ARM, timers: ebsa110 cleanup
Subsystems (8)
cover ARM, timers: ebsa110 cleanup
The ebsa110 platform is the last thing that uses CONFIG_ARCH_USES_GETTIMEOFFSET, and Russell has previously said that he thinks the platform can be retired now.
Removing it allows us clean up the timer code by throwing out all of the references to arch_gettimeoffset().
The am79c961a network driver can presumably also go, as no other platform references it.
I don't expect these to make it into the coming merge window, posting here as an RFC, and as a reference for the mildly related timer tick series.
Arnd
Arnd Bergmann (3):
ARM: remove ebsa110 platform
net: remove am79c961a driver
timekeeping: remove arch_gettimeoffset
.../time/modern-timekeeping/arch-support.txt | 33 - MAINTAINERS | 8 - arch/arm/Kconfig | 24 +- arch/arm/Kconfig.debug | 6 +- arch/arm/Makefile | 8 - arch/arm/configs/ebsa110_defconfig | 74 -- arch/arm/kernel/Makefile | 6 +- arch/arm/mach-ebsa110/Makefile | 8 - arch/arm/mach-ebsa110/Makefile.boot | 5 - arch/arm/mach-ebsa110/core.c | 323 -------- arch/arm/mach-ebsa110/core.h | 38 - .../mach-ebsa110/include/mach/entry-macro.S | 33 - arch/arm/mach-ebsa110/include/mach/hardware.h | 21 - arch/arm/mach-ebsa110/include/mach/io.h | 89 -- arch/arm/mach-ebsa110/include/mach/irqs.h | 17 - arch/arm/mach-ebsa110/include/mach/memory.h | 22 - .../mach-ebsa110/include/mach/uncompress.h | 41 - arch/arm/mach-ebsa110/io.c | 440 ---------- arch/arm/mach-ebsa110/leds.c | 71 -- drivers/Makefile | 2 - drivers/clocksource/Kconfig | 2 +- drivers/net/ethernet/amd/Kconfig | 10 +- drivers/net/ethernet/amd/Makefile | 1 - drivers/net/ethernet/amd/am79c961a.c | 763 ------------------ drivers/net/ethernet/amd/am79c961a.h | 143 ---- include/linux/time.h | 13 - kernel/time/Kconfig | 9 - kernel/time/clocksource.c | 8 - kernel/time/timekeeping.c | 25 +- kernel/trace/Kconfig | 2 - 30 files changed, 9 insertions(+), 2236 deletions(-) delete mode 100644 Documentation/features/time/modern-timekeeping/arch-support.txt delete mode 100644 arch/arm/configs/ebsa110_defconfig delete mode 100644 arch/arm/mach-ebsa110/Makefile delete mode 100644 arch/arm/mach-ebsa110/Makefile.boot delete mode 100644 arch/arm/mach-ebsa110/core.c delete mode 100644 arch/arm/mach-ebsa110/core.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/entry-macro.S delete mode 100644 arch/arm/mach-ebsa110/include/mach/hardware.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/io.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/irqs.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/memory.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/uncompress.h delete mode 100644 arch/arm/mach-ebsa110/io.c delete mode 100644 arch/arm/mach-ebsa110/leds.c delete mode 100644 drivers/net/ethernet/amd/am79c961a.c delete mode 100644 drivers/net/ethernet/amd/am79c961a.h
Signature · 1 line
2.27.0
PATCH 1/3 ARM: remove ebsa110 platform
Russell said that he is no longer using this machine, and it seems that nobody else has in a long time, so it's time to say goodbye to it.
As this is the last platform using CONFIG_ARCH_USES_GETTIMEOFFSET, there are some follow-up patches to remove that as well.
--- MAINTAINERS | 8 - arch/arm/Kconfig | 24 +- arch/arm/Kconfig.debug | 6 +- arch/arm/Makefile | 8 - arch/arm/configs/ebsa110_defconfig | 74 --- arch/arm/kernel/Makefile | 6 +- arch/arm/mach-ebsa110/Makefile | 8 - arch/arm/mach-ebsa110/Makefile.boot | 5 - arch/arm/mach-ebsa110/core.c | 323 ------------- arch/arm/mach-ebsa110/core.h | 38 -- .../mach-ebsa110/include/mach/entry-macro.S | 33 -- arch/arm/mach-ebsa110/include/mach/hardware.h | 21 - arch/arm/mach-ebsa110/include/mach/io.h | 89 ---- arch/arm/mach-ebsa110/include/mach/irqs.h | 17 - arch/arm/mach-ebsa110/include/mach/memory.h | 22 - .../mach-ebsa110/include/mach/uncompress.h | 41 -- arch/arm/mach-ebsa110/io.c | 440 ------------------ arch/arm/mach-ebsa110/leds.c | 71 --- 18 files changed, 6 insertions(+), 1228 deletions(-) delete mode 100644 arch/arm/configs/ebsa110_defconfig delete mode 100644 arch/arm/mach-ebsa110/Makefile delete mode 100644 arch/arm/mach-ebsa110/Makefile.boot delete mode 100644 arch/arm/mach-ebsa110/core.c delete mode 100644 arch/arm/mach-ebsa110/core.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/entry-macro.S delete mode 100644 arch/arm/mach-ebsa110/include/mach/hardware.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/io.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/irqs.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/memory.h delete mode 100644 arch/arm/mach-ebsa110/include/mach/uncompress.h delete mode 100644 arch/arm/mach-ebsa110/io.c delete mode 100644 arch/arm/mach-ebsa110/leds.c
MAINTAINERS +0 -8
@@ -1809,14 +1809,6 @@ F: drivers/firmware/turris-mox-rwtm.c F: drivers/gpio/gpio-moxtet.c F: include/linux/moxtet.h -ARM/EBSA110 MACHINE SUPPORT -M: Russell King <[email protected]> -L: [email protected] (moderated for non-subscribers) -S: Maintained -W: http://www.armlinux.org.uk/ -F: arch/arm/mach-ebsa110/ -F: drivers/net/ethernet/amd/am79c961a.* - ARM/ENERGY MICRO (SILICON LABS) EFM32 SUPPORT M: Uwe Kleine-König <[email protected]> R: Pengutronix Kernel Team <[email protected]>
arch/arm/Kconfig +3 -21
@@ -264,8 +264,7 @@ config PHYS_OFFSET hex "Physical address of main memory" if MMU depends on !ARM_PATCH_PHYS_VIRT default DRAM_BASE if !MMU - default 0x00000000 if ARCH_EBSA110 || \ - ARCH_FOOTBRIDGE || \ + default 0x00000000 if ARCH_FOOTBRIDGE || \ ARCH_INTEGRATOR || \ ARCH_REALVIEW default 0x10000000 if ARCH_OMAP1 || ARCH_RPC
@@ -341,20 +340,6 @@ config ARM_SINGLE_ARMV7M select SPARSE_IRQ select USE_OF -config ARCH_EBSA110 - bool "EBSA-110" - select ARCH_USES_GETTIMEOFFSET - select CPU_SA110 - select ISA - select NEED_MACH_IO_H - select NEED_MACH_MEMORY_H - select NO_IOPORT_MAP - help - This is an evaluation board for the StrongARM processor available - from Digital. It has limited hardware on-board, including an - Ethernet interface, two PCMCIA sockets, two serial ports and a - parallel port. - config ARCH_EP93XX bool "EP93xx-based" select ARCH_SPARSEMEM_ENABLE
@@ -1372,7 +1357,6 @@ config ARCH_NR_GPIO config HZ_FIXED int - default 200 if ARCH_EBSA110 default 128 if SOC_AT91RM9200 default 0
@@ -1588,9 +1572,7 @@ config FORCE_MAX_ZONEORDER a value of 11 means that the largest free memory block is 2^10 pages. config ALIGNMENT_TRAP - bool - depends on CPU_CP15_MMU - default y if !ARCH_EBSA110 + def_bool CPU_CP15_MMU select HAVE_PROC_CPU if PROC_FS help ARM processors cannot fetch/store information which is not
@@ -1799,7 +1781,7 @@ config CMDLINE string "Default kernel command string" default "" help - On some architectures (EBSA110 and CATS), there is currently no way + On some architectures (e.g. CATS), there is currently no way for the boot loader to pass arguments to the kernel. For these architectures, you should supply some command-line options at build time by entering them here. As a minimum, you should specify the
arch/arm/Kconfig.debug +2 -4
@@ -1610,7 +1610,7 @@ config DEBUG_UART_PL01X # Compatibility options for 8250 config DEBUG_UART_8250 - def_bool ARCH_EBSA110 || ARCH_IOP32X || ARCH_IXP4XX || ARCH_RPC + def_bool ARCH_IOP32X || ARCH_IXP4XX || ARCH_RPC config DEBUG_UART_PHYS hex "Physical base address of debug UART"
@@ -1713,7 +1713,6 @@ config DEBUG_UART_PHYS default 0xe8008000 if DEBUG_R7S72100_SCIF2 || DEBUG_R7S9210_SCIF2 default 0xe8009000 if DEBUG_R7S9210_SCIF4 default 0xf0000000 if DEBUG_DIGICOLOR_UA0 - default 0xf0000be0 if ARCH_EBSA110 default 0xf1012000 if DEBUG_MVEBU_UART0_ALTERNATE default 0xf1012100 if DEBUG_MVEBU_UART1_ALTERNATE default 0xf7fc9000 if DEBUG_BERLIN_UART
@@ -1760,7 +1759,6 @@ config DEBUG_UART_VIRT default 0xc8821000 if DEBUG_RV1108_UART1 default 0xc8912000 if DEBUG_RV1108_UART0 default 0xe0010fe0 if ARCH_RPC - default 0xf0000be0 if ARCH_EBSA110 default 0xf0010000 if DEBUG_ASM9260_UART default 0xf0100000 if DEBUG_DIGICOLOR_UA0 default 0xf01fb000 if DEBUG_NOMADIK_UART
@@ -1896,7 +1894,7 @@ config DEBUG_UART_8250_PALMCHIP config DEBUG_UART_8250_FLOW_CONTROL bool "Enable flow control for 8250 UART" depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250 - default y if ARCH_EBSA110 || DEBUG_FOOTBRIDGE_COM1 || DEBUG_GEMINI || ARCH_RPC + default y if DEBUG_FOOTBRIDGE_COM1 || DEBUG_GEMINI || ARCH_RPC config DEBUG_UNCOMPRESS bool "Enable decompressor debugging via DEBUG_LL output"
arch/arm/Makefile +0 -8
@@ -169,7 +169,6 @@ machine-$(CONFIG_ARCH_CNS3XXX) += cns3xxx machine-$(CONFIG_ARCH_DAVINCI) += davinci machine-$(CONFIG_ARCH_DIGICOLOR) += digicolor machine-$(CONFIG_ARCH_DOVE) += dove -machine-$(CONFIG_ARCH_EBSA110) += ebsa110 machine-$(CONFIG_ARCH_EFM32) += efm32 machine-$(CONFIG_ARCH_EP93XX) += ep93xx machine-$(CONFIG_ARCH_EXYNOS) += exynos
@@ -241,13 +240,6 @@ plat-$(CONFIG_PLAT_PXA) += pxa plat-$(CONFIG_PLAT_S3C24XX) += samsung plat-$(CONFIG_PLAT_VERSATILE) += versatile -ifeq ($(CONFIG_ARCH_EBSA110),y) -# This is what happens if you forget the IOCS16 line. -# PCMCIA cards stop working. -CFLAGS_3c589_cs.o :=-DISA_SIXTEEN_BIT_PERIPHERAL -export CFLAGS_3c589_cs.o -endif - # The byte offset of the kernel image in RAM from the start of RAM. TEXT_OFFSET := $(textofs-y)
arch/arm/configs/ebsa110_defconfig deleted +0 -74
74 lines not on this page · raw message
arch/arm/kernel/Makefile +1 -5
@@ -17,7 +17,7 @@ CFLAGS_REMOVE_return_address.o = -pg # Object file lists. obj-y := elf.o entry-common.o irq.o opcodes.o \ - process.o ptrace.o reboot.o \ + process.o ptrace.o reboot.o io.o \ setup.o signal.o sigreturn_codes.o \ stacktrace.o sys_arm.o time.o traps.o
@@ -83,10 +83,6 @@ AFLAGS_iwmmxt.o := -Wa,-mcpu=iwmmxt obj-$(CONFIG_ARM_CPU_TOPOLOGY) += topology.o obj-$(CONFIG_VDSO) += vdso.o obj-$(CONFIG_EFI) += efi.o - -ifneq ($(CONFIG_ARCH_EBSA110),y) - obj-y += io.o -endif obj-$(CONFIG_PARAVIRT) += paravirt.o head-y := head$(MMUEXT).o
arch/arm/mach-ebsa110/Makefile deleted +0 -8
@@ -1,8 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only -# -# Makefile for the linux kernel. -# - -# Object file lists. - -obj-y := core.o io.o leds.o
arch/arm/mach-ebsa110/Makefile.boot deleted +0 -5
@@ -1,5 +0,0 @@ -# SPDX-License-Identifier: GPL-2.0-only - zreladdr-y += 0x00008000 -params_phys-y := 0x00000400 -initrd_phys-y := 0x00800000 -
arch/arm/mach-ebsa110/core.c deleted +0 -323
323 lines not on this page · raw message
arch/arm/mach-ebsa110/core.h deleted +0 -38
@@ -1,38 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* -* Copyright (C) 1996-2000 Russell King. -* -* This file contains the core hardware definitions of the EBSA-110. -*/ -#ifndef CORE_H -#define CORE_H - -/* Physical addresses/sizes */ -#define ISAMEM_PHYS 0xe0000000 -#define ISAMEM_SIZE 0x10000000 - -#define ISAIO_PHYS 0xf0000000 -#define ISAIO_SIZE PGDIR_SIZE - -#define TRICK0_PHYS 0xf2000000 -#define TRICK0_SIZE PGDIR_SIZE -#define TRICK1_PHYS 0xf2400000 -#define TRICK1_SIZE PGDIR_SIZE -#define TRICK2_PHYS 0xf2800000 -#define TRICK3_PHYS 0xf2c00000 -#define TRICK3_SIZE PGDIR_SIZE -#define TRICK4_PHYS 0xf3000000 -#define TRICK4_SIZE PGDIR_SIZE -#define TRICK5_PHYS 0xf3400000 -#define TRICK6_PHYS 0xf3800000 -#define TRICK7_PHYS 0xf3c00000 - -/* Virtual addresses */ -#define PIT_BASE IOMEM(0xfc000000)/* trick 0 */ -#define SOFT_BASE IOMEM(0xfd000000)/* trick 1 */ -#define IRQ_MASK IOMEM(0xfe000000)/* trick 3 - read */ -#define IRQ_MSET IOMEM(0xfe000000)/* trick 3 - write */ -#define IRQ_STAT IOMEM(0xff000000)/* trick 4 - read */ -#define IRQ_MCLR IOMEM(0xff000000)/* trick 4 - write */ - -#endif
arch/arm/mach-ebsa110/include/mach/entry-macro.S deleted +0 -33
@@ -1,33 +0,0 @@ -/* - * arch/arm/mach-ebsa110/include/mach/entry-macro.S - * - * Low-level IRQ helper macros for ebsa110 platform. - * - * This file is licensed under the terms of the GNU General Public - * License version 2. This program is licensed "as is" without any - * warranty of any kind, whether express or implied. - */ - - - -#define IRQ_STAT 0xff000000 /* read */ - - .macro get_irqnr_preamble, base, tmp - mov \base, #IRQ_STAT - .endm - - .macro get_irqnr_and_base, irqnr, stat, base, tmp - ldrb \stat, [\base] @ get interrupts - mov \irqnr, #0 - tst \stat, #15 - addeq \irqnr, \irqnr, #4 - moveq \stat, \stat, lsr #4 - tst \stat, #3 - addeq \irqnr, \irqnr, #2 - moveq \stat, \stat, lsr #2 - tst \stat, #1 - addeq \irqnr, \irqnr, #1 - moveq \stat, \stat, lsr #1 - tst \stat, #1 @ bit 0 should be set - .endm -
arch/arm/mach-ebsa110/include/mach/hardware.h deleted +0 -21
@@ -1,21 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* -* arch/arm/mach-ebsa110/include/mach/hardware.h -* -* Copyright (C) 1996-2000 Russell King. -* -* This file contains the hardware definitions of the EBSA-110. -*/ -#ifndef __ASM_ARCH_HARDWARE_H -#define __ASM_ARCH_HARDWARE_H - -#define ISAMEM_BASE 0xe0000000 -#define ISAIO_BASE 0xf0000000 - -/* -* RAM definitions -*/ -#define UNCACHEABLE_ADDR 0xff000000/* IRQ_STAT */ - -#endif -
arch/arm/mach-ebsa110/include/mach/io.h deleted +0 -89
89 lines not on this page · raw message
arch/arm/mach-ebsa110/include/mach/irqs.h deleted +0 -17
@@ -1,17 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* -* arch/arm/mach-ebsa110/include/mach/irqs.h -* -* Copyright (C) 1996 Russell King -*/ - -#define NR_IRQS 8 - -#define IRQ_EBSA110_PRINTER 0 -#define IRQ_EBSA110_COM1 1 -#define IRQ_EBSA110_COM2 2 -#define IRQ_EBSA110_ETHERNET 3 -#define IRQ_EBSA110_TIMER0 4 -#define IRQ_EBSA110_TIMER1 5 -#define IRQ_EBSA110_PCMCIA 6 -#define IRQ_EBSA110_IMMEDIATE 7
arch/arm/mach-ebsa110/include/mach/memory.h deleted +0 -22
@@ -1,22 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* -* arch/arm/mach-ebsa110/include/mach/memory.h -* -* Copyright (C) 1996-1999 Russell King. -* -* Changelog: -* 20-Oct-1996 RMK Created -* 31-Dec-1997 RMK Fixed definitions to reduce warnings -* 21-Mar-1999 RMK Renamed to memory.h -* RMK Moved TASK_SIZE and PAGE_OFFSET here -*/ -#ifndef __ASM_ARCH_MEMORY_H -#define __ASM_ARCH_MEMORY_H - -/* -* Cache flushing area - SRAM -*/ -#define FLUSH_BASE_PHYS 0x40000000 -#define FLUSH_BASE 0xdf000000 - -#endif
arch/arm/mach-ebsa110/include/mach/uncompress.h deleted +0 -41
41 lines not on this page · raw message
arch/arm/mach-ebsa110/io.c deleted +0 -440
440 lines not on this page · raw message
arch/arm/mach-ebsa110/leds.c deleted +0 -71
71 lines not on this page · raw message
Signature · 1 line
2.27.0
PATCH 2/3 net: remove am79c961a driver
In reply to Arnd BergmannThis driver was only used on the EBSA110 platform, which is now getting removed, so the driver is no longer needed either.
--- drivers/net/ethernet/amd/Kconfig | 10 +- drivers/net/ethernet/amd/Makefile | 1 - drivers/net/ethernet/amd/am79c961a.c | 763 --------------------------- drivers/net/ethernet/amd/am79c961a.h | 143 ----- 4 files changed, 1 insertion(+), 916 deletions(-) delete mode 100644 drivers/net/ethernet/amd/am79c961a.c delete mode 100644 drivers/net/ethernet/amd/am79c961a.h
drivers/net/ethernet/amd/Kconfig +1 -9
@@ -8,7 +8,7 @@ config NET_VENDOR_AMD default y depends on DIO || MACH_DECSTATION || MVME147 || ATARI || SUN3 || \ SUN3X || SBUS || PCI || ZORRO || (ISA && ISA_DMA_API) || \ - (ARM && ARCH_EBSA110) || ISA || EISA || PCMCIA || ARM64 + ISA || EISA || PCMCIA || ARM64 help If you have a network (Ethernet) chipset belonging to this class, say Y.
@@ -75,14 +75,6 @@ config ARIADNE To compile this driver as a module, choose M here: the module will be called ariadne. -config ARM_AM79C961A - bool "ARM EBSA110 AM79C961A support" - depends on ARM && ARCH_EBSA110 - select CRC32 - help - If you wish to compile a kernel for the EBSA-110, then you should - always answer Y to this. - config ATARILANCE tristate "Atari LANCE support" depends on ATARI
drivers/net/ethernet/amd/Makefile +0 -1
@@ -5,7 +5,6 @@ obj-$(CONFIG_A2065) += a2065.o obj-$(CONFIG_AMD8111_ETH) += amd8111e.o -obj-$(CONFIG_ARM_AM79C961A) += am79c961a.o obj-$(CONFIG_ARIADNE) += ariadne.o obj-$(CONFIG_ATARILANCE) += atarilance.o obj-$(CONFIG_DECLANCE) += declance.o
drivers/net/ethernet/amd/am79c961a.c deleted +0 -763
763 lines not on this page · raw message
drivers/net/ethernet/amd/am79c961a.h deleted +0 -143
143 lines not on this page · raw message
Signature · 1 line
2.27.0
Re: [PATCH 2/3] net: remove am79c961a driver
Quoted 4 lines · On Thu, 8 Oct 2020 17:46:00 +0200 Arnd Bergmann wrote:
This driver was only used on the EBSA110 platform, which is now getting removed, so the driver is no longer needed either.
Signed-off-by Arnd Bergmann <[email protected]>
FWIW
PATCH 3/3 timekeeping: remove arch_gettimeoffset
In reply to Arnd BergmannWith Arm EBSA110 gone, nothing uses it any more, so the corresponding code and the Kconfig option can be removed.
--- .../time/modern-timekeeping/arch-support.txt | 33 ------------------- drivers/Makefile | 2 -- drivers/clocksource/Kconfig | 2 +- include/linux/time.h | 13 -------- kernel/time/Kconfig | 9 ----- kernel/time/clocksource.c | 8 ----- kernel/time/timekeeping.c | 25 +------------- kernel/trace/Kconfig | 2 -- 8 files changed, 2 insertions(+), 92 deletions(-) delete mode 100644 Documentation/features/time/modern-timekeeping/arch-support.txt
Documentation/features/time/modern-timekeeping/arch-support.txt deleted +0 -33
@@ -1,33 +0,0 @@ -# -# Feature name: modern-timekeeping -# Kconfig: !ARCH_USES_GETTIMEOFFSET -# description: arch does not use arch_gettimeoffset() anymore -# - ----------------------- - | arch |status| - ----------------------- - | alpha: | ok | - | arc: | ok | - | arm: | TODO | - | arm64: | ok | - | c6x: | ok | - | csky: | ok | - | h8300: | ok | - | hexagon: | ok | - | ia64: | ok | - | m68k: | ok | - | microblaze: | ok | - | mips: | ok | - | nds32: | ok | - | nios2: | ok | - | openrisc: | ok | - | parisc: | ok | - | powerpc: | ok | - | riscv: | ok | - | s390: | ok | - | sh: | ok | - | sparc: | ok | - | um: | ok | - | x86: | ok | - | xtensa: | ok | - -----------------------
drivers/Makefile +0 -2
@@ -135,9 +135,7 @@ obj-$(CONFIG_INFINIBAND) += infiniband/ obj-y += firmware/ obj-$(CONFIG_CRYPTO) += crypto/ obj-$(CONFIG_SUPERH) += sh/ -ifndef CONFIG_ARCH_USES_GETTIMEOFFSET obj-y += clocksource/ -endif obj-$(CONFIG_DCA) += dca/ obj-$(CONFIG_HID) += hid/ obj-$(CONFIG_PPC_PS3) += ps3/
drivers/clocksource/Kconfig +1 -1
@@ -563,7 +563,7 @@ config CLKSRC_QCOM config CLKSRC_VERSATILE bool "ARM Versatile (Express) reference platforms clock source" if COMPILE_TEST - depends on GENERIC_SCHED_CLOCK && !ARCH_USES_GETTIMEOFFSET + depends on GENERIC_SCHED_CLOCK select TIMER_OF default y if (ARCH_VEXPRESS || ARCH_VERSATILE) && ARM help
include/linux/time.h +0 -13
@@ -21,19 +21,6 @@ extern time64_t mktime64(const unsigned int year, const unsigned int mon, const unsigned int day, const unsigned int hour, const unsigned int min, const unsigned int sec); -/* Some architectures do not supply their own clocksource. -* This is mainly the case in architectures that get their -* inter-tick times by reading the counter on their interval -* timer. Since these timers wrap every tick, they're not really -* useful as clocksources. Wrapping them to act like one is possible -* but not very efficient. So we provide a callout these arches -* can implement for use with the jiffies clocksource to provide -* finer then tick granular time. -*/ -#ifdef CONFIG_ARCH_USES_GETTIMEOFFSET -extern u32 (*arch_gettimeoffset)(void ); -#endif -#ifdef CONFIG_POSIX_TIMERS extern void clear_itimer (void );#else
kernel/time/Kconfig +0 -9
@@ -26,10 +26,6 @@ config CLOCKSOURCE_VALIDATE_LAST_CYCLE config GENERIC_TIME_VSYSCALL bool -# Old style timekeeping -config ARCH_USES_GETTIMEOFFSET - bool - # The generic clock events infrastructure config GENERIC_CLOCKEVENTS bool
@@ -72,7 +68,6 @@ config TICK_ONESHOT config NO_HZ_COMMON bool - depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS select TICK_ONESHOT choice
@@ -87,7 +82,6 @@ config HZ_PERIODIC config NO_HZ_IDLE bool "Idle dynticks system (tickless idle)" - depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS select NO_HZ_COMMON help This option enables a tickless idle system: timer interrupts
@@ -99,7 +93,6 @@ config NO_HZ_IDLE config NO_HZ_FULL bool "Full dynticks system (tickless)" # NO_HZ_COMMON dependency - depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS # We need at least one periodic CPU for timekeeping depends on SMP depends on HAVE_CONTEXT_TRACKING
@@ -158,7 +151,6 @@ config CONTEXT_TRACKING_FORCE config NO_HZ bool "Old Idle dynticks config" - depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS help This is the old config entry that enables dynticks idle. We keep it around for a little while to enforce backward
@@ -166,7 +158,6 @@ config NO_HZ config HIGH_RES_TIMERS bool "High Resolution Timer Support" - depends on !ARCH_USES_GETTIMEOFFSET && GENERIC_CLOCKEVENTS select TICK_ONESHOT help This option enables high resolution timer support. If your
kernel/time/clocksource.c +0 -8
@@ -705,8 +705,6 @@ static inline void clocksource_update_max_deferment(struct clocksource *cs) &cs->max_cycles); } -#ifndef CONFIG_ARCH_USES_GETTIMEOFFSET -static struct clocksource *clocksource_find_best (bool oneshot,bool skipcur) {struct clocksource *cs;
@@ -798,12 +796,6 @@ static void clocksource_select_fallback(void)__clocksource_select (true); } -#else /* !CONFIG_ARCH_USES_GETTIMEOFFSET */ -static inline void clocksource_select (void ) { } -static inline void clocksource_select_fallback (void ) { } - -#endif -/* * clocksource_done_booting - Called near the end of core bootup *
kernel/time/timekeeping.c +1 -24
@@ -352,13 +352,6 @@ static void tk_setup_internals(struct timekeeper *tk, struct clocksource *clock)/* Timekeeper helper functions. */ -#ifdef CONFIG_ARCH_USES_GETTIMEOFFSET -static u32 default_arch_gettimeoffset (void ) {return 0 ; } -u32 (*arch_gettimeoffset)(void) = default_arch_gettimeoffset; -#else -static inline u32 arch_gettimeoffset (void ) {return 0 ; } -#endif -static inline u64 timekeeping_delta_to_ns (const struct tk_read_base *tkr,u64 delta) {u64 nsec;
@@ -366,8 +359,7 @@ static inline u64 timekeeping_delta_to_ns(const struct tk_read_base *tkr, u64 de nsec = delta * tkr->mult + tkr->xtime_nsec; nsec >>= tkr->shift; -/* If arch requires, add in get_arch_timeoffset() */ -return nsec +arch_gettimeoffset (); +return nsec; }static inline u64 timekeeping_get_ns (const struct tk_read_base *tkr)
@@ -707,16 +699,8 @@ static void timekeeping_forward_now(struct timekeeper *tk) tk->tkr_raw.cycle_last = cycle_now; tk->tkr_mono.xtime_nsec += delta * tk->tkr_mono.mult; - -/* If arch requires, add in get_arch_timeoffset() */ - tk->tkr_mono.xtime_nsec += (u64 )arch_gettimeoffset () << tk->tkr_mono.shift; - - tk->tkr_raw.xtime_nsec += delta * tk->tkr_raw.mult; -/* If arch requires, add in get_arch_timeoffset() */ - tk->tkr_raw.xtime_nsec += (u64 )arch_gettimeoffset () << tk->tkr_raw.shift; -tk_normalize_xtime (tk); }
@@ -2062,19 +2046,12 @@ static void timekeeping_advance(enum timekeeping_adv_mode mode)if (unlikely (timekeeping_suspended)) goto out; -#ifdef CONFIG_ARCH_USES_GETTIMEOFFSET - offset = real_tk->cycle_interval; - -if (mode !=TK_ADV_TICK ) - goto out; -#else offset =clocksource_delta (tk_clock_read (&tk->tkr_mono), tk->tkr_mono.cycle_last, tk->tkr_mono.mask);/* Check if there's really nothing to do */ if (offset < real_tk->cycle_interval && mode ==TK_ADV_TICK ) goto out; -#endif /* Do some additional sanity checking */ timekeeping_check_update (tk, offset);
kernel/trace/Kconfig +0 -2
@@ -253,7 +253,6 @@ config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n depends on TRACE_IRQFLAGS_SUPPORT - depends on !ARCH_USES_GETTIMEOFFSET select TRACE_IRQFLAGS select GENERIC_TRACER select TRACER_MAX_TRACE
@@ -277,7 +276,6 @@ config IRQSOFF_TRACER config PREEMPT_TRACER bool "Preemption-off Latency Tracer" default n - depends on !ARCH_USES_GETTIMEOFFSET depends on PREEMPTION select GENERIC_TRACER select TRACER_MAX_TRACE
Signature · 1 line
2.27.0
Re: [PATCH 3/3] timekeeping: remove arch_gettimeoffset
Quoted 2 lines · Arnd Bergmann
With Arm EBSA110 gone, nothing uses it any more, so the corresponding code and the
Kconfigoption can be removed.
Yay! It only took 11+ years to get rid of that.
Feel free to route it through your tree.
Re: [PATCH 3/3] timekeeping: remove arch_gettimeoffset
In reply to Arnd BergmannQuoted 4 lines · Arnd Bergmann
With Arm EBSA110 gone, nothing uses it any more, so the corresponding code and the
Kconfigoption can be removed.Signed-off-by Arnd Bergmann <[email protected]>
Very nice cleanup.
At some point we should do a retrospect about how hard it really is for us to get rid of some really annoying core legacy codepaths and the best strategy for doing so, even though this took 11+ years as Thomas says it still looks like a solid path to move over to a generic framework, only it requires enough gritty people (like you) to hang around and do the work piece by piece.
Yours,
Linus Walleij
Russell King - ARM Linux admin
maintainerRe: [PATCH 3/3] timekeeping: remove arch_gettimeoffset
Quoted 8 lines · Linus Walleij
Quoted 4 lines · Arnd Bergmann
With Arm EBSA110 gone, nothing uses it any more, so the corresponding code and the
Kconfigoption can be removed.Signed-off-by Arnd Bergmann <[email protected]>Very nice cleanup.
Quoted 8 lines
Reviewed-by Linus Walleij <[email protected]>At some point we should do a retrospect about how hard it really is for us to get rid of some really annoying core legacy codepaths and the best strategy for doing so, even though this took 11+ years as Thomas says it still looks like a solid path to move over to a generic framework, only it requires enough gritty people (like you) to hang around and do the work piece by piece.
Don't be misled. It was not a matter of "enough gritty people", it was a matter that EBSA110 was blocking it.
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
Re: [PATCH 3/3] timekeeping: remove arch_gettimeoffset
Quoted 2 lines · Russell King - ARM Linux admin
Don't be misled. It was not a matter of "enough gritty people", it was a matter that EBSA110 was blocking it.
I remember that EBSA110 was quite different in that it had this especially limited PIT timer, true that. At one point I even read up on the details of it and then forgot them :/
Yours,
Linus Walleij
Russell King - ARM Linux admin
maintainerRe: [PATCH 3/3] timekeeping: remove arch_gettimeoffset
Quoted 9 lines · Linus Walleij
Quoted 2 lines · Russell King - ARM Linux admin
Don't be misled. It was not a matter of "enough gritty people", it was a matter that EBSA110 was blocking it.
I remember that EBSA110 was quite different in that it had this especially limited PIT timer, true that. At one point I even read up on the details of it and then forgot them :/
Yes, it was so basic that it required software to reprogram it on every interrupt - it had no sensible periodic mode. The side effect of which was that unless done carefully, your timekeeping varied all over the place (so much that ntpd was not happy.) The final implementation made ntpd really quite happy on the machine despite the timer's weirdness, up until Linux stopped allowing interrupts to be nested. At that point, it was back to utterly crap timekeeping.
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
Quoted 9 lines · Arnd Bergmann
The ebsa110 platform is the last thing that uses
CONFIG_ARCH_USES_GETTIMEOFFSET, and Russell has previously said that he thinks the platform can be retired now.Removing it allows us clean up the timer code by throwing out all of the references to
arch_gettimeoffset().The am79c961a network driver can presumably also go, as no other platform references it.
As far as I know, that is correct; it's rather specific to the platform.
Quoted 2 lines
I don't expect these to make it into the coming merge window, posting here as an RFC, and as a reference for the mildly related timer tick series.
For the ebsa110 and am79c961a patches:
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!