Skip to contentSkip to conversation

Intel IOMMU updates for Linux v5.20

GIT PULL New Lu Baolu 2 participants 27 messages Jul 12, 2022 – Jul 15, 2022

Subsystems (10)
DRM drivers x86 Intel i915 Tracing +6 more subsystems
lore
Share & tools
lore

Lu Baolu

author maintainer
#0 lore raw

Hi Joerg,

This includes patches queued for v5.20. It includes:

- Make intel-iommu.h private
- Optimize the use of two locks
- Extend the driver to support large-scale platforms
- Cleanup some dead code

The whole series is also available at:
https://github.com/LuBaolu/intel-iommu/commits/vtd-update-for-v5.20

Please consider them for v5.20.

[The vt-d branch needs to include below fix patch:
4140d77a02210 ("iommu/vt-d: Fix RID2PASID setup/teardown failure") before applying the patches.]

Best regards,
Baolu

Lu Baolu (25):
iommu/vt-d: Move trace/events/intel_iommu.h under iommu agp/intel: Use per device iommu check
iommu/vt-d: Remove unnecessary exported symbol
drm/i915: Remove unnecessary include
KVM: x86: Remove unnecessary include
x86/boot/tboot: Move tboot_force_iommu() to Intel IOMMU iommu/vt-d: Move include/linux/intel-iommu.h under iommu iommu/vt-d: Remove unused iovad from dmar_domain iommu/vt-d: debugfs: Remove device_domain_lock usage iommu/vt-d: Remove clearing translation data in disable_dmar_iommu() iommu/vt-d: Use pci_get_domain_bus_and_slot() in pgtable_walk() iommu/vt-d: Unnecessary spinlock for root table alloc and free iommu/vt-d: Replace spin_lock_irqsave() with spin_lock() iommu/vt-d: Acquiring lock in domain ID allocation helpers iommu/vt-d: Acquiring lock in pasid manipulation helpers iommu/vt-d: Check device list of domain in domain free path iommu/vt-d: Fold __dmar_remove_one_dev_info() into its caller iommu/vt-d: Use device_domain_lock accurately
iommu/vt-d: Convert global spinlock into per domain lock iommu/vt-d: Remove unused domain_get_iommu()
iommu/vt-d: Use IDA interface to manage iommu sequence id iommu/vt-d: Refactor iommu information of each domain iommu/vt-d: Remove unnecessary check in intel_iommu_add() iommu/vt-d: Remove global g_iommus array
iommu/vt-d: Make DMAR_UNITS_SUPPORTED default 1024

 include/linux/dmar.h                          |   6 +-
 include/linux/tboot.h                         |   2 -
 .../iommu/intel/iommu.h                       |  35 +-
 drivers/iommu/intel/pasid.h                   |   1 +
 .../iommu/intel/trace.h                       |   7 +-
 arch/x86/kernel/tboot.c                       |  15 -
 arch/x86/kvm/x86.c                            |   1 -
 drivers/char/agp/intel-gtt.c                  |  17 +-
 drivers/gpu/drm/i915/display/intel_display.c  |   1 -
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    |   1 -
 drivers/iommu/intel/cap_audit.c               |   2 +-
 drivers/iommu/intel/debugfs.c                 |  51 +-
 drivers/iommu/intel/dmar.c                    |  39 +-
 drivers/iommu/intel/iommu.c                   | 446 ++++++------------
 drivers/iommu/intel/irq_remapping.c           |   2 +-
 drivers/iommu/intel/pasid.c                   | 107 +++--
 drivers/iommu/intel/perf.c                    |   2 +-
 drivers/iommu/intel/svm.c                     |  11 +-
 drivers/iommu/intel/trace.c                   |   2 +-
 MAINTAINERS                                   |   1 -
 20 files changed, 297 insertions(+), 452 deletions(-)
 rename include/linux/intel-iommu.h => drivers/iommu/intel/iommu.h (97%)
 rename include/trace/events/intel_iommu.h => drivers/iommu/intel/trace.h (93%)
Signature · 1 line
2.25.1
author maintainer
#1 lore raw

PATCH 1/25 iommu/vt-d: Move trace/events/intel_iommu.h under iommu

This header file is private to the Intel IOMMU driver. Move it to the driver folder.

---
 .../trace/events/intel_iommu.h => drivers/iommu/intel/trace.h | 4 ++++
 drivers/iommu/intel/dmar.c                                    | 2 +-
 drivers/iommu/intel/svm.c                                     | 2 +-
 drivers/iommu/intel/trace.c                                   | 2 +-
 4 files changed, 7 insertions(+), 3 deletions(-)
 rename include/trace/events/intel_iommu.h => drivers/iommu/intel/trace.h (94%)
drivers/iommu/intel/trace.h renamed from include/trace/events/intel_iommu.h +4 -0
similarity index 94%
rename from include/trace/events/intel_iommu.h
rename to drivers/iommu/intel/trace.h
index e5c1ca6d16ee..25cb7f88e1a2 100644
@@ -91,4 +91,8 @@ TRACE_EVENT(prq_report,
 #endif /* _TRACE_INTEL_IOMMU_H */
 
 /* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH ../../drivers/iommu/intel/
+#define TRACE_INCLUDE_FILE trace
 #include <trace/define_trace.h>
drivers/iommu/intel/dmar.c +1 -1
index 9699ca101c62..f91b45be1d92 100644
@@ -30,10 +30,10 @@
 #include <linux/numa.h>
 #include <linux/limits.h>
 #include <asm/irq_remapping.h>
-#include <trace/events/intel_iommu.h>
 
 #include "../irq_remapping.h"
 #include "perf.h"
+#include "trace.h"
 
 typedef int (*dmar_res_handler_t)(struct acpi_dmar_header *, void *);
 struct dmar_res_callback {
drivers/iommu/intel/svm.c +1 -1
index 7ee37d996e15..70b40d007a52 100644
@@ -21,11 +21,11 @@
 #include <linux/ioasid.h>
 #include <asm/page.h>
 #include <asm/fpu/api.h>
-#include <trace/events/intel_iommu.h>
 
 #include "pasid.h"
 #include "perf.h"
 #include "../iommu-sva-lib.h"
+#include "trace.h"
 
 static irqreturn_t prq_event_thread(int irq, void *d);
 static void intel_svm_drain_prq(struct device *dev, u32 pasid);
drivers/iommu/intel/trace.c +1 -1
index bfb6a6e37a88..117e626e3ea9 100644
@@ -11,4 +11,4 @@
 #include <linux/types.h>
 
 #define CREATE_TRACE_POINTS
-#include <trace/events/intel_iommu.h>
+#include "trace.h"
Signature · 1 line
2.25.1
author maintainer
#2 lore raw

PATCH 2/25 agp/intel: Use per device iommu check

In reply to Lu Baolu

The IOMMU subsystem has already provided an interface to query whether the IOMMU hardware is enabled for a specific device. This changes the check from Intel specific intel_iommu_gfx_mapped (globally exported by the Intel IOMMU driver) to probing the presence of IOMMU on a specific device using the generic device_iommu_mapped().

This follows commit cca084692394a ("drm/i915: Use per device iommu check") which converted drm/i915 driver to use device_iommu_mapped().

---
 drivers/char/agp/intel-gtt.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)
drivers/char/agp/intel-gtt.c +7 -10
index 79a1b65527c2..cfcb450e9b59 100644
@@ -20,7 +20,7 @@
 #include <linux/kernel.h>
 #include <linux/pagemap.h>
 #include <linux/agp_backend.h>
-#include <linux/intel-iommu.h>
+#include <linux/iommu.h>
 #include <linux/delay.h>
 #include <asm/smp.h>
 #include "agp.h"
@@ -573,18 +573,15 @@ static void intel_gtt_cleanup(void)
  */
 static inline int needs_ilk_vtd_wa(void)
 {
-#ifdef CONFIG_INTEL_IOMMU
 	const unsigned short gpu_devid = intel_private.pcidev->device;
 
-	/* Query intel_iommu to see if we need the workaround. Presumably that
-	 * was loaded first.
+	/*
+	 * Query iommu subsystem to see if we need the workaround. Presumably
+	 * that was loaded first.
 	 */
-	if ((gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_D_IG ||
-	     gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
-	     intel_iommu_gfx_mapped)
-		return 1;
-#endif
-	return 0;
+	return ((gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_D_IG ||
+		 gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) &&
+		device_iommu_mapped(&intel_private.pcidev->dev));
 }
 
 static bool intel_gtt_can_wc(void)
Signature · 1 line
2.25.1
author maintainer
#3 lore raw

PATCH 3/25 iommu/vt-d: Remove unnecessary exported symbol

In reply to Lu Baolu

The exported symbol intel_iommu_gfx_mapped is not used anywhere in the tree. Remove it to avoid dead code.

---
 include/linux/intel-iommu.h | 1 -
 drivers/iommu/intel/iommu.c | 6 ------
 2 files changed, 7 deletions(-)
include/linux/intel-iommu.h +0 -1
index 5fcf89faa31a..4ebf3c4da45e 100644
@@ -787,7 +787,6 @@ extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
 extern int dmar_disabled;
 extern int intel_iommu_enabled;
-extern int intel_iommu_gfx_mapped;
 #else
 static inline int iommu_calculate_agaw(struct intel_iommu *iommu)
 {
drivers/iommu/intel/iommu.c +0 -6
index 5c0dce78586a..6564af7ec0dd 100644
@@ -314,9 +314,6 @@ static int iommu_skip_te_disable;
 #define IDENTMAP_GFX		2
 #define IDENTMAP_AZALIA		4
 
-int intel_iommu_gfx_mapped;
-EXPORT_SYMBOL_GPL(intel_iommu_gfx_mapped);
-
 DEFINE_SPINLOCK(device_domain_lock);
 static LIST_HEAD(device_domain_list);
 
@@ -4093,9 +4090,6 @@ int __init intel_iommu_init(void)
 	if (list_empty(&dmar_satc_units))
 		pr_info("No SATC found\n");
 
-	if (dmar_map_gfx)
-		intel_iommu_gfx_mapped = 1;
-
 	init_no_remapping_devices();
 
 	ret = init_dmars();
Signature · 1 line
2.25.1
author maintainer
#4 lore raw

PATCH 4/25 drm/i915: Remove unnecessary include

In reply to Lu Baolu

intel-iommu.h is not needed in drm/i915 anymore. Remove its include.

---
 drivers/gpu/drm/i915/display/intel_display.c   | 1 -
 drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 1 -
 2 files changed, 2 deletions(-)
drivers/gpu/drm/i915/display/intel_display.c +0 -1
index 806d50b302ab..493421759afa 100644
@@ -27,7 +27,6 @@
 #include <acpi/video.h>
 #include <linux/i2c.h>
 #include <linux/input.h>
-#include <linux/intel-iommu.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/dma-resv.h>
drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +0 -1
index 30fe847c6664..193c7c83c70f 100644
@@ -6,7 +6,6 @@
 
 #include <linux/dma-resv.h>
 #include <linux/highmem.h>
-#include <linux/intel-iommu.h>
 #include <linux/sync_file.h>
 #include <linux/uaccess.h>
 
Signature · 1 line
2.25.1
author maintainer
#5 lore raw

PATCH 5/25 KVM: x86: Remove unnecessary include

In reply to Lu Baolu

intel-iommu.h is not needed in kvm/x86 anymore. Remove its include.

---
 arch/x86/kvm/x86.c | 1 -
 1 file changed, 1 deletion(-)
arch/x86/kvm/x86.c +0 -1
index 1910e1e78b15..f7ec1e39eec7 100644
@@ -41,7 +41,6 @@
 #include <linux/mman.h>
 #include <linux/highmem.h>
 #include <linux/iommu.h>
-#include <linux/intel-iommu.h>
 #include <linux/cpufreq.h>
 #include <linux/user-return-notifier.h>
 #include <linux/srcu.h>
Signature · 1 line
2.25.1
author maintainer
#6 lore raw

PATCH 6/25 x86/boot/tboot: Move tboot_force_iommu() to Intel IOMMU

In reply to Lu Baolu

tboot_force_iommu() is only called by the Intel IOMMU driver. Move the helper into that driver. No functional change intended.

---
 include/linux/tboot.h       |  2 --
 arch/x86/kernel/tboot.c     | 15 ---------------
 drivers/iommu/intel/iommu.c | 14 ++++++++++++++
 3 files changed, 14 insertions(+), 17 deletions(-)
include/linux/tboot.h +0 -2
index 5146d2574e85..d2279160ef39 100644
@@ -126,7 +126,6 @@ extern void tboot_probe(void);
 extern void tboot_shutdown(u32 shutdown_type);
 extern struct acpi_table_header *tboot_get_dmar_table(
 				      struct acpi_table_header *dmar_tbl);
-extern int tboot_force_iommu(void);
 
 #else
 
@@ -136,7 +135,6 @@ extern int tboot_force_iommu(void);
 #define tboot_sleep(sleep_state, pm1a_control, pm1b_control)	\
 					do { } while (0)
 #define tboot_get_dmar_table(dmar_tbl)	(dmar_tbl)
-#define tboot_force_iommu()		0
 
 #endif /* !CONFIG_INTEL_TXT */
 
arch/x86/kernel/tboot.c +0 -15
index 0c1154a1c403..3bacd935f840 100644
@@ -6,7 +6,6 @@
  * Copyright (c) 2006-2009, Intel Corporation
  */
 
-#include <linux/intel-iommu.h>
 #include <linux/init_task.h>
 #include <linux/spinlock.h>
 #include <linux/export.h>
@@ -516,17 +515,3 @@ struct acpi_table_header *tboot_get_dmar_table(struct acpi_table_header *dmar_tb
 
 	return dmar_tbl;
 }
-
-int tboot_force_iommu(void)
-{
-	if (!tboot_enabled())
-		return 0;
-
-	if (no_iommu || dmar_disabled)
-		pr_warn("Forcing Intel-IOMMU to enabled\n");
-
-	dmar_disabled = 0;
-	no_iommu = 0;
-
-	return 1;
-}
drivers/iommu/intel/iommu.c +14 -0
index 6564af7ec0dd..c70948021e8e 100644
@@ -4019,6 +4019,20 @@ static int __init probe_acpi_namespace_devices(void)
 	return 0;
 }
 
+static __init int tboot_force_iommu(void)
+{
+	if (!tboot_enabled())
+		return 0;
+
+	if (no_iommu || dmar_disabled)
+		pr_warn("Forcing Intel-IOMMU to enabled\n");
+
+	dmar_disabled = 0;
+	no_iommu = 0;
+
+	return 1;
+}
+
 int __init intel_iommu_init(void)
 {
 	int ret = -ENODEV;
Signature · 1 line
2.25.1
author maintainer
#7 lore raw

PATCH 7/25 iommu/vt-d: Move include/linux/intel-iommu.h under iommu

In reply to Lu Baolu

This header file is private to the Intel IOMMU driver. Move it to the driver folder.

---
 include/linux/intel-iommu.h => drivers/iommu/intel/iommu.h | 0
 drivers/iommu/intel/trace.h                                | 3 ++-
 drivers/iommu/intel/cap_audit.c                            | 2 +-
 drivers/iommu/intel/debugfs.c                              | 2 +-
 drivers/iommu/intel/dmar.c                                 | 2 +-
 drivers/iommu/intel/iommu.c                                | 2 +-
 drivers/iommu/intel/irq_remapping.c                        | 2 +-
 drivers/iommu/intel/pasid.c                                | 2 +-
 drivers/iommu/intel/perf.c                                 | 2 +-
 drivers/iommu/intel/svm.c                                  | 2 +-
 MAINTAINERS                                                | 1 -
 11 files changed, 10 insertions(+), 10 deletions(-)
 rename include/linux/intel-iommu.h => drivers/iommu/intel/iommu.h (100%)
drivers/iommu/intel/iommu.h renamed from include/linux/intel-iommu.h +0 -0
similarity index 100%
rename from include/linux/intel-iommu.h
rename to drivers/iommu/intel/iommu.h
drivers/iommu/intel/trace.h +2 -1
index 25cb7f88e1a2..93d96f93a89b 100644
@@ -13,7 +13,8 @@
 #define _TRACE_INTEL_IOMMU_H
 
 #include <linux/tracepoint.h>
-#include <linux/intel-iommu.h>
+
+#include "iommu.h"
 
 #define MSG_MAX		256
 
drivers/iommu/intel/cap_audit.c +1 -1
index 71596fc62822..3ee68393122f 100644
@@ -10,7 +10,7 @@
 
 #define pr_fmt(fmt)	"DMAR: " fmt
 
-#include <linux/intel-iommu.h>
+#include "iommu.h"
 #include "cap_audit.h"
 
 static u64 intel_iommu_cap_sanity;
drivers/iommu/intel/debugfs.c +1 -1
index ed796eea4581..d927ef10641b 100644
@@ -10,11 +10,11 @@
 
 #include <linux/debugfs.h>
 #include <linux/dmar.h>
-#include <linux/intel-iommu.h>
 #include <linux/pci.h>
 
 #include <asm/irq_remapping.h>
 
+#include "iommu.h"
 #include "pasid.h"
 #include "perf.h"
 
drivers/iommu/intel/dmar.c +1 -1
index f91b45be1d92..2a5e0f91e647 100644
@@ -19,7 +19,6 @@
 #include <linux/pci.h>
 #include <linux/dmar.h>
 #include <linux/iova.h>
-#include <linux/intel-iommu.h>
 #include <linux/timer.h>
 #include <linux/irq.h>
 #include <linux/interrupt.h>
@@ -31,6 +30,7 @@
 #include <linux/limits.h>
 #include <asm/irq_remapping.h>
 
+#include "iommu.h"
 #include "../irq_remapping.h"
 #include "perf.h"
 #include "trace.h"
drivers/iommu/intel/iommu.c +1 -1
index c70948021e8e..10bda4bec8fe 100644
@@ -17,7 +17,6 @@
 #include <linux/dma-direct.h>
 #include <linux/dma-iommu.h>
 #include <linux/dmi.h>
-#include <linux/intel-iommu.h>
 #include <linux/intel-svm.h>
 #include <linux/memory.h>
 #include <linux/pci.h>
@@ -26,6 +25,7 @@
 #include <linux/syscore_ops.h>
 #include <linux/tboot.h>
 
+#include "iommu.h"
 #include "../irq_remapping.h"
 #include "../iommu-sva-lib.h"
 #include "pasid.h"
drivers/iommu/intel/irq_remapping.c +1 -1
index a67319597884..2e9683e970f8 100644
@@ -10,7 +10,6 @@
 #include <linux/hpet.h>
 #include <linux/pci.h>
 #include <linux/irq.h>
-#include <linux/intel-iommu.h>
 #include <linux/acpi.h>
 #include <linux/irqdomain.h>
 #include <linux/crash_dump.h>
@@ -21,6 +20,7 @@
 #include <asm/irq_remapping.h>
 #include <asm/pci-direct.h>
 
+#include "iommu.h"
 #include "../irq_remapping.h"
 #include "cap_audit.h"
 
drivers/iommu/intel/pasid.c +1 -1
index 17cad7c1f62d..43f090381ec7 100644
@@ -12,13 +12,13 @@
 #include <linux/bitops.h>
 #include <linux/cpufeature.h>
 #include <linux/dmar.h>
-#include <linux/intel-iommu.h>
 #include <linux/iommu.h>
 #include <linux/memory.h>
 #include <linux/pci.h>
 #include <linux/pci-ats.h>
 #include <linux/spinlock.h>
 
+#include "iommu.h"
 #include "pasid.h"
 
 /*
drivers/iommu/intel/perf.c +1 -1
index 0e8e03252d92..94ee70ac38e3 100644
@@ -9,8 +9,8 @@
  */
 
 #include <linux/spinlock.h>
-#include <linux/intel-iommu.h>
 
+#include "iommu.h"
 #include "perf.h"
 
 static DEFINE_SPINLOCK(latency_lock);
drivers/iommu/intel/svm.c +1 -1
index 70b40d007a52..580713aa9e07 100644
@@ -5,7 +5,6 @@
  * Authors: David Woodhouse <[email protected]>
  */
 
-#include <linux/intel-iommu.h>
 #include <linux/mmu_notifier.h>
 #include <linux/sched.h>
 #include <linux/sched/mm.h>
@@ -22,6 +21,7 @@
 #include <asm/page.h>
 #include <asm/fpu/api.h>
 
+#include "iommu.h"
 #include "pasid.h"
 #include "perf.h"
 #include "../iommu-sva-lib.h"
MAINTAINERS +0 -1
index ead381fdfc5a..fa2cde7dc644 100644
@@ -9998,7 +9998,6 @@ L:	[email protected]
 S:	Supported
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu.git
 F:	drivers/iommu/intel/
-F:	include/linux/intel-iommu.h
 F:	include/linux/intel-svm.h
 
 INTEL IOP-ADMA DMA DRIVER
Signature · 1 line
2.25.1
author maintainer
#8 lore raw

PATCH 8/25 iommu/vt-d: Remove unused iovad from dmar_domain

In reply to Lu Baolu

Not used anywhere. Cleanup it to avoid dead code.

---
 drivers/iommu/intel/iommu.h | 1 -
 1 file changed, 1 deletion(-)
drivers/iommu/intel/iommu.h +0 -1
index 4ebf3c4da45e..8285fcfa5fea 100644
@@ -543,7 +543,6 @@ struct dmar_domain {
 	u8 set_pte_snp:1;
 
 	struct list_head devices;	/* all devices' list */
-	struct iova_domain iovad;	/* iova's that belong to this domain */
 
 	struct dma_pte	*pgd;		/* virtual address */
 	int		gaw;		/* max guest address width */
Signature · 1 line
2.25.1
author maintainer
#9 lore raw

PATCH 9/25 iommu/vt-d: debugfs: Remove device_domain_lock usage

In reply to Lu Baolu

The domain_translation_struct debugfs node is used to dump the DMAR page tables for the PCI devices. It potentially races with setting domains to devices. The existing code uses the global spinlock device_domain_lock to avoid the races.

This removes the use of device_domain_lock outside of iommu.c by replacing it with the group mutex lock. Using the group mutex lock is cleaner and more compatible to following cleanups.

---
 drivers/iommu/intel/iommu.h   |  1 -
 drivers/iommu/intel/debugfs.c | 43 +++++++++++++++++++++++++----------
 drivers/iommu/intel/iommu.c   |  2 +-
 3 files changed, 32 insertions(+), 14 deletions(-)
drivers/iommu/intel/iommu.h +0 -1
index 8285fcfa5fea..8deb745d8b36 100644
@@ -480,7 +480,6 @@ enum {
 #define VTD_FLAG_SVM_CAPABLE		(1 << 2)
 
 extern int intel_iommu_sm;
-extern spinlock_t device_domain_lock;
 
 #define sm_supported(iommu)	(intel_iommu_sm && ecap_smts((iommu)->ecap))
 #define pasid_supported(iommu)	(sm_supported(iommu) &&			\
drivers/iommu/intel/debugfs.c +31 -12

64 lines not on this page · raw message

drivers/iommu/intel/iommu.c +1 -1
index 10bda4bec8fe..3b6571681bdd 100644
@@ -314,7 +314,7 @@ static int iommu_skip_te_disable;
 #define IDENTMAP_GFX		2
 #define IDENTMAP_AZALIA		4
 
-DEFINE_SPINLOCK(device_domain_lock);
+static DEFINE_SPINLOCK(device_domain_lock);
 static LIST_HEAD(device_domain_list);
 
 const struct iommu_ops intel_iommu_ops;
Signature · 1 line
2.25.1
author maintainer
#10 lore raw

PATCH 10/25 iommu/vt-d: Remove clearing translation data in disable_dmar_iommu()

In reply to Lu Baolu

The disable_dmar_iommu() is called when IOMMU initialization fails or the IOMMU is hot-removed from the system. In both cases, there is no need to clear the IOMMU translation data structures for devices.

On the initialization path, the device probing only happens after the IOMMU is initialized successfully, hence there're no translation data structures.

On the hot-remove path, there is no real use case where the IOMMU is hot-removed, but the devices that it manages are still alive in the system. The translation data structures were torn down during device release, hence there's no need to repeat it in IOMMU hot-remove path either. This removes the unnecessary code and only leaves a check.

---
 drivers/iommu/intel/pasid.h |  1 +
 drivers/iommu/intel/iommu.c | 21 +++++++--------------
 2 files changed, 8 insertions(+), 14 deletions(-)
drivers/iommu/intel/pasid.h +1 -0
index bf5b937848b4..20c54e50f533 100644
@@ -39,6 +39,7 @@
  * only and pass-through transfer modes.
  */
 #define FLPT_DEFAULT_DID		1
+#define NUM_RESERVED_DID		2
 
 /*
  * The SUPERVISOR_MODE flag indicates a first level translation which
drivers/iommu/intel/iommu.c +7 -14
index 3b6571681bdd..43aaec5bdd84 100644
@@ -1716,23 +1716,16 @@ static int iommu_init_domains(struct intel_iommu *iommu)
 
 static void disable_dmar_iommu(struct intel_iommu *iommu)
 {
-	struct device_domain_info *info, *tmp;
-	unsigned long flags;
-
 	if (!iommu->domain_ids)
 		return;
 
-	spin_lock_irqsave(&device_domain_lock, flags);
-	list_for_each_entry_safe(info, tmp, &device_domain_list, global) {
-		if (info->iommu != iommu)
-			continue;
-
-		if (!info->dev || !info->domain)
-			continue;
-
-		__dmar_remove_one_dev_info(info);
-	}
-	spin_unlock_irqrestore(&device_domain_lock, flags);
+	/*
+	 * All iommu domains must have been detached from the devices,
+	 * hence there should be no domain IDs in use.
+	 */
+	if (WARN_ON(bitmap_weight(iommu->domain_ids, cap_ndoms(iommu->cap))
+		    > NUM_RESERVED_DID))
+		return;
 
 	if (iommu->gcmd & DMA_GCMD_TE)
 		iommu_disable_translation(iommu);
Signature · 1 line
2.25.1
author maintainer
#11 lore raw

PATCH 11/25 iommu/vt-d: Use pci_get_domain_bus_and_slot() in pgtable_walk()

In reply to Lu Baolu

Use pci_get_domain_bus_and_slot() instead of searching the global list to retrieve the pci device pointer. This also removes the global device_domain_list as there isn't any consumer anymore.

---
 drivers/iommu/intel/iommu.h |  1 -
 drivers/iommu/intel/iommu.c | 34 ++++++----------------------------
 2 files changed, 6 insertions(+), 29 deletions(-)
drivers/iommu/intel/iommu.h +0 -1
index 8deb745d8b36..198c6c822ef4 100644
@@ -609,7 +609,6 @@ struct intel_iommu {
 /* PCI domain-device relationship */
 struct device_domain_info {
 	struct list_head link;	/* link to domain siblings */
-	struct list_head global; /* link to global list */
 	u32 segment;		/* PCI segment number */
 	u8 bus;			/* PCI bus number */
 	u8 devfn;		/* PCI devfn number */
drivers/iommu/intel/iommu.c +6 -28
index 43aaec5bdd84..77915d61f7ec 100644
@@ -131,8 +131,6 @@ static struct intel_iommu **g_iommus;
 
 static void __init check_tylersburg_isoch(void);
 static int rwbf_quirk;
-static inline struct device_domain_info *
-dmar_search_domain_by_dev_info(int segment, int bus, int devfn);
 
 /*
  * set to 1 to panic kernel if can't successfully enable VT-d
@@ -315,8 +313,6 @@ static int iommu_skip_te_disable;
 #define IDENTMAP_AZALIA		4
 
 static DEFINE_SPINLOCK(device_domain_lock);
-static LIST_HEAD(device_domain_list);
-
 const struct iommu_ops intel_iommu_ops;
 
 static bool translation_pre_enabled(struct intel_iommu *iommu)
@@ -846,9 +842,14 @@ static void pgtable_walk(struct intel_iommu *iommu, unsigned long pfn, u8 bus, u
 	struct device_domain_info *info;
 	struct dma_pte *parent, *pte;
 	struct dmar_domain *domain;
+	struct pci_dev *pdev;
 	int offset, level;
 
-	info = dmar_search_domain_by_dev_info(iommu->segment, bus, devfn);
+	pdev = pci_get_domain_bus_and_slot(iommu->segment, bus, devfn);
+	if (!pdev)
+		return;
+
+	info = dev_iommu_priv_get(&pdev->dev);
 	if (!info || !info->domain) {
 		pr_info("device [%02x:%02x.%d] not probed\n",
 			bus, PCI_SLOT(devfn), PCI_FUNC(devfn));
@@ -2357,19 +2358,6 @@ static void domain_remove_dev_info(struct dmar_domain *domain)
 	spin_unlock_irqrestore(&device_domain_lock, flags);
 }
 
-static inline struct device_domain_info *
-dmar_search_domain_by_dev_info(int segment, int bus, int devfn)
-{
-	struct device_domain_info *info;
-
-	list_for_each_entry(info, &device_domain_list, global)
-		if (info->segment == segment && info->bus == bus &&
-		    info->devfn == devfn)
-			return info;
-
-	return NULL;
-}
-
 static int domain_setup_first_level(struct intel_iommu *iommu,
 				    struct dmar_domain *domain,
 				    struct device *dev,
@@ -4573,7 +4561,6 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
 	struct pci_dev *pdev = dev_is_pci(dev) ? to_pci_dev(dev) : NULL;
 	struct device_domain_info *info;
 	struct intel_iommu *iommu;
-	unsigned long flags;
 	u8 bus, devfn;
 
 	iommu = device_to_iommu(dev, &bus, &devfn);
@@ -4616,10 +4603,7 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
 		}
 	}
 
-	spin_lock_irqsave(&device_domain_lock, flags);
-	list_add(&info->global, &device_domain_list);
 	dev_iommu_priv_set(dev, info);
-	spin_unlock_irqrestore(&device_domain_lock, flags);
 
 	return &iommu->iommu;
 }
@@ -4627,15 +4611,9 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
 static void intel_iommu_release_device(struct device *dev)
 {
 	struct device_domain_info *info = dev_iommu_priv_get(dev);
-	unsigned long flags;
 
 	dmar_remove_one_dev_info(dev);
-
-	spin_lock_irqsave(&device_domain_lock, flags);
 	dev_iommu_priv_set(dev, NULL);
-	list_del(&info->global);
-	spin_unlock_irqrestore(&device_domain_lock, flags);
-
 	kfree(info);
 	set_dma_ops(dev, NULL);
 }
Signature · 1 line
2.25.1
author maintainer
#12 lore raw

PATCH 12/25 iommu/vt-d: Unnecessary spinlock for root table alloc and free

In reply to Lu Baolu

The IOMMU root table is allocated and freed in the IOMMU initialization code in static boot or hot-remove paths. There's no need for a spinlock.

---
 drivers/iommu/intel/iommu.c | 18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)
drivers/iommu/intel/iommu.c +5 -13
index 77915d61f7ec..ff49c9460ede 100644
@@ -809,14 +809,12 @@ static int device_context_mapped(struct intel_iommu *iommu, u8 bus, u8 devfn)
 
 static void free_context_table(struct intel_iommu *iommu)
 {
-	int i;
-	unsigned long flags;
 	struct context_entry *context;
+	int i;
+
+	if (!iommu->root_entry)
+		return;
 
-	spin_lock_irqsave(&iommu->lock, flags);
-	if (!iommu->root_entry) {
-		goto out;
-	}
 	for (i = 0; i < ROOT_ENTRY_NR; i++) {
 		context = iommu_context_addr(iommu, i, 0, 0);
 		if (context)
@@ -828,12 +826,10 @@ static void free_context_table(struct intel_iommu *iommu)
 		context = iommu_context_addr(iommu, i, 0x80, 0);
 		if (context)
 			free_pgtable_page(context);
-
 	}
+
 	free_pgtable_page(iommu->root_entry);
 	iommu->root_entry = NULL;
-out:
-	spin_unlock_irqrestore(&iommu->lock, flags);
 }
 
 #ifdef CONFIG_DMAR_DEBUG
@@ -1232,7 +1228,6 @@ static void domain_unmap(struct dmar_domain *domain, unsigned long start_pfn,
 static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 {
 	struct root_entry *root;
-	unsigned long flags;
 
 	root = (struct root_entry *)alloc_pgtable_page(iommu->node);
 	if (!root) {
@@ -1242,10 +1237,7 @@ static int iommu_alloc_root_entry(struct intel_iommu *iommu)
 	}
 
 	__iommu_flush_cache(iommu, root, ROOT_SIZE);
-
-	spin_lock_irqsave(&iommu->lock, flags);
 	iommu->root_entry = root;
-	spin_unlock_irqrestore(&iommu->lock, flags);
 
 	return 0;
 }
Signature · 1 line
2.25.1
author maintainer
#13 lore raw

PATCH 13/25 iommu/vt-d: Replace spin_lock_irqsave() with spin_lock()

In reply to Lu Baolu

The iommu->lock is used to protect changes in root/context/pasid tables and domain ID allocation. There's no use case to change these resources in any interrupt context. Therefore, it is unnecessary to disable the interrupts when the spinlock is held. The same thing happens on the device_domain_lock side, which protects the device domain attachment information. This replaces spin_lock/unlock_irqsave/irqrestore() calls with the normal spin_lock/unlock().

---
 drivers/iommu/intel/debugfs.c |  6 ++--
 drivers/iommu/intel/iommu.c   | 66 ++++++++++++++---------------------
 drivers/iommu/intel/svm.c     |  6 ++--
 3 files changed, 32 insertions(+), 46 deletions(-)
drivers/iommu/intel/debugfs.c +2 -4
index 6e1a3f88abc8..1f925285104e 100644
@@ -263,10 +263,9 @@ static void ctx_tbl_walk(struct seq_file *m, struct intel_iommu *iommu, u16 bus)
 
 static void root_tbl_walk(struct seq_file *m, struct intel_iommu *iommu)
 {
-	unsigned long flags;
 	u16 bus;
 
-	spin_lock_irqsave(&iommu->lock, flags);
+	spin_lock(&iommu->lock);
 	seq_printf(m, "IOMMU %s: Root Table Address: 0x%llx\n", iommu->name,
 		   (u64)virt_to_phys(iommu->root_entry));
 	seq_puts(m, "B.D.F\tRoot_entry\t\t\t\tContext_entry\t\t\t\tPASID\tPASID_table_entry\n");
@@ -278,8 +277,7 @@ static void root_tbl_walk(struct seq_file *m, struct intel_iommu *iommu)
 	 */
 	for (bus = 0; bus < 256; bus++)
 		ctx_tbl_walk(m, iommu, bus);
-
-	spin_unlock_irqrestore(&iommu->lock, flags);
+	spin_unlock(&iommu->lock);
 }
 
 static int dmar_translation_struct_show(struct seq_file *m, void *unused)
drivers/iommu/intel/iommu.c +27 -39

241 lines not on this page · raw message

drivers/iommu/intel/svm.c +3 -3
index 580713aa9e07..82288a50660d 100644
@@ -328,9 +328,9 @@ static struct iommu_sva *intel_svm_bind_mm(struct intel_iommu *iommu,
 					   unsigned int flags)
 {
 	struct device_domain_info *info = dev_iommu_priv_get(dev);
-	unsigned long iflags, sflags;
 	struct intel_svm_dev *sdev;
 	struct intel_svm *svm;
+	unsigned long sflags;
 	int ret = 0;
 
 	svm = pasid_private_find(mm->pasid);
@@ -394,10 +394,10 @@ static struct iommu_sva *intel_svm_bind_mm(struct intel_iommu *iommu,
 	sflags = (flags & SVM_FLAG_SUPERVISOR_MODE) ?
 			PASID_FLAG_SUPERVISOR_MODE : 0;
 	sflags |= cpu_feature_enabled(X86_FEATURE_LA57) ? PASID_FLAG_FL5LP : 0;
-	spin_lock_irqsave(&iommu->lock, iflags);
+	spin_lock(&iommu->lock);
 	ret = intel_pasid_setup_first_level(iommu, dev, mm->pgd, mm->pasid,
 					    FLPT_DEFAULT_DID, sflags);
-	spin_unlock_irqrestore(&iommu->lock, iflags);
+	spin_unlock(&iommu->lock);
 
 	if (ret)
 		goto free_sdev;
Signature · 1 line
2.25.1
author maintainer
#14 lore raw

PATCH 14/25 iommu/vt-d: Acquiring lock in domain ID allocation helpers

In reply to Lu Baolu

The iommu->lock is used to protect the per-IOMMU domain ID resource. Moving the lock into the ID alloc/free helpers makes the code more compact. At the same time, the device_domain_lock is irrelevant to the domain ID resource, remove its assertion as well.

---
 drivers/iommu/intel/iommu.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)
drivers/iommu/intel/iommu.c +9 -15
index 93f01082dce1..7f03576e72d7 100644
@@ -1779,16 +1779,13 @@ static struct dmar_domain *alloc_domain(unsigned int type)
 	return domain;
 }
 
-/* Must be called with iommu->lock */
 static int domain_attach_iommu(struct dmar_domain *domain,
 			       struct intel_iommu *iommu)
 {
 	unsigned long ndomains;
-	int num;
-
-	assert_spin_locked(&device_domain_lock);
-	assert_spin_locked(&iommu->lock);
+	int num, ret = 0;
 
+	spin_lock(&iommu->lock);
 	domain->iommu_refcnt[iommu->seq_id] += 1;
 	if (domain->iommu_refcnt[iommu->seq_id] == 1) {
 		ndomains = cap_ndoms(iommu->cap);
@@ -1797,7 +1794,8 @@ static int domain_attach_iommu(struct dmar_domain *domain,
 		if (num >= ndomains) {
 			pr_err("%s: No free domain ids\n", iommu->name);
 			domain->iommu_refcnt[iommu->seq_id] -= 1;
-			return -ENOSPC;
+			ret = -ENOSPC;
+			goto out_unlock;
 		}
 
 		set_bit(num, iommu->domain_ids);
@@ -1806,7 +1804,9 @@ static int domain_attach_iommu(struct dmar_domain *domain,
 		domain_update_iommu_cap(domain);
 	}
 
-	return 0;
+out_unlock:
+	spin_unlock(&iommu->lock);
+	return ret;
 }
 
 static void domain_detach_iommu(struct dmar_domain *domain,
@@ -1814,9 +1814,7 @@ static void domain_detach_iommu(struct dmar_domain *domain,
 {
 	int num;
 
-	assert_spin_locked(&device_domain_lock);
-	assert_spin_locked(&iommu->lock);
-
+	spin_lock(&iommu->lock);
 	domain->iommu_refcnt[iommu->seq_id] -= 1;
 	if (domain->iommu_refcnt[iommu->seq_id] == 0) {
 		num = domain->iommu_did[iommu->seq_id];
@@ -1824,6 +1822,7 @@ static void domain_detach_iommu(struct dmar_domain *domain,
 		domain_update_iommu_cap(domain);
 		domain->iommu_did[iommu->seq_id] = 0;
 	}
+	spin_unlock(&iommu->lock);
 }
 
 static inline int guestwidth_to_adjustwidth(int gaw)
@@ -2472,9 +2471,7 @@ static int domain_add_dev_info(struct dmar_domain *domain, struct device *dev)
 
 	spin_lock(&device_domain_lock);
 	info->domain = domain;
-	spin_lock(&iommu->lock);
 	ret = domain_attach_iommu(domain, iommu);
-	spin_unlock(&iommu->lock);
 	if (ret) {
 		spin_unlock(&device_domain_lock);
 		return ret;
@@ -4178,10 +4175,7 @@ static void __dmar_remove_one_dev_info(struct device_domain_info *info)
 	}
 
 	list_del(&info->link);
-
-	spin_lock(&iommu->lock);
 	domain_detach_iommu(domain, iommu);
-	spin_unlock(&iommu->lock);
 }
 
 static void dmar_remove_one_dev_info(struct device *dev)
Signature · 1 line
2.25.1
author maintainer
#15 lore raw

PATCH 15/25 iommu/vt-d: Acquiring lock in pasid manipulation helpers

In reply to Lu Baolu

The iommu->lock is used to protect the per-IOMMU pasid directory table and pasid table. Move the spinlock acquisition/release into the helpers to make the code self-contained.

---
 drivers/iommu/intel/iommu.c |   2 -
 drivers/iommu/intel/pasid.c | 103 +++++++++++++++++++-----------------
 drivers/iommu/intel/svm.c   |   3 --
 3 files changed, 55 insertions(+), 53 deletions(-)
drivers/iommu/intel/iommu.c +0 -2
index 7f03576e72d7..3d53de8c7634 100644
@@ -2489,7 +2489,6 @@ static int domain_add_dev_info(struct dmar_domain *domain, struct device *dev)
 		}
 
 		/* Setup the PASID entry for requests without PASID: */
-		spin_lock(&iommu->lock);
 		if (hw_pass_through && domain_type_is_si(domain))
 			ret = intel_pasid_setup_pass_through(iommu, domain,
 					dev, PASID_RID2PASID);
@@ -2499,7 +2498,6 @@ static int domain_add_dev_info(struct dmar_domain *domain, struct device *dev)
 		else
 			ret = intel_pasid_setup_second_level(iommu, domain,
 					dev, PASID_RID2PASID);
-		spin_unlock(&iommu->lock);
 		if (ret) {
 			dev_err(dev, "Setup RID2PASID failed\n");
 			dmar_remove_one_dev_info(dev);
drivers/iommu/intel/pasid.c +55 -48

181 lines not on this page · raw message

drivers/iommu/intel/svm.c +0 -3
index 82288a50660d..64072e628bbd 100644
@@ -394,11 +394,8 @@ static struct iommu_sva *intel_svm_bind_mm(struct intel_iommu *iommu,
 	sflags = (flags & SVM_FLAG_SUPERVISOR_MODE) ?
 			PASID_FLAG_SUPERVISOR_MODE : 0;
 	sflags |= cpu_feature_enabled(X86_FEATURE_LA57) ? PASID_FLAG_FL5LP : 0;
-	spin_lock(&iommu->lock);
 	ret = intel_pasid_setup_first_level(iommu, dev, mm->pgd, mm->pasid,
 					    FLPT_DEFAULT_DID, sflags);
-	spin_unlock(&iommu->lock);
-
 	if (ret)
 		goto free_sdev;
 
Signature · 1 line
2.25.1
author maintainer
#16 lore raw

PATCH 16/25 iommu/vt-d: Check device list of domain in domain free path

In reply to Lu Baolu

When the IOMMU domain is about to be freed, it should not be set on any device. Instead of silently dealing with some bug cases, it's better to trigger a warning to report and fix any potential bugs at the first time.

---
 drivers/iommu/intel/iommu.c | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)
drivers/iommu/intel/iommu.c +3 -15
index 3d53de8c7634..a94fb69e1f9a 100644
@@ -294,7 +294,6 @@ static LIST_HEAD(dmar_satc_units);
 /* bitmap for indexing intel_iommus */
 static int g_num_of_iommus;
 
-static void domain_remove_dev_info(struct dmar_domain *domain);
 static void dmar_remove_one_dev_info(struct device *dev);
 static void __dmar_remove_one_dev_info(struct device_domain_info *info);
 
@@ -1841,10 +1840,6 @@ static inline int guestwidth_to_adjustwidth(int gaw)
 
 static void domain_exit(struct dmar_domain *domain)
 {
-
-	/* Remove associated devices and clear attached or cached domains */
-	domain_remove_dev_info(domain);
-
 	if (domain->pgd) {
 		LIST_HEAD(freelist);
 
@@ -1852,6 +1847,9 @@ static void domain_exit(struct dmar_domain *domain)
 		put_pages_list(&freelist);
 	}
 
+	if (WARN_ON(!list_empty(&domain->devices)))
+		return;
+
 	kfree(domain);
 }
 
@@ -2333,16 +2331,6 @@ static void domain_context_clear_one(struct device_domain_info *info, u8 bus, u8
 	__iommu_flush_dev_iotlb(info, 0, MAX_AGAW_PFN_WIDTH);
 }
 
-static void domain_remove_dev_info(struct dmar_domain *domain)
-{
-	struct device_domain_info *info, *tmp;
-
-	spin_lock(&device_domain_lock);
-	list_for_each_entry_safe(info, tmp, &domain->devices, link)
-		__dmar_remove_one_dev_info(info);
-	spin_unlock(&device_domain_lock);
-}
-
 static int domain_setup_first_level(struct intel_iommu *iommu,
 				    struct dmar_domain *domain,
 				    struct device *dev,
Signature · 1 line
2.25.1
author maintainer
#17 lore raw

PATCH 17/25 iommu/vt-d: Fold __dmar_remove_one_dev_info() into its caller

In reply to Lu Baolu

Fold __dmar_remove_one_dev_info() into dmar_remove_one_dev_info() which is its only caller. Make the spin lock critical range only cover the device list change code and remove some unnecessary checks.

---
 drivers/iommu/intel/iommu.c | 32 ++++++++------------------------
 1 file changed, 8 insertions(+), 24 deletions(-)
drivers/iommu/intel/iommu.c +8 -24
index a94fb69e1f9a..ccddf634fae9 100644
@@ -295,7 +295,6 @@ static LIST_HEAD(dmar_satc_units);
 static int g_num_of_iommus;
 
 static void dmar_remove_one_dev_info(struct device *dev);
-static void __dmar_remove_one_dev_info(struct device_domain_info *info);
 
 int dmar_disabled = !IS_ENABLED(CONFIG_INTEL_IOMMU_DEFAULT_ON);
 int intel_iommu_sm = IS_ENABLED(CONFIG_INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON);
@@ -4137,20 +4136,12 @@ static void domain_context_clear(struct device_domain_info *info)
 			       &domain_context_clear_one_cb, info);
 }
 
-static void __dmar_remove_one_dev_info(struct device_domain_info *info)
+static void dmar_remove_one_dev_info(struct device *dev)
 {
-	struct dmar_domain *domain;
-	struct intel_iommu *iommu;
-
-	assert_spin_locked(&device_domain_lock);
-
-	if (WARN_ON(!info))
-		return;
-
-	iommu = info->iommu;
-	domain = info->domain;
+	struct device_domain_info *info = dev_iommu_priv_get(dev);
+	struct intel_iommu *iommu = info->iommu;
 
-	if (info->dev && !dev_is_real_dma_subdevice(info->dev)) {
+	if (!dev_is_real_dma_subdevice(info->dev)) {
 		if (dev_is_pci(info->dev) && sm_supported(iommu))
 			intel_pasid_tear_down_entry(iommu, info->dev,
 					PASID_RID2PASID, false);
@@ -4160,19 +4151,12 @@ static void __dmar_remove_one_dev_info(struct device_domain_info *info)
 		intel_pasid_free_table(info->dev);
 	}
 
-	list_del(&info->link);
-	domain_detach_iommu(domain, iommu);
-}
-
-static void dmar_remove_one_dev_info(struct device *dev)
-{
-	struct device_domain_info *info;
-
 	spin_lock(&device_domain_lock);
-	info = dev_iommu_priv_get(dev);
-	if (info)
-		__dmar_remove_one_dev_info(info);
+	list_del(&info->link);
 	spin_unlock(&device_domain_lock);
+
+	domain_detach_iommu(info->domain, iommu);
+	info->domain = NULL;
 }
 
 static int md_domain_init(struct dmar_domain *domain, int guest_width)
Signature · 1 line
2.25.1
author maintainer
#18 lore raw

PATCH 18/25 iommu/vt-d: Use device_domain_lock accurately

In reply to Lu Baolu

The device_domain_lock is used to protect the device tracking list of a domain. Remove unnecessary spin_lock/unlock()'s and move the necessary ones around the list access.

---
 drivers/iommu/intel/iommu.c | 57 +++++++++++++------------------------
 1 file changed, 19 insertions(+), 38 deletions(-)
drivers/iommu/intel/iommu.c +19 -38
index ccddf634fae9..ae42a9e43997 100644
@@ -535,15 +535,8 @@ static int domain_update_device_node(struct dmar_domain *domain)
 	struct device_domain_info *info;
 	int nid = NUMA_NO_NODE;
 
-	assert_spin_locked(&device_domain_lock);
-
-	if (list_empty(&domain->devices))
-		return NUMA_NO_NODE;
-
+	spin_lock(&device_domain_lock);
 	list_for_each_entry(info, &domain->devices, link) {
-		if (!info->dev)
-			continue;
-
 		/*
 		 * There could possibly be multiple device numa nodes as devices
 		 * within the same domain may sit behind different IOMMUs. There
@@ -554,6 +547,7 @@ static int domain_update_device_node(struct dmar_domain *domain)
 		if (nid != NUMA_NO_NODE)
 			break;
 	}
+	spin_unlock(&device_domain_lock);
 
 	return nid;
 }
@@ -1376,23 +1370,23 @@ static void __iommu_flush_iotlb(struct intel_iommu *iommu, u16 did,
 }
 
 static struct device_domain_info *
-iommu_support_dev_iotlb (struct dmar_domain *domain, struct intel_iommu *iommu,
-			 u8 bus, u8 devfn)
+iommu_support_dev_iotlb(struct dmar_domain *domain, struct intel_iommu *iommu,
+			u8 bus, u8 devfn)
 {
 	struct device_domain_info *info;
 
-	assert_spin_locked(&device_domain_lock);
-
 	if (!iommu->qi)
 		return NULL;
 
-	list_for_each_entry(info, &domain->devices, link)
+	spin_lock(&device_domain_lock);
+	list_for_each_entry(info, &domain->devices, link) {
 		if (info->iommu == iommu && info->bus == bus &&
 		    info->devfn == devfn) {
-			if (info->ats_supported && info->dev)
-				return info;
-			break;
+			spin_unlock(&device_domain_lock);
+			return info->ats_supported ? info : NULL;
 		}
+	}
+	spin_unlock(&device_domain_lock);
 
 	return NULL;
 }
@@ -1402,23 +1396,21 @@ static void domain_update_iotlb(struct dmar_domain *domain)
 	struct device_domain_info *info;
 	bool has_iotlb_device = false;
 
-	assert_spin_locked(&device_domain_lock);
-
-	list_for_each_entry(info, &domain->devices, link)
+	spin_lock(&device_domain_lock);
+	list_for_each_entry(info, &domain->devices, link) {
 		if (info->ats_enabled) {
 			has_iotlb_device = true;
 			break;
 		}
-
+	}
 	domain->has_iotlb_device = has_iotlb_device;
+	spin_unlock(&device_domain_lock);
 }
 
 static void iommu_enable_dev_iotlb(struct device_domain_info *info)
 {
 	struct pci_dev *pdev;
 
-	assert_spin_locked(&device_domain_lock);
-
 	if (!info || !dev_is_pci(info->dev))
 		return;
 
@@ -1464,8 +1456,6 @@ static void iommu_disable_dev_iotlb(struct device_domain_info *info)
 {
 	struct pci_dev *pdev;
 
-	assert_spin_locked(&device_domain_lock);
-
 	if (!dev_is_pci(info->dev))
 		return;
 
@@ -1906,9 +1896,10 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 				      struct pasid_table *table,
 				      u8 bus, u8 devfn)
 {
+	struct device_domain_info *info =
+			iommu_support_dev_iotlb(domain, iommu, bus, devfn);
 	u16 did = domain->iommu_did[iommu->seq_id];
 	int translation = CONTEXT_TT_MULTI_LEVEL;
-	struct device_domain_info *info = NULL;
 	struct context_entry *context;
 	int ret;
 
@@ -1922,9 +1913,7 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 
 	BUG_ON(!domain->pgd);
 
-	spin_lock(&device_domain_lock);
 	spin_lock(&iommu->lock);
-
 	ret = -ENOMEM;
 	context = iommu_context_addr(iommu, bus, devfn, 1);
 	if (!context)
@@ -1975,7 +1964,6 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 		 * Setup the Device-TLB enable bit and Page request
 		 * Enable bit:
 		 */
-		info = iommu_support_dev_iotlb(domain, iommu, bus, devfn);
 		if (info && info->ats_supported)
 			context_set_sm_dte(context);
 		if (info && info->pri_supported)
@@ -1998,7 +1986,6 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 					goto out_unlock;
 			}
 
-			info = iommu_support_dev_iotlb(domain, iommu, bus, devfn);
 			if (info && info->ats_supported)
 				translation = CONTEXT_TT_DEV_IOTLB;
 			else
@@ -2044,7 +2031,6 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
 
 out_unlock:
 	spin_unlock(&iommu->lock);
-	spin_unlock(&device_domain_lock);
 
 	return ret;
 }
@@ -2456,13 +2442,11 @@ static int domain_add_dev_info(struct dmar_domain *domain, struct device *dev)
 	if (!iommu)
 		return -ENODEV;
 
-	spin_lock(&device_domain_lock);
-	info->domain = domain;
 	ret = domain_attach_iommu(domain, iommu);
-	if (ret) {
-		spin_unlock(&device_domain_lock);
+	if (ret)
 		return ret;
-	}
+	info->domain = domain;
+	spin_lock(&device_domain_lock);
 	list_add(&info->link, &domain->devices);
 	spin_unlock(&device_domain_lock);
 
@@ -4638,9 +4622,7 @@ int intel_iommu_enable_pasid(struct intel_iommu *iommu, struct device *dev)
 	if (!domain)
 		return -EINVAL;
 
-	spin_lock(&device_domain_lock);
 	spin_lock(&iommu->lock);
-
 	ret = -EINVAL;
 	if (!info->pasid_supported)
 		goto out;
@@ -4670,7 +4652,6 @@ int intel_iommu_enable_pasid(struct intel_iommu *iommu, struct device *dev)
 
  out:
 	spin_unlock(&iommu->lock);
-	spin_unlock(&device_domain_lock);
 
 	return ret;
 }
Signature · 1 line
2.25.1
author maintainer
#19 lore raw

PATCH 19/25 iommu/vt-d: Convert global spinlock into per domain lock

In reply to Lu Baolu

Using a global device_domain_lock spinlock to protect per-domain device tracking lists is an inefficient way, especially considering this lock is also needed in the hot paths. This optimizes the locking mechanism by converting the global lock to per domain lock.

On the other hand, as the device tracking lists are never accessed in any interrupt context, there is no need to disable interrupts while spinning. Replace irqsave variant with spinlock calls.

---
 drivers/iommu/intel/iommu.h |  1 +
 drivers/iommu/intel/iommu.c | 42 ++++++++++++++++++-------------------
 2 files changed, 22 insertions(+), 21 deletions(-)
drivers/iommu/intel/iommu.h +1 -0
index 198c6c822ef4..df64d3d9c49a 100644
@@ -541,6 +541,7 @@ struct dmar_domain {
 	u8 force_snooping : 1;		/* Create IOPTEs with snoop control */
 	u8 set_pte_snp:1;
 
+	spinlock_t lock;		/* Protect device tracking lists */
 	struct list_head devices;	/* all devices' list */
 
 	struct dma_pte	*pgd;		/* virtual address */
drivers/iommu/intel/iommu.c +21 -21

143 lines not on this page · raw message

Signature · 1 line
2.25.1
author maintainer
#20 lore raw

PATCH 20/25 iommu/vt-d: Remove unused domain_get_iommu()

In reply to Lu Baolu

It is not used anywhere. Remove it to avoid dead code.

---
 drivers/iommu/intel/iommu.h |  1 -
 drivers/iommu/intel/iommu.c | 18 ------------------
 2 files changed, 19 deletions(-)
drivers/iommu/intel/iommu.h +0 -1
index df64d3d9c49a..56e0d8cd2102 100644
@@ -725,7 +725,6 @@ extern int dmar_ir_support(void);
 
 void *alloc_pgtable_page(int node);
 void free_pgtable_page(void *vaddr);
-struct intel_iommu *domain_get_iommu(struct dmar_domain *domain);
 void iommu_flush_write_buffer(struct intel_iommu *iommu);
 int intel_iommu_enable_pasid(struct intel_iommu *iommu, struct device *dev);
 struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devfn);
drivers/iommu/intel/iommu.c +0 -18
index 769e850b9767..55a055a8ed4b 100644
@@ -445,24 +445,6 @@ int iommu_calculate_agaw(struct intel_iommu *iommu)
 	return __iommu_calculate_agaw(iommu, DEFAULT_DOMAIN_ADDRESS_WIDTH);
 }
 
-/* This functionin only returns single iommu in a domain */
-struct intel_iommu *domain_get_iommu(struct dmar_domain *domain)
-{
-	int iommu_id;
-
-	/* si_domain and vm domain should not get here. */
-	if (WARN_ON(!iommu_is_dma_domain(&domain->domain)))
-		return NULL;
-
-	for_each_domain_iommu(iommu_id, domain)
-		break;
-
-	if (iommu_id < 0 || iommu_id >= g_num_of_iommus)
-		return NULL;
-
-	return g_iommus[iommu_id];
-}
-
 static inline bool iommu_paging_structure_coherency(struct intel_iommu *iommu)
 {
 	return sm_supported(iommu) ?
Signature · 1 line
2.25.1
author maintainer
#21 lore raw

PATCH 21/25 iommu/vt-d: Use IDA interface to manage iommu sequence id

In reply to Lu Baolu

Switch dmar unit sequence id allocation and release from bitmap to IDA interface.

---
 drivers/iommu/intel/dmar.c | 35 ++++++++---------------------------
 1 file changed, 8 insertions(+), 27 deletions(-)
drivers/iommu/intel/dmar.c +8 -27
index 2a5e0f91e647..6327b34f5aa7 100644
@@ -60,7 +60,7 @@ LIST_HEAD(dmar_drhd_units);
 
 struct acpi_table_header * __initdata dmar_tbl;
 static int dmar_dev_scope_status = 1;
-static unsigned long dmar_seq_ids[BITS_TO_LONGS(DMAR_UNITS_SUPPORTED)];
+static DEFINE_IDA(dmar_seq_ids);
 
 static int alloc_iommu(struct dmar_drhd_unit *drhd);
 static void free_iommu(struct intel_iommu *iommu);
@@ -1023,28 +1023,6 @@ static int map_iommu(struct intel_iommu *iommu, u64 phys_addr)
 	return err;
 }
 
-static int dmar_alloc_seq_id(struct intel_iommu *iommu)
-{
-	iommu->seq_id = find_first_zero_bit(dmar_seq_ids,
-					    DMAR_UNITS_SUPPORTED);
-	if (iommu->seq_id >= DMAR_UNITS_SUPPORTED) {
-		iommu->seq_id = -1;
-	} else {
-		set_bit(iommu->seq_id, dmar_seq_ids);
-		sprintf(iommu->name, "dmar%d", iommu->seq_id);
-	}
-
-	return iommu->seq_id;
-}
-
-static void dmar_free_seq_id(struct intel_iommu *iommu)
-{
-	if (iommu->seq_id >= 0) {
-		clear_bit(iommu->seq_id, dmar_seq_ids);
-		iommu->seq_id = -1;
-	}
-}
-
 static int alloc_iommu(struct dmar_drhd_unit *drhd)
 {
 	struct intel_iommu *iommu;
@@ -1062,11 +1040,14 @@ static int alloc_iommu(struct dmar_drhd_unit *drhd)
 	if (!iommu)
 		return -ENOMEM;
 
-	if (dmar_alloc_seq_id(iommu) < 0) {
+	iommu->seq_id = ida_alloc_range(&dmar_seq_ids, 0,
+					DMAR_UNITS_SUPPORTED - 1, GFP_KERNEL);
+	if (iommu->seq_id < 0) {
 		pr_err("Failed to allocate seq_id\n");
-		err = -ENOSPC;
+		err = iommu->seq_id;
 		goto error;
 	}
+	sprintf(iommu->name, "dmar%d", iommu->seq_id);
 
 	err = map_iommu(iommu, drhd->reg_base_addr);
 	if (err) {
@@ -1150,7 +1131,7 @@ static int alloc_iommu(struct dmar_drhd_unit *drhd)
 err_unmap:
 	unmap_iommu(iommu);
 error_free_seq_id:
-	dmar_free_seq_id(iommu);
+	ida_free(&dmar_seq_ids, iommu->seq_id);
 error:
 	kfree(iommu);
 	return err;
@@ -1183,7 +1164,7 @@ static void free_iommu(struct intel_iommu *iommu)
 	if (iommu->reg)
 		unmap_iommu(iommu);
 
-	dmar_free_seq_id(iommu);
+	ida_free(&dmar_seq_ids, iommu->seq_id);
 	kfree(iommu);
 }
 
Signature · 1 line
2.25.1
author maintainer
#22 lore raw

PATCH 22/25 iommu/vt-d: Refactor iommu information of each domain

In reply to Lu Baolu

When a DMA domain is attached to a device, it needs to allocate a domain ID from its IOMMU. Currently, the domain ID information is stored in two static arrays embedded in the domain structure. This can lead to memory waste when the driver is running on a small platform.

This optimizes these static arrays by replacing them with an xarray and consuming memory on demand.

---
 drivers/iommu/intel/iommu.h |  29 ++++++---
 drivers/iommu/intel/iommu.c | 123 +++++++++++++++++++++---------------
 drivers/iommu/intel/pasid.c |   2 +-
 drivers/iommu/intel/svm.c   |   2 +-
 4 files changed, 93 insertions(+), 63 deletions(-)
drivers/iommu/intel/iommu.h +20 -9
index 56e0d8cd2102..fae45bbb0c7f 100644
@@ -21,6 +21,7 @@
 #include <linux/dmar.h>
 #include <linux/ioasid.h>
 #include <linux/bitfield.h>
+#include <linux/xarray.h>
 
 #include <asm/cacheflush.h>
 #include <asm/iommu.h>
@@ -524,17 +525,17 @@ struct context_entry {
  */
 #define DOMAIN_FLAG_USE_FIRST_LEVEL		BIT(1)
 
-struct dmar_domain {
-	int	nid;			/* node id */
-
-	unsigned int iommu_refcnt[DMAR_UNITS_SUPPORTED];
-					/* Refcount of devices per iommu */
-
-
-	u16		iommu_did[DMAR_UNITS_SUPPORTED];
-					/* Domain ids per IOMMU. Use u16 since
+struct iommu_domain_info {
+	struct intel_iommu *iommu;
+	unsigned int refcnt;		/* Refcount of devices per iommu */
+	u16 did;			/* Domain ids per IOMMU. Use u16 since
 					 * domain ids are 16 bit wide according
 					 * to VT-d spec, section 9.3 */
+};
+
+struct dmar_domain {
+	int	nid;			/* node id */
+	struct xarray iommu_array;	/* Attached IOMMU array */
 
 	u8 has_iotlb_device: 1;
 	u8 iommu_coherency: 1;		/* indicate coherency of iommu access */
@@ -640,6 +641,16 @@ static inline struct dmar_domain *to_dmar_domain(struct iommu_domain *dom)
 	return container_of(dom, struct dmar_domain, domain);
 }
 
+/* Retrieve the domain ID which has allocated to the domain */
+static inline u16
+domain_id_iommu(struct dmar_domain *domain, struct intel_iommu *iommu)
+{
+	struct iommu_domain_info *info =
+			xa_load(&domain->iommu_array, iommu->seq_id);
+
+	return info->did;
+}
+
 /*
  * 0: readable
  * 1: writable
drivers/iommu/intel/iommu.c +71 -52

238 lines not on this page · raw message

drivers/iommu/intel/pasid.c +1 -1
index 7792a1b2ebc4..c5e7e8b020a5 100644
@@ -626,7 +626,7 @@ int intel_pasid_setup_second_level(struct intel_iommu *iommu,
 	}
 
 	pgd_val = virt_to_phys(pgd);
-	did = domain->iommu_did[iommu->seq_id];
+	did = domain_id_iommu(domain, iommu);
 
 	spin_lock(&iommu->lock);
 	pte = intel_pasid_get_entry(dev, pasid);
drivers/iommu/intel/svm.c +1 -1
index 64072e628bbd..8bcfb93dda56 100644
@@ -541,7 +541,7 @@ static void intel_svm_drain_prq(struct device *dev, u32 pasid)
 	domain = info->domain;
 	pdev = to_pci_dev(dev);
 	sid = PCI_DEVID(info->bus, info->devfn);
-	did = domain->iommu_did[iommu->seq_id];
+	did = domain_id_iommu(domain, iommu);
 	qdep = pci_ats_queue_depth(pdev);
 
 	/*
Signature · 1 line
2.25.1
author maintainer
#23 lore raw

PATCH 23/25 iommu/vt-d: Remove unnecessary check in intel_iommu_add()

In reply to Lu Baolu

The Intel IOMMU hot-add process starts from dmar_device_hotplug(). It uses the global dmar_global_lock to synchronize all the hot-add and hot-remove paths. In the hot-add path, the new IOMMU data structures are allocated firstly by dmar_parse_one_drhd() and then initialized by dmar_hp_add_drhd(). All the IOMMU units are allocated and initialized in the same synchronized path. There is no case where any IOMMU unit is created and then initialized for multiple times.

This removes the unnecessary check in intel_iommu_add() which is the last reference place of the global IOMMU array.

---
 drivers/iommu/intel/iommu.c | 3 ---
 1 file changed, 3 deletions(-)
drivers/iommu/intel/iommu.c +0 -3
index 36b3649a83e8..2f64ff6e6e75 100644
@@ -3460,9 +3460,6 @@ static int intel_iommu_add(struct dmar_drhd_unit *dmaru)
 	int sp, ret;
 	struct intel_iommu *iommu = dmaru->iommu;
 
-	if (g_iommus[iommu->seq_id])
-		return 0;
-
 	ret = intel_cap_audit(CAP_AUDIT_HOTPLUG_DMAR, iommu);
 	if (ret)
 		goto out;
Signature · 1 line
2.25.1
author maintainer
#24 lore raw

PATCH 24/25 iommu/vt-d: Remove global g_iommus array

In reply to Lu Baolu

The g_iommus and g_num_of_iommus is not used anywhere. Remove them to avoid dead code.

---
 drivers/iommu/intel/iommu.c | 44 -------------------------------------
 1 file changed, 44 deletions(-)
drivers/iommu/intel/iommu.c +0 -44
index 2f64ff6e6e75..3ba0d7e76efd 100644
@@ -126,9 +126,6 @@ static inline unsigned long virt_to_dma_pfn(void *p)
 	return page_to_dma_pfn(virt_to_page(p));
 }
 
-/* global iommu list, set NULL for ignored DMAR units */
-static struct intel_iommu **g_iommus;
-
 static void __init check_tylersburg_isoch(void);
 static int rwbf_quirk;
 
@@ -287,9 +284,6 @@ static LIST_HEAD(dmar_satc_units);
 #define for_each_rmrr_units(rmrr) \
 	list_for_each_entry(rmrr, &dmar_rmrr_units, list)
 
-/* bitmap for indexing intel_iommus */
-static int g_num_of_iommus;
-
 static void dmar_remove_one_dev_info(struct device *dev);
 
 int dmar_disabled = !IS_ENABLED(CONFIG_INTEL_IOMMU_DEFAULT_ON);
@@ -1694,8 +1688,6 @@ static void free_dmar_iommu(struct intel_iommu *iommu)
 		iommu->domain_ids = NULL;
 	}
 
-	g_iommus[iommu->seq_id] = NULL;
-
 	/* free context mapping */
 	free_context_table(iommu);
 
@@ -2898,36 +2890,6 @@ static int __init init_dmars(void)
 	struct intel_iommu *iommu;
 	int ret;
 
-	/*
-	 * for each drhd
-	 *    allocate root
-	 *    initialize and program root entry to not present
-	 * endfor
-	 */
-	for_each_drhd_unit(drhd) {
-		/*
-		 * lock not needed as this is only incremented in the single
-		 * threaded kernel __init code path all other access are read
-		 * only
-		 */
-		if (g_num_of_iommus < DMAR_UNITS_SUPPORTED) {
-			g_num_of_iommus++;
-			continue;
-		}
-		pr_err_once("Exceeded %d IOMMUs\n", DMAR_UNITS_SUPPORTED);
-	}
-
-	/* Preallocate enough resources for IOMMU hot-addition */
-	if (g_num_of_iommus < DMAR_UNITS_SUPPORTED)
-		g_num_of_iommus = DMAR_UNITS_SUPPORTED;
-
-	g_iommus = kcalloc(g_num_of_iommus, sizeof(struct intel_iommu *),
-			GFP_KERNEL);
-	if (!g_iommus) {
-		ret = -ENOMEM;
-		goto error;
-	}
-
 	ret = intel_cap_audit(CAP_AUDIT_STATIC_DMAR, NULL);
 	if (ret)
 		goto free_iommu;
@@ -2950,8 +2912,6 @@ static int __init init_dmars(void)
 						   intel_pasid_max_id);
 		}
 
-		g_iommus[iommu->seq_id] = iommu;
-
 		intel_iommu_init_qi(iommu);
 
 		ret = iommu_init_domains(iommu);
@@ -3077,9 +3037,6 @@ static int __init init_dmars(void)
 		free_dmar_iommu(iommu);
 	}
 
-	kfree(g_iommus);
-
-error:
 	return ret;
 }
 
@@ -3483,7 +3440,6 @@ static int intel_iommu_add(struct dmar_drhd_unit *dmaru)
 	if (iommu->gcmd & DMA_GCMD_TE)
 		iommu_disable_translation(iommu);
 
-	g_iommus[iommu->seq_id] = iommu;
 	ret = iommu_init_domains(iommu);
 	if (ret == 0)
 		ret = iommu_alloc_root_entry(iommu);
Signature · 1 line
2.25.1
author maintainer
#25 lore raw

PATCH 25/25 iommu/vt-d: Make DMAR_UNITS_SUPPORTED default 1024

In reply to Lu Baolu

If the available hardware exceeds DMAR_UNITS_SUPPORTED (previously set to MAX_IO_APICS, or 128), it causes these messages: "DMAR: Failed to allocate seq_id", "DMAR: Parse DMAR table failure.", and "x2apic: IRQ remapping doesn't support X2APIC mode x2apic disabled"; and the system fails to boot properly.

To support up to 64 sockets with 10 DMAR units each (640), make the value of DMAR_UNITS_SUPPORTED default 1024.

---
 include/linux/dmar.h | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
include/linux/dmar.h +1 -5
index cbd714a198a0..d81a51978d01 100644
@@ -18,11 +18,7 @@
 
 struct acpi_dmar_header;
 
-#ifdef	CONFIG_X86
-# define	DMAR_UNITS_SUPPORTED	MAX_IO_APICS
-#else
-# define	DMAR_UNITS_SUPPORTED	64
-#endif
+#define DMAR_UNITS_SUPPORTED	1024
 
 /* DMAR Flags */
 #define DMAR_INTR_REMAP		0x1
Signature · 1 line
2.25.1