* [PATCH v5 1/2] kexec: Consolidate machine_kexec_mask_interrupts() implementation
2024-11-30 20:11 [PATCH v5 0/2] Improve interrupt handling during machine kexec Eliav Farber
@ 2024-11-30 20:11 ` Eliav Farber
2024-12-03 11:04 ` Thomas Gleixner
2024-12-04 11:02 ` Jiri Slaby
2024-11-30 20:11 ` [PATCH v5 2/2] kexec: Prevent redundant IRQ masking by checking state before shutdown Eliav Farber
1 sibling, 2 replies; 5+ messages in thread
From: Eliav Farber @ 2024-11-30 20:11 UTC (permalink / raw)
To: linux, catalin.marinas, will, mpe, npiggin, christophe.leroy,
naveen, maddy, paul.walmsley, palmer, aou, tglx, akpm, bhe,
farbere, hbathini, sourabhjain, adityag, songshuaishuai, takakura,
linux-arm-kernel, linux-kernel, linuxppc-dev, linux-riscv
Cc: jonnyc
Consolidate the machine_kexec_mask_interrupts implementation into a common
function located in a new file: kernel/irq/kexec.c. This removes duplicate
implementations from architecture-specific files in arch/arm, arch/arm64,
arch/powerpc, and arch/riscv, reducing code duplication and improving
maintainability.
The new implementation retains architecture-specific behavior for
CONFIG_GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD, which was previously implemented
for ARM64. When enabled (currently for ARM64), it clears the active state
of interrupts forwarded to virtual machines (VMs) before handling other
interrupt masking operations.
Signed-off-by: Eliav Farber <farbere@amazon.com>
---
V4 -> V5:
- The function machine_kexec_mask_interrupts() has been moved
from kernel/kexec_core.c to a new file kernel/irq/kexec.c.
- A new configuration option, GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD, has been
added.
- The parameters for the machine_kexec_mask_interrupts() function have
been defined in reverse Christmas Tree style.
- The comment explaining the call to irq_set_irqchip_state() has been
improved for clarity.
arch/arm/kernel/machine_kexec.c | 23 ------------------
arch/arm64/Kconfig | 1 +
arch/arm64/kernel/machine_kexec.c | 31 ------------------------
arch/powerpc/include/asm/kexec.h | 1 -
arch/powerpc/kexec/core.c | 22 -----------------
arch/powerpc/kexec/core_32.c | 1 +
arch/riscv/kernel/machine_kexec.c | 23 ------------------
include/linux/irq.h | 3 +++
kernel/irq/Kconfig | 9 +++++++
kernel/irq/Makefile | 2 +-
kernel/irq/kexec.c | 40 +++++++++++++++++++++++++++++++
11 files changed, 55 insertions(+), 101 deletions(-)
create mode 100644 kernel/irq/kexec.c
diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c
index 80ceb5bd2680..dd430477e7c1 100644
--- a/arch/arm/kernel/machine_kexec.c
+++ b/arch/arm/kernel/machine_kexec.c
@@ -127,29 +127,6 @@ void crash_smp_send_stop(void)
cpus_stopped = 1;
}
-static void machine_kexec_mask_interrupts(void)
-{
- unsigned int i;
- struct irq_desc *desc;
-
- for_each_irq_desc(i, desc) {
- struct irq_chip *chip;
-
- chip = irq_desc_get_chip(desc);
- if (!chip)
- continue;
-
- if (chip->irq_eoi && irqd_irq_inprogress(&desc->irq_data))
- chip->irq_eoi(&desc->irq_data);
-
- if (chip->irq_mask)
- chip->irq_mask(&desc->irq_data);
-
- if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
- chip->irq_disable(&desc->irq_data);
- }
-}
-
void machine_crash_shutdown(struct pt_regs *regs)
{
local_irq_disable();
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index d743737bf9ce..359b9dcb35b9 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -150,6 +150,7 @@ config ARM64
select GENERIC_IDLE_POLL_SETUP
select GENERIC_IOREMAP
select GENERIC_IRQ_IPI
+ select GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD
select GENERIC_IRQ_PROBE
select GENERIC_IRQ_SHOW
select GENERIC_IRQ_SHOW_LEVEL
diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c
index 82e2203d86a3..6f121a0164a4 100644
--- a/arch/arm64/kernel/machine_kexec.c
+++ b/arch/arm64/kernel/machine_kexec.c
@@ -207,37 +207,6 @@ void machine_kexec(struct kimage *kimage)
BUG(); /* Should never get here. */
}
-static void machine_kexec_mask_interrupts(void)
-{
- unsigned int i;
- struct irq_desc *desc;
-
- for_each_irq_desc(i, desc) {
- struct irq_chip *chip;
- int ret;
-
- chip = irq_desc_get_chip(desc);
- if (!chip)
- continue;
-
- /*
- * First try to remove the active state. If this
- * fails, try to EOI the interrupt.
- */
- ret = irq_set_irqchip_state(i, IRQCHIP_STATE_ACTIVE, false);
-
- if (ret && irqd_irq_inprogress(&desc->irq_data) &&
- chip->irq_eoi)
- chip->irq_eoi(&desc->irq_data);
-
- if (chip->irq_mask)
- chip->irq_mask(&desc->irq_data);
-
- if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
- chip->irq_disable(&desc->irq_data);
- }
-}
-
/**
* machine_crash_shutdown - shutdown non-crashing cpus and save registers
*/
diff --git a/arch/powerpc/include/asm/kexec.h b/arch/powerpc/include/asm/kexec.h
index 270ee93a0f7d..601e569303e1 100644
--- a/arch/powerpc/include/asm/kexec.h
+++ b/arch/powerpc/include/asm/kexec.h
@@ -61,7 +61,6 @@ struct pt_regs;
extern void kexec_smp_wait(void); /* get and clear naca physid, wait for
master to copy new code to 0 */
extern void default_machine_kexec(struct kimage *image);
-extern void machine_kexec_mask_interrupts(void);
void relocate_new_kernel(unsigned long indirection_page, unsigned long reboot_code_buffer,
unsigned long start_address) __noreturn;
diff --git a/arch/powerpc/kexec/core.c b/arch/powerpc/kexec/core.c
index b8333a49ea5d..58a930a47422 100644
--- a/arch/powerpc/kexec/core.c
+++ b/arch/powerpc/kexec/core.c
@@ -22,28 +22,6 @@
#include <asm/setup.h>
#include <asm/firmware.h>
-void machine_kexec_mask_interrupts(void) {
- unsigned int i;
- struct irq_desc *desc;
-
- for_each_irq_desc(i, desc) {
- struct irq_chip *chip;
-
- chip = irq_desc_get_chip(desc);
- if (!chip)
- continue;
-
- if (chip->irq_eoi && irqd_irq_inprogress(&desc->irq_data))
- chip->irq_eoi(&desc->irq_data);
-
- if (chip->irq_mask)
- chip->irq_mask(&desc->irq_data);
-
- if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
- chip->irq_disable(&desc->irq_data);
- }
-}
-
#ifdef CONFIG_CRASH_DUMP
void machine_crash_shutdown(struct pt_regs *regs)
{
diff --git a/arch/powerpc/kexec/core_32.c b/arch/powerpc/kexec/core_32.c
index c95f96850c9e..deb28eb44f30 100644
--- a/arch/powerpc/kexec/core_32.c
+++ b/arch/powerpc/kexec/core_32.c
@@ -7,6 +7,7 @@
* Copyright (C) 2005 IBM Corporation.
*/
+#include <linux/irq.h>
#include <linux/kexec.h>
#include <linux/mm.h>
#include <linux/string.h>
diff --git a/arch/riscv/kernel/machine_kexec.c b/arch/riscv/kernel/machine_kexec.c
index 3c830a6f7ef4..2306ce3e5f22 100644
--- a/arch/riscv/kernel/machine_kexec.c
+++ b/arch/riscv/kernel/machine_kexec.c
@@ -114,29 +114,6 @@ void machine_shutdown(void)
#endif
}
-static void machine_kexec_mask_interrupts(void)
-{
- unsigned int i;
- struct irq_desc *desc;
-
- for_each_irq_desc(i, desc) {
- struct irq_chip *chip;
-
- chip = irq_desc_get_chip(desc);
- if (!chip)
- continue;
-
- if (chip->irq_eoi && irqd_irq_inprogress(&desc->irq_data))
- chip->irq_eoi(&desc->irq_data);
-
- if (chip->irq_mask)
- chip->irq_mask(&desc->irq_data);
-
- if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
- chip->irq_disable(&desc->irq_data);
- }
-}
-
/*
* machine_crash_shutdown - Prepare to kexec after a kernel crash
*
diff --git a/include/linux/irq.h b/include/linux/irq.h
index fa711f80957b..25f51bf3c351 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -694,6 +694,9 @@ extern int irq_chip_request_resources_parent(struct irq_data *data);
extern void irq_chip_release_resources_parent(struct irq_data *data);
#endif
+/* Disable or mask interrupts during a kernel kexec */
+extern void machine_kexec_mask_interrupts(void);
+
/* Handling of unhandled and spurious interrupts: */
extern void note_interrupt(struct irq_desc *desc, irqreturn_t action_ret);
diff --git a/kernel/irq/Kconfig b/kernel/irq/Kconfig
index 529adb1f5859..5967accb8e87 100644
--- a/kernel/irq/Kconfig
+++ b/kernel/irq/Kconfig
@@ -154,3 +154,12 @@ config DEPRECATED_IRQ_CPU_ONOFFLINE
bool
depends on CAVIUM_OCTEON_SOC
default CAVIUM_OCTEON_SOC
+
+config GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD
+ bool "Clear forwarded VM interrupts during kexec"
+ default n
+ help
+ When enabled, this option allows the kernel to clear the active state
+ of interrupts that are forwarded to virtual machines (VMs) during a
+ machine kexec. For interrupts that are not forwarded, if supported,
+ the kernel will attempt to trigger an End of Interrupt (EOI).
diff --git a/kernel/irq/Makefile b/kernel/irq/Makefile
index f19d3080bf11..c0f44c06d69d 100644
--- a/kernel/irq/Makefile
+++ b/kernel/irq/Makefile
@@ -1,6 +1,6 @@
# SPDX-License-Identifier: GPL-2.0
-obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o
+obj-y := irqdesc.o handle.o manage.o spurious.o resend.o chip.o dummychip.o devres.o kexec.o
obj-$(CONFIG_IRQ_TIMINGS) += timings.o
ifeq ($(CONFIG_TEST_IRQ_TIMINGS),y)
CFLAGS_timings.o += -DDEBUG
diff --git a/kernel/irq/kexec.c b/kernel/irq/kexec.c
new file mode 100644
index 000000000000..0f9548c1708d
--- /dev/null
+++ b/kernel/irq/kexec.c
@@ -0,0 +1,40 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/irqdesc.h>
+#include <linux/irqnr.h>
+
+#include "internals.h"
+
+void machine_kexec_mask_interrupts(void)
+{
+ struct irq_desc *desc;
+ unsigned int i;
+
+ for_each_irq_desc(i, desc) {
+ struct irq_chip *chip;
+ int check_eoi = 1;
+
+ chip = irq_desc_get_chip(desc);
+ if (!chip)
+ continue;
+
+ if (IS_ENABLED(CONFIG_GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD)) {
+ /*
+ * First try to remove the active state from an interrupt which is forwarded
+ * to a VM. If the interrupt is not forwarded, try to EOI the interrupt.
+ */
+ check_eoi = irq_set_irqchip_state(i, IRQCHIP_STATE_ACTIVE, false);
+ }
+
+ if (check_eoi && chip->irq_eoi && irqd_irq_inprogress(&desc->irq_data))
+ chip->irq_eoi(&desc->irq_data);
+
+ if (chip->irq_mask)
+ chip->irq_mask(&desc->irq_data);
+
+ if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
+ chip->irq_disable(&desc->irq_data);
+ }
+}
--
2.40.1
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH v5 2/2] kexec: Prevent redundant IRQ masking by checking state before shutdown
2024-11-30 20:11 [PATCH v5 0/2] Improve interrupt handling during machine kexec Eliav Farber
2024-11-30 20:11 ` [PATCH v5 1/2] kexec: Consolidate machine_kexec_mask_interrupts() implementation Eliav Farber
@ 2024-11-30 20:11 ` Eliav Farber
1 sibling, 0 replies; 5+ messages in thread
From: Eliav Farber @ 2024-11-30 20:11 UTC (permalink / raw)
To: linux, catalin.marinas, will, mpe, npiggin, christophe.leroy,
naveen, maddy, paul.walmsley, palmer, aou, tglx, akpm, bhe,
farbere, hbathini, sourabhjain, adityag, songshuaishuai, takakura,
linux-arm-kernel, linux-kernel, linuxppc-dev, linux-riscv
Cc: jonnyc
During machine kexec, the function machine_kexec_mask_interrupts() is
responsible for disabling or masking all interrupts. While the irq_disable
hook ensures that an already-disabled IRQ is not disabled again, the
current implementation unconditionally invokes the irq_mask() function for
every interrupt descriptor, even when the interrupt is already masked.
A specific issue was observed in the crash kernel flow after unbinding a
device (prior to kexec) that used a GPIO as an IRQ source. The warning was
triggered by the gpiochip_disable_irq() function, which attempted to clear
the FLAG_IRQ_IS_ENABLED flag when FLAG_USED_AS_IRQ was not set:
```
void gpiochip_disable_irq(struct gpio_chip *gc, unsigned int offset)
{
struct gpio_desc *desc = gpiochip_get_desc(gc, offset);
if (!IS_ERR(desc) &&
!WARN_ON(!test_bit(FLAG_USED_AS_IRQ, &desc->flags)))
clear_bit(FLAG_IRQ_IS_ENABLED, &desc->flags);
}
```
This issue surfaced after commit a8173820f441 ("gpio: gpiolib: Allow GPIO
IRQs to lazy disable") introduced lazy disablement for GPIO IRQs. It
replaced disable/enable hooks with mask/unmask hooks. Unlike the disable
hook, the mask hook doesn't handle already-masked IRQs.
When a GPIO-IRQ driver is unbound, the IRQ is released, triggering
__irq_disable() and irq_state_set_masked(). A subsequent call to
machine_kexec_mask_interrupts() re-invokes chip->irq_mask(). This results
in a call chain, including gpiochip_irq_mask() and gpiochip_disable_irq().
Since FLAG_USED_AS_IRQ was cleared earlier, a warning occurs.
Replace calls to irq_mask() and irq_disable() hooks with a simplified call
to irq_shutdown(), and check if the interrupt is started (irqd_is_started)
before calling the shutdown.
Signed-off-by: Eliav Farber <farbere@amazon.com>
---
V4 -> V5:
- The phrase 'This patch' has been removed from the commit message.
kernel/irq/kexec.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/kernel/irq/kexec.c b/kernel/irq/kexec.c
index 0f9548c1708d..1a3deffe6b5b 100644
--- a/kernel/irq/kexec.c
+++ b/kernel/irq/kexec.c
@@ -17,7 +17,7 @@ void machine_kexec_mask_interrupts(void)
int check_eoi = 1;
chip = irq_desc_get_chip(desc);
- if (!chip)
+ if (!chip || !irqd_is_started(&desc->irq_data))
continue;
if (IS_ENABLED(CONFIG_GENERIC_IRQ_KEXEC_CLEAR_VM_FORWARD)) {
@@ -31,10 +31,6 @@ void machine_kexec_mask_interrupts(void)
if (check_eoi && chip->irq_eoi && irqd_irq_inprogress(&desc->irq_data))
chip->irq_eoi(&desc->irq_data);
- if (chip->irq_mask)
- chip->irq_mask(&desc->irq_data);
-
- if (chip->irq_disable && !irqd_irq_disabled(&desc->irq_data))
- chip->irq_disable(&desc->irq_data);
+ irq_shutdown(desc);
}
}
--
2.40.1
^ permalink raw reply related [flat|nested] 5+ messages in thread