lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Tue, 24 May 2011 11:42:39 -0600
From:	Alex Williamson <alex.williamson@...hat.com>
To:	Yinghai Lu <yinghai@...nel.org>
Cc:	David Woodhouse <dwmw2@...radead.org>,
	Vinod Koul <vinod.koul@...el.com>, linux-pci@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Jesse Barnes <jbarnes@...tuousgeek.org>,
	iommu@...ts.linux-foundation.org,
	Dan Williams <dan.j.williams@...el.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] pci, dmar: Update dmar units devices list during
 hotplug

On Tue, 2011-05-24 at 11:58 +0100, David Woodhouse wrote:
> On Thu, 2011-05-19 at 16:15 -0600, Alex Williamson wrote:
> > I think I'd vote for saving some kind of representation of the bus
> > hierarchy, we probably don't need to list every possible individual
> > device.  Leaving a broken pointer around to be matched up and restored
> > later just seems like a continuation of an idea that was bad to begin
> > with.  Thanks, 
> 
> I agree. We should just process the original ATSR information in
> dmar_find_matched_drhd_unit(), rather than comparing with a list of
> possibly stale pointers.
> 
> I don't quite understand why the list of PCI devices was *ever* done
> like that.

Yinghai,

I thought I might be running into something similar so spent some time
taking a different slant coding up the bug you found.  Turns out I
should have tested your patch first because I wasn't hitting that bug at
all.  The patch below is a work-in-progress that I think fixes the bug
by providing a quick means of re-parsing the scope as needed to match
current struct pci_devs.  It needs testing and cleanup, but feel free to
run with it (or ignore).  Just figured its better to post than waste the
code if you end up doing something similar.  Thanks,

Alex


Not for commit

Signed-off-by: Alex Williamson <alex.williamson@...hat.com>
---

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index 9cbeb0b..1d7bc58 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -59,8 +59,8 @@ static void __init dmar_register_drhd_unit(struct dmar_drhd_unit *drhd)
 		list_add(&drhd->list, &dmar_drhd_units);
 }
 
-static int __init dmar_parse_one_dev_scope(struct acpi_dmar_device_scope *scope,
-					   struct pci_dev **dev, u16 segment)
+struct pci_dev *dmar_get_scope_dev(struct acpi_dmar_device_scope *scope,
+				   u16 segment)
 {
 	struct pci_bus *bus;
 	struct pci_dev *pdev = NULL;
@@ -72,7 +72,7 @@ static int __init dmar_parse_one_dev_scope(struct acpi_dmar_device_scope *scope,
 	count = (scope->length - sizeof(struct acpi_dmar_device_scope))
 		/ sizeof(struct acpi_dmar_pci_path);
 
-	while (count) {
+	for (; count; path++, count--, bus = pdev->subordinate) {
 		if (pdev)
 			pci_dev_put(pdev);
 		/*
@@ -80,82 +80,103 @@ static int __init dmar_parse_one_dev_scope(struct acpi_dmar_device_scope *scope,
 		 * ignore it
 		 */
 		if (!bus) {
-			printk(KERN_WARNING
-			PREFIX "Device scope bus [%d] not found\n",
-			scope->bus);
-			break;
+			printk(KERN_WARNING PREFIX
+			       "Device scope bus [%d] not found\n", scope->bus);
+			return NULL;
 		}
 		pdev = pci_get_slot(bus, PCI_DEVFN(path->dev, path->fn));
 		if (!pdev) {
 			printk(KERN_WARNING PREFIX
-			"Device scope device [%04x:%02x:%02x.%02x] not found\n",
-				segment, bus->number, path->dev, path->fn);
-			break;
+			       "Device scope device [%04x:%02x:%02x.%02x] not found\n",
+			       segment, bus->number, path->dev, path->fn);
+			return NULL;
 		}
-		path ++;
-		count --;
-		bus = pdev->subordinate;
 	}
-	if (!pdev) {
-		printk(KERN_WARNING PREFIX
-		"Device scope device [%04x:%02x:%02x.%02x] not found\n",
-		segment, scope->bus, path->dev, path->fn);
-		*dev = NULL;
+
+	return pdev;
+}
+
+static int dmar_match_scope_one(struct acpi_dmar_device_scope *scope,
+				struct pci_dev *dev, u16 segment)
+{
+	struct pci_dev *pdev;
+	int ret = 0;
+
+	if (segment != pci_domain_nr(dev->bus))
 		return 0;
+
+	pdev = dmar_get_scope_dev(scope, segment);
+	if (!pdev)
+		return 0;
+
+	if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT) {
+		if (dev == pdev)
+			ret = 1;
+	} else {
+		while (dev) {
+			if (dev == pdev) {
+				ret = 1;
+				break;
+			}
+			dev = dev->bus->self;
+		}
 	}
-	if ((scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT && \
-			pdev->subordinate) || (scope->entry_type == \
-			ACPI_DMAR_SCOPE_TYPE_BRIDGE && !pdev->subordinate)) {
-		pci_dev_put(pdev);
-		printk(KERN_WARNING PREFIX
-			"Device scope type does not match for %s\n",
-			 pci_name(pdev));
-		return -EINVAL;
+
+	pci_dev_put(pdev);
+
+	return ret;
+}
+
+int dmar_match_scope(struct acpi_dmar_device_scope **scopes, int cnt,
+		     struct pci_dev *dev, u16 segment)
+{
+	int i;
+
+	for (i = 0; i < cnt; i++) {
+		if (dmar_match_scope_one(scopes[i], dev, segment))
+			return 1;
 	}
-	*dev = pdev;
 	return 0;
 }
 
 static int __init dmar_parse_dev_scope(void *start, void *end, int *cnt,
-				       struct pci_dev ***devices, u16 segment)
+				       struct acpi_dmar_device_scope ***scopes)
 {
 	struct acpi_dmar_device_scope *scope;
-	void * tmp = start;
-	int index;
-	int ret;
+	void *tmp = start;
+	int index = 0;
 
 	*cnt = 0;
+
 	while (start < end) {
 		scope = start;
+
 		if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT ||
 		    scope->entry_type == ACPI_DMAR_SCOPE_TYPE_BRIDGE)
 			(*cnt)++;
 		else
 			printk(KERN_WARNING PREFIX
-				"Unsupported device scope\n");
+			       "Unsupported device scope\n");
+
 		start += scope->length;
 	}
+
 	if (*cnt == 0)
 		return 0;
 
-	*devices = kcalloc(*cnt, sizeof(struct pci_dev *), GFP_KERNEL);
-	if (!*devices)
+	*scopes = kcalloc(*cnt, sizeof(struct acpi_dmar_device_scope *),
+                          GFP_KERNEL);
+	if (!*scopes)
 		return -ENOMEM;
 
 	start = tmp;
-	index = 0;
 	while (start < end) {
 		scope = start;
+
 		if (scope->entry_type == ACPI_DMAR_SCOPE_TYPE_ENDPOINT ||
-		    scope->entry_type == ACPI_DMAR_SCOPE_TYPE_BRIDGE) {
-			ret = dmar_parse_one_dev_scope(scope,
-				&(*devices)[index], segment);
-			if (ret) {
-				kfree(*devices);
-				return ret;
-			}
-			index ++;
-		}
+		    scope->entry_type == ACPI_DMAR_SCOPE_TYPE_BRIDGE)
+			(*scopes)[index++] = scope;
+
 		start += scope->length;
 	}
 
@@ -204,9 +225,8 @@ static int __init dmar_parse_dev(struct dmar_drhd_unit *dmaru)
 		return 0;
 
 	ret = dmar_parse_dev_scope((void *)(drhd + 1),
-				((void *)drhd) + drhd->header.length,
-				&dmaru->devices_cnt, &dmaru->devices,
-				drhd->segment);
+				   ((void *)drhd) + drhd->header.length,
+				   &dmaru->scopes_cnt, &dmaru->scopes);
 	if (ret) {
 		list_del(&dmaru->list);
 		kfree(dmaru);
@@ -250,10 +270,10 @@ rmrr_parse_dev(struct dmar_rmrr_unit *rmrru)
 
 	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);
+				   ((void *)rmrr) + rmrr->header.length,
+				   &rmrru->scopes_cnt, &rmrru->scopes);
 
-	if (ret || (rmrru->devices_cnt == 0)) {
+	if (ret || (rmrru->scopes_cnt == 0)) {
 		list_del(&rmrru->list);
 		kfree(rmrru);
 	}
@@ -290,10 +310,9 @@ static int __init atsr_parse_dev(struct dmar_atsr_unit *atsru)
 
 	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) {
+				  (void *)atsr + atsr->header.length,
+				  &atsru->scopes_cnt, &atsru->scopes);
+	if (rc || !atsru->scopes_cnt) {
 		list_del(&atsru->list);
 		kfree(atsru);
 	}
@@ -307,6 +326,7 @@ int dmar_find_matched_atsr_unit(struct pci_dev *dev)
 	struct pci_bus *bus;
 	struct acpi_dmar_atsr *atsr;
 	struct dmar_atsr_unit *atsru;
+	struct pci_dev *pdev;
 
 	list_for_each_entry(atsru, &dmar_atsr_units, list) {
 		atsr = container_of(atsru->hdr, struct acpi_dmar_atsr, header);
@@ -325,10 +345,18 @@ found:
 			return 0;
 
 		if (bridge->pcie_type == PCI_EXP_TYPE_ROOT_PORT) {
-			for (i = 0; i < atsru->devices_cnt; i++)
-				if (atsru->devices[i] == bridge)
+			for (i = 0; i < atsru->scopes_cnt; i++) {
+				pdev = dmar_get_scope_dev(atsru->scopes[i],
+							  atsr->segment);
+				if (!pdev)
+					continue;
+
+				if (pdev == bridge) {
+					pci_dev_put(pdev);
 					return 1;
-			break;
+				}
+				pci_dev_put(pdev);
+			}
 		}
 	}
 
@@ -475,23 +503,6 @@ parse_dmar_table(void)
 	return ret;
 }
 
-int dmar_pci_device_match(struct pci_dev *devices[], int cnt,
-			  struct pci_dev *dev)
-{
-	int index;
-
-	while (dev) {
-		for (index = 0; index < cnt; index++)
-			if (dev == devices[index])
-				return 1;
-
-		/* Check our parent */
-		dev = dev->bus->self;
-	}
-
-	return 0;
-}
-
 struct dmar_drhd_unit *
 dmar_find_matched_drhd_unit(struct pci_dev *dev)
 {
@@ -504,11 +515,11 @@ dmar_find_matched_drhd_unit(struct pci_dev *dev)
 				    header);
 
 		if (dmaru->include_all &&
-		    drhd->segment == pci_domain_nr(dev->bus))
+		    dmaru->segment == pci_domain_nr(dev->bus))
 			return dmaru;
 
-		if (dmar_pci_device_match(dmaru->devices,
-					  dmaru->devices_cnt, dev))
+		if (dmar_match_scope(dmaru->scopes, dmaru->scopes_cnt,
+				     dev, dmaru->segment))
 			return dmaru;
 	}
 
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index e7ff5b8..7d0b095 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -541,32 +541,34 @@ static void domain_update_iommu_cap(struct dmar_domain *domain)
 
 static struct intel_iommu *device_to_iommu(int segment, u8 bus, u8 devfn)
 {
-	struct dmar_drhd_unit *drhd = NULL;
-	int i;
+	struct dmar_drhd_unit *dmaru = NULL;
+	struct pci_dev *pdev;
+	struct intel_iommu *found = NULL;
 
-	for_each_drhd_unit(drhd) {
-		if (drhd->ignored)
+	pdev = pci_get_domain_bus_and_slot(segment, bus, devfn);
+
+	for_each_drhd_unit(dmaru) {
+		if (dmaru->ignored)
 			continue;
-		if (segment != drhd->segment)
+		if (segment != dmaru->segment)
 			continue;
 
-		for (i = 0; i < drhd->devices_cnt; i++) {
-			if (drhd->devices[i] &&
-			    drhd->devices[i]->bus->number == bus &&
-			    drhd->devices[i]->devfn == devfn)
-				return drhd->iommu;
-			if (drhd->devices[i] &&
-			    drhd->devices[i]->subordinate &&
-			    drhd->devices[i]->subordinate->number <= bus &&
-			    drhd->devices[i]->subordinate->subordinate >= bus)
-				return drhd->iommu;
+		if (dmaru->include_all) {
+			found = dmaru->iommu;
+			break;
+		}
+
+		if (dmar_match_scope(dmaru->scopes, dmaru->scopes_cnt,
+				     pdev, dmaru->segment)) {
+			found = dmaru->iommu;
+			break;
 		}
 
-		if (drhd->include_all)
-			return drhd->iommu;
 	}
 
-	return NULL;
+	pci_dev_put(pdev);
+
+	return found;
 }
 
 static void domain_flush_cache(struct dmar_domain *domain,
@@ -2250,7 +2252,7 @@ int __init init_dmars(void)
 	struct dmar_rmrr_unit *rmrr;
 	struct pci_dev *pdev;
 	struct intel_iommu *iommu;
-	int i, ret;
+	int ret;
 
 	/*
 	 * for each drhd
@@ -2397,18 +2399,22 @@ int __init init_dmars(void)
 	 */
 	printk(KERN_INFO "IOMMU: Setting RMRR:\n");
 	for_each_rmrr_units(rmrr) {
-		for (i = 0; i < rmrr->devices_cnt; i++) {
-			pdev = rmrr->devices[i];
-			/*
-			 * some BIOS lists non-exist devices in DMAR
-			 * table.
-			 */
+		struct acpi_dmar_reserved_memory *rmrrh;
+		int i;
+
+		rmrrh = container_of(rmrr->hdr,
+				     struct acpi_dmar_reserved_memory, header);
+
+		for (i = 0; i < rmrr->scopes_cnt; i++) {
+			pdev = dmar_get_scope_dev(rmrr->scopes[i],
+						  rmrrh->segment);
 			if (!pdev)
 				continue;
-			ret = iommu_prepare_rmrr_dev(rmrr, pdev);
-			if (ret)
+
+			if (iommu_prepare_rmrr_dev(rmrr, pdev))
 				printk(KERN_ERR
 				       "IOMMU: mapping reserved region failed\n");
+			pci_dev_put(pdev);
 		}
 	}
 
@@ -3078,15 +3084,21 @@ DECLARE_PCI_FIXUP_ENABLE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IOAT_SNB, quir
 static void __init init_no_remapping_devices(void)
 {
 	struct dmar_drhd_unit *drhd;
+	struct pci_dev *pdev;
 
 	for_each_drhd_unit(drhd) {
 		if (!drhd->include_all) {
 			int i;
-			for (i = 0; i < drhd->devices_cnt; i++)
-				if (drhd->devices[i] != NULL)
+			for (i = 0; i < drhd->scopes_cnt; i++) {
+				pdev = dmar_get_scope_dev(drhd->scopes[i],
+							  drhd->segment);
+				if (pdev) {
+					pci_dev_put(pdev);
 					break;
+				}
+			}
 			/* ignore DMAR unit if no pci devices exist */
-			if (i == drhd->devices_cnt)
+			if (i == drhd->scopes_cnt)
 				drhd->ignored = 1;
 		}
 	}
@@ -3099,20 +3111,28 @@ static void __init init_no_remapping_devices(void)
 		if (drhd->ignored || drhd->include_all)
 			continue;
 
-		for (i = 0; i < drhd->devices_cnt; i++)
-			if (drhd->devices[i] &&
-				!IS_GFX_DEVICE(drhd->devices[i]))
+		for (i = 0; i < drhd->scopes_cnt; i++) {
+			pdev = dmar_get_scope_dev(drhd->scopes[i],
+						  drhd->segment);
+			if (pdev && !IS_GFX_DEVICE(pdev)) {
+				pci_dev_put(pdev);
 				break;
+			}
+			pci_dev_put(pdev);
+		}
 
-		if (i < drhd->devices_cnt)
+		if (i < drhd->scopes_cnt)
 			continue;
 
 		/* bypass IOMMU if it is just for gfx devices */
 		drhd->ignored = 1;
-		for (i = 0; i < drhd->devices_cnt; i++) {
-			if (!drhd->devices[i])
+		for (i = 0; i < drhd->scopes_cnt; i++) {
+			pdev = dmar_get_scope_dev(drhd->scopes[i],
+						  drhd->segment);
+			if (!pdev)
 				continue;
-			drhd->devices[i]->dev.archdata.iommu = DUMMY_DEVICE_DOMAIN_INFO;
+			pdev->dev.archdata.iommu = DUMMY_DEVICE_DOMAIN_INFO;
+			pci_dev_put(pdev);
 		}
 	}
 }
diff --git a/include/linux/dmar.h b/include/linux/dmar.h
index 69a6fba..70b7e3c 100644
--- a/include/linux/dmar.h
+++ b/include/linux/dmar.h
@@ -32,8 +32,8 @@ struct dmar_drhd_unit {
 	struct list_head list;		/* list of drhd units	*/
 	struct  acpi_dmar_header *hdr;	/* ACPI header		*/
 	u64	reg_base_addr;		/* register base address*/
-	struct	pci_dev **devices; 	/* target device array	*/
-	int	devices_cnt;		/* target device count	*/
+	struct	acpi_dmar_device_scope **scopes; /* target scope array	*/
+	int	scopes_cnt;		/* target scope count	*/
 	u16	segment;		/* PCI domain		*/
 	u8	ignored:1; 		/* ignore drhd		*/
 	u8	include_all:1;
@@ -55,6 +55,9 @@ extern struct list_head dmar_drhd_units;
 
 extern int dmar_table_init(void);
 extern int dmar_dev_scope_init(void);
+extern int dmar_match_scope(struct acpi_dmar_device_scope **, int,
+			    struct pci_dev *, u16);
+extern struct pci_dev *dmar_get_scope_dev(struct acpi_dmar_device_scope *, u16);
 
 /* Intel IOMMU detection */
 extern void detect_intel_iommu(void);
@@ -72,6 +75,20 @@ static inline int dmar_table_init(void)
 {
 	return -ENODEV;
 }
+
+static inline int dmar_match_scope(struct acpi_dmar_device_scope **scopes,
+				   int cnt, struct pci_dev *dev, u16 segment)
+{
+	return 0;
+}
+
+static inline struct pci_dev *dmar_get_scope_dev(
+					struct acpi_dmar_device_scope *scope,
+					u16 segment)
+{
+	return NULL;
+}
+
 static inline int enable_drhd_fault_handling(void)
 {
 	return -1;
@@ -203,8 +220,8 @@ struct dmar_rmrr_unit {
 	struct acpi_dmar_header *hdr;	/* ACPI header		*/
 	u64	base_address;		/* reserved base address*/
 	u64	end_address;		/* reserved end address */
-	struct pci_dev **devices;	/* target devices */
-	int	devices_cnt;		/* target device count */
+	struct	acpi_dmar_device_scope **scopes; /* target scope array */
+	int	scopes_cnt;		/* target scope count */
 };
 
 #define for_each_rmrr_units(rmrr) \
@@ -213,8 +230,8 @@ struct dmar_rmrr_unit {
 struct dmar_atsr_unit {
 	struct list_head list;		/* list of ATSR units */
 	struct acpi_dmar_header *hdr;	/* ACPI header */
-	struct pci_dev **devices;	/* target devices */
-	int devices_cnt;		/* target device count */
+	struct acpi_dmar_device_scope **scopes;	/* target scope array */
+	int scopes_cnt;		/* target scope count */
 	u8 include_all:1;		/* include all ports */
 };
 


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ