[patch 0/8] intel iommu, irq-remap, x2apic bug fixes and cleanups

[patch 0/8] intel iommu, irq-remap, x2apic bug fixes and cleanups

[Suresh Siddha]

x2apic opt out patch and the iommu parse device scope patch were held up
on the request of some code cleanups in this area.

Refer http://marc.info/?l=linux-kernel&m=131261409128226&w=2
and http://marc.info/?l=linux-pci&m=131157843015411&w=2

This series attempts to address the bug fixes aswell as the code cleanups.

thanks,
suresh

[patch 1/8] x86, x2apic: enable the bios request for x2apic optout

[Suresh Siddha]

[-- Attachment #1: x2apic_optout_20110823.patch --]
[-- Type: text/plain, Size: 8628 bytes --]

On the platforms which are x2apic and interrupt-remapping capable, Linux
kernel is enabling x2apic even if the BIOS doesn't. This is to take
advantage of the features that x2apic brings in.

Some of the OEM platforms are running into issues because of this, as their
bios is not x2apic aware. For example, this was resulting in interrupt migration
issues on one of the platforms. Also if the BIOS SMI handling uses APIC
interface to send SMI's, then the BIOS need to be aware of x2apic mode
that OS has enabled.

On some of these platforms, BIOS doesn't have a HW mechanism to turnoff
the x2apic feature to prevent OS from enabling it.

To resolve this mess, recent changes to the VT-d2 specification
(http://download.intel.com/technology/computing/vptech/Intel(r)_VT_for_Direct_IO.pdf)
includes a mechanism that provides BIOS a way to request system software
to opt out of enabling x2apic mode.

Look at the x2apic optout flag in the DMAR tables before enabling the x2apic
mode in the platform. Also print a warning that we have disabled x2apic
based on the BIOS request.

Kernel boot parameter "intremap=no_x2apic_optout" can be used to override
the BIOS x2apic optout request.

Signed-off-by: Youquan Song <youquan.song@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 Documentation/kernel-parameters.txt |    3 +-
 arch/x86/kernel/apic/apic.c         |   31 ++++++++++++-------------
 drivers/iommu/dmar.c                |    2 -
 drivers/iommu/intr_remapping.c      |   44 +++++++++++++++++++++++++++++-------
 include/linux/dmar.h                |   14 +++++++++--
 5 files changed, 66 insertions(+), 28 deletions(-)

Index: linux-2.6-tip/Documentation/kernel-parameters.txt
===================================================================
--- linux-2.6-tip.orig/Documentation/kernel-parameters.txt
+++ linux-2.6-tip/Documentation/kernel-parameters.txt
@@ -1027,10 +1027,11 @@ bytes respectively. Such letter suffixes
 			has the capability. With this option, super page will
 			not be supported.
 	intremap=	[X86-64, Intel-IOMMU]
-			Format: { on (default) | off | nosid }
 			on	enable Interrupt Remapping (default)
 			off	disable Interrupt Remapping
 			nosid	disable Source ID checking
+			no_x2apic_optout
+				BIOS x2APIC opt-out request will be ignored
 
 	inttest=	[IA-64]
 
Index: linux-2.6-tip/arch/x86/kernel/apic/apic.c
===================================================================
--- linux-2.6-tip.orig/arch/x86/kernel/apic/apic.c
+++ linux-2.6-tip/arch/x86/kernel/apic/apic.c
@@ -1440,24 +1440,18 @@ int __init enable_IR(void)
 #ifdef CONFIG_INTR_REMAP
 	if (!intr_remapping_supported()) {
 		pr_debug("intr-remapping not supported\n");
-		return 0;
+		return -1;
 	}
 
 	if (!x2apic_preenabled && skip_ioapic_setup) {
 		pr_info("Skipped enabling intr-remap because of skipping "
 			"io-apic setup\n");
-		return 0;
+		return -1;
 	}
 
-	if (enable_intr_remapping(x2apic_supported()))
-		return 0;
-
-	pr_info("Enabled Interrupt-remapping\n");
-
-	return 1;
-
+	return enable_intr_remapping();
 #endif
-	return 0;
+	return -1;
 }
 
 void __init enable_IR_x2apic(void)
@@ -1481,11 +1475,11 @@ void __init enable_IR_x2apic(void)
 	mask_ioapic_entries();
 
 	if (dmar_table_init_ret)
-		ret = 0;
+		ret = -1;
 	else
 		ret = enable_IR();
 
-	if (!ret) {
+	if (ret < 0) {
 		/* IR is required if there is APIC ID > 255 even when running
 		 * under KVM
 		 */
@@ -1499,6 +1493,9 @@ void __init enable_IR_x2apic(void)
 		x2apic_force_phys();
 	}
 
+	if (ret == IRQ_REMAP_XAPIC_MODE)
+		goto nox2apic;
+
 	x2apic_enabled = 1;
 
 	if (x2apic_supported() && !x2apic_mode) {
@@ -1508,19 +1505,21 @@ void __init enable_IR_x2apic(void)
 	}
 
 nox2apic:
-	if (!ret) /* IR enabling failed */
+	if (ret < 0) /* IR enabling failed */
 		restore_ioapic_entries();
 	legacy_pic->restore_mask();
 	local_irq_restore(flags);
 
 out:
-	if (x2apic_enabled)
+	if (x2apic_enabled || !x2apic_supported())
 		return;
 
 	if (x2apic_preenabled)
 		panic("x2apic: enabled by BIOS but kernel init failed.");
-	else if (cpu_has_x2apic)
-		pr_info("Not enabling x2apic, Intr-remapping init failed.\n");
+	else if (ret == IRQ_REMAP_XAPIC_MODE)
+		pr_info("x2apic not enabled, IRQ remapping is in xapic mode\n");
+	else if (ret < 0)
+		pr_info("x2apic not enabled, IRQ remapping init failed\n");
 }
 
 #ifdef CONFIG_X86_64
Index: linux-2.6-tip/drivers/iommu/dmar.c
===================================================================
--- linux-2.6-tip.orig/drivers/iommu/dmar.c
+++ linux-2.6-tip/drivers/iommu/dmar.c
@@ -46,7 +46,7 @@
  */
 LIST_HEAD(dmar_drhd_units);
 
-static struct acpi_table_header * __initdata dmar_tbl;
+struct acpi_table_header * __initdata dmar_tbl;
 static acpi_size dmar_tbl_size;
 
 static void __init dmar_register_drhd_unit(struct dmar_drhd_unit *drhd)
Index: linux-2.6-tip/drivers/iommu/intr_remapping.c
===================================================================
--- linux-2.6-tip.orig/drivers/iommu/intr_remapping.c
+++ linux-2.6-tip/drivers/iommu/intr_remapping.c
@@ -21,6 +21,7 @@ int intr_remapping_enabled;
 
 static int disable_intremap;
 static int disable_sourceid_checking;
+static int no_x2apic_optout;
 
 static __init int setup_nointremap(char *str)
 {
@@ -34,12 +35,20 @@ static __init int setup_intremap(char *s
 	if (!str)
 		return -EINVAL;
 
-	if (!strncmp(str, "on", 2))
-		disable_intremap = 0;
-	else if (!strncmp(str, "off", 3))
-		disable_intremap = 1;
-	else if (!strncmp(str, "nosid", 5))
-		disable_sourceid_checking = 1;
+	while (*str) {
+		if (!strncmp(str, "on", 2))
+			disable_intremap = 0;
+		else if (!strncmp(str, "off", 3))
+			disable_intremap = 1;
+		else if (!strncmp(str, "nosid", 5))
+			disable_sourceid_checking = 1;
+		else if (!strncmp(str, "no_x2apic_optout", 16))
+			no_x2apic_optout = 1;
+
+		str += strcspn(str, ",");
+		while (*str == ',')
+			str++;
+	}
 
 	return 0;
 }
@@ -501,6 +510,15 @@ end:
 	spin_unlock_irqrestore(&iommu->register_lock, flags);
 }
 
+static int __init dmar_x2apic_optout(void)
+{
+	struct acpi_table_dmar *dmar;
+	dmar = (struct acpi_table_dmar *)dmar_tbl;
+	if (!dmar || no_x2apic_optout)
+		return 0;
+	return dmar->flags & DMAR_X2APIC_OPT_OUT;
+}
+
 int __init intr_remapping_supported(void)
 {
 	struct dmar_drhd_unit *drhd;
@@ -521,16 +539,25 @@ int __init intr_remapping_supported(void
 	return 1;
 }
 
-int __init enable_intr_remapping(int eim)
+int __init enable_intr_remapping(void)
 {
 	struct dmar_drhd_unit *drhd;
 	int setup = 0;
+	int eim = 0;
 
 	if (parse_ioapics_under_ir() != 1) {
 		printk(KERN_INFO "Not enable interrupt remapping\n");
 		return -1;
 	}
 
+	if (x2apic_supported()) {
+		eim = !dmar_x2apic_optout();
+		WARN(!eim, KERN_WARNING
+			   "Your BIOS is broken and requested that x2apic be disabled\n"
+			   "This will leave your machine vulnerable to irq-injection attacks\n"
+			   "Use 'intremap=no_x2apic_optout' to override BIOS request\n");
+	}
+
 	for_each_drhd_unit(drhd) {
 		struct intel_iommu *iommu = drhd->iommu;
 
@@ -606,8 +633,9 @@ int __init enable_intr_remapping(int eim
 		goto error;
 
 	intr_remapping_enabled = 1;
+	pr_info("Enabled IRQ remapping in %s mode\n", eim ? "x2apic" : "xapic");
 
-	return 0;
+	return eim ? IRQ_REMAP_X2APIC_MODE : IRQ_REMAP_XAPIC_MODE;
 
 error:
 	/*
Index: linux-2.6-tip/include/linux/dmar.h
===================================================================
--- linux-2.6-tip.orig/include/linux/dmar.h
+++ linux-2.6-tip/include/linux/dmar.h
@@ -26,8 +26,13 @@
 #include <linux/msi.h>
 #include <linux/irqreturn.h>
 
+/* DMAR Flags */
+#define DMAR_INTR_REMAP		0x1
+#define DMAR_X2APIC_OPT_OUT	0x2
+
 struct intel_iommu;
 #if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP)
+extern struct acpi_table_header *dmar_tbl;
 struct dmar_drhd_unit {
 	struct list_head list;		/* list of drhd units	*/
 	struct  acpi_dmar_header *hdr;	/* ACPI header		*/
@@ -110,7 +115,7 @@ struct irte {
 #ifdef CONFIG_INTR_REMAP
 extern int intr_remapping_enabled;
 extern int intr_remapping_supported(void);
-extern int enable_intr_remapping(int);
+extern int enable_intr_remapping(void);
 extern void disable_intr_remapping(void);
 extern int reenable_intr_remapping(int);
 
@@ -177,7 +182,7 @@ static inline int set_msi_sid(struct irt
 
 #define intr_remapping_enabled		(0)
 
-static inline int enable_intr_remapping(int eim)
+static inline int enable_intr_remapping(void)
 {
 	return -1;
 }
@@ -192,6 +197,11 @@ static inline int reenable_intr_remappin
 }
 #endif
 
+enum {
+	IRQ_REMAP_XAPIC_MODE,
+	IRQ_REMAP_X2APIC_MODE,
+};
+
 /* Can't use the common MSI interrupt functions
  * since DMAR is not a pci device
  */

[patch 2/8] intr_remap: call dmar_dev_scope_init() explicitly

[Suresh Siddha]

[-- Attachment #1: init_dev_scope_from_intr_remap.patch --]
[-- Type: text/plain, Size: 3346 bytes --]

Both DMA-remapping aswell as Interrupt-remapping depend on the dmar dev scope
to be initialized. When both DMA and IRQ-remapping are enabled, we depend
on DMA-remapping init code to call dmar_dev_scope_init(). This resulted
in not doing this init when DMA-remapping was turned off but
interrupt-remapping turned on in the kernel config.

This caused interrupt routing to break with CONFIG_INTR_REMAP=y
and CONFIG_DMAR=n.

This issue was introduced by this commit:

   commit 9d5ce73a64be2be8112147a3e0b551ad9cd1247b
   Author: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
   Date:   Tue Nov 10 19:46:16 2009 +0900
 
	x86: intel-iommu: Convert detect_intel_iommu to use iommu_init hook

Fix this by calling dmar_dev_scope_init() explicitly from the interrupt
remapping code too.

Reported-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 drivers/iommu/dmar.c           |   15 ++++++++++++---
 drivers/iommu/intel-iommu.c    |    6 +-----
 drivers/iommu/intr_remapping.c |    9 +++++++++
 3 files changed, 22 insertions(+), 8 deletions(-)

Index: tree/drivers/iommu/dmar.c
===================================================================
--- tree.orig/drivers/iommu/dmar.c
+++ tree/drivers/iommu/dmar.c
@@ -557,13 +557,17 @@ dmar_find_matched_drhd_unit(struct pci_d
 
 int __init dmar_dev_scope_init(void)
 {
+	static int dmar_dev_scope_initialized;
 	struct dmar_drhd_unit *drhd, *drhd_n;
 	int ret = -ENODEV;
 
+	if (dmar_dev_scope_initialized)
+		return dmar_dev_scope_initialized;
+
 	list_for_each_entry_safe(drhd, drhd_n, &dmar_drhd_units, list) {
 		ret = dmar_parse_dev(drhd);
 		if (ret)
-			return ret;
+			goto fail;
 	}
 
 #ifdef CONFIG_DMAR
@@ -574,17 +578,22 @@ int __init dmar_dev_scope_init(void)
 		list_for_each_entry_safe(rmrr, rmrr_n, &dmar_rmrr_units, list) {
 			ret = rmrr_parse_dev(rmrr);
 			if (ret)
-				return ret;
+				goto fail;
 		}
 
 		list_for_each_entry_safe(atsr, atsr_n, &dmar_atsr_units, list) {
 			ret = atsr_parse_dev(atsr);
 			if (ret)
-				return ret;
+				goto fail;
 		}
 	}
 #endif
 
+	dmar_dev_scope_initialized = 1;
+	return 0;
+
+fail:
+	dmar_dev_scope_initialized = ret;
 	return ret;
 }
 
Index: tree/drivers/iommu/intr_remapping.c
===================================================================
--- tree.orig/drivers/iommu/intr_remapping.c
+++ tree/drivers/iommu/intr_remapping.c
@@ -773,6 +773,15 @@ int __init parse_ioapics_under_ir(void)
 	return ir_supported;
 }
 
+int ir_dev_scope_init(void)
+{
+	if (!intr_remapping_enabled)
+		return 0;
+
+	return dmar_dev_scope_init();
+}
+rootfs_initcall(ir_dev_scope_init);
+
 void disable_intr_remapping(void)
 {
 	struct dmar_drhd_unit *drhd;
Index: tree/drivers/iommu/intel-iommu.c
===================================================================
--- tree.orig/drivers/iommu/intel-iommu.c
+++ tree/drivers/iommu/intel-iommu.c
@@ -3439,16 +3439,12 @@ int __init intel_iommu_init(void)
 		return 	-ENODEV;
 	}
 
-	if (dmar_dev_scope_init()) {
+	if (dmar_dev_scope_init() < 0) {
 		if (force_on)
 			panic("tboot: Failed to initialize DMAR device scope\n");
 		return 	-ENODEV;
 	}
 
-	/*
-	 * Check the need for DMA-remapping initialization now.
-	 * Above initialization will also be used by Interrupt-remapping.
-	 */
 	if (no_iommu || dmar_disabled)
 		return -ENODEV;
 

[patch 3/8] iommu: move IOMMU specific code to intel-iommu.c

[Suresh Siddha]

[-- Attachment #1: mv-iommu-routines-to-intel-iommu.patch --]
[-- Type: text/plain, Size: 11745 bytes --]

Move the IOMMU specific routines to intel-iommu.c leaving the dmar.c
to the common ACPI dmar code shared between DMA-remapping and
Interrupt-remapping.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 drivers/iommu/dmar.c          |  171 +-----------------------------------------
 drivers/iommu/intel-iommu.c   |  151 +++++++++++++++++++++++++++++++++++++
 include/linux/dma_remapping.h |    5 -
 include/linux/dmar.h          |   17 ++++
 4 files changed, 180 insertions(+), 164 deletions(-)

Index: tree/drivers/iommu/dmar.c
===================================================================
--- tree.orig/drivers/iommu/dmar.c
+++ tree/drivers/iommu/dmar.c
@@ -118,8 +118,8 @@ static int __init dmar_parse_one_dev_sco
 	return 0;
 }
 
-static int __init dmar_parse_dev_scope(void *start, void *end, int *cnt,
-				       struct pci_dev ***devices, u16 segment)
+int __init dmar_parse_dev_scope(void *start, void *end, int *cnt,
+				struct pci_dev ***devices, u16 segment)
 {
 	struct acpi_dmar_device_scope *scope;
 	void * tmp = start;
@@ -217,133 +217,6 @@ static int __init dmar_parse_dev(struct 
 	return ret;
 }
 
-#ifdef CONFIG_DMAR
-LIST_HEAD(dmar_rmrr_units);
-
-static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
-{
-	list_add(&rmrr->list, &dmar_rmrr_units);
-}
-
-
-static int __init
-dmar_parse_one_rmrr(struct acpi_dmar_header *header)
-{
-	struct acpi_dmar_reserved_memory *rmrr;
-	struct dmar_rmrr_unit *rmrru;
-
-	rmrru = kzalloc(sizeof(*rmrru), GFP_KERNEL);
-	if (!rmrru)
-		return -ENOMEM;
-
-	rmrru->hdr = header;
-	rmrr = (struct acpi_dmar_reserved_memory *)header;
-	rmrru->base_address = rmrr->base_address;
-	rmrru->end_address = rmrr->end_address;
-
-	dmar_register_rmrr_unit(rmrru);
-	return 0;
-}
-
-static int __init
-rmrr_parse_dev(struct dmar_rmrr_unit *rmrru)
-{
-	struct acpi_dmar_reserved_memory *rmrr;
-	int ret;
-
-	rmrr = (struct acpi_dmar_reserved_memory *) rmrru->hdr;
-	ret = dmar_parse_dev_scope((void *)(rmrr + 1),
-		((void *)rmrr) + rmrr->header.length,
-		&rmrru->devices_cnt, &rmrru->devices, rmrr->segment);
-
-	if (ret || (rmrru->devices_cnt == 0)) {
-		list_del(&rmrru->list);
-		kfree(rmrru);
-	}
-	return ret;
-}
-
-static LIST_HEAD(dmar_atsr_units);
-
-static int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
-{
-	struct acpi_dmar_atsr *atsr;
-	struct dmar_atsr_unit *atsru;
-
-	atsr = container_of(hdr, struct acpi_dmar_atsr, header);
-	atsru = kzalloc(sizeof(*atsru), GFP_KERNEL);
-	if (!atsru)
-		return -ENOMEM;
-
-	atsru->hdr = hdr;
-	atsru->include_all = atsr->flags & 0x1;
-
-	list_add(&atsru->list, &dmar_atsr_units);
-
-	return 0;
-}
-
-static int __init atsr_parse_dev(struct dmar_atsr_unit *atsru)
-{
-	int rc;
-	struct acpi_dmar_atsr *atsr;
-
-	if (atsru->include_all)
-		return 0;
-
-	atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
-	rc = dmar_parse_dev_scope((void *)(atsr + 1),
-				(void *)atsr + atsr->header.length,
-				&atsru->devices_cnt, &atsru->devices,
-				atsr->segment);
-	if (rc || !atsru->devices_cnt) {
-		list_del(&atsru->list);
-		kfree(atsru);
-	}
-
-	return rc;
-}
-
-int dmar_find_matched_atsr_unit(struct pci_dev *dev)
-{
-	int i;
-	struct pci_bus *bus;
-	struct acpi_dmar_atsr *atsr;
-	struct dmar_atsr_unit *atsru;
-
-	dev = pci_physfn(dev);
-
-	list_for_each_entry(atsru, &dmar_atsr_units, list) {
-		atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
-		if (atsr->segment == pci_domain_nr(dev->bus))
-			goto found;
-	}
-
-	return 0;
-
-found:
-	for (bus = dev->bus; bus; bus = bus->parent) {
-		struct pci_dev *bridge = bus->self;
-
-		if (!bridge || !pci_is_pcie(bridge) ||
-		    bridge->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE)
-			return 0;
-
-		if (bridge->pcie_type == PCI_EXP_TYPE_ROOT_PORT) {
-			for (i = 0; i < atsru->devices_cnt; i++)
-				if (atsru->devices[i] == bridge)
-					return 1;
-			break;
-		}
-	}
-
-	if (atsru->include_all)
-		return 1;
-
-	return 0;
-}
-#endif
-
 #ifdef CONFIG_ACPI_NUMA
 static int __init
 dmar_parse_one_rhsa(struct acpi_dmar_header *header)
@@ -484,14 +357,10 @@ parse_dmar_table(void)
 			ret = dmar_parse_one_drhd(entry_header);
 			break;
 		case ACPI_DMAR_TYPE_RESERVED_MEMORY:
-#ifdef CONFIG_DMAR
 			ret = dmar_parse_one_rmrr(entry_header);
-#endif
 			break;
 		case ACPI_DMAR_TYPE_ATSR:
-#ifdef CONFIG_DMAR
 			ret = dmar_parse_one_atsr(entry_header);
-#endif
 			break;
 		case ACPI_DMAR_HARDWARE_AFFINITY:
 #ifdef CONFIG_ACPI_NUMA
@@ -564,30 +433,18 @@ int __init dmar_dev_scope_init(void)
 	if (dmar_dev_scope_initialized)
 		return dmar_dev_scope_initialized;
 
+	if (list_empty(&dmar_drhd_units))
+		goto fail;
+
 	list_for_each_entry_safe(drhd, drhd_n, &dmar_drhd_units, list) {
 		ret = dmar_parse_dev(drhd);
 		if (ret)
 			goto fail;
 	}
 
-#ifdef CONFIG_DMAR
-	{
-		struct dmar_rmrr_unit *rmrr, *rmrr_n;
-		struct dmar_atsr_unit *atsr, *atsr_n;
-
-		list_for_each_entry_safe(rmrr, rmrr_n, &dmar_rmrr_units, list) {
-			ret = rmrr_parse_dev(rmrr);
-			if (ret)
-				goto fail;
-		}
-
-		list_for_each_entry_safe(atsr, atsr_n, &dmar_atsr_units, list) {
-			ret = atsr_parse_dev(atsr);
-			if (ret)
-				goto fail;
-		}
-	}
-#endif
+	ret = dmar_parse_rmrr_atsr_dev();
+	if (ret)
+		goto fail;
 
 	dmar_dev_scope_initialized = 1;
 	return 0;
@@ -620,14 +477,6 @@ int __init dmar_table_init(void)
 		return -ENODEV;
 	}
 
-#ifdef CONFIG_DMAR
-	if (list_empty(&dmar_rmrr_units))
-		printk(KERN_INFO PREFIX "No RMRR found\n");
-
-	if (list_empty(&dmar_atsr_units))
-		printk(KERN_INFO PREFIX "No ATSR found\n");
-#endif
-
 	return 0;
 }
 
@@ -767,7 +616,6 @@ int alloc_iommu(struct dmar_drhd_unit *d
 		goto err_unmap;
 	}
 
-#ifdef CONFIG_DMAR
 	agaw = iommu_calculate_agaw(iommu);
 	if (agaw < 0) {
 		printk(KERN_ERR
@@ -782,7 +630,6 @@ int alloc_iommu(struct dmar_drhd_unit *d
 			iommu->seq_id);
 		goto err_unmap;
 	}
-#endif
 	iommu->agaw = agaw;
 	iommu->msagaw = msagaw;
 
@@ -826,9 +673,7 @@ void free_iommu(struct intel_iommu *iomm
 	if (!iommu)
 		return;
 
-#ifdef CONFIG_DMAR
 	free_dmar_iommu(iommu);
-#endif
 
 	if (iommu->reg)
 		iounmap(iommu->reg);
Index: tree/drivers/iommu/intel-iommu.c
===================================================================
--- tree.orig/drivers/iommu/intel-iommu.c
+++ tree/drivers/iommu/intel-iommu.c
@@ -3390,6 +3390,151 @@ static void __init init_iommu_pm_ops(voi
 static inline void init_iommu_pm_ops(void) {}
 #endif	/* CONFIG_PM */
 
+LIST_HEAD(dmar_rmrr_units);
+
+static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
+{
+	list_add(&rmrr->list, &dmar_rmrr_units);
+}
+
+
+int __init dmar_parse_one_rmrr(struct acpi_dmar_header *header)
+{
+	struct acpi_dmar_reserved_memory *rmrr;
+	struct dmar_rmrr_unit *rmrru;
+
+	rmrru = kzalloc(sizeof(*rmrru), GFP_KERNEL);
+	if (!rmrru)
+		return -ENOMEM;
+
+	rmrru->hdr = header;
+	rmrr = (struct acpi_dmar_reserved_memory *)header;
+	rmrru->base_address = rmrr->base_address;
+	rmrru->end_address = rmrr->end_address;
+
+	dmar_register_rmrr_unit(rmrru);
+	return 0;
+}
+
+static int __init
+rmrr_parse_dev(struct dmar_rmrr_unit *rmrru)
+{
+	struct acpi_dmar_reserved_memory *rmrr;
+	int ret;
+
+	rmrr = (struct acpi_dmar_reserved_memory *) rmrru->hdr;
+	ret = dmar_parse_dev_scope((void *)(rmrr + 1),
+		((void *)rmrr) + rmrr->header.length,
+		&rmrru->devices_cnt, &rmrru->devices, rmrr->segment);
+
+	if (ret || (rmrru->devices_cnt == 0)) {
+		list_del(&rmrru->list);
+		kfree(rmrru);
+	}
+	return ret;
+}
+
+static LIST_HEAD(dmar_atsr_units);
+
+int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
+{
+	struct acpi_dmar_atsr *atsr;
+	struct dmar_atsr_unit *atsru;
+
+	atsr = container_of(hdr, struct acpi_dmar_atsr, header);
+	atsru = kzalloc(sizeof(*atsru), GFP_KERNEL);
+	if (!atsru)
+		return -ENOMEM;
+
+	atsru->hdr = hdr;
+	atsru->include_all = atsr->flags & 0x1;
+
+	list_add(&atsru->list, &dmar_atsr_units);
+
+	return 0;
+}
+
+static int __init atsr_parse_dev(struct dmar_atsr_unit *atsru)
+{
+	int rc;
+	struct acpi_dmar_atsr *atsr;
+
+	if (atsru->include_all)
+		return 0;
+
+	atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
+	rc = dmar_parse_dev_scope((void *)(atsr + 1),
+				(void *)atsr + atsr->header.length,
+				&atsru->devices_cnt, &atsru->devices,
+				atsr->segment);
+	if (rc || !atsru->devices_cnt) {
+		list_del(&atsru->list);
+		kfree(atsru);
+	}
+
+	return rc;
+}
+
+int dmar_find_matched_atsr_unit(struct pci_dev *dev)
+{
+	int i;
+	struct pci_bus *bus;
+	struct acpi_dmar_atsr *atsr;
+	struct dmar_atsr_unit *atsru;
+
+	dev = pci_physfn(dev);
+
+	list_for_each_entry(atsru, &dmar_atsr_units, list) {
+		atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
+		if (atsr->segment == pci_domain_nr(dev->bus))
+			goto found;
+	}
+
+	return 0;
+
+found:
+	for (bus = dev->bus; bus; bus = bus->parent) {
+		struct pci_dev *bridge = bus->self;
+
+		if (!bridge || !pci_is_pcie(bridge) ||
+		    bridge->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE)
+			return 0;
+
+		if (bridge->pcie_type == PCI_EXP_TYPE_ROOT_PORT) {
+			for (i = 0; i < atsru->devices_cnt; i++)
+				if (atsru->devices[i] == bridge)
+					return 1;
+			break;
+		}
+	}
+
+	if (atsru->include_all)
+		return 1;
+
+	return 0;
+}
+
+int dmar_parse_rmrr_atsr_dev(void)
+{
+	struct dmar_rmrr_unit *rmrr, *rmrr_n;
+	struct dmar_atsr_unit *atsr, *atsr_n;
+	int ret = 0;
+
+	list_for_each_entry_safe(rmrr, rmrr_n, &dmar_rmrr_units, list) {
+		ret = rmrr_parse_dev(rmrr);
+		if (ret)
+			return ret;
+	}
+
+	list_for_each_entry_safe(atsr, atsr_n, &dmar_atsr_units, list) {
+		ret = atsr_parse_dev(atsr);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
 /*
  * Here we only respond to action of unbound device from driver.
  *
@@ -3454,6 +3599,12 @@ int __init intel_iommu_init(void)
 		return 	-ENODEV;
 	}
 
+	if (list_empty(&dmar_rmrr_units))
+		printk(KERN_INFO "DMAR: No RMRR found\n");
+
+	if (list_empty(&dmar_atsr_units))
+		printk(KERN_INFO "DMAR: No ATSR found\n");
+
 	if (dmar_init_reserved_ranges()) {
 		if (force_on)
 			panic("tboot: Failed to reserve iommu ranges\n");
Index: tree/include/linux/dma_remapping.h
===================================================================
--- tree.orig/include/linux/dma_remapping.h
+++ tree/include/linux/dma_remapping.h
@@ -25,9 +25,9 @@ struct intel_iommu;
 struct dmar_domain;
 struct root_entry;
 
-extern void free_dmar_iommu(struct intel_iommu *iommu);
 
 #ifdef CONFIG_DMAR
+extern void free_dmar_iommu(struct intel_iommu *iommu);
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
 #else
@@ -39,6 +39,9 @@ static inline int iommu_calculate_max_sa
 {
 	return 0;
 }
+static inline void free_dmar_iommu(struct intel_iommu *iommu)
+{
+}
 #endif
 
 extern int dmar_disabled;
Index: tree/include/linux/dmar.h
===================================================================
--- tree.orig/include/linux/dmar.h
+++ tree/include/linux/dmar.h
@@ -237,9 +237,26 @@ struct dmar_atsr_unit {
 	u8 include_all:1;		/* include all ports */
 };
 
+int dmar_parse_rmrr_atsr_dev(void);
+extern int dmar_parse_one_rmrr(struct acpi_dmar_header *header);
+extern int dmar_parse_one_atsr(struct acpi_dmar_header *header);
+extern int dmar_parse_dev_scope(void *start, void *end, int *cnt,
+				struct pci_dev ***devices, u16 segment);
 extern int intel_iommu_init(void);
 #else /* !CONFIG_DMAR: */
 static inline int intel_iommu_init(void) { return -ENODEV; }
+static inline int dmar_parse_one_rmrr(struct acpi_dmar_header *header)
+{
+	return 0;
+}
+static inline int dmar_parse_one_atsr(struct acpi_dmar_header *header)
+{
+	return 0;
+}
+static inline int dmar_parse_rmrr_atsr_dev(void)
+{
+	return 0;
+}
 #endif /* CONFIG_DMAR */
 
 #endif /* __DMAR_H__ */

[patch 4/8] iommu: no need to set dmar_disabled in check_zero_address()

[Suresh Siddha]

[-- Attachment #1: remove_dmar_disabled_in_check_address.patch --]
[-- Type: text/plain, Size: 844 bytes --]

Before the restructruing of the x86 IOMMU code, intel_iommu_init()
was getting called directly from pci_iommu_init() and hence needed
to explicitly set dmar_disabled to 1 for the failure conditions of
check_zero_address().

Recent changes don't call intel_iommu_init() if the intel iommu
detection fails as a result of failure in check_zero_address().

So no need for this ifdef and the code inside it.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 drivers/iommu/dmar.c |    3 ---
 1 file changed, 3 deletions(-)

Index: tree/drivers/iommu/dmar.c
===================================================================
--- tree.orig/drivers/iommu/dmar.c
+++ tree/drivers/iommu/dmar.c
@@ -540,9 +540,6 @@ int __init check_zero_address(void)
 	return 1;
 
 failed:
-#ifdef CONFIG_DMAR
-	dmar_disabled = 1;
-#endif
 	return 0;
 }
 

[patch 5/8] iommu: cleanup ifdefs in detect_intel_iommu()

[Suresh Siddha]

[-- Attachment #1: add_dmar_irq_remapping_disabled_routines.patch --]
[-- Type: text/plain, Size: 2708 bytes --]

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 arch/ia64/include/asm/iommu.h |    6 ++++--
 drivers/iommu/dmar.c          |   13 ++++++-------
 include/linux/dma_remapping.h |    3 ++-
 3 files changed, 12 insertions(+), 10 deletions(-)

Index: linux-2.6-tip/drivers/iommu/dmar.c
===================================================================
--- linux-2.6-tip.orig/drivers/iommu/dmar.c
+++ linux-2.6-tip/drivers/iommu/dmar.c
@@ -551,22 +551,21 @@ int __init detect_intel_iommu(void)
 	if (ret)
 		ret = check_zero_address();
 	{
-#ifdef CONFIG_INTR_REMAP
 		struct acpi_table_dmar *dmar;
 
 		dmar = (struct acpi_table_dmar *) dmar_tbl;
-		if (ret && cpu_has_x2apic && dmar->flags & 0x1)
+
+		if (ret && intr_remapping_enabled && cpu_has_x2apic &&
+		    dmar->flags & 0x1)
 			printk(KERN_INFO
-			       "Queued invalidation will be enabled to support "
-			       "x2apic and Intr-remapping.\n");
-#endif
-#ifdef CONFIG_DMAR
+			       "Queued invalidation will be enabled to support x2apic and Intr-remapping.\n");
+
 		if (ret && !no_iommu && !iommu_detected && !dmar_disabled) {
 			iommu_detected = 1;
 			/* Make sure ACS will be enabled */
 			pci_request_acs();
 		}
-#endif
+
 #ifdef CONFIG_X86
 		if (ret)
 			x86_init.iommu.iommu_init = intel_iommu_init;
Index: linux-2.6-tip/include/linux/dma_remapping.h
===================================================================
--- linux-2.6-tip.orig/include/linux/dma_remapping.h
+++ linux-2.6-tip/include/linux/dma_remapping.h
@@ -30,6 +30,7 @@ struct root_entry;
 extern void free_dmar_iommu(struct intel_iommu *iommu);
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
+extern int dmar_disabled;
 #else
 static inline int iommu_calculate_agaw(struct intel_iommu *iommu)
 {
@@ -42,8 +43,8 @@ static inline int iommu_calculate_max_sa
 static inline void free_dmar_iommu(struct intel_iommu *iommu)
 {
 }
+#define dmar_disabled	(1)
 #endif
 
-extern int dmar_disabled;
 
 #endif
Index: linux-2.6-tip/arch/ia64/include/asm/iommu.h
===================================================================
--- linux-2.6-tip.orig/arch/ia64/include/asm/iommu.h
+++ linux-2.6-tip/arch/ia64/include/asm/iommu.h
@@ -7,12 +7,14 @@
 
 extern void pci_iommu_shutdown(void);
 extern void no_iommu_init(void);
-extern int force_iommu, no_iommu;
-extern int iommu_detected;
 #ifdef	CONFIG_DMAR
+extern int force_iommu, no_iommu;
 extern int iommu_pass_through;
+extern int iommu_detected;
 #else
 #define iommu_pass_through	(0)
+#define no_iommu		(1)
+#define iommu_detected		(0)
 #endif
 extern void iommu_dma_init(void);
 extern void machvec_init(const char *name);

[patch 6/8] x86, msi, intr-remap: use the ioapic set affinity routine

[Suresh Siddha]

[-- Attachment #1: remove_ir_msi_set_affinity.patch --]
[-- Type: text/plain, Size: 3316 bytes --]

IRQ set affinity routine is same for the IO-APIC IRQ's aswell as the MSI
IRQ's in the presence of interrupt-remapping. This is because we modify
the interrupt-remapping table entry and doesn't touch the IO-APIC RTE or the
MSI entry.

So remove the ir_msi_set_affinity() and re-use the ir_ioapic_set_affinity()

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 arch/x86/kernel/apic/io_apic.c |   54 +++++++++--------------------------------
 1 file changed, 12 insertions(+), 42 deletions(-)

Index: tree/arch/x86/kernel/apic/io_apic.c
===================================================================
--- tree.orig/arch/x86/kernel/apic/io_apic.c
+++ tree/arch/x86/kernel/apic/io_apic.c
@@ -2267,6 +2267,9 @@ ioapic_set_affinity(struct irq_data *dat
  * updated vector information), by using a virtual vector (io-apic pin number).
  * Real vector that is used for interrupting cpu will be coming from
  * the interrupt-remapping table entry.
+ *
+ * As the migration is a simple atomic update of IRTE, the same mechanism
+ * is used to migrate MSI irq's in the presence of interrupt-remapping.
  */
 static int
 ir_ioapic_set_affinity(struct irq_data *data, const struct cpumask *mask,
@@ -2291,10 +2294,16 @@ ir_ioapic_set_affinity(struct irq_data *
 	irte.dest_id = IRTE_DEST(dest);
 
 	/*
-	 * Modified the IRTE and flushes the Interrupt entry cache.
+	 * Atomically updates the IRTE with the new destination, vector
+	 * and flushes the interrupt entry cache.
 	 */
 	modify_irte(irq, &irte);
 
+	/*
+	 * After this point, all the interrupts will start arriving
+	 * at the new destination. So, time to cleanup the previous
+	 * vector allocation.
+	 */
 	if (cfg->move_in_progress)
 		send_cleanup_vector(cfg);
 
@@ -3144,45 +3153,6 @@ msi_set_affinity(struct irq_data *data, 
 
 	return 0;
 }
-#ifdef CONFIG_INTR_REMAP
-/*
- * Migrate the MSI irq to another cpumask. This migration is
- * done in the process context using interrupt-remapping hardware.
- */
-static int
-ir_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
-		    bool force)
-{
-	struct irq_cfg *cfg = data->chip_data;
-	unsigned int dest, irq = data->irq;
-	struct irte irte;
-
-	if (get_irte(irq, &irte))
-		return -1;
-
-	if (__ioapic_set_affinity(data, mask, &dest))
-		return -1;
-
-	irte.vector = cfg->vector;
-	irte.dest_id = IRTE_DEST(dest);
-
-	/*
-	 * atomically update the IRTE with the new destination and vector.
-	 */
-	modify_irte(irq, &irte);
-
-	/*
-	 * After this point, all the interrupts will start arriving
-	 * at the new destination. So, time to cleanup the previous
-	 * vector allocation.
-	 */
-	if (cfg->move_in_progress)
-		send_cleanup_vector(cfg);
-
-	return 0;
-}
-
-#endif
 #endif /* CONFIG_SMP */
 
 /*
@@ -3207,7 +3177,7 @@ static struct irq_chip msi_ir_chip = {
 #ifdef CONFIG_INTR_REMAP
 	.irq_ack		= ir_ack_apic_edge,
 #ifdef CONFIG_SMP
-	.irq_set_affinity	= ir_msi_set_affinity,
+	.irq_set_affinity	= ir_ioapic_set_affinity,
 #endif
 #endif
 	.irq_retrigger		= ioapic_retrigger_irq,
@@ -3416,7 +3386,7 @@ static struct irq_chip ir_hpet_msi_type 
 #ifdef CONFIG_INTR_REMAP
 	.irq_ack		= ir_ack_apic_edge,
 #ifdef CONFIG_SMP
-	.irq_set_affinity	= ir_msi_set_affinity,
+	.irq_set_affinity	= ir_ioapic_set_affinity,
 #endif
 #endif
 	.irq_retrigger		= ioapic_retrigger_irq,

[patch 7/8] x86, ioapic: define irq_remap_modify_chip_defaults()

[Suresh Siddha]

[-- Attachment #1: ioapic_ifdef_intr_remap_cleanups.patch --]
[-- Type: text/plain, Size: 4623 bytes --]

define irq_remap_modify_chip_defaults() and remove the duplicate code,
cleanup the unnecessary ifdefs.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 arch/x86/include/asm/irq_remapping.h |    4 ++
 arch/x86/kernel/apic/io_apic.c       |   64 ++++++++++-------------------------
 2 files changed, 23 insertions(+), 45 deletions(-)

Index: tree/arch/x86/include/asm/irq_remapping.h
===================================================================
--- tree.orig/arch/x86/include/asm/irq_remapping.h
+++ tree/arch/x86/include/asm/irq_remapping.h
@@ -4,6 +4,7 @@
 #define IRTE_DEST(dest) ((x2apic_mode) ? dest : dest << 8)
 
 #ifdef CONFIG_INTR_REMAP
+static void irq_remap_modify_chip_defaults(struct irq_chip *chip);
 static inline void prepare_irte(struct irte *irte, int vector,
 			        unsigned int dest)
 {
@@ -36,6 +37,9 @@ static inline bool irq_remapped(struct i
 {
 	return false;
 }
+static inline void irq_remap_modify_chip_defaults(struct irq_chip *chip)
+{
+}
 #endif
 
 #endif	/* _ASM_X86_IRQ_REMAPPING_H */
Index: tree/arch/x86/kernel/apic/io_apic.c
===================================================================
--- tree.orig/arch/x86/kernel/apic/io_apic.c
+++ tree/arch/x86/kernel/apic/io_apic.c
@@ -1202,7 +1202,6 @@ void __setup_vector_irq(int cpu)
 }
 
 static struct irq_chip ioapic_chip;
-static struct irq_chip ir_ioapic_chip;
 
 #ifdef CONFIG_X86_32
 static inline int IO_APIC_irq_trigger(int irq)
@@ -1246,7 +1245,7 @@ static void ioapic_register_intr(unsigne
 
 	if (irq_remapped(cfg)) {
 		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-		chip = &ir_ioapic_chip;
+		irq_remap_modify_chip_defaults(chip);
 		fasteoi = trigger != 0;
 	}
 
@@ -2572,6 +2571,22 @@ static void ir_ack_apic_level(struct irq
 	ack_APIC_irq();
 	eoi_ioapic_irq(data->irq, data->chip_data);
 }
+
+static void ir_print_prefix(struct irq_data *data, struct seq_file *p)
+{
+	seq_printf(p, " IR-%s", data->chip->name);
+}
+
+static void irq_remap_modify_chip_defaults(struct irq_chip *chip)
+{
+	chip->irq_print_chip = ir_print_prefix;
+	chip->irq_ack = ir_ack_apic_edge;
+	chip->irq_eoi = ir_ack_apic_level;
+
+#ifdef CONFIG_SMP
+	chip->irq_set_affinity = ir_ioapic_set_affinity;
+#endif
+}
 #endif /* CONFIG_INTR_REMAP */
 
 static struct irq_chip ioapic_chip __read_mostly = {
@@ -2587,21 +2602,6 @@ static struct irq_chip ioapic_chip __rea
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
-static struct irq_chip ir_ioapic_chip __read_mostly = {
-	.name			= "IR-IO-APIC",
-	.irq_startup		= startup_ioapic_irq,
-	.irq_mask		= mask_ioapic_irq,
-	.irq_unmask		= unmask_ioapic_irq,
-#ifdef CONFIG_INTR_REMAP
-	.irq_ack		= ir_ack_apic_edge,
-	.irq_eoi		= ir_ack_apic_level,
-#ifdef CONFIG_SMP
-	.irq_set_affinity	= ir_ioapic_set_affinity,
-#endif
-#endif
-	.irq_retrigger		= ioapic_retrigger_irq,
-};
-
 static inline void init_IO_APIC_traps(void)
 {
 	struct irq_cfg *cfg;
@@ -3170,19 +3170,6 @@ static struct irq_chip msi_chip = {
 	.irq_retrigger		= ioapic_retrigger_irq,
 };
 
-static struct irq_chip msi_ir_chip = {
-	.name			= "IR-PCI-MSI",
-	.irq_unmask		= unmask_msi_irq,
-	.irq_mask		= mask_msi_irq,
-#ifdef CONFIG_INTR_REMAP
-	.irq_ack		= ir_ack_apic_edge,
-#ifdef CONFIG_SMP
-	.irq_set_affinity	= ir_ioapic_set_affinity,
-#endif
-#endif
-	.irq_retrigger		= ioapic_retrigger_irq,
-};
-
 /*
  * Map the PCI dev to the corresponding remapping hardware unit
  * and allocate 'nvec' consecutive interrupt-remapping table entries
@@ -3225,7 +3212,7 @@ static int setup_msi_irq(struct pci_dev 
 
 	if (irq_remapped(irq_get_chip_data(irq))) {
 		irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
-		chip = &msi_ir_chip;
+		irq_remap_modify_chip_defaults(chip);
 	}
 
 	irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
@@ -3379,19 +3366,6 @@ static int hpet_msi_set_affinity(struct 
 
 #endif /* CONFIG_SMP */
 
-static struct irq_chip ir_hpet_msi_type = {
-	.name			= "IR-HPET_MSI",
-	.irq_unmask		= hpet_msi_unmask,
-	.irq_mask		= hpet_msi_mask,
-#ifdef CONFIG_INTR_REMAP
-	.irq_ack		= ir_ack_apic_edge,
-#ifdef CONFIG_SMP
-	.irq_set_affinity	= ir_ioapic_set_affinity,
-#endif
-#endif
-	.irq_retrigger		= ioapic_retrigger_irq,
-};
-
 static struct irq_chip hpet_msi_type = {
 	.name = "HPET_MSI",
 	.irq_unmask = hpet_msi_unmask,
@@ -3428,7 +3402,7 @@ int arch_setup_hpet_msi(unsigned int irq
 	hpet_msi_write(irq_get_handler_data(irq), &msg);
 	irq_set_status_flags(irq, IRQ_MOVE_PCNTXT);
 	if (irq_remapped(irq_get_chip_data(irq)))
-		chip = &ir_hpet_msi_type;
+		irq_remap_modify_chip_defaults(chip);
 
 	irq_set_chip_and_handler_name(irq, chip, handle_edge_irq, "edge");
 	return 0;

[patch 8/8] iommu: rename the DMAR and INTR_REMAP config options

[Suresh Siddha]

[-- Attachment #1: change_dmar_intr_remap_config_options.patch --]
[-- Type: text/plain, Size: 18623 bytes --]

Change the CONFIG_DMAR to CONFIG_INTEL_IOMMU to be consistent with the other
IOMMU options.

Rename the CONFIG_INTR_REMAP to CONFIG_IRQ_REMAP to match the irq
subsystem name.

And define the CONFIG_DMAR_TABLE for the common ACPI DMAR routines
shared by both CONFIG_INTEL_IOMMU and CONFIG_IRQ_REMAP.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
---
 arch/ia64/configs/generic_defconfig  |    2 +-
 arch/ia64/dig/Makefile               |    2 +-
 arch/ia64/include/asm/device.h       |    2 +-
 arch/ia64/include/asm/iommu.h        |    2 +-
 arch/ia64/include/asm/pci.h          |    2 +-
 arch/ia64/kernel/Makefile            |    2 +-
 arch/ia64/kernel/acpi.c              |    4 ++--
 arch/ia64/kernel/msi_ia64.c          |    4 ++--
 arch/ia64/kernel/pci-dma.c           |    2 +-
 arch/x86/Kconfig                     |    6 +++---
 arch/x86/configs/x86_64_defconfig    |    4 ++--
 arch/x86/include/asm/device.h        |    2 +-
 arch/x86/include/asm/hw_irq.h        |    2 +-
 arch/x86/include/asm/irq_remapping.h |    2 +-
 arch/x86/kernel/apic/apic.c          |    2 +-
 arch/x86/kernel/apic/io_apic.c       |    8 ++++----
 drivers/char/agp/intel-gtt.c         |    4 ++--
 drivers/iommu/Kconfig                |   25 +++++++++++++++----------
 drivers/iommu/Makefile               |    5 +++--
 drivers/iommu/intel-iommu.c          |   10 +++++-----
 drivers/pci/quirks.c                 |    2 +-
 include/linux/dma_remapping.h        |    2 +-
 include/linux/dmar.h                 |   12 ++++++------
 include/linux/intel-iommu.h          |    6 +++---
 24 files changed, 60 insertions(+), 54 deletions(-)

Index: tree/arch/x86/kernel/apic/io_apic.c
===================================================================
--- tree.orig/arch/x86/kernel/apic/io_apic.c
+++ tree/arch/x86/kernel/apic/io_apic.c
@@ -2254,7 +2254,7 @@ ioapic_set_affinity(struct irq_data *dat
 	return ret;
 }
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 
 /*
  * Migrate the IO-APIC irq in the presence of intr-remapping.
@@ -2560,7 +2560,7 @@ static void ack_apic_level(struct irq_da
 	}
 }
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 static void ir_ack_apic_edge(struct irq_data *data)
 {
 	ack_APIC_irq();
@@ -2587,7 +2587,7 @@ static void irq_remap_modify_chip_defaul
 	chip->irq_set_affinity = ir_ioapic_set_affinity;
 #endif
 }
-#endif /* CONFIG_INTR_REMAP */
+#endif /* CONFIG_IRQ_REMAP */
 
 static struct irq_chip ioapic_chip __read_mostly = {
 	.name			= "IO-APIC",
@@ -3285,7 +3285,7 @@ void native_teardown_msi_irq(unsigned in
 	destroy_irq(irq);
 }
 
-#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
+#ifdef CONFIG_DMAR_TABLE
 #ifdef CONFIG_SMP
 static int
 dmar_msi_set_affinity(struct irq_data *data, const struct cpumask *mask,
Index: tree/drivers/iommu/Kconfig
===================================================================
--- tree.orig/drivers/iommu/Kconfig
+++ tree/drivers/iommu/Kconfig
@@ -59,10 +59,14 @@ config AMD_IOMMU_STATS
 	  If unsure, say N.
 
 # Intel IOMMU support
-config DMAR
-	bool "Support for DMA Remapping Devices"
+config DMAR_TABLE
+	bool
+
+config INTEL_IOMMU
+	bool "Support for Intel IOMMU using DMA Remapping Devices"
 	depends on PCI_MSI && ACPI && (X86 || IA64_GENERIC)
 	select IOMMU_API
+	select DMAR_TABLE
 	help
 	  DMA remapping (DMAR) devices support enables independent address
 	  translations for Direct Memory Access (DMA) from devices.
@@ -70,18 +74,18 @@ config DMAR
 	  and include PCI device scope covered by these DMA
 	  remapping devices.
 
-config DMAR_DEFAULT_ON
+config INTEL_IOMMU_DEFAULT_ON
 	def_bool y
-	prompt "Enable DMA Remapping Devices by default"
-	depends on DMAR
+	prompt "Enable Intel DMA Remapping Devices by default"
+	depends on INTEL_IOMMU
 	help
 	  Selecting this option will enable a DMAR device at boot time if
 	  one is found. If this option is not selected, DMAR support can
 	  be enabled by passing intel_iommu=on to the kernel.
 
-config DMAR_BROKEN_GFX_WA
+config INTEL_IOMMU_BROKEN_GFX_WA
 	bool "Workaround broken graphics drivers (going away soon)"
-	depends on DMAR && BROKEN && X86
+	depends on INTEL_IOMMU && BROKEN && X86
 	---help---
 	  Current Graphics drivers tend to use physical address
 	  for DMA and avoid using DMA APIs. Setting this config
@@ -90,18 +94,19 @@ config DMAR_BROKEN_GFX_WA
 	  to use physical addresses for DMA, at least until this
 	  option is removed in the 2.6.32 kernel.
 
-config DMAR_FLOPPY_WA
+config INTEL_IOMMU_FLOPPY_WA
 	def_bool y
-	depends on DMAR && X86
+	depends on INTEL_IOMMU && X86
 	---help---
 	  Floppy disk drivers are known to bypass DMA API calls
 	  thereby failing to work when IOMMU is enabled. This
 	  workaround will setup a 1:1 mapping for the first
 	  16MiB to make floppy (an ISA device) work.
 
-config INTR_REMAP
+config IRQ_REMAP
 	bool "Support for Interrupt Remapping (EXPERIMENTAL)"
 	depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI && EXPERIMENTAL
+	select DMAR_TABLE
 	---help---
 	  Supports Interrupt remapping for IO-APIC and MSI devices.
 	  To use x2apic mode in the CPU's which support x2APIC enhancements or
Index: tree/drivers/iommu/Makefile
===================================================================
--- tree.orig/drivers/iommu/Makefile
+++ tree/drivers/iommu/Makefile
@@ -1,5 +1,6 @@
 obj-$(CONFIG_IOMMU_API) += iommu.o
 obj-$(CONFIG_MSM_IOMMU) += msm_iommu.o msm_iommu_dev.o
 obj-$(CONFIG_AMD_IOMMU) += amd_iommu.o amd_iommu_init.o
-obj-$(CONFIG_DMAR) += dmar.o iova.o intel-iommu.o
-obj-$(CONFIG_INTR_REMAP) += dmar.o intr_remapping.o
+obj-$(CONFIG_DMAR_TABLE) += dmar.o
+obj-$(CONFIG_INTEL_IOMMU) += iova.o intel-iommu.o
+obj-$(CONFIG_IRQ_REMAP) += intr_remapping.o
Index: tree/include/linux/dmar.h
===================================================================
--- tree.orig/include/linux/dmar.h
+++ tree/include/linux/dmar.h
@@ -31,7 +31,7 @@
 #define DMAR_X2APIC_OPT_OUT	0x2
 
 struct intel_iommu;
-#if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP)
+#ifdef CONFIG_DMAR_TABLE
 extern struct acpi_table_header *dmar_tbl;
 struct dmar_drhd_unit {
 	struct list_head list;		/* list of drhd units	*/
@@ -81,7 +81,7 @@ static inline int enable_drhd_fault_hand
 {
 	return -1;
 }
-#endif /* !CONFIG_DMAR && !CONFIG_INTR_REMAP */
+#endif /* !CONFIG_DMAR_TABLE */
 
 struct irte {
 	union {
@@ -112,7 +112,7 @@ struct irte {
 	};
 };
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 extern int intr_remapping_enabled;
 extern int intr_remapping_supported(void);
 extern int enable_intr_remapping(void);
@@ -214,7 +214,7 @@ extern int dmar_set_interrupt(struct int
 extern irqreturn_t dmar_fault(int irq, void *dev_id);
 extern int arch_setup_dmar_msi(unsigned int irq);
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 extern int iommu_detected, no_iommu;
 extern struct list_head dmar_rmrr_units;
 struct dmar_rmrr_unit {
@@ -243,7 +243,7 @@ extern int dmar_parse_one_atsr(struct ac
 extern int dmar_parse_dev_scope(void *start, void *end, int *cnt,
 				struct pci_dev ***devices, u16 segment);
 extern int intel_iommu_init(void);
-#else /* !CONFIG_DMAR: */
+#else /* !CONFIG_INTEL_IOMMU: */
 static inline int intel_iommu_init(void) { return -ENODEV; }
 static inline int dmar_parse_one_rmrr(struct acpi_dmar_header *header)
 {
@@ -257,6 +257,6 @@ static inline int dmar_parse_rmrr_atsr_d
 {
 	return 0;
 }
-#endif /* CONFIG_DMAR */
+#endif /* CONFIG_INTEL_IOMMU */
 
 #endif /* __DMAR_H__ */
Index: tree/arch/x86/include/asm/device.h
===================================================================
--- tree.orig/arch/x86/include/asm/device.h
+++ tree/arch/x86/include/asm/device.h
@@ -8,7 +8,7 @@ struct dev_archdata {
 #ifdef CONFIG_X86_64
 struct dma_map_ops *dma_ops;
 #endif
-#if defined(CONFIG_DMAR) || defined(CONFIG_AMD_IOMMU)
+#if defined(CONFIG_INTEL_IOMMU) || defined(CONFIG_AMD_IOMMU)
 	void *iommu; /* hook for IOMMU specific extension */
 #endif
 };
Index: tree/drivers/iommu/intel-iommu.c
===================================================================
--- tree.orig/drivers/iommu/intel-iommu.c
+++ tree/drivers/iommu/intel-iommu.c
@@ -393,11 +393,11 @@ static long list_size;
 
 static void domain_remove_dev_info(struct dmar_domain *domain);
 
-#ifdef CONFIG_DMAR_DEFAULT_ON
+#ifdef CONFIG_INTEL_IOMMU_DEFAULT_ON
 int dmar_disabled = 0;
 #else
 int dmar_disabled = 1;
-#endif /*CONFIG_DMAR_DEFAULT_ON*/
+#endif /*CONFIG_INTEL_IOMMU_DEFAULT_ON*/
 
 static int dmar_map_gfx = 1;
 static int dmar_forcedac;
@@ -2150,7 +2150,7 @@ static inline int iommu_prepare_rmrr_dev
 		rmrr->end_address);
 }
 
-#ifdef CONFIG_DMAR_FLOPPY_WA
+#ifdef CONFIG_INTEL_IOMMU_FLOPPY_WA
 static inline void iommu_prepare_isa(void)
 {
 	struct pci_dev *pdev;
@@ -2173,7 +2173,7 @@ static inline void iommu_prepare_isa(voi
 {
 	return;
 }
-#endif /* !CONFIG_DMAR_FLPY_WA */
+#endif /* !CONFIG_INTEL_IOMMU_FLPY_WA */
 
 static int md_domain_init(struct dmar_domain *domain, int guest_width);
 
@@ -2484,7 +2484,7 @@ static int __init init_dmars(void)
 	if (iommu_pass_through)
 		iommu_identity_mapping |= IDENTMAP_ALL;
 
-#ifdef CONFIG_DMAR_BROKEN_GFX_WA
+#ifdef CONFIG_INTEL_IOMMU_BROKEN_GFX_WA
 	iommu_identity_mapping |= IDENTMAP_GFX;
 #endif
 
Index: tree/include/linux/dma_remapping.h
===================================================================
--- tree.orig/include/linux/dma_remapping.h
+++ tree/include/linux/dma_remapping.h
@@ -26,7 +26,7 @@ struct dmar_domain;
 struct root_entry;
 
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 extern void free_dmar_iommu(struct intel_iommu *iommu);
 extern int iommu_calculate_agaw(struct intel_iommu *iommu);
 extern int iommu_calculate_max_sagaw(struct intel_iommu *iommu);
Index: tree/include/linux/intel-iommu.h
===================================================================
--- tree.orig/include/linux/intel-iommu.h
+++ tree/include/linux/intel-iommu.h
@@ -279,7 +279,7 @@ struct q_inval {
 	int             free_cnt;
 };
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 /* 1MB - maximum possible interrupt remapping table size */
 #define INTR_REMAP_PAGE_ORDER	8
 #define INTR_REMAP_TABLE_REG_SIZE	0xf
@@ -318,7 +318,7 @@ struct intel_iommu {
 	unsigned int 	irq;
 	unsigned char 	name[13];    /* Device Name */
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 	unsigned long 	*domain_ids; /* bitmap of domains */
 	struct dmar_domain **domains; /* ptr to domains */
 	spinlock_t	lock; /* protect context, domain ids */
@@ -329,7 +329,7 @@ struct intel_iommu {
 	struct q_inval  *qi;            /* Queued invalidation info */
 	u32 *iommu_state; /* Store iommu states between suspend and resume.*/
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 	struct ir_table *ir_table;	/* Interrupt remapping info */
 #endif
 	int		node;
Index: tree/arch/x86/include/asm/hw_irq.h
===================================================================
--- tree.orig/arch/x86/include/asm/hw_irq.h
+++ tree/arch/x86/include/asm/hw_irq.h
@@ -119,7 +119,7 @@ struct irq_cfg {
 	cpumask_var_t		old_domain;
 	u8			vector;
 	u8			move_in_progress : 1;
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 	struct irq_2_iommu	irq_2_iommu;
 #endif
 };
Index: tree/arch/x86/include/asm/irq_remapping.h
===================================================================
--- tree.orig/arch/x86/include/asm/irq_remapping.h
+++ tree/arch/x86/include/asm/irq_remapping.h
@@ -3,7 +3,7 @@
 
 #define IRTE_DEST(dest) ((x2apic_mode) ? dest : dest << 8)
 
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 static void irq_remap_modify_chip_defaults(struct irq_chip *chip);
 static inline void prepare_irte(struct irte *irte, int vector,
 			        unsigned int dest)
Index: tree/drivers/char/agp/intel-gtt.c
===================================================================
--- tree.orig/drivers/char/agp/intel-gtt.c
+++ tree/drivers/char/agp/intel-gtt.c
@@ -30,10 +30,10 @@
 /*
  * If we have Intel graphics, we're not going to have anything other than
  * an Intel IOMMU. So make the correct use of the PCI DMA API contingent
- * on the Intel IOMMU support (CONFIG_DMAR).
+ * on the Intel IOMMU support (CONFIG_INTEL_IOMMU).
  * Only newer chipsets need to bother with this, of course.
  */
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 #define USE_PCI_DMA_API 1
 #else
 #define USE_PCI_DMA_API 0
Index: tree/drivers/pci/quirks.c
===================================================================
--- tree.orig/drivers/pci/quirks.c
+++ tree/drivers/pci/quirks.c
@@ -2788,7 +2788,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_RI
 DECLARE_PCI_FIXUP_RESUME_EARLY(PCI_VENDOR_ID_RICOH, PCI_DEVICE_ID_RICOH_R5CE823, ricoh_mmc_fixup_r5c832);
 #endif /*CONFIG_MMC_RICOH_MMC*/
 
-#if defined(CONFIG_DMAR) || defined(CONFIG_INTR_REMAP)
+#ifdef CONFIG_DMAR_TABLE
 #define VTUNCERRMSK_REG	0x1ac
 #define VTD_MSK_SPEC_ERRORS	(1 << 31)
 /*
Index: tree/arch/x86/kernel/apic/apic.c
===================================================================
--- tree.orig/arch/x86/kernel/apic/apic.c
+++ tree/arch/x86/kernel/apic/apic.c
@@ -1437,7 +1437,7 @@ void enable_x2apic(void)
 
 int __init enable_IR(void)
 {
-#ifdef CONFIG_INTR_REMAP
+#ifdef CONFIG_IRQ_REMAP
 	if (!intr_remapping_supported()) {
 		pr_debug("intr-remapping not supported\n");
 		return -1;
Index: tree/arch/ia64/configs/generic_defconfig
===================================================================
--- tree.orig/arch/ia64/configs/generic_defconfig
+++ tree/arch/ia64/configs/generic_defconfig
@@ -234,4 +234,4 @@ CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_ANSI_CPRNG is not set
 CONFIG_CRC_T10DIF=y
 CONFIG_MISC_DEVICES=y
-CONFIG_DMAR=y
+CONFIG_INTEL_IOMMU=y
Index: tree/arch/ia64/dig/Makefile
===================================================================
--- tree.orig/arch/ia64/dig/Makefile
+++ tree/arch/ia64/dig/Makefile
@@ -6,7 +6,7 @@
 #
 
 obj-y := setup.o
-ifeq ($(CONFIG_DMAR), y)
+ifeq ($(CONFIG_INTEL_IOMMU), y)
 obj-$(CONFIG_IA64_GENERIC) += machvec.o machvec_vtd.o
 else
 obj-$(CONFIG_IA64_GENERIC) += machvec.o
Index: tree/arch/ia64/include/asm/device.h
===================================================================
--- tree.orig/arch/ia64/include/asm/device.h
+++ tree/arch/ia64/include/asm/device.h
@@ -10,7 +10,7 @@ struct dev_archdata {
 #ifdef CONFIG_ACPI
 	void	*acpi_handle;
 #endif
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 	void *iommu; /* hook for IOMMU specific extension */
 #endif
 };
Index: tree/arch/ia64/include/asm/iommu.h
===================================================================
--- tree.orig/arch/ia64/include/asm/iommu.h
+++ tree/arch/ia64/include/asm/iommu.h
@@ -7,7 +7,7 @@
 
 extern void pci_iommu_shutdown(void);
 extern void no_iommu_init(void);
-#ifdef	CONFIG_DMAR
+#ifdef	CONFIG_INTEL_IOMMU
 extern int force_iommu, no_iommu;
 extern int iommu_pass_through;
 extern int iommu_detected;
Index: tree/arch/ia64/include/asm/pci.h
===================================================================
--- tree.orig/arch/ia64/include/asm/pci.h
+++ tree/arch/ia64/include/asm/pci.h
@@ -139,7 +139,7 @@ static inline int pci_get_legacy_ide_irq
 	return channel ? isa_irq_to_vector(15) : isa_irq_to_vector(14);
 }
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 extern void pci_iommu_alloc(void);
 #endif
 #endif /* _ASM_IA64_PCI_H */
Index: tree/arch/ia64/kernel/Makefile
===================================================================
--- tree.orig/arch/ia64/kernel/Makefile
+++ tree/arch/ia64/kernel/Makefile
@@ -43,7 +43,7 @@ obj-$(CONFIG_IA64_ESI)		+= esi.o
 ifneq ($(CONFIG_IA64_ESI),)
 obj-y				+= esi_stub.o	# must be in kernel proper
 endif
-obj-$(CONFIG_DMAR)		+= pci-dma.o
+obj-$(CONFIG_INTEL_IOMMU)	+= pci-dma.o
 obj-$(CONFIG_SWIOTLB)		+= pci-swiotlb.o
 
 obj-$(CONFIG_BINFMT_ELF)	+= elfcore.o
Index: tree/arch/ia64/kernel/acpi.c
===================================================================
--- tree.orig/arch/ia64/kernel/acpi.c
+++ tree/arch/ia64/kernel/acpi.c
@@ -88,7 +88,7 @@ acpi_get_sysname(void)
 	struct acpi_table_rsdp *rsdp;
 	struct acpi_table_xsdt *xsdt;
 	struct acpi_table_header *hdr;
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 	u64 i, nentries;
 #endif
 
@@ -125,7 +125,7 @@ acpi_get_sysname(void)
 		return "xen";
 	}
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 	/* Look for Intel IOMMU */
 	nentries = (hdr->length - sizeof(*hdr)) /
 			 sizeof(xsdt->table_offset_entry[0]);
Index: tree/arch/ia64/kernel/msi_ia64.c
===================================================================
--- tree.orig/arch/ia64/kernel/msi_ia64.c
+++ tree/arch/ia64/kernel/msi_ia64.c
@@ -131,7 +131,7 @@ void arch_teardown_msi_irq(unsigned int 
 	return ia64_teardown_msi_irq(irq);
 }
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 #ifdef CONFIG_SMP
 static int dmar_msi_set_affinity(struct irq_data *data,
 				 const struct cpumask *mask, bool force)
@@ -210,5 +210,5 @@ int arch_setup_dmar_msi(unsigned int irq
 				      "edge");
 	return 0;
 }
-#endif /* CONFIG_DMAR */
+#endif /* CONFIG_INTEL_IOMMU */
 
Index: tree/arch/ia64/kernel/pci-dma.c
===================================================================
--- tree.orig/arch/ia64/kernel/pci-dma.c
+++ tree/arch/ia64/kernel/pci-dma.c
@@ -14,7 +14,7 @@
 
 #include <asm/system.h>
 
-#ifdef CONFIG_DMAR
+#ifdef CONFIG_INTEL_IOMMU
 
 #include <linux/kernel.h>
 
Index: tree/arch/x86/configs/x86_64_defconfig
===================================================================
--- tree.orig/arch/x86/configs/x86_64_defconfig
+++ tree/arch/x86/configs/x86_64_defconfig
@@ -67,8 +67,8 @@ CONFIG_CPU_FREQ_GOV_PERFORMANCE=y
 CONFIG_CPU_FREQ_GOV_ONDEMAND=y
 CONFIG_X86_ACPI_CPUFREQ=y
 CONFIG_PCI_MMCONFIG=y
-CONFIG_DMAR=y
-# CONFIG_DMAR_DEFAULT_ON is not set
+CONFIG_INTEL_IOMMU=y
+# CONFIG_INTEL_IOMMU_DEFAULT_ON is not set
 CONFIG_PCIEPORTBUS=y
 CONFIG_PCCARD=y
 CONFIG_YENTA=y
Index: tree/arch/x86/Kconfig
===================================================================
--- tree.orig/arch/x86/Kconfig
+++ tree/arch/x86/Kconfig
@@ -130,7 +130,7 @@ config SBUS
 	bool
 
 config NEED_DMA_MAP_STATE
-       def_bool (X86_64 || DMAR || DMA_API_DEBUG)
+       def_bool (X86_64 || INTEL_IOMMU || DMA_API_DEBUG)
 
 config NEED_SG_DMA_LENGTH
 	def_bool y
@@ -220,7 +220,7 @@ config ARCH_SUPPORTS_DEBUG_PAGEALLOC
 
 config HAVE_INTEL_TXT
 	def_bool y
-	depends on EXPERIMENTAL && DMAR && ACPI
+	depends on EXPERIMENTAL && INTEL_IOMMU && ACPI
 
 config X86_32_SMP
 	def_bool y
@@ -287,7 +287,7 @@ config SMP
 
 config X86_X2APIC
 	bool "Support x2apic"
-	depends on X86_LOCAL_APIC && X86_64 && INTR_REMAP
+	depends on X86_LOCAL_APIC && X86_64 && IRQ_REMAP
 	---help---
 	  This enables x2apic support on CPUs that have this feature.
 

Re: [patch 8/8] iommu: rename the DMAR and INTR_REMAP config options

[Yinghai Lu]

On 08/23/2011 05:05 PM, Suresh Siddha wrote:
> Change the CONFIG_DMAR to CONFIG_INTEL_IOMMU to be consistent with the other
> IOMMU options.
> 
> Rename the CONFIG_INTR_REMAP to CONFIG_IRQ_REMAP to match the irq
> subsystem name.
> 
> And define the CONFIG_DMAR_TABLE for the common ACPI DMAR routines
> shared by both CONFIG_INTEL_IOMMU and CONFIG_IRQ_REMAP.
> 

is there something like IRQ_REMAP from AMD?

Yinghai

Re: [patch 6/8] x86, msi, intr-remap: use the ioapic set affinity routine

[Yinghai Lu]

On 08/23/2011 05:05 PM, Suresh Siddha wrote:
> IRQ set affinity routine is same for the IO-APIC IRQ's aswell as the MSI
> IRQ's in the presence of interrupt-remapping. This is because we modify
> the interrupt-remapping table entry and doesn't touch the IO-APIC RTE or the
> MSI entry.
> 
> So remove the ir_msi_set_affinity() and re-use the ir_ioapic_set_affinity()
> 

would be better to rename it to ir_set_affinity() ?

Yinghai

Re: [patch 8/8] iommu: rename the DMAR and INTR_REMAP config options

[Roedel, Joerg]

On Wed, Aug 24, 2011 at 02:47:55AM -0400, Yinghai Lu wrote:
> On 08/23/2011 05:05 PM, Suresh Siddha wrote:
> > Change the CONFIG_DMAR to CONFIG_INTEL_IOMMU to be consistent with the other
> > IOMMU options.
> > 
> > Rename the CONFIG_INTR_REMAP to CONFIG_IRQ_REMAP to match the irq
> > subsystem name.
> > 
> > And define the CONFIG_DMAR_TABLE for the common ACPI DMAR routines
> > shared by both CONFIG_INTEL_IOMMU and CONFIG_IRQ_REMAP.
> 
> is there something like IRQ_REMAP from AMD?

Yes there is. And I digged already through the existing VT-d code to see
how it can be implemented side by side. Doesn't seem easy but should be
doable. And it is on my list for the not too-distant future.

	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632

Re: [patch 3/8] iommu: move IOMMU specific code to intel-iommu.c

[Roedel, Joerg]

On Tue, Aug 23, 2011 at 08:05:20PM -0400, Suresh Siddha wrote:
> +static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
> +int __init dmar_parse_one_rmrr(struct acpi_dmar_header *header)
> +int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
> +int dmar_find_matched_atsr_unit(struct pci_dev *dev)
> +int dmar_parse_rmrr_atsr_dev(void)

If these functions are not DMAR specific, how about replacing the dmar_
prefix too while at it?

	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632

Re: [patch 8/8] iommu: rename the DMAR and INTR_REMAP config options

[Suresh Siddha]

On Wed, 2011-08-24 at 01:28 -0700, Roedel, Joerg wrote:
> On Wed, Aug 24, 2011 at 02:47:55AM -0400, Yinghai Lu wrote:
> > On 08/23/2011 05:05 PM, Suresh Siddha wrote:
> > > Change the CONFIG_DMAR to CONFIG_INTEL_IOMMU to be consistent with the other
> > > IOMMU options.
> > > 
> > > Rename the CONFIG_INTR_REMAP to CONFIG_IRQ_REMAP to match the irq
> > > subsystem name.
> > > 
> > > And define the CONFIG_DMAR_TABLE for the common ACPI DMAR routines
> > > shared by both CONFIG_INTEL_IOMMU and CONFIG_IRQ_REMAP.
> > 
> > is there something like IRQ_REMAP from AMD?
> 
> Yes there is. And I digged already through the existing VT-d code to see
> how it can be implemented side by side. Doesn't seem easy but should be
> doable. And it is on my list for the not too-distant future.

yeah I was not sure how different AMD's implementation is and so left it
as generic IRQ_REMAP. We can revisit this when the AMD's implementation
comes alive.

thanks,
suresh

Re: [patch 3/8] iommu: move IOMMU specific code to intel-iommu.c

[Suresh Siddha]

On Wed, 2011-08-24 at 02:27 -0700, Roedel, Joerg wrote:
> On Tue, Aug 23, 2011 at 08:05:20PM -0400, Suresh Siddha wrote:
> > +static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
> > +int __init dmar_parse_one_rmrr(struct acpi_dmar_header *header)
> > +int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
> > +int dmar_find_matched_atsr_unit(struct pci_dev *dev)
> > +int dmar_parse_rmrr_atsr_dev(void)
> 
> If these functions are not DMAR specific, how about replacing the dmar_
> prefix too while at it?

These parsing functions are APCI DMAR table specific and are used only
by intel-iommu routines. Because of this I didn't change their names but
just moved to intel-iommu.c to cleanup some of the ifdef's in the common
dmar.c code (specific to handling common ACPI DMAR tables etc) shared by
both DMA-remapping (iommu) and IRQ-remapping.

thanks,
suresh

Re: [patch 3/8] iommu: move IOMMU specific code to intel-iommu.c

[Roedel, Joerg]

On Wed, Aug 24, 2011 at 02:31:40PM -0400, Suresh Siddha wrote:
> On Wed, 2011-08-24 at 02:27 -0700, Roedel, Joerg wrote:
> > On Tue, Aug 23, 2011 at 08:05:20PM -0400, Suresh Siddha wrote:
> > > +static void __init dmar_register_rmrr_unit(struct dmar_rmrr_unit *rmrr)
> > > +int __init dmar_parse_one_rmrr(struct acpi_dmar_header *header)
> > > +int __init dmar_parse_one_atsr(struct acpi_dmar_header *hdr)
> > > +int dmar_find_matched_atsr_unit(struct pci_dev *dev)
> > > +int dmar_parse_rmrr_atsr_dev(void)
> > 
> > If these functions are not DMAR specific, how about replacing the dmar_
> > prefix too while at it?
> 
> These parsing functions are APCI DMAR table specific and are used only
> by intel-iommu routines. Because of this I didn't change their names but
> just moved to intel-iommu.c to cleanup some of the ifdef's in the common
> dmar.c code (specific to handling common ACPI DMAR tables etc) shared by
> both DMA-remapping (iommu) and IRQ-remapping.

Okay, should be fine then.

Thanks,

	Joerg

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632