Re: [PATCH] drm/i915: Use per device iommu check

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 11/11/21 11:06 PM, Tvrtko Ursulin wrote:

On 10/11/2021 12:35, Lu Baolu wrote:
On 2021/11/10 20:08, Tvrtko Ursulin wrote:

On 10/11/2021 12:04, Lu Baolu wrote:
On 2021/11/10 17:30, Tvrtko Ursulin wrote:

On 10/11/2021 07:12, Lu Baolu wrote:
Hi Tvrtko,

On 2021/11/9 20:17, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin<tvrtko.ursulin@xxxxxxxxx>

On igfx + dgfx setups, it appears that intel_iommu=igfx_off option only disables the igfx iommu. Stop relying on global intel_iommu_gfx_mapped and probe presence of iommu domain per device to accurately reflect its
status.

Signed-off-by: Tvrtko Ursulin<tvrtko.ursulin@xxxxxxxxx>
Cc: Lu Baolu<baolu.lu@xxxxxxxxxxxxxxx>
---
Baolu, is my understanding here correct? Maybe I am confused by both
intel_iommu_gfx_mapped and dmar_map_gfx being globals in the intel_iommu driver. But it certainly appears the setup can assign some iommu ops (and assign the discrete i915 to iommu group) when those two are set to off.

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e967cd08f23e..9fb38a54f1fe 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1763,26 +1763,27 @@ static inline bool run_as_guest(void)
  #define HAS_D12_PLANE_MINIMIZATION(dev_priv) (IS_ROCKETLAKE(dev_priv) || \
                            IS_ALDERLAKE_S(dev_priv))

-static inline bool intel_vtd_active(void)
+static inline bool intel_vtd_active(struct drm_i915_private *i915)
  {
-#ifdef CONFIG_INTEL_IOMMU
-    if (intel_iommu_gfx_mapped)
+    if (iommu_get_domain_for_dev(i915->drm.dev))
          return true;
-#endif

      /* Running as a guest, we assume the host is enforcing VT'd */
      return run_as_guest();
  }

Have you verified this change? I am afraid that
iommu_get_domain_for_dev() always gets a valid iommu domain even
intel_iommu_gfx_mapped == 0.

Yes it seems to work as is:

default:

# grep -i iommu /sys/kernel/debug/dri/*/i915_capabilities
/sys/kernel/debug/dri/0/i915_capabilities:iommu: enabled
/sys/kernel/debug/dri/1/i915_capabilities:iommu: enabled

intel_iommu=igfx_off:

# grep -i iommu /sys/kernel/debug/dri/*/i915_capabilities
/sys/kernel/debug/dri/0/i915_capabilities:iommu: disabled
/sys/kernel/debug/dri/1/i915_capabilities:iommu: enabled

On my system dri device 0 is integrated graphics and 1 is discrete.

The drm device 0 has a dedicated iommu. When the user request igfx not
mapped, the VT-d implementation will turn it off to save power. But for
shared iommu, you definitely will get it enabled.

Sorry I am not following, what exactly do you mean? Is there a platform with integrated graphics without a dedicated iommu, in which case intel_iommu=igfx_off results in intel_iommu_gfx_mapped == 0 and iommu_get_domain_for_dev returning non-NULL?

Your code always work for an igfx with a dedicated iommu. This might be
always true on today's platforms. But from driver's point of view, we
should not make such assumption.

For example, if the iommu implementation decides not to turn off the
graphic iommu (perhaps due to some hw quirk or for graphic
virtualization), your code will be broken.

I tried your suggestion (checking for __IOMMU_DOMAIN_PAGING) and it works better, however I have observed one odd behaviour (for me at least).

In short - why does the DMAR mode for the discrete device change depending on igfx_off parameter?

Consider the laptop has these two graphics cards:

# cat /sys/kernel/debug/dri/0/name
i915 dev=0000:00:02.0 unique=0000:00:02.0 # integrated

# cat /sys/kernel/debug/dri/1/name
i915 dev=0000:03:00.0 unique=0000:03:00.0 # discrete

Booting with different options:
===============================

default / intel_iommu=on
------------------------

# cat /sys/class/iommu/dmar0/devices/0000:00:02.0/iommu_group/type
DMA-FQ
# cat /sys/class/iommu/dmar2/devices/0000:03:00.0/iommu_group/type
DMA-FQ

# grep -i iommu /sys/kernel/debug/dri/*/i915_capabilities
/sys/kernel/debug/dri/0/i915_capabilities:iommu: enabled
/sys/kernel/debug/dri/1/i915_capabilities:iommu: enabled

All good.

intel_iommu=igfx_off
--------------------

## no dmar0 in sysfs
# cat /sys/class/iommu/dmar2/devices/0000:03:00.0/iommu_group/type
identity

Unexpected!?

# grep -i iommu /sys/kernel/debug/dri/*/i915_capabilities
/sys/kernel/debug/dri/0/i915_capabilities:iommu: disabled
/sys/kernel/debug/dri/1/i915_capabilities:iommu: disabled # At least the i915 patch detects it correctly.

intel_iommu=off
---------------

## no dmar0 in sysfs
## no dmar2 in sysfs

# grep -i iommu /sys/kernel/debug/dri/*/i915_capabilities
/sys/kernel/debug/dri/0/i915_capabilities:iommu: disabled
/sys/kernel/debug/dri/1/i915_capabilities:iommu: disabled

All good.

The fact discrete graphics changes from translated to pass-through when igfx_off is set is surprising to me. Is this a bug?

The existing VT-d implementation doesn't distinguish igfx from dgfx. It
only checks whether the device is of a display class:

#define IS_GFX_DEVICE(pdev) ((pdev->class >> 16) == PCI_BASE_CLASS_DISPLAY)

When igfx_off is specified, all graphic devices will put into pass-
through (the same meaning as identity mapping) mode. For igfx, since
the iommu is always dedicated, hence it further turn off the iommu
(hence there's no iommu domain) to save power.

Best regards,
baolu



[Index of Archives]     [AMD Graphics]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux