From 4c5502b1c5744b2090414e1b80ca6388d5c46e06 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:53 -0700
Subject: x86, x2apic: fix lock ordering during IRQ migration

Impact: fix potential deadlock on x2apic

fix "hard-safe -> hard-unsafe lock order detected" with irq_2_ir_lock

On x2apic enabled system:
   [ INFO: hard-safe -> hard-unsafe lock order detected ]
   2.6.27-03151-g4480f15b #1
   ------------------------------------------------------
   swapper/1 [HC0[0]:SC0[0]:HE0:SE1] is trying to acquire:
    (irq_2_ir_lock){--..}, at: [<ffffffff8038ebc0>] get_irte+0x2f/0x95

   and this task is already holding:
    (&irq_desc_lock_class){+...}, at: [<ffffffff802649ed>] setup_irq+0x67/0x281
   which would create a new lock dependency:
    (&irq_desc_lock_class){+...} -> (irq_2_ir_lock){--..}

   but this new dependency connects a hard-irq-safe lock:
    (&irq_desc_lock_class){+...}
   ... which became hard-irq-safe at:
     [<ffffffffffffffff>] 0xffffffffffffffff

   to a hard-irq-unsafe lock:
    (irq_2_ir_lock){--..}
   ... which became hard-irq-unsafe at:
   ...  [<ffffffff802547b5>] __lock_acquire+0x571/0x706
     [<ffffffff8025499f>] lock_acquire+0x55/0x71
     [<ffffffff8062f2c4>] _spin_lock+0x2c/0x38
     [<ffffffff8038ee50>] alloc_irte+0x8a/0x14b
     [<ffffffff8021f733>] setup_IO_APIC_irq+0x119/0x30e
     [<ffffffff8090860e>] setup_IO_APIC+0x146/0x6e5
     [<ffffffff809058fc>] native_smp_prepare_cpus+0x24e/0x2e9
     [<ffffffff808f982c>] kernel_init+0x5a/0x176
     [<ffffffff8020c289>] child_rip+0xa/0x11
     [<ffffffffffffffff>] 0xffffffffffffffff

Fix this theoretical lock order issue by using spin_lock_irqsave() instead of
spin_lock()

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/intr_remapping.c | 58 +++++++++++++++++++++++++-------------------
 1 file changed, 33 insertions(+), 25 deletions(-)

(limited to 'drivers')

diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 8e44db040db7..5ffa65fffb6a 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -117,21 +117,22 @@ int get_irte(int irq, struct irte *entry)
 {
 	int index;
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
 	if (!entry)
 		return -1;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 	*entry = *(irq_iommu->iommu->ir_table->base + index);
 
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 	return 0;
 }
 
@@ -141,6 +142,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 	struct irq_2_iommu *irq_iommu;
 	u16 index, start_index;
 	unsigned int mask = 0;
+	unsigned long flags;
 	int i;
 
 	if (!count)
@@ -170,7 +172,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 		return -1;
 	}
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	do {
 		for (i = index; i < index + count; i++)
 			if  (table->base[i].present)
@@ -182,7 +184,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 		index = (index + count) % INTR_REMAP_TABLE_ENTRIES;
 
 		if (index == start_index) {
-			spin_unlock(&irq_2_ir_lock);
+			spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 			printk(KERN_ERR "can't allocate an IRTE\n");
 			return -1;
 		}
@@ -193,7 +195,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 
 	irq_iommu = irq_2_iommu_alloc(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		printk(KERN_ERR "can't allocate irq_2_iommu\n");
 		return -1;
 	}
@@ -203,7 +205,7 @@ int alloc_irte(struct intel_iommu *iommu, int irq, u16 count)
 	irq_iommu->sub_handle = 0;
 	irq_iommu->irte_mask = mask;
 
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return index;
 }
@@ -223,30 +225,32 @@ int map_irq_to_irte_handle(int irq, u16 *sub_handle)
 {
 	int index;
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
 	*sub_handle = irq_iommu->sub_handle;
 	index = irq_iommu->irte_index;
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 	return index;
 }
 
 int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 {
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 
 	irq_iommu = irq_2_iommu_alloc(irq);
 
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		printk(KERN_ERR "can't allocate irq_2_iommu\n");
 		return -1;
 	}
@@ -256,7 +260,7 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 	irq_iommu->sub_handle = subhandle;
 	irq_iommu->irte_mask = 0;
 
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return 0;
 }
@@ -264,11 +268,12 @@ int set_irte_irq(int irq, struct intel_iommu *iommu, u16 index, u16 subhandle)
 int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
 {
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
@@ -277,7 +282,7 @@ int clear_irte_irq(int irq, struct intel_iommu *iommu, u16 index)
 	irq_iommu->sub_handle = 0;
 	irq_2_iommu(irq)->irte_mask = 0;
 
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return 0;
 }
@@ -289,11 +294,12 @@ int modify_irte(int irq, struct irte *irte_modified)
 	struct irte *irte;
 	struct intel_iommu *iommu;
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
@@ -306,7 +312,7 @@ int modify_irte(int irq, struct irte *irte_modified)
 	__iommu_flush_cache(iommu, irte, sizeof(*irte));
 
 	rc = qi_flush_iec(iommu, index, 0);
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return rc;
 }
@@ -317,11 +323,12 @@ int flush_irte(int irq)
 	int index;
 	struct intel_iommu *iommu;
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
@@ -330,7 +337,7 @@ int flush_irte(int irq)
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 
 	rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return rc;
 }
@@ -363,11 +370,12 @@ int free_irte(int irq)
 	struct irte *irte;
 	struct intel_iommu *iommu;
 	struct irq_2_iommu *irq_iommu;
+	unsigned long flags;
 
-	spin_lock(&irq_2_ir_lock);
+	spin_lock_irqsave(&irq_2_ir_lock, flags);
 	irq_iommu = valid_irq_2_iommu(irq);
 	if (!irq_iommu) {
-		spin_unlock(&irq_2_ir_lock);
+		spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 		return -1;
 	}
 
@@ -387,7 +395,7 @@ int free_irte(int irq)
 	irq_iommu->sub_handle = 0;
 	irq_iommu->irte_mask = 0;
 
-	spin_unlock(&irq_2_ir_lock);
+	spin_unlock_irqrestore(&irq_2_ir_lock, flags);
 
 	return rc;
 }
-- 
cgit v1.2.3-59-g8ed1b


From 0ac2491f57af5644f88383d28809760902d6f4d7 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:54 -0700
Subject: x86, dmar: move page fault handling code to dmar.c

Impact: code movement

Move page fault handling code to dmar.c
This will be shared both by DMA-remapping and Intr-remapping code.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/dmar.c        | 191 ++++++++++++++++++++++++++++++++++++++++++++++
 drivers/pci/intel-iommu.c | 188 ---------------------------------------------
 2 files changed, 191 insertions(+), 188 deletions(-)

(limited to 'drivers')

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 5f333403c2ea..75d34bf2db50 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -31,6 +31,8 @@
 #include <linux/iova.h>
 #include <linux/intel-iommu.h>
 #include <linux/timer.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
 
 #undef PREFIX
 #define PREFIX "DMAR:"
@@ -812,3 +814,192 @@ int dmar_enable_qi(struct intel_iommu *iommu)
 
 	return 0;
 }
+
+/* iommu interrupt handling. Most stuff are MSI-like. */
+
+static const char *fault_reason_strings[] =
+{
+	"Software",
+	"Present bit in root entry is clear",
+	"Present bit in context entry is clear",
+	"Invalid context entry",
+	"Access beyond MGAW",
+	"PTE Write access is not set",
+	"PTE Read access is not set",
+	"Next page table ptr is invalid",
+	"Root table address invalid",
+	"Context table ptr is invalid",
+	"non-zero reserved fields in RTP",
+	"non-zero reserved fields in CTP",
+	"non-zero reserved fields in PTE",
+};
+#define MAX_FAULT_REASON_IDX 	(ARRAY_SIZE(fault_reason_strings) - 1)
+
+const char *dmar_get_fault_reason(u8 fault_reason)
+{
+	if (fault_reason > MAX_FAULT_REASON_IDX)
+		return "Unknown";
+	else
+		return fault_reason_strings[fault_reason];
+}
+
+void dmar_msi_unmask(unsigned int irq)
+{
+	struct intel_iommu *iommu = get_irq_data(irq);
+	unsigned long flag;
+
+	/* unmask it */
+	spin_lock_irqsave(&iommu->register_lock, flag);
+	writel(0, iommu->reg + DMAR_FECTL_REG);
+	/* Read a reg to force flush the post write */
+	readl(iommu->reg + DMAR_FECTL_REG);
+	spin_unlock_irqrestore(&iommu->register_lock, flag);
+}
+
+void dmar_msi_mask(unsigned int irq)
+{
+	unsigned long flag;
+	struct intel_iommu *iommu = get_irq_data(irq);
+
+	/* mask it */
+	spin_lock_irqsave(&iommu->register_lock, flag);
+	writel(DMA_FECTL_IM, iommu->reg + DMAR_FECTL_REG);
+	/* Read a reg to force flush the post write */
+	readl(iommu->reg + DMAR_FECTL_REG);
+	spin_unlock_irqrestore(&iommu->register_lock, flag);
+}
+
+void dmar_msi_write(int irq, struct msi_msg *msg)
+{
+	struct intel_iommu *iommu = get_irq_data(irq);
+	unsigned long flag;
+
+	spin_lock_irqsave(&iommu->register_lock, flag);
+	writel(msg->data, iommu->reg + DMAR_FEDATA_REG);
+	writel(msg->address_lo, iommu->reg + DMAR_FEADDR_REG);
+	writel(msg->address_hi, iommu->reg + DMAR_FEUADDR_REG);
+	spin_unlock_irqrestore(&iommu->register_lock, flag);
+}
+
+void dmar_msi_read(int irq, struct msi_msg *msg)
+{
+	struct intel_iommu *iommu = get_irq_data(irq);
+	unsigned long flag;
+
+	spin_lock_irqsave(&iommu->register_lock, flag);
+	msg->data = readl(iommu->reg + DMAR_FEDATA_REG);
+	msg->address_lo = readl(iommu->reg + DMAR_FEADDR_REG);
+	msg->address_hi = readl(iommu->reg + DMAR_FEUADDR_REG);
+	spin_unlock_irqrestore(&iommu->register_lock, flag);
+}
+
+static int dmar_fault_do_one(struct intel_iommu *iommu, int type,
+		u8 fault_reason, u16 source_id, unsigned long long addr)
+{
+	const char *reason;
+
+	reason = dmar_get_fault_reason(fault_reason);
+
+	printk(KERN_ERR
+		"DMAR:[%s] Request device [%02x:%02x.%d] "
+		"fault addr %llx \n"
+		"DMAR:[fault reason %02d] %s\n",
+		(type ? "DMA Read" : "DMA Write"),
+		(source_id >> 8), PCI_SLOT(source_id & 0xFF),
+		PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason);
+	return 0;
+}
+
+#define PRIMARY_FAULT_REG_LEN (16)
+static irqreturn_t dmar_fault(int irq, void *dev_id)
+{
+	struct intel_iommu *iommu = dev_id;
+	int reg, fault_index;
+	u32 fault_status;
+	unsigned long flag;
+
+	spin_lock_irqsave(&iommu->register_lock, flag);
+	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
+
+	/* TBD: ignore advanced fault log currently */
+	if (!(fault_status & DMA_FSTS_PPF))
+		goto clear_overflow;
+
+	fault_index = dma_fsts_fault_record_index(fault_status);
+	reg = cap_fault_reg_offset(iommu->cap);
+	while (1) {
+		u8 fault_reason;
+		u16 source_id;
+		u64 guest_addr;
+		int type;
+		u32 data;
+
+		/* highest 32 bits */
+		data = readl(iommu->reg + reg +
+				fault_index * PRIMARY_FAULT_REG_LEN + 12);
+		if (!(data & DMA_FRCD_F))
+			break;
+
+		fault_reason = dma_frcd_fault_reason(data);
+		type = dma_frcd_type(data);
+
+		data = readl(iommu->reg + reg +
+				fault_index * PRIMARY_FAULT_REG_LEN + 8);
+		source_id = dma_frcd_source_id(data);
+
+		guest_addr = dmar_readq(iommu->reg + reg +
+				fault_index * PRIMARY_FAULT_REG_LEN);
+		guest_addr = dma_frcd_page_addr(guest_addr);
+		/* clear the fault */
+		writel(DMA_FRCD_F, iommu->reg + reg +
+			fault_index * PRIMARY_FAULT_REG_LEN + 12);
+
+		spin_unlock_irqrestore(&iommu->register_lock, flag);
+
+		dmar_fault_do_one(iommu, type, fault_reason,
+				source_id, guest_addr);
+
+		fault_index++;
+		if (fault_index > cap_num_fault_regs(iommu->cap))
+			fault_index = 0;
+		spin_lock_irqsave(&iommu->register_lock, flag);
+	}
+clear_overflow:
+	/* clear primary fault overflow */
+	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
+	if (fault_status & DMA_FSTS_PFO)
+		writel(DMA_FSTS_PFO, iommu->reg + DMAR_FSTS_REG);
+
+	spin_unlock_irqrestore(&iommu->register_lock, flag);
+	return IRQ_HANDLED;
+}
+
+int dmar_set_interrupt(struct intel_iommu *iommu)
+{
+	int irq, ret;
+
+	irq = create_irq();
+	if (!irq) {
+		printk(KERN_ERR "IOMMU: no free vectors\n");
+		return -EINVAL;
+	}
+
+	set_irq_data(irq, iommu);
+	iommu->irq = irq;
+
+	ret = arch_setup_dmar_msi(irq);
+	if (ret) {
+		set_irq_data(irq, NULL);
+		iommu->irq = 0;
+		destroy_irq(irq);
+		return 0;
+	}
+
+	/* Force fault register is cleared */
+	dmar_fault(irq, iommu);
+
+	ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu);
+	if (ret)
+		printk(KERN_ERR "IOMMU: can't request irq\n");
+	return ret;
+}
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index f3f686581a90..4a4ab651b709 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1004,194 +1004,6 @@ static int iommu_disable_translation(struct intel_iommu *iommu)
 	return 0;
 }
 
-/* iommu interrupt handling. Most stuff are MSI-like. */
-
-static const char *fault_reason_strings[] =
-{
-	"Software",
-	"Present bit in root entry is clear",
-	"Present bit in context entry is clear",
-	"Invalid context entry",
-	"Access beyond MGAW",
-	"PTE Write access is not set",
-	"PTE Read access is not set",
-	"Next page table ptr is invalid",
-	"Root table address invalid",
-	"Context table ptr is invalid",
-	"non-zero reserved fields in RTP",
-	"non-zero reserved fields in CTP",
-	"non-zero reserved fields in PTE",
-};
-#define MAX_FAULT_REASON_IDX 	(ARRAY_SIZE(fault_reason_strings) - 1)
-
-const char *dmar_get_fault_reason(u8 fault_reason)
-{
-	if (fault_reason > MAX_FAULT_REASON_IDX)
-		return "Unknown";
-	else
-		return fault_reason_strings[fault_reason];
-}
-
-void dmar_msi_unmask(unsigned int irq)
-{
-	struct intel_iommu *iommu = get_irq_data(irq);
-	unsigned long flag;
-
-	/* unmask it */
-	spin_lock_irqsave(&iommu->register_lock, flag);
-	writel(0, iommu->reg + DMAR_FECTL_REG);
-	/* Read a reg to force flush the post write */
-	readl(iommu->reg + DMAR_FECTL_REG);
-	spin_unlock_irqrestore(&iommu->register_lock, flag);
-}
-
-void dmar_msi_mask(unsigned int irq)
-{
-	unsigned long flag;
-	struct intel_iommu *iommu = get_irq_data(irq);
-
-	/* mask it */
-	spin_lock_irqsave(&iommu->register_lock, flag);
-	writel(DMA_FECTL_IM, iommu->reg + DMAR_FECTL_REG);
-	/* Read a reg to force flush the post write */
-	readl(iommu->reg + DMAR_FECTL_REG);
-	spin_unlock_irqrestore(&iommu->register_lock, flag);
-}
-
-void dmar_msi_write(int irq, struct msi_msg *msg)
-{
-	struct intel_iommu *iommu = get_irq_data(irq);
-	unsigned long flag;
-
-	spin_lock_irqsave(&iommu->register_lock, flag);
-	writel(msg->data, iommu->reg + DMAR_FEDATA_REG);
-	writel(msg->address_lo, iommu->reg + DMAR_FEADDR_REG);
-	writel(msg->address_hi, iommu->reg + DMAR_FEUADDR_REG);
-	spin_unlock_irqrestore(&iommu->register_lock, flag);
-}
-
-void dmar_msi_read(int irq, struct msi_msg *msg)
-{
-	struct intel_iommu *iommu = get_irq_data(irq);
-	unsigned long flag;
-
-	spin_lock_irqsave(&iommu->register_lock, flag);
-	msg->data = readl(iommu->reg + DMAR_FEDATA_REG);
-	msg->address_lo = readl(iommu->reg + DMAR_FEADDR_REG);
-	msg->address_hi = readl(iommu->reg + DMAR_FEUADDR_REG);
-	spin_unlock_irqrestore(&iommu->register_lock, flag);
-}
-
-static int iommu_page_fault_do_one(struct intel_iommu *iommu, int type,
-		u8 fault_reason, u16 source_id, unsigned long long addr)
-{
-	const char *reason;
-
-	reason = dmar_get_fault_reason(fault_reason);
-
-	printk(KERN_ERR
-		"DMAR:[%s] Request device [%02x:%02x.%d] "
-		"fault addr %llx \n"
-		"DMAR:[fault reason %02d] %s\n",
-		(type ? "DMA Read" : "DMA Write"),
-		(source_id >> 8), PCI_SLOT(source_id & 0xFF),
-		PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason);
-	return 0;
-}
-
-#define PRIMARY_FAULT_REG_LEN (16)
-static irqreturn_t iommu_page_fault(int irq, void *dev_id)
-{
-	struct intel_iommu *iommu = dev_id;
-	int reg, fault_index;
-	u32 fault_status;
-	unsigned long flag;
-
-	spin_lock_irqsave(&iommu->register_lock, flag);
-	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
-
-	/* TBD: ignore advanced fault log currently */
-	if (!(fault_status & DMA_FSTS_PPF))
-		goto clear_overflow;
-
-	fault_index = dma_fsts_fault_record_index(fault_status);
-	reg = cap_fault_reg_offset(iommu->cap);
-	while (1) {
-		u8 fault_reason;
-		u16 source_id;
-		u64 guest_addr;
-		int type;
-		u32 data;
-
-		/* highest 32 bits */
-		data = readl(iommu->reg + reg +
-				fault_index * PRIMARY_FAULT_REG_LEN + 12);
-		if (!(data & DMA_FRCD_F))
-			break;
-
-		fault_reason = dma_frcd_fault_reason(data);
-		type = dma_frcd_type(data);
-
-		data = readl(iommu->reg + reg +
-				fault_index * PRIMARY_FAULT_REG_LEN + 8);
-		source_id = dma_frcd_source_id(data);
-
-		guest_addr = dmar_readq(iommu->reg + reg +
-				fault_index * PRIMARY_FAULT_REG_LEN);
-		guest_addr = dma_frcd_page_addr(guest_addr);
-		/* clear the fault */
-		writel(DMA_FRCD_F, iommu->reg + reg +
-			fault_index * PRIMARY_FAULT_REG_LEN + 12);
-
-		spin_unlock_irqrestore(&iommu->register_lock, flag);
-
-		iommu_page_fault_do_one(iommu, type, fault_reason,
-				source_id, guest_addr);
-
-		fault_index++;
-		if (fault_index > cap_num_fault_regs(iommu->cap))
-			fault_index = 0;
-		spin_lock_irqsave(&iommu->register_lock, flag);
-	}
-clear_overflow:
-	/* clear primary fault overflow */
-	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
-	if (fault_status & DMA_FSTS_PFO)
-		writel(DMA_FSTS_PFO, iommu->reg + DMAR_FSTS_REG);
-
-	spin_unlock_irqrestore(&iommu->register_lock, flag);
-	return IRQ_HANDLED;
-}
-
-int dmar_set_interrupt(struct intel_iommu *iommu)
-{
-	int irq, ret;
-
-	irq = create_irq();
-	if (!irq) {
-		printk(KERN_ERR "IOMMU: no free vectors\n");
-		return -EINVAL;
-	}
-
-	set_irq_data(irq, iommu);
-	iommu->irq = irq;
-
-	ret = arch_setup_dmar_msi(irq);
-	if (ret) {
-		set_irq_data(irq, NULL);
-		iommu->irq = 0;
-		destroy_irq(irq);
-		return 0;
-	}
-
-	/* Force fault register is cleared */
-	iommu_page_fault(irq, iommu);
-
-	ret = request_irq(irq, iommu_page_fault, 0, iommu->name, iommu);
-	if (ret)
-		printk(KERN_ERR "IOMMU: can't request irq\n");
-	return ret;
-}
 
 static int iommu_init_domains(struct intel_iommu *iommu)
 {
-- 
cgit v1.2.3-59-g8ed1b


From 9d783ba042771284fb4ee5013c3d94220755ae7f Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:55 -0700
Subject: x86, x2apic: enable fault handling for intr-remapping

Impact: interface augmentation (not yet used)

Enable fault handling flow for intr-remapping aswell. Fault handling
code now shared by both dma-remapping and intr-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 arch/x86/include/asm/msidef.h   |   1 +
 arch/x86/kernel/apic/io_apic.c  |   9 +++-
 arch/x86/kernel/apic/probe_64.c |   9 ++++
 drivers/pci/dmar.c              | 102 +++++++++++++++++++++++++++++++++-------
 drivers/pci/intel-iommu.c       |   3 +-
 drivers/pci/intr_remapping.c    |   2 +-
 include/linux/dmar.h            |   5 +-
 include/linux/intel-iommu.h     |   4 +-
 8 files changed, 107 insertions(+), 28 deletions(-)

(limited to 'drivers')

diff --git a/arch/x86/include/asm/msidef.h b/arch/x86/include/asm/msidef.h
index 6706b3006f13..4cc48af23fef 100644
--- a/arch/x86/include/asm/msidef.h
+++ b/arch/x86/include/asm/msidef.h
@@ -47,6 +47,7 @@
 #define	 MSI_ADDR_DEST_ID_MASK		0x00ffff0
 #define  MSI_ADDR_DEST_ID(dest)		(((dest) << MSI_ADDR_DEST_ID_SHIFT) & \
 					 MSI_ADDR_DEST_ID_MASK)
+#define MSI_ADDR_EXT_DEST_ID(dest)	((dest) & 0xffffff00)
 
 #define MSI_ADDR_IR_EXT_INT		(1 << 4)
 #define MSI_ADDR_IR_SHV			(1 << 3)
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 00e6071cefc4..b18a7734d689 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -3294,7 +3294,12 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms
 	} else
 #endif
 	{
-		msg->address_hi = MSI_ADDR_BASE_HI;
+		if (x2apic_enabled())
+			msg->address_hi = MSI_ADDR_BASE_HI |
+					  MSI_ADDR_EXT_DEST_ID(dest);
+		else
+			msg->address_hi = MSI_ADDR_BASE_HI;
+
 		msg->address_lo =
 			MSI_ADDR_BASE_LO |
 			((apic->irq_dest_mode == 0) ?
@@ -3528,7 +3533,7 @@ void arch_teardown_msi_irq(unsigned int irq)
 	destroy_irq(irq);
 }
 
-#ifdef CONFIG_DMAR
+#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
 #ifdef CONFIG_SMP
 static void dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
 {
diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c
index 8d7748efe6a8..8297c2b8ed20 100644
--- a/arch/x86/kernel/apic/probe_64.c
+++ b/arch/x86/kernel/apic/probe_64.c
@@ -68,6 +68,15 @@ void __init default_setup_apic_routing(void)
 			apic = &apic_physflat;
 		printk(KERN_INFO "Setting APIC routing to %s\n", apic->name);
 	}
+
+#ifdef CONFIG_X86_X2APIC
+	/*
+	 * Now that apic routing model is selected, configure the
+	 * fault handling for intr remapping.
+	 */
+	if (intr_remapping_enabled)
+		enable_drhd_fault_handling();
+#endif
 }
 
 /* Same for both flat and physical. */
diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 75d34bf2db50..bb4ed985f9c7 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -511,6 +511,7 @@ int alloc_iommu(struct dmar_drhd_unit *drhd)
 		return -ENOMEM;
 
 	iommu->seq_id = iommu_allocated++;
+	sprintf (iommu->name, "dmar%d", iommu->seq_id);
 
 	iommu->reg = ioremap(drhd->reg_base_addr, VTD_PAGE_SIZE);
 	if (!iommu->reg) {
@@ -817,7 +818,13 @@ int dmar_enable_qi(struct intel_iommu *iommu)
 
 /* iommu interrupt handling. Most stuff are MSI-like. */
 
-static const char *fault_reason_strings[] =
+enum faulttype {
+	DMA_REMAP,
+	INTR_REMAP,
+	UNKNOWN,
+};
+
+static const char *dma_remap_fault_reasons[] =
 {
 	"Software",
 	"Present bit in root entry is clear",
@@ -833,14 +840,33 @@ static const char *fault_reason_strings[] =
 	"non-zero reserved fields in CTP",
 	"non-zero reserved fields in PTE",
 };
+
+static const char *intr_remap_fault_reasons[] =
+{
+	"Detected reserved fields in the decoded interrupt-remapped request",
+	"Interrupt index exceeded the interrupt-remapping table size",
+	"Present field in the IRTE entry is clear",
+	"Error accessing interrupt-remapping table pointed by IRTA_REG",
+	"Detected reserved fields in the IRTE entry",
+	"Blocked a compatibility format interrupt request",
+	"Blocked an interrupt request due to source-id verification failure",
+};
+
 #define MAX_FAULT_REASON_IDX 	(ARRAY_SIZE(fault_reason_strings) - 1)
 
-const char *dmar_get_fault_reason(u8 fault_reason)
+const char *dmar_get_fault_reason(u8 fault_reason, int *fault_type)
 {
-	if (fault_reason > MAX_FAULT_REASON_IDX)
+	if (fault_reason >= 0x20 && (fault_reason <= 0x20 +
+				     ARRAY_SIZE(intr_remap_fault_reasons))) {
+		*fault_type = INTR_REMAP;
+		return intr_remap_fault_reasons[fault_reason - 0x20];
+	} else if (fault_reason < ARRAY_SIZE(dma_remap_fault_reasons)) {
+		*fault_type = DMA_REMAP;
+		return dma_remap_fault_reasons[fault_reason];
+	} else {
+		*fault_type = UNKNOWN;
 		return "Unknown";
-	else
-		return fault_reason_strings[fault_reason];
+	}
 }
 
 void dmar_msi_unmask(unsigned int irq)
@@ -897,16 +923,25 @@ static int dmar_fault_do_one(struct intel_iommu *iommu, int type,
 		u8 fault_reason, u16 source_id, unsigned long long addr)
 {
 	const char *reason;
+	int fault_type;
 
-	reason = dmar_get_fault_reason(fault_reason);
+	reason = dmar_get_fault_reason(fault_reason, &fault_type);
 
-	printk(KERN_ERR
-		"DMAR:[%s] Request device [%02x:%02x.%d] "
-		"fault addr %llx \n"
-		"DMAR:[fault reason %02d] %s\n",
-		(type ? "DMA Read" : "DMA Write"),
-		(source_id >> 8), PCI_SLOT(source_id & 0xFF),
-		PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason);
+	if (fault_type == INTR_REMAP)
+		printk(KERN_ERR "INTR-REMAP: Request device [[%02x:%02x.%d] "
+		       "fault index %llx\n"
+			"INTR-REMAP:[fault reason %02d] %s\n",
+			(source_id >> 8), PCI_SLOT(source_id & 0xFF),
+			PCI_FUNC(source_id & 0xFF), addr >> 48,
+			fault_reason, reason);
+	else
+		printk(KERN_ERR
+		       "DMAR:[%s] Request device [%02x:%02x.%d] "
+		       "fault addr %llx \n"
+		       "DMAR:[fault reason %02d] %s\n",
+		       (type ? "DMA Read" : "DMA Write"),
+		       (source_id >> 8), PCI_SLOT(source_id & 0xFF),
+		       PCI_FUNC(source_id & 0xFF), addr, fault_reason, reason);
 	return 0;
 }
 
@@ -920,10 +955,13 @@ static irqreturn_t dmar_fault(int irq, void *dev_id)
 
 	spin_lock_irqsave(&iommu->register_lock, flag);
 	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
+	if (fault_status)
+		printk(KERN_ERR "DRHD: handling fault status reg %x\n",
+		       fault_status);
 
 	/* TBD: ignore advanced fault log currently */
 	if (!(fault_status & DMA_FSTS_PPF))
-		goto clear_overflow;
+		goto clear_rest;
 
 	fault_index = dma_fsts_fault_record_index(fault_status);
 	reg = cap_fault_reg_offset(iommu->cap);
@@ -964,11 +1002,10 @@ static irqreturn_t dmar_fault(int irq, void *dev_id)
 			fault_index = 0;
 		spin_lock_irqsave(&iommu->register_lock, flag);
 	}
-clear_overflow:
-	/* clear primary fault overflow */
+clear_rest:
+	/* clear all the other faults */
 	fault_status = readl(iommu->reg + DMAR_FSTS_REG);
-	if (fault_status & DMA_FSTS_PFO)
-		writel(DMA_FSTS_PFO, iommu->reg + DMAR_FSTS_REG);
+	writel(fault_status, iommu->reg + DMAR_FSTS_REG);
 
 	spin_unlock_irqrestore(&iommu->register_lock, flag);
 	return IRQ_HANDLED;
@@ -978,6 +1015,12 @@ int dmar_set_interrupt(struct intel_iommu *iommu)
 {
 	int irq, ret;
 
+	/*
+	 * Check if the fault interrupt is already initialized.
+	 */
+	if (iommu->irq)
+		return 0;
+
 	irq = create_irq();
 	if (!irq) {
 		printk(KERN_ERR "IOMMU: no free vectors\n");
@@ -1003,3 +1046,26 @@ int dmar_set_interrupt(struct intel_iommu *iommu)
 		printk(KERN_ERR "IOMMU: can't request irq\n");
 	return ret;
 }
+
+int __init enable_drhd_fault_handling(void)
+{
+	struct dmar_drhd_unit *drhd;
+
+	/*
+	 * Enable fault control interrupt.
+	 */
+	for_each_drhd_unit(drhd) {
+		int ret;
+		struct intel_iommu *iommu = drhd->iommu;
+		ret = dmar_set_interrupt(iommu);
+
+		if (ret) {
+			printk(KERN_ERR "DRHD %Lx: failed to enable fault, "
+			       " interrupt, ret %d\n",
+			       (unsigned long long)drhd->reg_base_addr, ret);
+			return -1;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 4a4ab651b709..25fc1df486bb 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1799,7 +1799,7 @@ static int __init init_dmars(void)
 	struct dmar_rmrr_unit *rmrr;
 	struct pci_dev *pdev;
 	struct intel_iommu *iommu;
-	int i, ret, unit = 0;
+	int i, ret;
 
 	/*
 	 * for each drhd
@@ -1921,7 +1921,6 @@ static int __init init_dmars(void)
 		if (drhd->ignored)
 			continue;
 		iommu = drhd->iommu;
-		sprintf (iommu->name, "dmar%d", unit++);
 
 		iommu_flush_write_buffer(iommu);
 
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 5ffa65fffb6a..c38e3f437a81 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -308,7 +308,7 @@ int modify_irte(int irq, struct irte *irte_modified)
 	index = irq_iommu->irte_index + irq_iommu->sub_handle;
 	irte = &iommu->ir_table->base[index];
 
-	set_64bit((unsigned long *)irte, irte_modified->low | (1 << 1));
+	set_64bit((unsigned long *)irte, irte_modified->low);
 	__iommu_flush_cache(iommu, irte, sizeof(*irte));
 
 	rc = qi_flush_iec(iommu, index, 0);
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index f28440784cf0..c7768330c11d 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -49,6 +49,7 @@ extern int dmar_dev_scope_init(void);
 
 /* Intel IOMMU detection */
 extern void detect_intel_iommu(void);
+extern int enable_drhd_fault_handling(void);
 
 
 extern int parse_ioapics_under_ir(void);
@@ -116,9 +117,6 @@ extern struct intel_iommu *map_ioapic_to_ir(int apic);
 #define intr_remapping_enabled		(0)
 #endif
 
-#ifdef CONFIG_DMAR
-extern const char *dmar_get_fault_reason(u8 fault_reason);
-
 /* Can't use the common MSI interrupt functions
  * since DMAR is not a pci device
  */
@@ -129,6 +127,7 @@ extern void dmar_msi_write(int irq, struct msi_msg *msg);
 extern int dmar_set_interrupt(struct intel_iommu *iommu);
 extern int arch_setup_dmar_msi(unsigned int irq);
 
+#ifdef CONFIG_DMAR
 extern int iommu_detected, no_iommu;
 extern struct list_head dmar_rmrr_units;
 struct dmar_rmrr_unit {
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index d2e3cbfba14f..a9563840644b 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -292,6 +292,8 @@ struct intel_iommu {
 	spinlock_t	register_lock; /* protect register handling */
 	int		seq_id;	/* sequence id of the iommu */
 	int		agaw; /* agaw of this iommu */
+	unsigned int 	irq;
+	unsigned char 	name[13];    /* Device Name */
 
 #ifdef CONFIG_DMAR
 	unsigned long 	*domain_ids; /* bitmap of domains */
@@ -299,8 +301,6 @@ struct intel_iommu {
 	spinlock_t	lock; /* protect context, domain ids */
 	struct root_entry *root_entry; /* virtual address */
 
-	unsigned int irq;
-	unsigned char name[7];    /* Device Name */
 	struct iommu_flush flush;
 #endif
 	struct q_inval  *qi;            /* Queued invalidation info */
-- 
cgit v1.2.3-59-g8ed1b


From eba67e5da6e971993b2899d2cdf459ce77d3dbc5 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:56 -0700
Subject: x86, dmar: routines for disabling queued invalidation and intr
 remapping

Impact: new interfaces (not yet used)

Routines for disabling queued invalidation and interrupt remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/dmar.c           | 36 ++++++++++++++++++++++++++++++++++++
 drivers/pci/intr_remapping.c | 27 +++++++++++++++++++++++++++
 include/linux/intel-iommu.h  |  1 +
 3 files changed, 64 insertions(+)

(limited to 'drivers')

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index bb4ed985f9c7..932e5e3930fc 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -753,6 +753,42 @@ int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
 	return qi_submit_sync(&desc, iommu);
 }
 
+/*
+ * Disable Queued Invalidation interface.
+ */
+void dmar_disable_qi(struct intel_iommu *iommu)
+{
+	unsigned long flags;
+	u32 sts;
+	cycles_t start_time = get_cycles();
+
+	if (!ecap_qis(iommu->ecap))
+		return;
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+
+	sts =  dmar_readq(iommu->reg + DMAR_GSTS_REG);
+	if (!(sts & DMA_GSTS_QIES))
+		goto end;
+
+	/*
+	 * Give a chance to HW to complete the pending invalidation requests.
+	 */
+	while ((readl(iommu->reg + DMAR_IQT_REG) !=
+		readl(iommu->reg + DMAR_IQH_REG)) &&
+		(DMAR_OPERATION_TIMEOUT > (get_cycles() - start_time)))
+		cpu_relax();
+
+	iommu->gcmd &= ~DMA_GCMD_QIE;
+
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
+
+	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG, readl,
+		      !(sts & DMA_GSTS_QIES), sts);
+end:
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+}
+
 /*
  * Enable Queued Invalidation interface. This is a must to support
  * interrupt-remapping. Also used by DMA-remapping, which replaces
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index c38e3f437a81..0d202d73a1ac 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -467,6 +467,33 @@ static int setup_intr_remapping(struct intel_iommu *iommu, int mode)
 	return 0;
 }
 
+/*
+ * Disable Interrupt Remapping.
+ */
+static void disable_intr_remapping(struct intel_iommu *iommu)
+{
+	unsigned long flags;
+	u32 sts;
+
+	if (!ecap_ir_support(iommu->ecap))
+		return;
+
+	spin_lock_irqsave(&iommu->register_lock, flags);
+
+	sts = dmar_readq(iommu->reg + DMAR_GSTS_REG);
+	if (!(sts & DMA_GSTS_IRES))
+		goto end;
+
+	iommu->gcmd &= ~DMA_GCMD_IRE;
+	writel(iommu->gcmd, iommu->reg + DMAR_GCMD_REG);
+
+	IOMMU_WAIT_OP(iommu, DMAR_GSTS_REG,
+		      readl, !(sts & DMA_GSTS_IRES), sts);
+
+end:
+	spin_unlock_irqrestore(&iommu->register_lock, flags);
+}
+
 int __init enable_intr_remapping(int eim)
 {
 	struct dmar_drhd_unit *drhd;
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index a9563840644b..78c1262e8704 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -321,6 +321,7 @@ extern struct dmar_drhd_unit * dmar_find_matched_drhd_unit(struct pci_dev *dev);
 extern int alloc_iommu(struct dmar_drhd_unit *drhd);
 extern void free_iommu(struct intel_iommu *iommu);
 extern int dmar_enable_qi(struct intel_iommu *iommu);
+extern void dmar_disable_qi(struct intel_iommu *iommu);
 extern void qi_global_iec(struct intel_iommu *iommu);
 
 extern int qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid,
-- 
cgit v1.2.3-59-g8ed1b


From 1531a6a6b81a4e6f9eec9a5608758a6ea14b96e0 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:57 -0700
Subject: x86, dmar: start with sane state while enabling dma and
 interrupt-remapping

Impact: cleanup/sanitization

Start from a sane state while enabling dma and interrupt-remapping, by
clearing the previous recorded faults and disabling previously
enabled queued invalidation and interrupt-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/dmar.c           |  5 +----
 drivers/pci/intel-iommu.c    | 29 +++++++++++++++++++++++++++++
 drivers/pci/intr_remapping.c | 17 +++++++++++++++++
 include/linux/dmar.h         |  2 ++
 4 files changed, 49 insertions(+), 4 deletions(-)

(limited to 'drivers')

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 932e5e3930fc..f1805002e436 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -982,7 +982,7 @@ static int dmar_fault_do_one(struct intel_iommu *iommu, int type,
 }
 
 #define PRIMARY_FAULT_REG_LEN (16)
-static irqreturn_t dmar_fault(int irq, void *dev_id)
+irqreturn_t dmar_fault(int irq, void *dev_id)
 {
 	struct intel_iommu *iommu = dev_id;
 	int reg, fault_index;
@@ -1074,9 +1074,6 @@ int dmar_set_interrupt(struct intel_iommu *iommu)
 		return 0;
 	}
 
-	/* Force fault register is cleared */
-	dmar_fault(irq, iommu);
-
 	ret = request_irq(irq, dmar_fault, 0, iommu->name, iommu);
 	if (ret)
 		printk(KERN_ERR "IOMMU: can't request irq\n");
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 25fc1df486bb..ef167b8b047d 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -1855,11 +1855,40 @@ static int __init init_dmars(void)
 		}
 	}
 
+	/*
+	 * Start from the sane iommu hardware state.
+	 */
 	for_each_drhd_unit(drhd) {
 		if (drhd->ignored)
 			continue;
 
 		iommu = drhd->iommu;
+
+		/*
+		 * If the queued invalidation is already initialized by us
+		 * (for example, while enabling interrupt-remapping) then
+		 * we got the things already rolling from a sane state.
+		 */
+		if (iommu->qi)
+			continue;
+
+		/*
+		 * Clear any previous faults.
+		 */
+		dmar_fault(-1, iommu);
+		/*
+		 * Disable queued invalidation if supported and already enabled
+		 * before OS handover.
+		 */
+		dmar_disable_qi(iommu);
+	}
+
+	for_each_drhd_unit(drhd) {
+		if (drhd->ignored)
+			continue;
+
+		iommu = drhd->iommu;
+
 		if (dmar_enable_qi(iommu)) {
 			/*
 			 * Queued Invalidate not enabled, use Register Based
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index 0d202d73a1ac..a84686b2478b 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -499,6 +499,23 @@ int __init enable_intr_remapping(int eim)
 	struct dmar_drhd_unit *drhd;
 	int setup = 0;
 
+	for_each_drhd_unit(drhd) {
+		struct intel_iommu *iommu = drhd->iommu;
+
+		/*
+		 * Clear previous faults.
+		 */
+		dmar_fault(-1, iommu);
+
+		/*
+		 * Disable intr remapping and queued invalidation, if already
+		 * enabled prior to OS handover.
+		 */
+		disable_intr_remapping(iommu);
+
+		dmar_disable_qi(iommu);
+	}
+
 	/*
 	 * check for the Interrupt-remapping support
 	 */
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index c7768330c11d..8a035aec14a9 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -24,6 +24,7 @@
 #include <linux/acpi.h>
 #include <linux/types.h>
 #include <linux/msi.h>
+#include <linux/irqreturn.h>
 
 #if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP)
 struct intel_iommu;
@@ -125,6 +126,7 @@ extern void dmar_msi_mask(unsigned int irq);
 extern void dmar_msi_read(int irq, struct msi_msg *msg);
 extern void dmar_msi_write(int irq, struct msi_msg *msg);
 extern int dmar_set_interrupt(struct intel_iommu *iommu);
+extern irqreturn_t dmar_fault(int irq, void *dev_id);
 extern int arch_setup_dmar_msi(unsigned int irq);
 
 #ifdef CONFIG_DMAR
-- 
cgit v1.2.3-59-g8ed1b


From 2e93456f5c069cf889c0c3acd1246ee88c49ae5c Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:04:58 -0700
Subject: x86, intr-remapping: fix free_irte() to clear all the IRTE entries

Impact: fix interrupt table entry leak

Fix the typo which was not clearing all the interrupt remapping table
entries corresponding to an irq.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/intr_remapping.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'drivers')

diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index a84686b2478b..f7ecd85e2104 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -386,7 +386,7 @@ int free_irte(int irq)
 
 	if (!irq_iommu->sub_handle) {
 		for (i = 0; i < (1 << irq_iommu->irte_mask); i++)
-			set_64bit((unsigned long *)irte, 0);
+			set_64bit((unsigned long *)(irte + i), 0);
 		rc = qi_flush_iec(iommu, index, irq_iommu->irte_mask);
 	}
 
-- 
cgit v1.2.3-59-g8ed1b


From fa4b57cc045d6134b9862b2873f9c8ba9ed53ffe Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 16 Mar 2009 17:05:05 -0700
Subject: x86, dmar: use atomic allocations for QI and Intr-remapping init

Impact: invalid use of GFP_KERNEL in interrupt context

Queued invalidation and interrupt-remapping will get initialized with
interrupts disabled (while enabling interrupt-remapping). So use
GFP_ATOMIC instead of GFP_KERNEL for memory alloacations.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
---
 drivers/pci/dmar.c           | 6 +++---
 drivers/pci/intr_remapping.c | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'drivers')

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index f1805002e436..d313039e2fdf 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -809,20 +809,20 @@ int dmar_enable_qi(struct intel_iommu *iommu)
 	if (iommu->qi)
 		return 0;
 
-	iommu->qi = kmalloc(sizeof(*qi), GFP_KERNEL);
+	iommu->qi = kmalloc(sizeof(*qi), GFP_ATOMIC);
 	if (!iommu->qi)
 		return -ENOMEM;
 
 	qi = iommu->qi;
 
-	qi->desc = (void *)(get_zeroed_page(GFP_KERNEL));
+	qi->desc = (void *)(get_zeroed_page(GFP_ATOMIC));
 	if (!qi->desc) {
 		kfree(qi);
 		iommu->qi = 0;
 		return -ENOMEM;
 	}
 
-	qi->desc_status = kmalloc(QI_LENGTH * sizeof(int), GFP_KERNEL);
+	qi->desc_status = kmalloc(QI_LENGTH * sizeof(int), GFP_ATOMIC);
 	if (!qi->desc_status) {
 		free_page((unsigned long) qi->desc);
 		kfree(qi);
diff --git a/drivers/pci/intr_remapping.c b/drivers/pci/intr_remapping.c
index f7ecd85e2104..bc5b6976f918 100644
--- a/drivers/pci/intr_remapping.c
+++ b/drivers/pci/intr_remapping.c
@@ -447,12 +447,12 @@ static int setup_intr_remapping(struct intel_iommu *iommu, int mode)
 	struct page *pages;
 
 	ir_table = iommu->ir_table = kzalloc(sizeof(struct ir_table),
-					     GFP_KERNEL);
+					     GFP_ATOMIC);
 
 	if (!iommu->ir_table)
 		return -ENOMEM;
 
-	pages = alloc_pages(GFP_KERNEL | __GFP_ZERO, INTR_REMAP_PAGE_ORDER);
+	pages = alloc_pages(GFP_ATOMIC | __GFP_ZERO, INTR_REMAP_PAGE_ORDER);
 
 	if (!pages) {
 		printk(KERN_ERR "failed to allocate pages of order %d\n",
-- 
cgit v1.2.3-59-g8ed1b