OpenCloudOS-Kernel/arch/arm/mach-spear/hotplug.c

101 lines
1.9 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* linux/arch/arm/mach-spear13xx/hotplug.c
*
* Copyright (C) 2012 ST Microelectronics Ltd.
* Deepak Sikri <deepak.sikri@st.com>
*
* based upon linux/arch/arm/mach-realview/hotplug.c
*/
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/smp.h>
#include <asm/cp15.h>
#include <asm/smp_plat.h>
ARM: smp: remove arch-provided "pen_release" Consolidating the "pen_release" stuff amongst the various SoC implementations gives credence to having a CPU holding pen for secondary CPUs. However, this is far from the truth. Many SoC implementations cargo-cult copied various bits of the pen release implementation from the initial Realview/Versatile Express implementation without understanding what it was or why it existed. The reason it existed is because these are _development_ platforms, and some board firmware is unable to individually control the startup of secondary CPUs. Moreover, they do not have a way to power down or reset secondary CPUs for hot-unplug. Hence, the pen_release implementation was designed for ARM Ltd's development platforms to provide a working implementation, even though it is very far from what is required. It was decided a while back to reduce the duplication by consolidating the "pen_release" variable, but this only made the situation worse - we have ended up with several implementations that read this variable but do not write it - again, showing the cargo-cult mentality at work, lack of proper review of new code, and in some cases a lack of testing. While it would be preferable to remove pen_release entirely from the kernel, this is not possible without help from the SoC maintainers, which seems to be lacking. However, I want to remove pen_release from arch code to remove the credence that having it gives. This patch removes pen_release from the arch code entirely, adding private per-SoC definitions for it instead, and explicitly stating that write_pen_release() is cargo-cult copied and should not be copied any further. Rename write_pen_release() in a similar fashion as well. Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
2018-12-13 22:02:48 +08:00
#include "generic.h"
static inline void cpu_enter_lowpower(void)
{
unsigned int v;
asm volatile(
" mcr p15, 0, %1, c7, c5, 0\n"
" dsb\n"
/*
* Turn off coherency
*/
" mrc p15, 0, %0, c1, c0, 1\n"
" bic %0, %0, #0x20\n"
" mcr p15, 0, %0, c1, c0, 1\n"
" mrc p15, 0, %0, c1, c0, 0\n"
" bic %0, %0, %2\n"
" mcr p15, 0, %0, c1, c0, 0\n"
: "=&r" (v)
: "r" (0), "Ir" (CR_C)
: "cc", "memory");
}
static inline void cpu_leave_lowpower(void)
{
unsigned int v;
asm volatile("mrc p15, 0, %0, c1, c0, 0\n"
" orr %0, %0, %1\n"
" mcr p15, 0, %0, c1, c0, 0\n"
" mrc p15, 0, %0, c1, c0, 1\n"
" orr %0, %0, #0x20\n"
" mcr p15, 0, %0, c1, c0, 1\n"
: "=&r" (v)
: "Ir" (CR_C)
: "cc");
}
static inline void spear13xx_do_lowpower(unsigned int cpu, int *spurious)
{
for (;;) {
wfi();
ARM: smp: remove arch-provided "pen_release" Consolidating the "pen_release" stuff amongst the various SoC implementations gives credence to having a CPU holding pen for secondary CPUs. However, this is far from the truth. Many SoC implementations cargo-cult copied various bits of the pen release implementation from the initial Realview/Versatile Express implementation without understanding what it was or why it existed. The reason it existed is because these are _development_ platforms, and some board firmware is unable to individually control the startup of secondary CPUs. Moreover, they do not have a way to power down or reset secondary CPUs for hot-unplug. Hence, the pen_release implementation was designed for ARM Ltd's development platforms to provide a working implementation, even though it is very far from what is required. It was decided a while back to reduce the duplication by consolidating the "pen_release" variable, but this only made the situation worse - we have ended up with several implementations that read this variable but do not write it - again, showing the cargo-cult mentality at work, lack of proper review of new code, and in some cases a lack of testing. While it would be preferable to remove pen_release entirely from the kernel, this is not possible without help from the SoC maintainers, which seems to be lacking. However, I want to remove pen_release from arch code to remove the credence that having it gives. This patch removes pen_release from the arch code entirely, adding private per-SoC definitions for it instead, and explicitly stating that write_pen_release() is cargo-cult copied and should not be copied any further. Rename write_pen_release() in a similar fashion as well. Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
2018-12-13 22:02:48 +08:00
if (spear_pen_release == cpu) {
/*
* OK, proper wakeup, we're done
*/
break;
}
/*
* Getting here, means that we have come out of WFI without
* having been woken up - this shouldn't happen
*
* Just note it happening - when we're woken, we can report
* its occurrence.
*/
(*spurious)++;
}
}
/*
* platform-specific code to shutdown a CPU
*
* Called with IRQs disabled
*/
void spear13xx_cpu_die(unsigned int cpu)
{
int spurious = 0;
/*
* we're ready for shutdown now, so do it
*/
cpu_enter_lowpower();
spear13xx_do_lowpower(cpu, &spurious);
/*
* bring this CPU back into the world of cache
* coherency, and then restore interrupts
*/
cpu_leave_lowpower();
if (spurious)
pr_warn("CPU%u: %u spurious wakeup calls\n", cpu, spurious);
}