFMAC_SDIO depends on (MMC = y || MMC = BRCMFMAC) depends on BRCMFMAC select BRCMFMAC_PROTO_BCDC - select FW_LOADER default y ---help--- This option enables the SDIO bus interface support for Broadcom
@@ -50,7 +48,6 @@ config BRCMFMAC_USB depends on (USB = y || USB = BRCMFMAC) depends on BRCMFMAC select BRCMFMAC_PROTO_BCDC - select FW_LOADER ---help--- This option enables the USB bus interface support for Broadcom IEEE802.11n embedded FullMAC WLAN driver. Say Y if you want to
@@ -62,7 +59,6 @@ config BRCMFMAC_PCIE depends on PCI depends on HAS_DMA select BRCMFMAC_PROTO_MSGBUF - select FW_LOADER ---help--- This option enables the PCIE bus interface support for Broadcom IEEE802.11ac embedded FullMAC WLAN driver. Say Y if you want to
drivers/net/wireless/ipw2x00/Kconfig +0 -2
@@ -8,7 +8,6 @@ config IPW2100 select WIRELESS_EXT select WEXT_SPY select WEXT_PRIV - select FW_LOADER select LIB80211 select LIBIPW ---help---
@@ -70,7 +69,6 @@ config IPW2200 select WIRELESS_EXT select WEXT_SPY select WEXT_PRIV - select FW_LOADER select LIB80211 select LIBIPW ---help---
drivers/net/wireless/iwlegacy/Kconfig +0 -1
@@ -1,6 +1,5 @@ config IWLEGACY tristate - select FW_LOADER select NEW_LEDS select LEDS_CLASS select LEDS_TRIGGERS
drivers/net/wireless/iwlwifi/Kconfig +0 -1
@@ -1,7 +1,6 @@ config IWLWIFI tristate "Intel Wireless WiFi Next Gen AGN - Wireless-N/Advanced-N/Ultimate-N (iwlwifi) " depends on PCI && MAC80211 && HAS_IOMEM - select FW_LOADER ---help--- Select to build the driver supporting the:
drivers/net/wireless/libertas/Kconfig +0 -1
@@ -4,7 +4,6 @@ config LIBERTAS select WIRELESS_EXT select WEXT_SPY select LIB80211 - select FW_LOADER ---help--- A library for Marvell Libertas 8xxx devices.
drivers/net/wireless/mwifiex/Kconfig +0 -3
@@ -11,7 +11,6 @@ config MWIFIEX config MWIFIEX_SDIO tristate "Marvell WiFi-Ex Driver for SD8786/SD8787/SD8797/SD8887/SD8897" depends on MWIFIEX && MMC - select FW_LOADER ---help--- This adds support for wireless adapters based on Marvell 8786/8787/8797/8887/8897 chipsets with SDIO interface.
@@ -22,7 +21,6 @@ config MWIFIEX_SDIO config MWIFIEX_PCIE tristate "Marvell WiFi-Ex Driver for PCIE 8766/8897" depends on MWIFIEX && PCI - select FW_LOADER ---help--- This adds support for wireless adapters based on Marvell 8766/8897 chipsets with PCIe interface.
@@ -33,7 +31,6 @@ config MWIFIEX_PCIE config MWIFIEX_USB tristate "Marvell WiFi-Ex Driver for USB8766/8797/8897" depends on MWIFIEX && USB - select FW_LOADER ---help--- This adds support for wireless adapters based on Marvell 8797/8897 chipset with USB interface.
drivers/net/wireless/orinoco/Kconfig +0 -2
@@ -6,7 +6,6 @@ config HERMES select WIRELESS_EXT select WEXT_SPY select WEXT_PRIV - select FW_LOADER select CRYPTO select CRYPTO_MICHAEL_MIC ---help---
@@ -137,6 +136,5 @@ config PCMCIA_SPECTRUM config ORINOCO_USB tristate "Agere Orinoco USB support" depends on USB && HERMES - select FW_LOADER ---help--- This driver is for USB versions of the Agere Orinoco card.
drivers/net/wireless/p54/Kconfig +0 -1
@@ -1,7 +1,6 @@ config P54_COMMON tristate "Softmac Prism54 support" depends on MAC80211 - select FW_LOADER select CRC_CCITT ---help--- This is common code for isl38xx/stlc45xx based modules.
drivers/net/wireless/rt2x00/Kconfig +0 -1
@@ -244,7 +244,6 @@ config RT2X00_LIB config RT2X00_LIB_FIRMWARE bool - select FW_LOADER config RT2X00_LIB_CRYPTO bool
drivers/net/wireless/rtlwifi/Kconfig +0 -1
@@ -119,7 +119,6 @@ config RTL8192CU config RTLWIFI tristate - select FW_LOADER config RTLWIFI_PCI tristate
drivers/net/wireless/ti/wl1251/Kconfig +0 -1
@@ -1,7 +1,6 @@ menuconfig WL1251 tristate "TI wl1251 driver support" depends on MAC80211 - select FW_LOADER select CRC7 ---help--- This will enable TI wl1251 driver support. The drivers make
drivers/net/wireless/ti/wlcore/Kconfig +0 -1
@@ -1,7 +1,6 @@ config WLCORE tristate "TI wlcore support" depends on WL_TI && MAC80211 - select FW_LOADER ---help--- This module contains the main code for TI WLAN chips. It abstracts hardware-specific differences among different chipset families.
drivers/net/wireless/zd1211rw/Kconfig +0 -1
@@ -1,7 +1,6 @@ config ZD1211RW tristate "ZyDAS ZD1211/ZD1211B USB-wireless support" depends on USB && MAC80211 - select FW_LOADER ---help--- This is a driver for the ZyDAS ZD1211/ZD1211B wireless chip, present in many USB-wireless adapters.
drivers/pcmcia/Kconfig +0 -1
@@ -37,7 +37,6 @@ config PCMCIA config PCMCIA_LOAD_CIS bool "Load CIS updates from userspace" depends on PCMCIA - select FW_LOADER default y help Some PCMCIA cards require an updated Card Information Structure (CIS)
drivers/remoteproc/Kconfig +0 -1
@@ -5,7 +5,6 @@ config REMOTEPROC tristate depends on HAS_DMA select CRC32 - select FW_LOADER select VIRTIO select VIRTUALIZATION
drivers/scsi/Kconfig +0 -1
@@ -1091,7 +1091,6 @@ config SCSI_SYM53C8XX_MMIO config SCSI_IPR tristate "IBM Power Linux RAID adapter support" depends on PCI && SCSI && ATA - select FW_LOADER ---help--- This driver supports the IBM Power Linux family RAID adapters. This includes IBM pSeries 5712, 5703, 5709, and 570A, as well
drivers/scsi/aic94xx/Kconfig +0 -1
@@ -28,7 +28,6 @@ config SCSI_AIC94XX tristate "Adaptec AIC94xx SAS/SATA support" depends on PCI select SCSI_SAS_LIBSAS - select FW_LOADER help This driver supports Adaptec's SAS/SATA 3Gb/s 64 bit PCI-X AIC94xx chip based host adapters.
drivers/scsi/csiostor/Kconfig +0 -1
@@ -2,7 +2,6 @@ config SCSI_CHELSIO_FCOE tristate "Chelsio Communications FCoE support" depends on PCI && SCSI depends on SCSI_FC_ATTRS - select FW_LOADER help This driver supports FCoE Offload functionality over Chelsio T4-based 10Gb Converged Network Adapters.
drivers/scsi/mvsas/Kconfig +0 -1
@@ -29,7 +29,6 @@ config SCSI_MVSAS tristate "Marvell 88SE64XX/88SE94XX SAS/SATA support" depends on PCI select SCSI_SAS_LIBSAS - select FW_LOADER help This driver supports Marvell's SAS/SATA 3Gb/s PCI-E 88SE64XX and 6Gb/s PCI-E 88SE94XX chip based host adapters.
drivers/scsi/qla2xxx/Kconfig +0 -1
@@ -2,7 +2,6 @@ config SCSI_QLA_FC tristate "QLogic QLA2XXX Fibre Channel Support" depends on PCI && SCSI depends on SCSI_FC_ATTRS - select FW_LOADER ---help--- This qla2xxx driver supports all QLogic Fibre Channel PCI and PCIe host adapters.
drivers/staging/rtl8192e/rtl8192e/Kconfig +0 -1
@@ -5,5 +5,4 @@ config RTL8192E select WIRELESS_EXT select WEXT_PRIV select CRYPTO - select FW_LOADER ---help---
drivers/staging/rtl8712/Kconfig +0 -1
@@ -3,7 +3,6 @@ config R8712U depends on WLAN && USB select WIRELESS_EXT select WEXT_PRIV - select FW_LOADER ---help--- This option adds the Realtek RTL8712 USB device such as the D-Link DWA-130. If built as a module, it will be called r8712u.
drivers/staging/vt6656/Kconfig +0 -1
@@ -1,7 +1,6 @@ config VT6656 tristate "VIA Technologies VT6656 support" depends on MAC80211 && USB && WLAN && m - select FW_LOADER ---help--- This is a vendor-written driver for VIA VT6656.
drivers/tty/Kconfig +0 -3
@@ -210,7 +210,6 @@ config ROCKETPORT config CYCLADES tristate "Cyclades async mux support" depends on SERIAL_NONSTANDARD && (PCI || ISA) - select FW_LOADER ---help--- This driver supports Cyclades Z and Y multiserial boards. You would need something like this to connect more than two modems to
@@ -239,7 +238,6 @@ config CYZ_INTR config MOXA_INTELLIO tristate "Moxa Intellio support" depends on SERIAL_NONSTANDARD && (ISA || EISA || PCI) - select FW_LOADER help Say Y here if you have a Moxa Intellio multiport serial card.
@@ -307,7 +305,6 @@ config NOZOMI config ISI tristate "Multi-Tech multiport card support" depends on SERIAL_NONSTANDARD && PCI - select FW_LOADER help This is a driver for the Multi-Tech cards which provide several serial ports. The driver is experimental and can currently only be
drivers/tty/serial/Kconfig +0 -2
@@ -931,7 +931,6 @@ config SERIAL_ICOM tristate "IBM Multiport Serial Adapter" depends on PCI && PPC_PSERIES select SERIAL_CORE - select FW_LOADER help This driver is for a family of multiport serial adapters including 2 port RVX, 2 port internal modem, 4 port internal
@@ -1158,7 +1157,6 @@ config SERIAL_QE tristate "Freescale QUICC Engine serial port support" depends on QUICC_ENGINE select SERIAL_CORE - select FW_LOADER default n help This driver supports the QE serial ports on Freescale embedded
drivers/usb/atm/Kconfig +0 -3
@@ -19,7 +19,6 @@ if USB_ATM config USB_SPEEDTOUCH tristate "Speedtouch USB support" - select FW_LOADER help Say Y here if you have an SpeedTouch USB or SpeedTouch 330 modem. In order to use your modem you will need to install the
@@ -31,7 +30,6 @@ config USB_SPEEDTOUCH config USB_CXACRU tristate "Conexant AccessRunner USB support" - select FW_LOADER help Say Y here if you have an ADSL USB modem based on the Conexant AccessRunner chipset. In order to use your modem you will need to
@@ -43,7 +41,6 @@ config USB_CXACRU config USB_UEAGLEATM tristate "ADI 930 and eagle USB DSL modem" - select FW_LOADER help Say Y here if you have an ADSL USB modem based on the ADI 930 or eagle chipset. In order to use your modem you will need to
drivers/usb/misc/Kconfig +0 -1
@@ -214,7 +214,6 @@ config USB_EHSET_TEST_FIXTURE config USB_ISIGHTFW tristate "iSight firmware loading support" - select FW_LOADER help This driver loads firmware for USB Apple iSight cameras, allowing them to be driven by the USB video class driver available at
drivers/uwb/Kconfig +0 -1
@@ -58,7 +58,6 @@ config UWB_WHCI config UWB_I1480U tristate "Support for Intel Wireless UWB Link 1480 HWA" depends on UWB_HWA - select FW_LOADER help This driver enables support for the i1480 when connected via USB. It consists of a firmware uploader that will enable it
sound/drivers/Kconfig +0 -1
@@ -14,7 +14,6 @@ config SND_OPL4_LIB config SND_VX_LIB tristate - select FW_LOADER select SND_HWDEP select SND_PCM
sound/isa/Kconfig +0 -5
@@ -383,7 +383,6 @@ config SND_SBAWE config SND_SB16_CSP bool "Sound Blaster 16/AWE CSP support" depends on (SND_SB16 || SND_SBAWE) && (BROKEN || !PPC) - select FW_LOADER help Say Y here to include support for the CSP core. This special coprocessor can do variable tasks like various compression and
@@ -393,7 +392,6 @@ config SND_SSCAPE tristate "Ensoniq SoundScape driver" select SND_MPU401_UART select SND_WSS_LIB - select FW_LOADER help Say Y here to include support for Ensoniq SoundScape and Ensoniq OEM soundcards.
@@ -412,7 +410,6 @@ config SND_SSCAPE config SND_WAVEFRONT tristate "Turtle Beach Maui,Tropez,Tropez+ (Wavefront)" - select FW_LOADER select SND_OPL3_LIB select SND_MPU401_UART select SND_WSS_LIB
@@ -426,7 +423,6 @@ config SND_WAVEFRONT config SND_MSND_PINNACLE tristate "Turtle Beach MultiSound Pinnacle/Fiji driver" depends on X86 - select FW_LOADER select SND_MPU401_UART select SND_PCM help
@@ -439,7 +435,6 @@ config SND_MSND_PINNACLE config SND_MSND_CLASSIC tristate "Support for Turtle Beach MultiSound Classic, Tahiti, Monterey" depends on X86 - select FW_LOADER select SND_MPU401_UART select SND_PCM help
sound/pci/Kconfig +0 -21
@@ -63,7 +63,6 @@ config SND_ALI5451 config SND_ASIHPI tristate "AudioScience ASIxxxx" depends on X86 - select FW_LOADER select SND_PCM select SND_HWDEP help
@@ -257,7 +256,6 @@ config SND_CS46XX tristate "Cirrus Logic (Sound Fusion) CS4280/CS461x/CS462x/CS463x" select SND_RAWMIDI select SND_AC97_CODEC - select FW_LOADER help Say Y here to include support for Cirrus Logic CS4610/CS4612/ CS4614/CS4615/CS4622/CS4624/CS4630/CS4280 chips.
@@ -314,7 +312,6 @@ config SND_CTXFI config SND_DARLA20 tristate "(Echoaudio) Darla20" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Darla.
@@ -324,7 +321,6 @@ config SND_DARLA20 config SND_GINA20 tristate "(Echoaudio) Gina20" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Gina.
@@ -334,7 +330,6 @@ config SND_GINA20 config SND_LAYLA20 tristate "(Echoaudio) Layla20" - select FW_LOADER select SND_RAWMIDI select SND_PCM help
@@ -345,7 +340,6 @@ config SND_LAYLA20 config SND_DARLA24 tristate "(Echoaudio) Darla24" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Darla24.
@@ -355,7 +349,6 @@ config SND_DARLA24 config SND_GINA24 tristate "(Echoaudio) Gina24" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Gina24.
@@ -365,7 +358,6 @@ config SND_GINA24 config SND_LAYLA24 tristate "(Echoaudio) Layla24" - select FW_LOADER select SND_RAWMIDI select SND_PCM help
@@ -376,7 +368,6 @@ config SND_LAYLA24 config SND_MONA tristate "(Echoaudio) Mona" - select FW_LOADER select SND_RAWMIDI select SND_PCM help
@@ -387,7 +378,6 @@ config SND_MONA config SND_MIA tristate "(Echoaudio) Mia" - select FW_LOADER select SND_RAWMIDI select SND_PCM help
@@ -398,7 +388,6 @@ config SND_MIA config SND_ECHO3G tristate "(Echoaudio) 3G cards" - select FW_LOADER select SND_RAWMIDI select SND_PCM help
@@ -409,7 +398,6 @@ config SND_ECHO3G config SND_INDIGO tristate "(Echoaudio) Indigo" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Indigo.
@@ -419,7 +407,6 @@ config SND_INDIGO config SND_INDIGOIO tristate "(Echoaudio) Indigo IO" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Indigo IO.
@@ -429,7 +416,6 @@ config SND_INDIGOIO config SND_INDIGODJ tristate "(Echoaudio) Indigo DJ" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Indigo DJ.
@@ -439,7 +425,6 @@ config SND_INDIGODJ config SND_INDIGOIOX tristate "(Echoaudio) Indigo IOx" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Indigo IOx.
@@ -449,7 +434,6 @@ config SND_INDIGOIOX config SND_INDIGODJX tristate "(Echoaudio) Indigo DJx" - select FW_LOADER select SND_PCM help Say 'Y' or 'M' to include support for Echoaudio Indigo DJx.
@@ -459,7 +443,6 @@ config SND_INDIGODJX config SND_EMU10K1 tristate "Emu10k1 (SB Live!, Audigy, E-mu APS)" - select FW_LOADER select SND_HWDEP select SND_RAWMIDI select SND_AC97_CODEC
@@ -581,7 +564,6 @@ config SND_FM801_TEA575X_BOOL config SND_HDSP tristate "RME Hammerfall DSP Audio" - select FW_LOADER select SND_HWDEP select SND_RAWMIDI select SND_PCM
@@ -720,7 +702,6 @@ config SND_MAESTRO3_INPUT config SND_MIXART tristate "Digigram miXart" - select FW_LOADER select SND_HWDEP select SND_PCM help
@@ -741,7 +722,6 @@ config SND_NM256 config SND_PCXHR tristate "Digigram PCXHR" - select FW_LOADER select SND_PCM select SND_HWDEP help
@@ -752,7 +732,6 @@ config SND_PCXHR config SND_RIPTIDE tristate "Conexant Riptide" - select FW_LOADER select SND_OPL3_LIB select SND_MPU401_UART select SND_AC97_CODEC
sound/pci/hda/Kconfig +0 -2
@@ -86,7 +86,6 @@ config SND_HDA_INPUT_BEEP_MODE config SND_HDA_PATCH_LOADER bool "Support initialization patch loading for HD-audio" - select FW_LOADER select SND_HDA_RECONFIG help Say Y here to allow the HD-audio driver to load a pseudo
@@ -193,7 +192,6 @@ config SND_HDA_CODEC_CA0132_DSP bool "Support new DSP code for CA0132 codec" depends on SND_HDA_CODEC_CA0132 select SND_HDA_DSP_LOADER - select FW_LOADER help Say Y here to enable the DSP for Creative CA0132 for extended features like equalizer or echo cancellation.
sound/soc/sh/Kconfig +0 -1
@@ -32,7 +32,6 @@ config SND_SOC_SH4_SIU select DMA_ENGINE select DMADEVICES select SH_DMAE - select FW_LOADER config SND_SOC_RCAR tristate "R-Car series SRU/SCU/SSIU/SSI support"
sound/usb/Kconfig +0 -1
@@ -103,7 +103,6 @@ config SND_USB_US122L config SND_USB_6FIRE tristate "TerraTec DMX 6Fire USB" - select FW_LOADER select BITREVERSE select SND_RAWMIDI select SND_PCM
Signature · 1 line
2.3.2.209.gd67f9d5.dirty
Borislav Petkov
maintainerQuoted 6 lines · Luis R. Rodriguez
The above history is just *one* set of reasons to remove all kernel "select
FW_LOADER" entries. Another simple reason is that sinceFW_LOADERcan only be disabled by an EXPERT and it defaults to y its now a superfluous select. We will ensure things build for you if using EXPERT but if you have selected EXPERT you should know what run time requirements you need.
Couple of notes only:
So the "depends" thing is a step backwards IMO because people would have to go and enable FW_LOADER first. For the microcode loader, at least, "select" is perfectly fine.
There's also the other aspect that this FW_LOADER thing ends up enabled on 99.9% (!?) of the distro installations anyway, right? I.e., there's always something needing firmware from userspace.
So the current "default y" behind EXPERT is probably the correct thing anyway.
So yeah, send me the x86 microcode loader patch, I'll take it.
Thanks.
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--Borislav Petkov
maintainerQuoted 9 lines · Borislav Petkov
Quoted 6 lines · Luis R. Rodriguez
The above history is just *one* set of reasons to remove all kernel "select
FW_LOADER" entries. Another simple reason is that sinceFW_LOADERcan only be disabled by an EXPERT and it defaults to y its now a superfluous select. We will ensure things build for you if using EXPERT but if you have selected EXPERT you should know what run time requirements you need.Couple of notes only:
One thing I forgot last night: what about randconfigs? All that functionality which selects FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even with FW_LOADER unset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
Yap, just triggered one:
$ make randconfig; grep -E "(MICROCODE|FW_LOADER)" .config scripts/kconfig/conf --randconfig Kconfig KCONFIG_SEED=0xE14FBE97 warning: (USB_OTG_FSM && FSL_USB2_OTG && USB_MV_OTG) selects USB_OTG which has unmet direct dependencies (USB_SUPPORT && USB && PM) warning: (USB_OTG_FSM && FSL_USB2_OTG && USB_MV_OTG) selects USB_OTG which has unmet direct dependencies (USB_SUPPORT && USB && PM) warning: (SND_SOC_INTEL_BYTCR_RT5640_MACH && SND_SOC_INTEL_CHT_BSW_RT5672_MACH && SND_SOC_INTEL_CHT_BSW_RT5645_MACH) selects SND_SST_IPC_ACPI which has unmet direct dependencies (SOUND && !M68K && !UML && SND && SND_SOC && ACPI) # # configuration written to .config # CONFIG_MICROCODE=m CONFIG_MICROCODE_INTEL=y CONFIG_MICROCODE_AMD=y CONFIG_MICROCODE_OLD_INTERFACE=y # CONFIG_FW_LOADER is not set
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--Hi Boris,
Quoted 9 lines · Borislav Petkov
Quoted 8 lines · Borislav Petkov
Quoted 6 lines · Luis R. Rodriguez
The above history is just *one* set of reasons to remove all kernel "select
FW_LOADER" entries. Another simple reason is that sinceFW_LOADERcan only be disabled by an EXPERT and it defaults to y its now a superfluous select. We will ensure things build for you if using EXPERT but if you have selected EXPERT you should know what run time requirements you need.
Quoted 5 lines
Quoted 1 line
Couple of notes only:
So the "depends" thing is a step backwards IMO because people would have to go and enable
FW_LOADERfirst. For the microcode loader, at least, "select" is perfectly fine.
That's what the EXPERT protection is for.FW_LOADER will always be enabled, unless people think they're smarter than they are, and enable EXPERT, and disable FW_LOADER.
Quoted 5 lines
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
If you expect to _boot_ randconfig kernels, you better disable EXPERT. Still, your kernel can fail to boot due to some other critical dependency (does your kernel support your CPU?).
If you want to boot randconfig kernels, create an allrandomconfig file listing all options that must have a certain value:
$ cat allrandomconfig CONFIG_EXPERT=n CONFIG_MY_CRITICAL_FEATURE=y ... $
and use "make randconfig KCONFIG_ALLCONFIG=1".
Quoted 1 line
ECO tip #101: Trim your mails when you reply.
You trimmed too much, so I had to readd it manually ;-)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Borislav Petkov
maintainerQuoted 3 lines · Geert Uytterhoeven
That's what the EXPERT protection is for.
FW_LOADERwill always be enabled, unless people think they're smarter than they are, and enable EXPERT, and disableFW_LOADER.
I know. I was just making the negative aspect of "depends" vs "select" in that particular case, explicit. Just so that it is mentioned in the conversation.
Quoted 10 lines
If you want to boot randconfig kernels, create an allrandomconfig file listing all options that must have a certain value:
$ cat allrandomconfig CONFIG_EXPERT=n CONFIG_MY_CRITICAL_FEATURE=y ... $and use "make randconfig KCONFIG_ALLCONFIG=1".
I do that but does everyone who builds randconfigs? I had to go and look what variables Kbuild honors and prep an all.config and yadda yadda...
So maybe we should put that as a note somewhere for randconfig-building people.
Quoted 1 line
You trimmed too much, so I had to readd it manually ;-)
Yeah, that's what mail threads are for.
People tend to reply to a huuge mail with one sentence, bury it somewhere on page gazillion and one and not trim the rest. Antisocial behavior, that. :-P
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--Quoted 9 lines · Borislav Petkov
Quoted 9 lines
If you want to boot randconfig kernels, create an allrandomconfig file listing all options that must have a certain value:
$ cat allrandomconfig CONFIG_EXPERT=n CONFIG_MY_CRITICAL_FEATURE=y ... $
Quoted 4 lines
Quoted 1 line
and use "make randconfig KCONFIG_ALLCONFIG=1".
I do that but does everyone who builds randconfigs? I had to go and look what variables
Kbuildhonors and prep an all.config and yadda yadda...
The ones who just build them probably not (unless they care about a specific case), which increases build coverage, so that's good.
The ones who boot them probably do, else they'll find out the hard way...
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Borislav Petkov
maintainerQuoted 2 lines · Geert Uytterhoeven
The ones who just build them probably not (unless they care about a specific case), which increases build coverage, so that's good.
Right.
Quoted 2 lines
The ones who boot them probably do, else they'll find out the hard way...
Right, and this is what I wanted to avoid or at least save people the trouble. Maybe randconfig builders should be expert enough to be able to figure that out on their own.
:-)
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
Luis also tried to explain to me that disabling FW_LOADER shouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)
Quoted 7 lines
If you expect to _boot_ randconfig kernels, you better disable EXPERT. Still, your kernel can fail to boot due to some other critical dependency (does your kernel support your CPU?).
If you want to boot randconfig kernels, create an allrandomconfig file listing all options that must have a certain value:
Quoted 7 lines
$ cat allrandomconfig CONFIG_EXPERT=n CONFIG_MY_CRITICAL_FEATURE=y ... $and use "make randconfig KCONFIG_ALLCONFIG=1".
Booting (truly) randconfig builds is like booting allyesconfig or allnoconfig builds: no one sane should do that, right?
Paul Bolle
Quoted 2 lines · Paul Bolle
Booting (truly) randconfig builds is like booting allyesconfig or allnoconfig builds: no one sane should do that, right?
allyesconfig should boot, modulo size limitations in boot loaders and early kernel setup.
(No, commit 486df8bc4627bdfc ("m68k: Increase initial mapping to 8 or 16 MiB
if possible") is not good enough ;-)
$ size linux-m68k-allyesconfig/vmlinux text data bss dec hex filename 42540590 2438292 1645336 46624218 2c76dda linux-m68k-allyesconfig/vmlinux $
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Borislav Petkov
maintainerQuoted 2 lines · Paul Bolle
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail.
It doesn't. There are #else provisions in the header which define empty functions returning -EINVAL and so on.
Quoted 2 lines
Booting (truly) randconfig builds is like booting allyesconfig or allnoconfig builds: no one sane should do that, right?
Why not? It doesn't hurt to try to boot them in a guest. It sometimes triggers valid issues.
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)
But FW_LOADER is a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.
Takashi
Quoted 10 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 5 lines · Paul Bolle
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.
include/linux/firmware.h contains:
#if defined(CONFIG_FW_LOADER) || (defined(CONFIG_FW_LOADER_MODULE) && defined(MODULE)
int request_firmware(const struct firmware **fw, const char *name,
struct device *device);
[...]
#else
static inline int request_firmware(const struct firmware **fw,
const char *name,
struct device *device)
{
return -EINVAL;
}
[...]
#endifSo I _think_ the build should be fine.
But, in case of built-in users of request_firmware() and friends, actually using that build might not go as expected. But if you set EXPERT and disable FW_LOADER, or as you point out, set it to 'm', you own the pieces when things break, don't you?
But now I'm doing Luis' job. I didn't sign off on that patch! So let's hope Luis thought of all the corner cases.
Paul Bolle
At Fri, 22 May 2015 11:06:37 +0200,
Quoted 28 lines · Paul Bolle
Quoted 10 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 5 lines · Paul Bolle
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.
include/linux/firmware.hcontains:#if defined(CONFIG_FW_LOADER) || (defined(CONFIG_FW_LOADER_MODULE) && defined(MODULE) int request_firmware(const struct firmware **fw, const char *name, struct device *device); [...] #else static inline int request_firmware(const struct firmware **fw, const char *name, struct device *device) { return -EINVAL; } [...] #endif
Quoted 1 line
So I _think_ the build should be fine.
Ah, OK, that would work.
Quoted 4 lines
But, in case of built-in users of
request_firmware()and friends, actually using that build might not go as expected. But if you set EXPERT and disableFW_LOADER, or as you point out, set it to 'm', you own the pieces when things break, don't you?
Yeah, I'm not against it, too. My concern is only about the silent breakage.
Thinking of this again, I also concluded that removing the mostly superfluous "select FW_LOADER" would make things easier in the end.
Quoted 2 lines
But now I'm doing Luis' job. I didn't sign off on that patch! So let's hope Luis thought of all the corner cases.
Heh, let's hope we all reach to a happy end.
thanks,
Takashi
Quoted 11 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
Quoted 8 lines
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.
We could go both ways, either remove the "select" or replace it with "depends on". As you note keeping the "depends on" ensures run time sanity for the possible tristate mismatches, but this is an EXPERT concern. The crux of what option to go with is:
Should we concern ourselves with run time configuration issues when
folks enable EXPERT?
Without EXPERT all run time configurations are vetted to run as FW_LOADER defaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool for FW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot with FW_LOADER?
Luis
Dmitry Torokhov
maintainerQuoted 21 lines · Luis R. Rodriguez
Quoted 11 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...Quoted 8 lines
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.
Quoted 7 lines
We could go both ways, either remove the "select" or replace it with "depends on". As you note keeping the "depends on" ensures run time sanity for the possible tristate mismatches, but this is an EXPERT concern. The crux of what option to go with is:
Should we concern ourselves with run time configuration issues when
folks enable EXPERT?
Yes.
dtor@dtor-ws:~/kernel/master$ grep -r CONFIG_EXPERT /boot/config* /boot/config-3.13.0-49-generic:CONFIG_EXPERT=y /boot/config-3.13.0-52-generic:CONFIG_EXPERT=y
This is distro config and that is what many people use as a base for their own configs.
Quoted 11 lines
Without EXPERT all run time configurations are vetted to run as
FW_LOADERdefaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool forFW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot withFW_LOADER?
I am afraid that we are moving into wrong direction here. Why don't we look into Kconfig to teach it the difference between forced selection and dependency instead?
Thanks.
Signature · 1 line
Dmitry
Quoted 30 lines · Dmitry Torokhov
Quoted 21 lines · Luis R. Rodriguez
Quoted 11 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...Quoted 8 lines
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.Quoted 7 lines
We could go both ways, either remove the "select" or replace it with "depends on". As you note keeping the "depends on" ensures run time sanity for the possible tristate mismatches, but this is an EXPERT concern. The crux of what option to go with is:
Should we concern ourselves with run time configuration issues when
folks enable EXPERT?
Quoted 8 lines
Yes.
dtor@dtor-ws:~/kernel/master$ grep -r CONFIG_EXPERT /boot/config* /boot/config-3.13.0-49-generic:CONFIG_EXPERT=y /boot/config-3.13.0-52-generic:CONFIG_EXPERT=yThis is distro config and that is what many people use as a base for their own configs.
Smells Ubuntu'ish, and hence ChromeOS'ish :)
Either way SUSE kernels also have EXPERT=y as well. Would not be surprised if other major distributions also have EXPERT=y.
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need. Also, are there no other runtime configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.
Following on topic: such distro configs would never have FW_LOADER as n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
Quoted 11 lines
Quoted 10 lines
Without EXPERT all run time configurations are vetted to run as
FW_LOADERdefaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool forFW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot withFW_LOADER?
Quoted 3 lines
I am afraid that we are moving into wrong direction here. Why don't we look into
Kconfigto teach it the difference between forced selection and dependency instead?
That was what I originally hinted we should try to do. See the implications of the issue right now and my explanation of a whitelist [0], granted this is just a description, I tried looking at the code and quickly gave up, it was not clear how such a thing could be implemented :( I gave up after Paul assured me this could not be an issue, or rather that the logic on kconfig was correct.
FWIW, I think my original description of implications if this issue is not a real bug is a bit more clearer than the ones me and Paul ended up reviewing. The examples me and Paul reviewed are examples one can experiment if trying to address and fix the issue at hand in the most simplest cases.
[0] https://lkml.org/lkml/2015/5/12/597
Luis
Dmitry Torokhov
maintainerQuoted 33 lines · Luis R. Rodriguez
Quoted 32 lines · Dmitry Torokhov
Quoted 21 lines · Luis R. Rodriguez
Quoted 11 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...Quoted 8 lines
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.Quoted 7 lines
We could go both ways, either remove the "select" or replace it with "depends on". As you note keeping the "depends on" ensures run time sanity for the possible tristate mismatches, but this is an EXPERT concern. The crux of what option to go with is:
Should we concern ourselves with run time configuration issues when
folks enable EXPERT?Yes.
Quoted 8 lines
Quoted 6 lines
dtor@dtor-ws:~/kernel/master$ grep -r CONFIG_EXPERT /boot/config* /boot/config-3.13.0-49-generic:CONFIG_EXPERT=y /boot/config-3.13.0-52-generic:CONFIG_EXPERT=yThis is distro config and that is what many people use as a base for their own configs.
Smells Ubuntu'ish, and hence ChromeOS'ish :)
Yes, this coming from Ubuntu, but ChromeOS is closer to Gentoo if anything I'd say ;)
Quoted 3 lines
Either way SUSE kernels also have EXPERT=y as well. Would not be surprised if other major distributions also have EXPERT=y.
Right, I believe Fedora has it set as well.
Quoted 4 lines
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need.
Not major, rather tweaking the driver sets. Like expert V4L or HID configs.
Quoted 9 lines
Also, are there no other runtime
configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.Following on topic: such distro configs would never have
FW_LOADERas n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?
But the user who is not expecrt might drop it. The premise was "only experts would have CONFIG_EXPERT and thus we do not care about breakage". But if people use distro configs as a starting point they all are "experts".
Quoted 28 lines
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
Quoted 11 lines
Quoted 10 lines
Without EXPERT all run time configurations are vetted to run as
FW_LOADERdefaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool forFW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot withFW_LOADER?Quoted 3 lines
I am afraid that we are moving into wrong direction here. Why don't we look into
Kconfigto teach it the difference between forced selection and dependency instead?That was what I originally hinted we should try to do. See the implications of the issue right now and my explanation of a whitelist [0], granted this is just a description, I tried looking at the code and quickly gave up, it was not clear how such a thing could be implemented :( I gave up after Paul assured me this could not be an issue, or rather that the logic on kconfig was correct.
Quoted 4 lines
FWIW, I think my original description of implications if this issue is not a real bug is a bit more clearer than the ones me and Paul ended up reviewing. The examples me and Paul reviewed are examples one can experiment if trying to address and fix the issue at hand in the most simplest cases.
So I see your GYM and LOCKER example, or better, the original FW & CRYPTO warning and I wonder why do we actually go and check dependencies of selected symbols when resolving given symbol. I.e. should we not "cut" the select branches off and evaluate them on their own? I.e. when we try to select a symbol we note the symbols it tries to select and evaluate them on their own (selects still don't cascade down - right?) and then evaluate "depends on" symbols?
Thanks.
Signature · 1 line
Dmitry
Adding Josh, as I know he seems to be on an EXPERT crusade to tuck away system calls for tinyconfig, so likely can chime in for some insane EXPERT runtime issue that users can run into.
Quoted 7 lines · Dmitry Torokhov
Quoted 3 lines · Luis R. Rodriguez
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need.
Not major, rather tweaking the driver sets. Like expert V4L or HID configs.
Driver tweaks seems rather harmless compared to "trust me I know what I'm doing" even for all run time cases.
So the question still stands -- are there any kconfig options which depend on EXPERT which enable folks to screw up run time (not just driver tweaks? If so, would it be good to separate the two types of expert config options, one for tweaks, another for run time possible conflicts?
Reason I ask this is -- IMO we should not treat the possible EXPERT run time conflicts of FW_LOADER specially and try to be consistent, if we are ensuring EXPERT run time conflicts will work with FW_LOADER why not for others? If we want the option to let an EXPERT kconfig option enable run time conflicts maybe we need to split things up further.
Quoted 6 lines
Quoted 6 lines
Also, are there no other runtime
configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.
Quoted 8 lines
Quoted 3 lines
Following on topic: such distro configs would never have
FW_LOADERas n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?But the user who is not expert might drop it. The premise was "only experts would have
CONFIG_EXPERTand thus we do not care about breakage". But if people use distro configs as a starting point they all are "experts".
Replacing "select FW_LOADER" with "depends on FW_LOADER", although still pegged with EXPERT, would not allow EXPERT users (most folks we've determined now) to disable FW_LOADER if a driver was selected that needed FW_LOADER, in fact I could not also create a tristate conflict of say MICROCODE set to y and FW_LOADER to m. Please let me know if I've missed something with testing this though.
So unless I'm missing a corner case, it would seem replacing "select FW_LOADER" with "depends on FW_LOADER" is a reasonable option so far. The subtleties lie then now in since EXPERT is used by most distributions, everyone is an expert, and:
a) Do we want to be consistent about ensuring run time sanity of EXPERT?
b) Should we split EXPERT into tweak / possible run time break options?
Quoted 27 lines
Quoted 27 lines
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
Quoted 11 lines
Quoted 10 lines
Without EXPERT all run time configurations are vetted to run as
FW_LOADERdefaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool forFW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot withFW_LOADER?Quoted 3 lines
I am afraid that we are moving into wrong direction here. Why don't we look into
Kconfigto teach it the difference between forced selection and dependency instead?That was what I originally hinted we should try to do. See the implications of the issue right now and my explanation of a whitelist [0], granted this is just a description, I tried looking at the code and quickly gave up, it was not clear how such a thing could be implemented :( I gave up after Paul assured me this could not be an issue, or rather that the logic on kconfig was correct.
Quoted 8 lines
Quoted 4 lines
FWIW, I think my original description of implications if this issue is not a real bug is a bit more clearer than the ones me and Paul ended up reviewing. The examples me and Paul reviewed are examples one can experiment if trying to address and fix the issue at hand in the most simplest cases.
So I see your GYM and LOCKER example, or better, the original FW & CRYPTO warning and I wonder why do we actually go and check dependencies of selected symbols when resolving given symbol.
I'm no kconfig expert and I wondered the same.
Perhaps Paul can elaborate.
Quoted 5 lines
I.e. should we not
"cut" the select branches off and evaluate them on their own? I.e. when we try to select a symbol we note the symbols it tries to select and evaluate them on their own (selects still don't cascade down - right?) and then evaluate "depends on" symbols?
Dmitry Torokhov
maintainerQuoted 89 lines · Luis R. Rodriguez
Adding Josh, as I know he seems to be on an EXPERT crusade to tuck away system calls for tinyconfig, so likely can chime in for some insane EXPERT runtime issue that users can run into.
Quoted 7 lines · Dmitry Torokhov
Quoted 3 lines · Luis R. Rodriguez
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need.
Not major, rather tweaking the driver sets. Like expert V4L or HID configs.
Driver tweaks seems rather harmless compared to "trust me I know what I'm doing" even for all run time cases.
So the question still stands -- are there any kconfig options which depend on EXPERT which enable folks to screw up run time (not just driver tweaks? If so, would it be good to separate the two types of expert config options, one for tweaks, another for run time possible conflicts?
Reason I ask this is -- IMO we should not treat the possible EXPERT run time conflicts of
FW_LOADERspecially and try to be consistent, if we are ensuring EXPERT run time conflicts will work withFW_LOADERwhy not for others? If we want the option to let an EXPERT kconfig option enable run time conflicts maybe we need to split things up further.Quoted 6 lines
Quoted 6 lines
Also, are there no other runtime
configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.Quoted 8 lines
Quoted 3 lines
Following on topic: such distro configs would never have
FW_LOADERas n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?But the user who is not expert might drop it. The premise was "only experts would have
CONFIG_EXPERTand thus we do not care about breakage". But if people use distro configs as a starting point they all are "experts".Replacing "select
FW_LOADER" with "depends onFW_LOADER", although still pegged with EXPERT, would not allow EXPERT users (most folks we've determined now) to disableFW_LOADERif a driver was selected that neededFW_LOADER, in fact I could not also create a tristate conflict of say MICROCODE set to y andFW_LOADERto m. Please let me know if I've missed something with testing this though.So unless I'm missing a corner case, it would seem replacing "select
FW_LOADER" with "depends onFW_LOADER" is a reasonable option so far. The subtleties lie then now in since EXPERT is used by most distributions, everyone is an expert, and:a) Do we want to be consistent about ensuring run time sanity of EXPERT?
b) Should we split EXPERT into tweak / possible run time break options?Quoted 32 lines
Quoted 31 lines
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
Quoted 11 lines
Quoted 10 lines
Without EXPERT all run time configurations are vetted to run as
FW_LOADERdefaults to y. If we go down the path of removing the select completely we'd be taking a position that we could at least ensure EXPERT builds will work, but we cannot vet for not run time sanity of such build. I favor simplicity so would prefer to nuke the select completely and if we're really concerned about EXPERT users tristate mismatch misconfiguration why not just replace tristate with bool forFW_LOADER. That would do us the service of simplifying that code a bit, and leave only in place one way for folks that enable EXPERT to shoot themselves in the foot withFW_LOADER?Quoted 3 lines
I am afraid that we are moving into wrong direction here. Why don't we look into
Kconfigto teach it the difference between forced selection and dependency instead?That was what I originally hinted we should try to do. See the implications of the issue right now and my explanation of a whitelist [0], granted this is just a description, I tried looking at the code and quickly gave up, it was not clear how such a thing could be implemented :( I gave up after Paul assured me this could not be an issue, or rather that the logic on kconfig was correct.
FWIW, I think my original description of implications if this issue is not a real bug is a bit more clearer than the ones me and Paul ended up reviewing. The examples me and Paul reviewed are examples one can experiment if trying to address and fix the issue at hand in the most simplest cases.
Quoted 5 lines
Quoted 3 lines
So I see your GYM and LOCKER example, or better, the original FW & CRYPTO warning and I wonder why do we actually go and check dependencies of selected symbols when resolving given symbol.
I'm no kconfig expert
Me neither ;)
Quoted 9 lines
and I wondered the same.
Perhaps Paul can elaborate.
Quoted 5 lines
I.e. should we not
"cut" the select branches off and evaluate them on their own? I.e. when we try to select a symbol we note the symbols it tries to select and evaluate them on their own (selects still don't cascade down - right?) and then evaluate "depends on" symbols?
By the way, to resolve your FW loader issue can you simply select CRYPTO instead of depending on it? I believe that certain core facilities (such as FW loaders, Crypto, etc) should be normally selected rather than depended upon, just because of workflow. If user works with menuconfig and has core facility disable he won't event be presented with an option. So one has to perform several "passes" over menuconfig to slect everything that is needed.
Thanks.
Signature · 1 line
Dmitry
Quoted 3 lines · Luis R. Rodriguez
Adding Josh, as I know he seems to be on an EXPERT crusade to tuck away system calls for tinyconfig, so likely can chime in for some insane EXPERT runtime issue that users can run into.
Thanks.
Quoted 17 lines · Dmitry Torokhov
Quoted 7 lines · Luis R. Rodriguez
Quoted 3 lines
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need.
Not major, rather tweaking the driver sets. Like expert V4L or HID configs.
Driver tweaks seems rather harmless compared to "trust me I know what I'm doing" even for all run time cases.
So the question still stands -- are there any kconfig options which depend on EXPERT which enable folks to screw up run time (not just driver tweaks? If so, would it be good to separate the two types of expert config options, one for tweaks, another for run time possible conflicts?
Quoted 5 lines
Reason I ask this is -- IMO we should not treat the possible EXPERT run time conflicts of
FW_LOADERspecially and try to be consistent, if we are ensuring EXPERT run time conflicts will work withFW_LOADERwhy not for others? If we want the option to let an EXPERT kconfig option enable run time conflicts maybe we need to split things up further.
I don't think it should be possible, even with EXPERT, to disable an option that a driver needs at runtime without disabling that driver. If the driver *optionally* needs something, but can work without it, it shouldn't have "depends" or "select".
Quoted 22 lines
Quoted 6 lines
Quoted 6 lines
Also, are there no other runtime
configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.Quoted 8 lines
Quoted 3 lines
Following on topic: such distro configs would never have
FW_LOADERas n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?But the user who is not expert might drop it. The premise was "only experts would have
CONFIG_EXPERTand thus we do not care about breakage". But if people use distro configs as a starting point they all are "experts".Replacing "select
FW_LOADER" with "depends onFW_LOADER", although still pegged with EXPERT, would not allow EXPERT users (most folks we've determined now) to disableFW_LOADERif a driver was selected that neededFW_LOADER, in fact I could not also create a tristate conflict of say MICROCODE set to y andFW_LOADERto m. Please let me know if I've missed something with testing this though.
Quoted 7 lines
So unless I'm missing a corner case, it would seem replacing "select
FW_LOADER" with "depends onFW_LOADER" is a reasonable option so far. The subtleties lie then now in since EXPERT is used by most distributions, everyone is an expert, and:a) Do we want to be consistent about ensuring run time sanity of EXPERT?
b) Should we split EXPERT into tweak / possible run time break options?
"depends on FW_LOADER" and "select FW_LOADER" are equivalent from the point of view of what configs they allow; they differ in how the dependency is shown in UIs. "select" means you can always see the drivers, and you can't turn off FW_LOADER until you turn off all of the drivers. "depends" means that if you disable FW_LOADER all the drivers that depend on it disappear, and you can't turn them on unless you turn on FW_LOADER (which however is always on when !EXPERT).
Either way, if you have FW_LOADER and a pile of drivers using it enabled, you have to disable all the drivers before disabling FW_LOADER. The behavior only differs when you have FW_LOADER disabled, which is a sufficiently uncommon case that it doesn't matter deeply which approach you take. Personally, if FW_LOADER doesn't have any dependencies, I'd favor the select, since it avoids the case of "wait, where is the driver I expected and why can't I find it?".
However, from the sounds of things, you want to add a dependency to FW_LOADER. So in that case, yeah, switch to depends rather than rewriting Kconfig, unless you're feeling really ambitious.
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
By "recursive dependency issue", I'm guessing that someone wants to make FW_LOADER depend on something else, which breaks because "select" doesn't know how to recurse?
Ideally, someone should teach Kconfig to handle recursive dependencies, but in the meantime using "depends" makes sense.
- Josh Triplett
Quoted 6 lines · On Fri, May 22, 2015 at 02:43:57PM -0700, [email protected] wrote:
Quoted 3 lines · Luis R. Rodriguez
So the question still stands -- are there any kconfig options which depend on EXPERT which enable folks to screw up run time (not just driver tweaks?
I don't think it should be possible
OK, thanks!
Quoted 8 lines
Quoted 7 lines
So unless I'm missing a corner case, it would seem replacing "select
FW_LOADER" with "depends onFW_LOADER" is a reasonable option so far. The subtleties lie then now in since EXPERT is used by most distributions, everyone is an expert, and:a) Do we want to be consistent about ensuring run time sanity of EXPERT?
b) Should we split EXPERT into tweak / possible run time break options?
Quoted 6 lines
"depends on
FW_LOADER" and "selectFW_LOADER" are equivalent from the point of view of what configs they allow; they differ in how the dependency is shown in UIs.However, from the sounds of things, you want to add a dependency to
FW_LOADER.
Nope the recursive case was that a new component which depends on FW_LOADER ended up selecting CRYPTO, and it turns out another driver which depends on FW_LOADER also depended on CRYPTO so that in turn creates a recursive dependency.
Quoted 2 lines
So in that case, yeah, switch to depends rather than rewriting
Kconfig, unless you're feeling really ambitious.
OK so switching to "depends on" is the winning contender so far.
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Note that not changing this means we *will* eventually run into the recursive dependency issue later, either with my FW API change patches or some other future feature. Likewise for any other kconfig option with similar semantics.
By "recursive dependency issue", I'm guessing that someone wants to make
FW_LOADERdepend on something else, which breaks because "select" doesn't know how to recurse?
Kind of, the issue actually was a new component which depends on FW_LOADER and has crypto dependencies. Since the qat crypto driver selects FW_LOADER but also has a set of crypto dependencies that creates a recursive dependency loop.
Quoted 2 lines
Ideally, someone should teach
Kconfigto handle recursive dependencies, but in the meantime using "depends" makes sense.
Doesn't sound like anyone is that enthusiastic about that.
So... shall we just switch to "depends on" tree-wide then for FW_LOADER?
Luis
Herbert Xu
maintainerQuoted 4 lines · Luis R. Rodriguez
Kind of, the issue actually was a new component which depends on
FW_LOADERand has crypto dependencies. Since the qat crypto driver selectsFW_LOADERbut also has a set of crypto dependencies that creates a recursive dependency loop.
Actually, how about making FW_SIG select FW_LOADER instead of depending on it? I think this should break the cycle.
Cheers,
Signature · 3 lines
Email: Herbert Xu <[email protected]> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Quoted 8 lines · Herbert Xu
Quoted 4 lines · Luis R. Rodriguez
Kind of, the issue actually was a new component which depends on
FW_LOADERand has crypto dependencies. Since the qat crypto driver selectsFW_LOADERbut also has a set of crypto dependencies that creates a recursive dependency loop.Actually, how about making
FW_SIGselectFW_LOADERinstead of depending on it? I think this should break the cycle.
Indeed, it does. Kind of odd, but works - and well if others run into the recursive issue then we have two diverging solutions now:
a) Either swap all "select FOO" to "depends on FOO" or,
b) Change the offending "depends on FOO" to "select FOO"
So sticking to one seems to make Kconfig happy for recursive dependency solving for now...
Luis
Quoted 10 lines · Luis R. Rodriguez
Quoted 8 lines · Herbert Xu
Quoted 4 lines · Luis R. Rodriguez
Kind of, the issue actually was a new component which depends on
FW_LOADERand has crypto dependencies. Since the qat crypto driver selectsFW_LOADERbut also has a set of crypto dependencies that creates a recursive dependency loop.Actually, how about making
FW_SIGselectFW_LOADERinstead of depending on it? I think this should break the cycle.
Quoted 8 lines
Indeed, it does. Kind of odd, but works - and well if others run into the recursive issue then we have two diverging solutions now:
a) Either swap all "select FOO" to "depends on FOO" or,
b) Change the offending "depends on FOO" to "select FOO"So sticking to one seems to make
Kconfighappy for recursive dependency solving for now...
For simplicity and avoidance of massive tree-wide patches, (b) seems preferable in this case.
Long-term, I think ideally we should have *every* visible Kconfig option always pulled in by "depends on" rather than "select", with visibility and recursion handled by smarter tools. That said, meddle not in the internals of Kconfig, for it has many unshorn yaks (and yaccs).
- Josh Triplett
Quoted 10 lines · Luis R. Rodriguez
Quoted 8 lines · Herbert Xu
Quoted 4 lines · Luis R. Rodriguez
Kind of, the issue actually was a new component which depends on
FW_LOADERand has crypto dependencies. Since the qat crypto driver selectsFW_LOADERbut also has a set of crypto dependencies that creates a recursive dependency loop.Actually, how about making
FW_SIGselectFW_LOADERinstead of depending on it? I think this should break the cycle.
Quoted 8 lines
Indeed, it does. Kind of odd, but works - and well if others run into the recursive issue then we have two diverging solutions now:
a) Either swap all "select FOO" to "depends on FOO" or,
b) Change the offending "depends on FOO" to "select FOO"So sticking to one seems to make
Kconfighappy for recursive dependency solving for now...
Yes, that works, too.
Perhaps Kconfig should start warning if it encounters a symbol that is both selected and being depended on, even without detecting an explicit recursive dependency?
That's gonna give tons of warnings, though...
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,
I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Paul Bolle
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
I don't mean cyclic dependencies (for which Kconfig should just report an error, ideally including the full list of symbols forming the cycle).
I mean that Kconfig should do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.
As an intermediate measure, it'd be *really* handy to be able to browse in the curses UI directly from a symbol to the symbols it depends on to quickly enable/disable them, rather than having to look at the list of dependencies of a symbol, search for that symbol, remember the path the search showed, and browse there manually.
- Josh Triplett
Quoted 4 lines · Josh Triplett
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution.
(My English might be letting me down here. Or my grasp of elementary logic, for that matter. Recursive and cyclic are not really interchangeable? Should Kconfig perhaps report a "Cyclic dependency error"? )
Quoted 3 lines
If B
depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.
And how should kconfig handle, say:
- B depends on (A || D)
- C depends on B
Should (B && D) be enabled or (B && A)? This is not meant as a rhetorical question. But I do fear the complications for the choices this idea might encounter are, at best, not worth the effort.
Quoted 5 lines
As an intermediate measure, it'd be *really* handy to be able to browse in the curses UI directly from a symbol to the symbols it depends on to quickly enable/disable them, rather than having to look at the list of dependencies of a symbol, search for that symbol, remember the path the search showed, and browse there manually.
Probably. I'd rather not think about the UI involved when the symbol(s) that should be enabled is (are) two or more levels down the chain. And the odd "stacked" UI for searches in the curses tool adds to the fun. But then again, I'm a the-glass-is-half-empty person.
Paul Bolle
Quoted 10 lines · Paul Bolle
Quoted 4 lines · Josh Triplett
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution.(My English might be letting me down here. Or my grasp of elementary logic, for that matter. Recursive and cyclic are not really interchangeable? Should
Kconfigperhaps report a "Cyclic dependency error"? )
I do think that'd be a clearer message, yes.
For clarity: "recursive dependency resolution" as in "recursive resolution of dependencies", not "resolution of recursive dependencies".
Quoted 11 lines
Quoted 3 lines
If B
depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.And how should kconfig handle, say:
- B depends on (A || D)
- C depends on BShould (B && D) be enabled or (B && A)? This is not meant as a rhetorical question. But I do fear the complications for the choices this idea might encounter are, at best, not worth the effort.
That's exactly the kind of problem that makes this difficult to do, and has thus stopped anyone from working on it. A fully general solution requires a system that can also solve arbitrary logic problems, which is not necessarily a feature.
The typical choice made by package management systems is to enable B and A, preferring the leftmost dependency of B. That's what Debian's apt does, in the absence of conflicts.
For an initial solution, I'd suggest just recursing through "select" dependencies; if C has "select B", and B has "select A", enabling C should enable B and A. Since you can't "select A || D", that seems straightforward enough. You can "select A if X", but for that, just bail out if the value of the expression X is changed by any of the select statements.
Quoted 10 lines
Quoted 5 lines
As an intermediate measure, it'd be *really* handy to be able to browse in the curses UI directly from a symbol to the symbols it depends on to quickly enable/disable them, rather than having to look at the list of dependencies of a symbol, search for that symbol, remember the path the search showed, and browse there manually.
Probably. I'd rather not think about the UI involved when the symbol(s) that should be enabled is (are) two or more levels down the chain. And the odd "stacked" UI for searches in the curses tool adds to the fun. But then again, I'm a the-glass-is-half-empty person.
I'm more thinking that if you're looking at an option, you should be able to hit a key to see all the symbols that option depends on in a list, then navigate to one of those symbols, and be taken to the menu containing that symbol with the cursor on that symbol. (You could then hit a key to go "back" to where you were.) It's not perfect, but it's an improvement.
I'd also suggest that the search mechanism should produce a list that can be browsed with the arrow keys, with enter taking you to the symbol.
- Josh Triplett
Quoted 4 lines · Josh Triplett
Quoted 1 line · Paul Bolle
On Mon, 2015-05-25 at 10:54 -0700, Josh Triplett wrote:
For clarity: "recursive dependency resolution" as in "recursive resolution of dependencies", not "resolution of recursive dependencies".
I feel a headache coming up.
Quoted 6 lines
For an initial solution, I'd suggest just recursing through "select" dependencies; if C has "select B", and B has "select A", enabling C should enable B and A. Since you can't "select A || D", that seems straightforward enough. You can "select A if X", but for that, just bail out if the value of the expression X is changed by any of the select statements.
Actually, kconfig already recurses through select statements. (I haven't checked, but I'd expect it to handle the "select A if X" case correctly while doing the recursive select dance.)
Thanks,
Paul Bolle
Quoted 8 lines · Josh Triplett
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Quoted 6 lines
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.
That's really hard in practise you have to make any symbol that selects something depend on the dependencies of the selected symbol. You can't do this without involving a SAT solver. A guy promised to do this a couple of years ago, but the patches never materialised. However, they may exist somewhere if someone wants to take a look at completing it.
James
Quoted 7 lines
As an intermediate measure, it'd be *really* handy to be able to browse in the curses UI directly from a symbol to the symbols it depends on to quickly enable/disable them, rather than having to look at the list of dependencies of a symbol, search for that symbol, remember the path the search showed, and browse there manually.
- Josh Triplett
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Quoted 16 lines · James Bottomley
Quoted 8 lines · Josh Triplett
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Quoted 6 lines
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.
Quoted 5 lines
That's really hard in practise you have to make any symbol that selects something depend on the dependencies of the selected symbol. You can't do this without involving a SAT solver. A guy promised to do this a couple of years ago, but the patches never materialised. However, they may exist somewhere if someone wants to take a look at completing it.
You can't do this in the completely general case without a SAT solver. However, I think it's possible to help the user a bit more than than Kconfig currently does without introducing a general dependency solver.
In particular, it'd be nice to have an easy way to see at a glance "can't enable C because it depends on B", together with an easy way to get to B to enable it to get to C. User-guided dependency resolution seems like an improvement over no dependency resolution.
- Josh Triplett
Quoted 23 lines · Josh Triplett
Quoted 16 lines · James Bottomley
Quoted 8 lines · Josh Triplett
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Quoted 6 lines
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.Quoted 5 lines
That's really hard in practise you have to make any symbol that selects something depend on the dependencies of the selected symbol. You can't do this without involving a SAT solver. A guy promised to do this a couple of years ago, but the patches never materialised. However, they may exist somewhere if someone wants to take a look at completing it.
Quoted 3 lines
You can't do this in the completely general case without a SAT solver. However, I think it's possible to help the user a bit more than than
Kconfigcurrently does without introducing a general dependency solver.
Well the logically simplest way of doing that it to require that a selected symbol can have no dependencies (but may select others) and to make this what Kconfig enforces ... of course, our current Kconfig tree violates this all over the place.
Quoted 4 lines
In particular, it'd be nice to have an easy way to see at a glance "can't enable C because it depends on B", together with an easy way to get to B to enable it to get to C. User-guided dependency resolution seems like an improvement over no dependency resolution.
Sure, but it's mostly people like us doing it and find . -name Kconfig\* -exec grep -3 <symbol> {} /dev/null \; usually does the trick for me.
James
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Quoted 24 lines · James Bottomley
Quoted 23 lines · Josh Triplett
Quoted 21 lines · James Bottomley
Quoted 8 lines · Josh Triplett
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Quoted 6 lines
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.That's really hard in practise you have to make any symbol that selects something depend on the dependencies of the selected symbol. You can't do this without involving a SAT solver. A guy promised to do this a couple of years ago, but the patches never materialised. However, they may exist somewhere if someone wants to take a look at completing it.
Quoted 8 lines
Quoted 3 lines
You can't do this in the completely general case without a SAT solver. However, I think it's possible to help the user a bit more than than
Kconfigcurrently does without introducing a general dependency solver.Well the logically simplest way of doing that it to require that a selected symbol can have no dependencies (but may select others) and to make this what
Kconfigenforces ... of course, our currentKconfigtree violates this all over the place.
Or enforce that it has no disabled dependencies.
Quoted 7 lines
Quoted 4 lines
In particular, it'd be nice to have an easy way to see at a glance "can't enable C because it depends on B", together with an easy way to get to B to enable it to get to C. User-guided dependency resolution seems like an improvement over no dependency resolution.
Sure, but it's mostly people like us doing it and find . -name Kconfig\* -exec grep -3 <symbol> {}
/dev/null\; usually does the trick for me.
That's not particularly convenient from the middle of menuconfig when configuring a kernel. It'd be nice to be able to navigate in one step from a symbol to its dependencies within the menuconfig interface.
- Josh Triplett
Quoted 29 lines · Josh Triplett
Quoted 28 lines · James Bottomley
Quoted 26 lines · Josh Triplett
Quoted 16 lines · James Bottomley
Quoted 8 lines · Josh Triplett
Quoted 6 lines · Paul Bolle
Quoted 1 line · On Fri, 2015-05-22 at 14:43 -0700, [email protected] wrote:
Ideally, someone should teach
Kconfigto handle recursive dependencies,I'm probably reading too much in this remark, but how should it handle that other than returning an error because the configuration it's fed makes it run in circles?
Quoted 6 lines
I don't mean cyclic dependencies (for which
Kconfigshould just report an error, ideally including the full list of symbols forming the cycle).I mean that
Kconfigshould do recursive dependency resolution. If B depends on A, and C depends on B, I should be able to turn on C directly and have B and A enabled.Quoted 5 lines
That's really hard in practise you have to make any symbol that selects something depend on the dependencies of the selected symbol. You can't do this without involving a SAT solver. A guy promised to do this a couple of years ago, but the patches never materialised. However, they may exist somewhere if someone wants to take a look at completing it.
You can't do this in the completely general case without a SAT solver. However, I think it's possible to help the user a bit more than than
Kconfigcurrently does without introducing a general dependency solver.
Quoted 6 lines
Quoted 4 lines
Well the logically simplest way of doing that it to require that a selected symbol can have no dependencies (but may select others) and to make this what
Kconfigenforces ... of course, our currentKconfigtree violates this all over the place.Or enforce that it has no disabled dependencies.
That's what doesn't work: once you depend on a symbol that selects something else, you have to evaluate the satisfiability condition again recursively ... and that's a SAT solver. Unfortunately, you need either a heuristic that makes the logic chains simple or resign yourself to doing it properly.
Quoted 11 lines
Quoted 7 lines
Quoted 4 lines
In particular, it'd be nice to have an easy way to see at a glance "can't enable C because it depends on B", together with an easy way to get to B to enable it to get to C. User-guided dependency resolution seems like an improvement over no dependency resolution.
Sure, but it's mostly people like us doing it and find . -name Kconfig\* -exec grep -3 <symbol> {}
/dev/null\; usually does the trick for me.That's not particularly convenient from the middle of menuconfig when configuring a kernel. It'd be nice to be able to navigate in one step from a symbol to its dependencies within the menuconfig interface.
Well, it works for me but, sure, do something you prefer ... I use make config rather than menuconfig anyway.
James
���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m����������?�I�
Quoted 2 lines · James Bottomley
Sure, but it's mostly people like us doing it and find . -name Kconfig\* -exec grep -3 <symbol> {}
/dev/null\; usually does the trick for me.
git grep -w3 <symbol> -- "*Kconf*"
And life becomes even easier with the following in your ~/.gitconfig:
[alias]
kgrep = !git grep $* -- \"*Kconf*\"
git kgrep -w3 <symbol>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoted 7 lines · Luis R. Rodriguez
Quoted 3 lines
So I see your GYM and LOCKER example, or better, the original FW & CRYPTO warning and I wonder why do we actually go and check dependencies of selected symbols when resolving given symbol.
I'm no kconfig expert and I wondered the same.
Perhaps Paul can elaborate.
Because both the dependencies of a symbol and the select statements on that symbol influence the set of values that is possible for that symbol.
Paul Bolle
Quoted 38 lines · Dmitry Torokhov
Quoted 37 lines · Luis R. Rodriguez
Quoted 36 lines · Dmitry Torokhov
Quoted 21 lines · Luis R. Rodriguez
Quoted 11 lines · Takashi Iwai
At Fri, 22 May 2015 10:17:48 +0200,
Quoted 9 lines · Paul Bolle
Quoted 6 lines · Geert Uytterhoeven
Quoted 5 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...Quoted 8 lines
Quoted 5 lines
Luis also tried to explain to me that disabling
FW_LOADERshouldn't make the build fail. (And, of course, we could decide to not care about randconfig builds that have EXPERT set. Maybe we could even special case EXPERT in randconfig. But that would make randconfig builds less useful. That's a separate issue, anyhow.)But
FW_LOADERis a tristate, so it might be inconsistent if selected randomly? Luis' patch doesn't add depends but just removes select.Quoted 7 lines
We could go both ways, either remove the "select" or replace it with "depends on". As you note keeping the "depends on" ensures run time sanity for the possible tristate mismatches, but this is an EXPERT concern. The crux of what option to go with is:
Should we concern ourselves with run time configuration issues when
folks enable EXPERT?Yes.
dtor@dtor-ws:~/kernel/master$ grep -r CONFIG_EXPERT /boot/config* /boot/config-3.13.0-49-generic:CONFIG_EXPERT=y /boot/config-3.13.0-52-generic:CONFIG_EXPERT=y
Quoted 8 lines
Quoted 4 lines
Quoted 2 lines
This is distro config and that is what many people use as a base for their own configs.
Smells Ubuntu'ish, and hence ChromeOS'ish :)
Yes, this coming from Ubuntu, but ChromeOS is closer to Gentoo if anything I'd say ;)
Technically, ChromeOS _is_ Gentoo, just with a bunch of third party patches and configuration layered on top. IMHO though, they really do have a good excuse for enabling CONFIG_EXPERT (namely, they have absolute control over pretty much the entire system, and run on specialized hardware).
Quoted 6 lines
Quoted 3 lines
Either way SUSE kernels also have EXPERT=y as well. Would not be surprised if other major distributions also have EXPERT=y.
Right, I believe Fedora has it set as well.
I'm pretty certain that Fedora does, and that Arch does as well.
Quoted 8 lines
Quoted 4 lines
OK so we obviously care about EXPERT run time issues then, seems to kind of defeat the purpose of EXPERT though, no? Makes me wonder what major options are under EXPERT which most distros need.
Not major, rather tweaking the driver sets. Like expert V4L or HID configs.
Of the people I know personally who build their own kernels, about 80% have EXPERT=y for exactly this reason, namely, they want finer control over driver options, and don't do anything to the options under the EXPERT menu. Personally this is one of two reasons that I use EXPERT=y, the other being disabling the handful of deprecated syscalls at the top of the list (and disabling AIO when I'm building embedded systems).
Quoted 6 lines
Quoted 6 lines
Also, are there no other runtime
configuration options tucked under EXPERT that we *do not* resolve right now? Or should all be taken care of? If not then we are being inconsistent. Both of these are side topics -- but since I have a feeling this might come up again, it may be worth evaluation.
Quoted 9 lines
Quoted 3 lines
Following on topic: such distro configs would never have
FW_LOADERas n or m, though right? Is that sufficient for us to drop the select and not apply the "depends on" replacement ? Or do we want to stick to the "depend on"?But the user who is not expecrt might drop it. The premise was "only experts would have
CONFIG_EXPERTand thus we do not care about breakage". But if people use distro configs as a starting point they all are "experts".
Based on this, maybe the EXPERT stuff should be subdivided into driver options, and core API options under separate config options.
Quoted 7 lines · Austin S Hemmelgarn
Quoted 6 lines
Quoted 3 lines
Either way SUSE kernels also have EXPERT=y as well. Would not be surprised if
other major distributions also have EXPERT=y.Right, I believe Fedora has it set as well.
I'm pretty certain that Fedora does, and that Arch does as well.
Perhaps they just use allmodconfig, which enables EXPERT ;-)
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Quoted 9 lines · Geert Uytterhoeven
Quoted 7 lines · Austin S Hemmelgarn
Quoted 6 lines
Quoted 3 lines
Either way SUSE kernels also have EXPERT=y as well. Would not be surprised if
other major distributions also have EXPERT=y.Right, I believe Fedora has it set as well.
I'm pretty certain that Fedora does, and that Arch does as well.
Fedora doesn't set EXPERT, and as far as I can see, never has. Same goes for EMBEDDED, which is the predecessor of EXPERT.
If EXPERT really turns out to be set on most distributions it will have gone the way of EXPERIMENTAL: a warning sign that lost its impact through overuse. Perhaps that's the way these things go. But I don't think that has happened with EXPERT, not yet.
Paul Bolle
Quoted 1 line · Dmitry Torokhov
selects still don't cascade down - right?
select statements trigger the select statements of the selected symbol, if that's what you mean.
Paul Bolle
Quoted 6 lines · Borislav Petkov
One thing I forgot last night: what about randconfigs? All that functionality which selects
FW_LOADER, won't boot anymore, right? I mean, there are provisions to build fine even withFW_LOADERunset but if you want to boot-test those kernels, you will artificially fail due to missing request_firmware* things...
Quoted 8 lines
Yap, just triggered one:
$ make randconfig; grep -E "(MICROCODE|FW_LOADER)" .config scripts/kconfig/conf --randconfig Kconfig KCONFIG_SEED=0xE14FBE97 warning: (USB_OTG_FSM && FSL_USB2_OTG && USB_MV_OTG) selects USB_OTG which has unmet direct dependencies (USB_SUPPORT && USB && PM) warning: (USB_OTG_FSM && FSL_USB2_OTG && USB_MV_OTG) selects USB_OTG which has unmet direct dependencies (USB_SUPPORT && USB && PM) warning: (SND_SOC_INTEL_BYTCR_RT5640_MACH && SND_SOC_INTEL_CHT_BSW_RT5672_MACH && SND_SOC_INTEL_CHT_BSW_RT5645_MACH) selects SND_SST_IPC_ACPI which has unmet direct dependencies (SOUND && !M68K && !UML && SND && SND_SOC && ACPI)
Those missing dependencies look unrelated to not having FW_LOADER set. Did I miss something here?
Thanks,
Paul Bolle
Borislav Petkov
maintainerQuoted 2 lines · Paul Bolle
Those missing dependencies look unrelated to not having
FW_LOADERset. Did I miss something here?
I just pasted what appeared on the console here. You can ignore them.
Signature · 5 lines
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--Quoted 4 lines · Luis R. Rodriguez
Paul also insists that *there is no reason* to think that current logic of the kconfig tools, as it is currently implemented, is flawed [2].
(It should be noted that Russell King taught me the approach I took there. Before Russell did that I found dependency errors involving recursive selects, ie most of them, rather baffling.
Any flaws in that message are entirely mine, of course.)
Quoted 1 line
"insist" is pushing things a bit too far. "claims" or "stated"?
Paul Bolle