[PATCH] iommu: Force iommu shutdown on panic.

[PATCH] iommu: Force iommu shutdown on panic.

[John Donnelly]

This eliminates intermitten kdump hangs reported on AMD/x86 servers 
with iommu enabled (iommu=on) regardless if iommu features are
used or not, on 5.4 kernels.  When kdump fails to initialize and run, no vmcore
(crash dumps) are captured.

Derived from the upsteam commit: iommu/vt-d: Turn off translations at shutdown.


Signed-off-by: John Donnelly <john.p.donnelly@oracle.com>
Reviewed-by: Jack Vogel <jack.vogel@oracle.com>
---
 arch/x86/include/asm/x86_init.h | 2 +-
 arch/x86/kernel/amd_gart_64.c   | 2 +-
 arch/x86/kernel/reboot.c        | 2 +-
 arch/x86/kernel/x86_init.c      | 2 +-
 drivers/iommu/amd_iommu_init.c  | 8 ++++----
 drivers/iommu/intel-iommu.c     | 9 ++++++---
 include/linux/dmar.h            | 4 ++--
 kernel/panic.c                  | 7 +++++++
 8 files changed, 23 insertions(+), 13 deletions(-)

diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h
index 96d9cd208610..06233f57b7c9 100644
--- a/arch/x86/include/asm/x86_init.h
+++ b/arch/x86/include/asm/x86_init.h
@@ -270,7 +270,7 @@ struct x86_platform_ops {
 	unsigned long (*calibrate_tsc)(void);
 	void (*get_wallclock)(struct timespec64 *ts);
 	int (*set_wallclock)(const struct timespec64 *ts);
-	void (*iommu_shutdown)(void);
+	void (*iommu_shutdown)(int panic);
 	bool (*is_untracked_pat_range)(u64 start, u64 end);
 	void (*nmi_init)(void);
 	unsigned char (*get_nmi_reason)(void);
diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c
index 4e5f50236048..2f805cc073ad 100644
--- a/arch/x86/kernel/amd_gart_64.c
+++ b/arch/x86/kernel/amd_gart_64.c
@@ -681,7 +681,7 @@ static const struct dma_map_ops gart_dma_ops = {
 	.get_required_mask		= dma_direct_get_required_mask,
 };
 
-static void gart_iommu_shutdown(void)
+static void gart_iommu_shutdown(int panic)
 {
 	struct pci_dev *dev;
 	int i;
diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c
index 0cc7c0b106bb..fd3c88a4f2e7 100644
--- a/arch/x86/kernel/reboot.c
+++ b/arch/x86/kernel/reboot.c
@@ -708,7 +708,7 @@ void native_machine_shutdown(void)
 #endif
 
 #ifdef CONFIG_X86_64
-	x86_platform.iommu_shutdown();
+	x86_platform.iommu_shutdown(0);
 #endif
 }
 
diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c
index 85f1a90c55cd..e63719710097 100644
--- a/arch/x86/kernel/x86_init.c
+++ b/arch/x86/kernel/x86_init.c
@@ -28,7 +28,7 @@
 void x86_init_noop(void) { }
 void __init x86_init_uint_noop(unsigned int unused) { }
 static int __init iommu_init_noop(void) { return 0; }
-static void iommu_shutdown_noop(void) { }
+static void iommu_shutdown_noop(int panic) { }
 bool __init bool_x86_init_noop(void) { return false; }
 void x86_op_int_noop(int cpu) { }
 static __init int set_rtc_noop(const struct timespec64 *now) { return -EINVAL; }
diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c
index 2759a8d57b7f..a1afc2e2a68c 100644
--- a/drivers/iommu/amd_iommu_init.c
+++ b/drivers/iommu/amd_iommu_init.c
@@ -2361,7 +2361,7 @@ static void enable_iommus(void)
 	enable_iommus_v2();
 }
 
-static void disable_iommus(void)
+static void disable_iommus(int panic)
 {
 	struct amd_iommu *iommu;
 
@@ -2395,7 +2395,7 @@ static void amd_iommu_resume(void)
 static int amd_iommu_suspend(void)
 {
 	/* disable IOMMUs to go out of the way for BIOS */
-	disable_iommus();
+	disable_iommus(0);
 
 	return 0;
 }
@@ -2612,7 +2612,7 @@ static int __init early_amd_iommu_init(void)
 
 	/* Disable any previously enabled IOMMUs */
 	if (!is_kdump_kernel() || amd_iommu_disabled)
-		disable_iommus();
+		disable_iommus(0);
 
 	if (amd_iommu_irq_remap)
 		amd_iommu_irq_remap = check_ioapic_information();
@@ -2762,7 +2762,7 @@ static int __init state_next(void)
 	if (ret) {
 		free_dma_resources();
 		if (!irq_remapping_enabled) {
-			disable_iommus();
+			disable_iommus(0);
 			free_iommu_resources();
 		} else {
 			struct amd_iommu *iommu;
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 9dc37672bf89..8bc95f4e7d3e 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4903,7 +4903,7 @@ static void intel_disable_iommus(void)
 		iommu_disable_translation(iommu);
 }
 
-void intel_iommu_shutdown(void)
+void intel_iommu_shutdown(int panic)
 {
 	struct dmar_drhd_unit *drhd;
 	struct intel_iommu *iommu = NULL;
@@ -4911,7 +4911,8 @@ void intel_iommu_shutdown(void)
 	if (no_iommu || dmar_disabled)
 		return;
 
-	down_write(&dmar_global_lock);
+	if (!panic)
+		down_write(&dmar_global_lock);
 
 	/* Disable PMRs explicitly here. */
 	for_each_iommu(iommu, drhd)
@@ -4920,7 +4921,9 @@ void intel_iommu_shutdown(void)
 	/* Make sure the IOMMUs are switched off */
 	intel_disable_iommus();
 
-	up_write(&dmar_global_lock);
+	if (!panic)
+		up_write(&dmar_global_lock);
+	
 }
 
 static inline struct intel_iommu *dev_to_intel_iommu(struct device *dev)
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index f64ca27dc210..c4f24bc41169 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -129,7 +129,7 @@ static inline int dmar_res_noop(struct acpi_dmar_header *hdr, void *arg)
 #ifdef CONFIG_INTEL_IOMMU
 extern int iommu_detected, no_iommu;
 extern int intel_iommu_init(void);
-extern void intel_iommu_shutdown(void);
+extern void intel_iommu_shutdown(int panic);
 extern int dmar_parse_one_rmrr(struct acpi_dmar_header *header, void *arg);
 extern int dmar_parse_one_atsr(struct acpi_dmar_header *header, void *arg);
 extern int dmar_check_one_atsr(struct acpi_dmar_header *hdr, void *arg);
@@ -138,7 +138,7 @@ extern int dmar_iommu_hotplug(struct dmar_drhd_unit *dmaru, bool insert);
 extern int dmar_iommu_notify_scope_dev(struct dmar_pci_notify_info *info);
 #else /* !CONFIG_INTEL_IOMMU: */
 static inline int intel_iommu_init(void) { return -ENODEV; }
-static inline void intel_iommu_shutdown(void) { }
+static inline void intel_iommu_shutdown(int panic) { }
 
 #define	dmar_parse_one_rmrr		dmar_res_noop
 #define	dmar_parse_one_atsr		dmar_res_noop
diff --git a/kernel/panic.c b/kernel/panic.c
index b69ee9e76cb2..ee81462d5b1f 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -32,6 +32,7 @@
 #include <linux/ratelimit.h>
 #include <linux/debugfs.h>
 #include <asm/sections.h>
+#include <linux/dmar.h>
 
 #define PANIC_TIMER_STEP 100
 #define PANIC_BLINK_SPD 18
@@ -213,6 +214,12 @@ void panic(const char *fmt, ...)
 		buf[len - 1] = '\0';
 
 	pr_emerg("Kernel panic - not syncing: %s\n", buf);
+
+#ifdef CONFIG_X86
+	pr_emerg("Shutting down iommu.\n");
+	x86_platform.iommu_shutdown(1); 
+#endif
+
 #ifdef CONFIG_DEBUG_BUGVERBOSE
 	/*
 	 * Avoid nested stack-dumping if a panic occurs during oops processing
-- 
2.20.1