i8253: Fix PIT shutdown quirk on Hyper-V
v2 cover i8253: Fix PIT shutdown quirk on Hyper-V
pit_shutdown() doesn't work on Hyper-V because of a quirk in the PIT emulation. This problem exists in all versions of Hyper-V and had not been noticed previously. When the counter register is set to zero, the emulated PIT continues to interrupt @18.2 HZ.
Account for this quirk by adding a global variable in the i8253 code that controls whether the counter register is zero'ed. Then in Hyper-V initialization code, override the default setting so the counter register is not zero'ed.
Changes in v2:
* Instead of a function call to check if running on Hyper-V, use a global variable to control whether the counter register is zero'ed. [Juergen Gross & Thomas Gleixner]
Michael Kelley (2):
i8253: Add support for PIT shutdown quirk
x86/hyper-v: Enable PIT shutdown quirk
arch/x86/kernel/cpu/mshyperv.c | 11 +++++++++++ drivers/clocksource/i8253.c | 14 ++++++++++++-- include/linux/i8253.h | 1 + 3 files changed, 24 insertions(+), 2 deletions(-)
Signature · 1 line
1.8.3.1
PATCH v2 1/2 i8253: Add support for PIT shutdown quirk
Add support for platforms where pit_shutdown() doesn't work because of a quirk in the PIT emulation. On these platforms setting the counter register to zero causes the PIT to start running again, negating the shutdown. Provide a global variable that controls whether the counter register is zero'ed, which platform specific code can override.
--- drivers/clocksource/i8253.c | 14 ++++++++++++-- include/linux/i8253.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-)
drivers/clocksource/i8253.c +12 -2
@@ -20,6 +20,13 @@DEFINE_RAW_SPINLOCK (i8253_lock);EXPORT_SYMBOL (i8253_lock); +/* +* Handle PIT quirk in pit_shutdown() where zeroing the counter register +* restarts the PIT, negating the shutdown. On platforms with the quirk, +* platform specific code can set this to false. +*/ +bool i8253_clear_counter__ro_after_init = true; +#ifdef CONFIG_CLKSRC_I8253 /* * Since the PIT overflows every tick, its not very useful
@@ -109,8 +116,11 @@ static int pit_shutdown(struct clock_event_device *evt)raw_spin_lock (&i8253_lock);outb_p (0x30 ,PIT_MODE ); -outb_p (0 ,PIT_CH0 ); -outb_p (0 ,PIT_CH0 ); + +if (i8253_clear_counter) { +outb_p (0 ,PIT_CH0 ); +outb_p (0 ,PIT_CH0 ); + }raw_spin_unlock (&i8253_lock);return 0 ;
include/linux/i8253.h +1 -0
@@ -21,6 +21,7 @@#define PIT_LATCH ((PIT_TICK_RATE + HZ/2) / HZ)extern raw_spinlock_t i8253_lock; +extern bool i8253_clear_counter;extern struct clock_event_device i8253_clockevent;extern void clockevent_i8253_init (bool oneshot);
Signature · 1 line
1.8.3.1
tip-bot for Michael Kelley
merge bot[tip:x86/urgent] clockevents/drivers/i8253: Add support for PIT shutdown quirk
clockevents/drivers/i8253: Add support for PIT shutdown quirk
Commit-ID: 35b69a420bfb56b7b74cb635ea903db05e357bec Gitweb: https://git.kernel.org/tip/35b69a420bfb56b7b74cb635ea903db05e357bec Author: Michael Kelley <[email protected]> AuthorDate: Sun, 4 Nov 2018 03:48:54 +0000 Committer: Thomas Gleixner <[email protected]> CommitDate: Sun, 4 Nov 2018 11:04:46 +0100
clockevents/drivers/i8253: Add support for PIT shutdown quirk
Add support for platforms where pit_shutdown() doesn't work because of a quirk in the PIT emulation. On these platforms setting the counter register to zero causes the PIT to start running again, negating the shutdown.
Provide a global variable that controls whether the counter register is zero'ed, which platform specific code can override.
--- drivers/clocksource/i8253.c | 14 ++++++++++++-- include/linux/i8253.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-)
drivers/clocksource/i8253.c +12 -2
@@ -20,6 +20,13 @@DEFINE_RAW_SPINLOCK (i8253_lock);EXPORT_SYMBOL (i8253_lock); +/* +* Handle PIT quirk in pit_shutdown() where zeroing the counter register +* restarts the PIT, negating the shutdown. On platforms with the quirk, +* platform specific code can set this to false. +*/ +bool i8253_clear_counter_on_shutdown__ro_after_init = true; +#ifdef CONFIG_CLKSRC_I8253 /* * Since the PIT overflows every tick, its not very useful
@@ -109,8 +116,11 @@ static int pit_shutdown(struct clock_event_device *evt)raw_spin_lock (&i8253_lock);outb_p (0x30 ,PIT_MODE ); -outb_p (0 ,PIT_CH0 ); -outb_p (0 ,PIT_CH0 ); + +if (i8253_clear_counter_on_shutdown) { +outb_p (0 ,PIT_CH0 ); +outb_p (0 ,PIT_CH0 ); + }raw_spin_unlock (&i8253_lock);return 0 ;
include/linux/i8253.h +1 -0
@@ -21,6 +21,7 @@#define PIT_LATCH ((PIT_TICK_RATE + HZ/2) / HZ)extern raw_spinlock_t i8253_lock; +extern bool i8253_clear_counter_on_shutdown;extern struct clock_event_device i8253_clockevent;extern void clockevent_i8253_init (bool oneshot);
tip-bot for Michael Kelley
merge bot[tip:x86/urgent] clockevents/drivers/i8253: Add support for PIT shutdown quirk
In reply to Michael Kelleyclockevents/drivers/i8253: Add support for PIT shutdown quirk
Commit-ID: 35b69a420bfb56b7b74cb635ea903db05e357bec Gitweb: https://git.kernel.org/tip/35b69a420bfb56b7b74cb635ea903db05e357bec Author: Michael Kelley <[email protected]> AuthorDate: Sun, 4 Nov 2018 03:48:54 +0000 Committer: Thomas Gleixner <[email protected]> CommitDate: Sun, 4 Nov 2018 11:04:46 +0100
clockevents/drivers/i8253: Add support for PIT shutdown quirk
Add support for platforms where pit_shutdown() doesn't work because of a quirk in the PIT emulation. On these platforms setting the counter register to zero causes the PIT to start running again, negating the shutdown.
Provide a global variable that controls whether the counter register is zero'ed, which platform specific code can override.
--- drivers/clocksource/i8253.c | 14 ++++++++++++-- include/linux/i8253.h | 1 + 2 files changed, 13 insertions(+), 2 deletions(-)
drivers/clocksource/i8253.c +12 -2
@@ -20,6 +20,13 @@DEFINE_RAW_SPINLOCK (i8253_lock);EXPORT_SYMBOL (i8253_lock); +/* +* Handle PIT quirk in pit_shutdown() where zeroing the counter register +* restarts the PIT, negating the shutdown. On platforms with the quirk, +* platform specific code can set this to false. +*/ +bool i8253_clear_counter_on_shutdown__ro_after_init = true; +#ifdef CONFIG_CLKSRC_I8253 /* * Since the PIT overflows every tick, its not very useful
@@ -109,8 +116,11 @@ static int pit_shutdown(struct clock_event_device *evt)raw_spin_lock (&i8253_lock);outb_p (0x30 ,PIT_MODE ); -outb_p (0 ,PIT_CH0 ); -outb_p (0 ,PIT_CH0 ); + +if (i8253_clear_counter_on_shutdown) { +outb_p (0 ,PIT_CH0 ); +outb_p (0 ,PIT_CH0 ); + }raw_spin_unlock (&i8253_lock);return 0 ;
include/linux/i8253.h +1 -0
@@ -21,6 +21,7 @@#define PIT_LATCH ((PIT_TICK_RATE + HZ/2) / HZ)extern raw_spinlock_t i8253_lock; +extern bool i8253_clear_counter_on_shutdown;extern struct clock_event_device i8253_clockevent;extern void clockevent_i8253_init (bool oneshot);
PATCH v2 2/2 x86/hyper-v: Enable PIT shutdown quirk
In reply to Michael KelleyHyper-V emulation of the PIT has a quirk such that the normal PIT shutdown path doesn't work. Enable the PIT code that handles this quirk.
--- arch/x86/kernel/cpu/mshyperv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
arch/x86/kernel/cpu/mshyperv.c +11 -0
@@ -20,6 +20,7 @@#include <linux/interrupt.h> #include <linux/irq.h> #include <linux/kexec.h> +#include <linux/i8253.h> #include <asm/processor.h> #include <asm/hypervisor.h> #include <asm/hyperv-tlfs.h>
@@ -295,6 +296,16 @@ static void __init ms_hyperv_init_platform(void)if (efi_enabled (EFI_BOOT )) x86_platform.get_nmi_reason = hv_get_nmi_reason; +/* +* Hyper-V VMs have a PIT emulation quirk such that zeroing the +* counter register during PIT shutdown restarts the PIT. So it +* continues to interrupt @18.2 HZ. Setting i8253_clear_counter +* to false tells pit_shutdown() not to zero the counter so that +* the PIT really is shutdown. Generation 2 VMs don't have a PIT, +* and setting this value has no effect. +*/ + i8253_clear_counter = false; +#if IS_ENABLED (CONFIG_HYPERV )/* * Setup the hook to get control post apic initialization.
Signature · 1 line
1.8.3.1
tip-bot for Michael Kelley
merge bot[tip:x86/urgent] x86/hyper-v: Enable PIT shutdown quirk
x86/hyper-v: Enable PIT shutdown quirk
Commit-ID: 1de72c706488b7be664a601cf3843bd01e327e58 Gitweb: https://git.kernel.org/tip/1de72c706488b7be664a601cf3843bd01e327e58 Author: Michael Kelley <[email protected]> AuthorDate: Sun, 4 Nov 2018 03:48:57 +0000 Committer: Thomas Gleixner <[email protected]> CommitDate: Sun, 4 Nov 2018 11:04:46 +0100
x86/hyper-v: Enable PIT shutdown quirk
Hyper-V emulation of the PIT has a quirk such that the normal PIT shutdown path doesn't work, because clearing the counter register restarts the timer.
Disable the counter clearing on PIT shutdown.
--- arch/x86/kernel/cpu/mshyperv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
arch/x86/kernel/cpu/mshyperv.c +11 -0
@@ -20,6 +20,7 @@#include <linux/interrupt.h> #include <linux/irq.h> #include <linux/kexec.h> +#include <linux/i8253.h> #include <asm/processor.h> #include <asm/hypervisor.h> #include <asm/hyperv-tlfs.h>
@@ -295,6 +296,16 @@ static void __init ms_hyperv_init_platform(void)if (efi_enabled (EFI_BOOT )) x86_platform.get_nmi_reason = hv_get_nmi_reason; +/* +* Hyper-V VMs have a PIT emulation quirk such that zeroing the +* counter register during PIT shutdown restarts the PIT. So it +* continues to interrupt @18.2 HZ. Setting i8253_clear_counter +* to false tells pit_shutdown() not to zero the counter so that +* the PIT really is shutdown. Generation 2 VMs don't have a PIT, +* and setting this value has no effect. +*/ + i8253_clear_counter_on_shutdown = false; +#if IS_ENABLED (CONFIG_HYPERV )/* * Setup the hook to get control post apic initialization.
tip-bot for Michael Kelley
merge bot[tip:x86/urgent] x86/hyper-v: Enable PIT shutdown quirk
In reply to Michael Kelleyx86/hyper-v: Enable PIT shutdown quirk
Commit-ID: 1de72c706488b7be664a601cf3843bd01e327e58 Gitweb: https://git.kernel.org/tip/1de72c706488b7be664a601cf3843bd01e327e58 Author: Michael Kelley <[email protected]> AuthorDate: Sun, 4 Nov 2018 03:48:57 +0000 Committer: Thomas Gleixner <[email protected]> CommitDate: Sun, 4 Nov 2018 11:04:46 +0100
x86/hyper-v: Enable PIT shutdown quirk
Hyper-V emulation of the PIT has a quirk such that the normal PIT shutdown path doesn't work, because clearing the counter register restarts the timer.
Disable the counter clearing on PIT shutdown.
--- arch/x86/kernel/cpu/mshyperv.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
arch/x86/kernel/cpu/mshyperv.c +11 -0
@@ -20,6 +20,7 @@#include <linux/interrupt.h> #include <linux/irq.h> #include <linux/kexec.h> +#include <linux/i8253.h> #include <asm/processor.h> #include <asm/hypervisor.h> #include <asm/hyperv-tlfs.h>
@@ -295,6 +296,16 @@ static void __init ms_hyperv_init_platform(void)if (efi_enabled (EFI_BOOT )) x86_platform.get_nmi_reason = hv_get_nmi_reason; +/* +* Hyper-V VMs have a PIT emulation quirk such that zeroing the +* counter register during PIT shutdown restarts the PIT. So it +* continues to interrupt @18.2 HZ. Setting i8253_clear_counter +* to false tells pit_shutdown() not to zero the counter so that +* the PIT really is shutdown. Generation 2 VMs don't have a PIT, +* and setting this value has no effect. +*/ + i8253_clear_counter_on_shutdown = false; +#if IS_ENABLED (CONFIG_HYPERV )/* * Setup the hook to get control post apic initialization.