Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1172027

[PATCH 3.16.y-ckt 50/71] iommu/vt-d: Allow RMRR on graphics devices too

From Luis Henriques <luis.henriques@canonical.com>
Newsgroups linux.kernel
Subject [PATCH 3.16.y-ckt 50/71] iommu/vt-d: Allow RMRR on graphics devices too
Date 2015-06-25 12:10 +0200
Message-ID <pFcgy-5Bx-43@gated-at.bofh.it> (permalink)
References <pFcgx-5Bx-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

3.16.7-ckt14 -stable review patch.  If anyone has any objections, please let me know.

------------------

From: David Woodhouse <David.Woodhouse@intel.com>

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.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Luis Henriques <luis.henriques@canonical.com>
---
 drivers/iommu/intel-iommu.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index b36bd2886940..afbde3730b11 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -51,6 +51,7 @@
 #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)
 
@@ -2538,6 +2539,10 @@ static bool device_has_rmrr(struct device *dev)
  * 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)
 {
@@ -2547,7 +2552,7 @@ static bool device_is_rmrr_locked(struct device *dev)
 	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;
 	}
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[3.16.y-ckt stable] Linux 3.16.7-ckt14 stable review Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 47/71] btrfs: cleanup orphans while looking up default subvolume Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 62/71] iser-target: Fix variable-length response error completion Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 13/71] ozwpan: Use unsigned ints to prevent heap overflow Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 01/71] iio: adis16400: Report pressure channel scale Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 15/71] ozwpan: unchecked signed subtraction leads to DoS Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 70/71] drm/radeon: fix freeze for laptop with Turks/Thames GPU. Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 61/71] block: fix ext_dev_lock lockdep report Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 14/71] ozwpan: divide-by-zero leading to panic Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 50/71] iommu/vt-d: Allow RMRR on graphics devices too Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 56/71] Input: synaptics - add min/max quirk for Lenovo S540 Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 10/71] usb: dwc3: gadget: Fix incorrect DEPCMD and DGCMD status macros Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 46/71] btrfs: incorrect handling for fiemap_fill_next_extent return Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 55/71] pata_octeon_cf: fix broken build Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 65/71] drm/mgag200: Reject non-character-cell-aligned mode widths Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 67/71] crypto: caam - improve initalization for context state saves Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 57/71] drm/i915: Fix DDC probe for passive adapters Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 51/71] iommu/vt-d: Fix passthrough mode with translation-disabled devices Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 68/71] crypto: caam - fix RNG buffer cache alignment Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 12/71] ozwpan: Use proper check to prevent heap overflow Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 63/71] iser-target: release stale iser connections Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 54/71] irqchip: sunxi-nmi: Fix off-by-one error in irq iterator Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 71/71] Revert "tools/vm: fix page-flags build" Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 11/71] ALSA: usb-audio: Add mic volume fix quirk for Logitech Quickcam Fusion Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 69/71] tracing: Have filter check for balanced ops Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 59/71] mm/memory_hotplug.c: set zone->wait_table to null after freeing it Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 66/71] crypto: caam - fix uninitialized state->buf_dma field Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 52/71] ata: ahci_mvebu: Fix wrongly set base address for the MBus window setting Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 60/71] ring-buffer-benchmark: Fix the wrong sched_priority of producer Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 64/71] ALSA: hda - adding a DAC/pin preference map for a HP Envy TS machine Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:10 +0200
  [PATCH 3.16.y-ckt 35/71] net: dp83640: fix broken calibration routine. Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 36/71] net: dp83640: reinforce locking rules. Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 29/71] scripts/sortextable: suppress warning: `relocs_size' may be used uninitialized Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 30/71] thermal: step_wise: Revert optimization Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 20/71] Input: elantech - fix detection of touchpads where the revision matches a known rate Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 37/71] unix/caif: sk_socket can disappear when state is unlocked Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 24/71] USB: cp210x: add ID for HubZ dual ZigBee and Z-Wave dongle Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 21/71] ALSA: hda/realtek - Add a fixup for another Acer Aspire 9420 Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 28/71] kconfig: Fix warning "‘jump’ may be used uninitialized" Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 23/71] ALSA: usb-audio: fix missing input volume controls in MAYA44 USB(+) Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 31/71] MIPS: KVM: Do not sign extend on unsigned MMIO load Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 38/71] xen/netback: Properly initialize credit_bytes Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 27/71] xfrm: fix a race in xfrm_state_lookup_byspi Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 42/71] bridge: disable softirqs around br_fdb_update to avoid lockup Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 39/71] udp: fix behavior of wrong checksums Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 22/71] ALSA: usb-audio: add MAYA44 USB+ mixer control names Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 41/71] ipv4/udp: Verify multicast group is ours in upd_v4_early_demux() Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 40/71] xen: netback: read hotplug script once at start of day. Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 43/71] drm/i915: Assume dual channel LVDS if pixel clock necessitates it Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 44/71] Btrfs: send, add missing check for dead clone root Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 33/71] net: core: Correct an over-stringent device loop detection. Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 34/71] net: phy: Allow EEE for all RGMII variants Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 32/71] arch/x86/kvm/mmu.c: work around gcc-4.4.4 bug Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 18/71] x86/asm/irq: Stop relying on magic JMP behavior for early_idt_handlers Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 48/71] Drivers: hv: vmbus: Add support for VMBus panic notifier handler Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 25/71] Input: elantech - add new icbody type Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 26/71] MIPS: Fix enabling of DEBUG_STACKOVERFLOW Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200
  [PATCH 3.16.y-ckt 45/71] Btrfs: send, don't leave without decrementing clone root's send_progress Luis Henriques <luis.henriques@canonical.com> - 2015-06-25 12:20 +0200

csiph-web