Skip to contentSkip to conversation

Patch "iommu/vt-d: Allow RMRR on graphics devices too" has been added to the 3.19-stable tree

PATCH [email protected] 2 participants 5 messages May 2, 2015 – Jun 2, 2015

Subsystems (2)
lore
Share & tools
lore

[email protected]

author maintainer Applied
#0 lore raw

This is a note to let you know that I've just added the patch titled

iommu/vt-d: Allow RMRR on graphics devices too

to the 3.19-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
iommu-vt-d-allow-rmrr-on-graphics-devices-too.patch
and it can be found in the queue-3.19 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.

Quoted 1 line

From 18436afdc11a00ac881990b454cfb2eae81d6003 Mon Sep 17 00:00:00 2001

From: David Woodhouse <[email protected]>
Date: Wed, 25 Mar 2015 15:05:47 +0000
Subject: iommu/vt-d: Allow RMRR on graphics devices too

From: David Woodhouse <[email protected]>

commit 18436afdc11a00ac881990b454cfb2eae81d6003 upstream.

Commit c875d2c1 ("iommu/vt-d: Exclude devices using RMRRs from IOMMU API domains") prevents certain options for devices with RMRRs. This even prevents those devices from getting a 1:1 mapping with 'iommu=pt', because we don't have the code to handle *preserving* the RMRR regions when moving the device between domains.

There's already an exclusion for USB devices, because we know the only reason for RMRRs there is a misguided desire to keep legacy keyboard/mouse emulation running in some theoretical OS which doesn't have support for USB in its own right... but which *does* enable the IOMMU.

Add an exclusion for graphics devices too, so that 'iommu=pt' works there. We should be able to successfully assign graphics devices to guests too, as long as the initial handling of stolen memory is reconfigured appropriately. This has certainly worked in the past.

---
 drivers/iommu/intel-iommu.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/iommu/intel-iommu.c +6 -1
@@ -50,6 +50,7 @@
 #define CONTEXT_SIZE		VTD_PAGE_SIZE
 
 #define IS_GFX_DEVICE(pdev) ((pdev->class >> 16) == PCI_BASE_CLASS_DISPLAY)
+#define IS_USB_DEVICE(pdev) ((pdev->class >> 8) == PCI_CLASS_SERIAL_USB)
 #define IS_ISA_DEVICE(pdev) ((pdev->class >> 8) == PCI_CLASS_BRIDGE_ISA)
 #define IS_AZALIA(pdev) ((pdev)->vendor == 0x8086 && (pdev)->device == 0x3a3e)
 
@@ -2561,6 +2562,10 @@ static bool device_has_rmrr(struct devic
  * In both cases we assume that PCI USB devices with RMRRs have them largely
  * for historical reasons and that the RMRR space is not actively used post
  * boot.  This exclusion may change if vendors begin to abuse it.
+ *
+ * The same exception is made for graphics devices, with the requirement that
+ * any use of the RMRR regions will be torn down before assigning the device
+ * to a guest.
  */
 static bool device_is_rmrr_locked(struct device *dev)
 {
@@ -2570,7 +2575,7 @@ static bool device_is_rmrr_locked(struct
 	if (dev_is_pci(dev)) {
 		struct pci_dev *pdev = to_pci_dev(dev);
 
-		if ((pdev->class >> 8) == PCI_CLASS_SERIAL_USB)
+		if (IS_USB_DEVICE(pdev) || IS_GFX_DEVICE(pdev))
 			return false;
 	}
 

Patches currently in stable-queue which might be from [email protected] are

queue-3.19/iommu-vt-d-allow-rmrr-on-graphics-devices-too.patch

#1 lore raw
Quoted 8 lines · On Sat, 2015-05-02 at 19:59 +0200, [email protected] wrote:

This is a note to let you know that I've just added the patch titled

iommu/vt-d: Allow RMRR on graphics devices too

to the 3.19-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable
-queue.git;a=summary

Quoted 7 lines

The filename of the patch is:
iommu-vt-d-allow-rmrr-on-graphics-devices-too.patch
and it can be found in the queue-3.19 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.

Actually, please wait. You're going to want this too...

Signature · 4 lines
                  Sent with Evolution's ActiveSync support.

David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation
Quoted 12 lines · On Mon, 2015-04-27 at 15:27 +0800, [email protected] wrote:

commit 18436afdc11a00ac881990b454cfb2eae81d6003 iommu/vt-d: Allow RMRR on graphics devices too

cause kernel panic with Z77+vt-d and kernel parameters:
intel_iommu=igfx_off iommu=pt
the panic path is:
dev_prepare_static_identity_mapping()
domain_add_dev_info()

    dmar_insert_dev_info()
      find_domain()
        info = dev->archdata.iommu; 
        if (info) 
Quoted 6 lines

-> return info->domain;

igfx_off cause <igfx@0:2.0>.archdata.iommu assigned to
DUMMY_DEVICE_DOMAIN_INFO.
Above patch cause iommu_should_identity_map() return 1, and attempt to setup identity mapping for igfx.

Can you try this please?

drivers/iommu/intel-iommu.c +8 -5
index 68d43be..2ffe589 100644
@@ -696,6 +696,11 @@ static inline struct context_entry *iommu_context_addr(struct intel_iommu *iommu
        return &context[devfn];
 }
 
+static int iommu_dummy(struct device *dev)
+{
+       return dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO;
+}
+
 static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devfn)
 {
        struct dmar_drhd_unit *drhd = NULL;
@@ -705,6 +710,9 @@ static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devf
        u16 segment = 0;
        int i;
 
+       if (iommu_dummy(dev))
+               return NULL;
+
        if (dev_is_pci(dev)) {
                pdev = to_pci_dev(dev);
                segment = pci_domain_nr(pdev->bus);
@@ -2969,11 +2977,6 @@ static inline struct dmar_domain *get_valid_domain_for_dev(struct device *dev)
        return __get_valid_domain_for_dev(dev);
 }
 
-static int iommu_dummy(struct device *dev)
-{
-       return dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO;
-}
-
 /* Check if the dev needs to go through non-identity map and unmap process.*/
 static int iommu_no_mapping(struct device *dev)
 {
Signature · 2 lines
David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation
author maintainer
#2 lore raw
Quoted 13 lines · David
Quoted 12 lines · On Sat, 2015-05-02 at 19:59 +0200, [email protected] wrote:

This is a note to let you know that I've just added the patch titled

iommu/vt-d: Allow RMRR on graphics devices too

to the 3.19-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable
-queue.git;a=summary

The filename of the patch is:
iommu-vt-d-allow-rmrr-on-graphics-devices-too.patch
and it can be found in the queue-3.19 subdirectory.

Quoted 5 lines
Quoted 3 lines

If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.

Actually, please wait. You're going to want this too...

Ok, I'll drop it for now until this gets into Linus's tree.

thanks,

greg k-h

author maintainer
#3 lore raw
In reply to Woodhouse, David
Quoted 84 lines · David
Quoted 8 lines · On Sat, 2015-05-02 at 19:59 +0200, [email protected] wrote:

This is a note to let you know that I've just added the patch titled

iommu/vt-d: Allow RMRR on graphics devices too

to the 3.19-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable
-queue.git;a=summary

Quoted 7 lines

The filename of the patch is:
iommu-vt-d-allow-rmrr-on-graphics-devices-too.patch
and it can be found in the queue-3.19 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.

Actually, please wait. You're going to want this too...

Signature · 11 lines
                  Sent with Evolution's ActiveSync support.

David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation


Date: Mon, 27 Apr 2015 22:18:10 +0100
From: David Woodhouse <[email protected]>
To: "[email protected]" <[email protected]>
Cc: [email protected]
Subject: Re: Intel-iommu panic on kernel v4.1-rc1
Quoted 12 lines · On Mon, 2015-04-27 at 15:27 +0800, [email protected] wrote:

commit 18436afdc11a00ac881990b454cfb2eae81d6003 iommu/vt-d: Allow RMRR on graphics devices too

cause kernel panic with Z77+vt-d and kernel parameters:
intel_iommu=igfx_off iommu=pt
the panic path is:
dev_prepare_static_identity_mapping()
domain_add_dev_info()

    dmar_insert_dev_info()
      find_domain()
        info = dev->archdata.iommu; 
        if (info) 
Quoted 6 lines

-> return info->domain;

igfx_off cause <igfx@0:2.0>.archdata.iommu assigned to
DUMMY_DEVICE_DOMAIN_INFO.
Above patch cause iommu_should_identity_map() return 1, and attempt to setup identity mapping for igfx.

Can you try this please?

drivers/iommu/intel-iommu.c +8 -0
index 68d43be..2ffe589 100644
@@ -696,6 +696,11 @@ static inline struct context_entry *iommu_context_addr(struct intel_iommu *iommu
        return &context[devfn];
 }
 
+static int iommu_dummy(struct device *dev)
+{
+       return dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO;
+}
+
 static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devfn)
 {
        struct dmar_drhd_unit *drhd = NULL;
@@ -705,6 +710,9 @@ static struct intel_iommu *device_to_iommu(struct device *dev, u8 *bus, u8 *devf
        u16 segment = 0;
        int i;
 
+       if (iommu_dummy(dev))
+               return NULL;
+
        if (dev_is_pci(dev)) {
                pdev = to_pci_dev(dev);
                segment = pci_domain_nr(pdev->bus);
@@ -2969,11 +2977,6 @@ static inline struct dmar_domain *get_valid_domain_for_dev(struct device *dev)
        return __get_valid_domain_for_dev(dev);

drivers/iommu/intel-iommu.c static inline struct dmar_domain *get_valid_domain_for_dev(struct device *dev) · line 2982

Quoted 10 lines
drivers/iommu/intel-iommu.c +0 -5
 }
 
-static int iommu_dummy(struct device *dev)
-{
-       return dev->archdata.iommu == DUMMY_DEVICE_DOMAIN_INFO;
-}
-
 /* Check if the dev needs to go through non-identity map and unmap process.*/
 static int iommu_no_mapping(struct device *dev)
 {

Did this patch ever end up in Linus's tree? I can't seem to find it...

#4 lore raw
Quoted 3 lines · On Tue, 2015-06-02 at 15:31 +0900, [email protected] wrote:

Did this patch ever end up in Linus's tree? I can't seem to find it...

Er, no. It's 4ed6a540fab in my linux-iommu.git tree; I need to send a pull request but have been somewhat distracted. Thanks for the reminder.

Signature · 4 lines
                  Sent with Evolution's ActiveSync support.

David Woodhouse                            Open Source Technology Centre
[email protected]                              Intel Corporation