You are not logged in.
Pages: 1
Since I successfully turned on IOMMU with intel_iommu=on, I've noticed a lot of the following entry in the kernel log.
[ 194.809429] DMAR: DRHD: handling fault status reg 3
[ 194.809432] DMAR: [DMA Read] Request device [03:00.0] fault addr fffe0000 [fault reason 06] PTE Read access is not set
[ 195.150607] DMAR: DRHD: handling fault status reg 3
[ 195.150613] DMAR: [DMA Read] Request device [03:00.0] fault addr fffe0000 [fault reason 06] PTE Read access is not set
[ 195.150770] DMAR: DRHD: handling fault status reg 3
[ 195.150774] DMAR: [DMA Read] Request device [03:00.0] fault addr fffe0000 [fault reason 06] PTE Read access is not set
[ 195.491960] DMAR: DRHD: handling fault status reg 3
[ 195.491967] DMAR: [DMA Read] Request device [03:00.0] fault addr fffe0000 [fault reason 06] PTE Read access is not set
[ 195.492115] DMAR: DRHD: handling fault status reg 3
[ 195.492119] DMAR: [DMA Read] Request device [03:00.0] fault addr fffe0000 [fault reason 06] PTE Read access is not set
A quick check with lspci yields:
03:00.0 Multimedia audio controller: Creative Labs EMU10k2/CA0100/CA0102/CA10200 [Sound Blaster Audigy Series] (rev 04)
It's annoying that this happens whenever there is sound played. And the amount of the entries is huge.
So the question is how can I fix this?
Offline
Successfully turn iommu off again, this btw. has lately been all over the board due to an overly optimisitic move by kernel devs.
https://bbs.archlinux.org/viewtopic.php?id=230362
Offline
Iommu creates issues for intel graphics. https://bugs.freedesktop.org/show_bug.cgi?id=89360
Best solution is to enable iommu but disable it for graphic device with:
intel_iommu=on,igfx_off
Offline
@Uriel_Bernhard48 but that will leave IOMMU enabled for the audio controller which is causing the DMAR faults you believe the iGPU if there is one is
somehow having an impact on the audio controller causing those errors that will be prevented by blocking the iGPU from IOMMU?
edit:
spelling causing not cuasing.
Last edited by loqs (2018-01-01 19:14:24)
Offline
Successfully turn iommu off again, this btw. has lately been all over the board due to an overly optimisitic move by kernel devs.
https://bbs.archlinux.org/viewtopic.php?id=230362
I turned it on because I need to passthrough the GPU. Is there any option like the one suggested above to only turn on IOMMU for the GPU?
Offline
intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option
on
Enable intel iommu driver.
off
Disable intel iommu driver.
igfx_off [Default Off]
By default, gfx is mapped as normal device. If a gfx
device has a dedicated DMAR unit, the DMAR unit is
bypassed by not enabling DMAR with this option. In
this case, gfx device will use physical address for
DMA.
forcedac [x86_64]
With this option iommu will not optimize to look
for io virtual address below 32-bit forcing dual
address cycle on pci bus for cards supporting greater
than 32-bit addressing. The default is to look
for translation below 32-bit and if not available
then look in the higher range.
strict [Default Off]
With this option on every unmap_single operation will
result in a hardware IOTLB flush operation as opposed
to batching them for performance.
sp_off [Default Off]
By default, super page will be supported if Intel IOMMU
has the capability. With this option, super page will
not be supported.
ecs_off [Default Off]
By default, extended context tables will be supported if
the hardware advertises that it has support both for the
extended tables themselves, and also PASID support. With
this option set, extended tables will not be used even
on hardware which claims to support them.
tboot_noforce [Default Off]
Do not force the Intel IOMMU enabled under tboot.
By default, tboot will force Intel IOMMU on, which
could harm performance of some high-throughput
devices like 40GBit network cards, even if identity
mapping is enabled.
Note that using this option lowers the security
provided by tboot because it makes the system
vulnerable to DMA attacks.
forcedac might work.
Did you try "iommu=soft" (NOT "intel_iommu")
Offline
intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option on Enable intel iommu driver. off Disable intel iommu driver. igfx_off [Default Off] By default, gfx is mapped as normal device. If a gfx device has a dedicated DMAR unit, the DMAR unit is bypassed by not enabling DMAR with this option. In this case, gfx device will use physical address for DMA. forcedac [x86_64] With this option iommu will not optimize to look for io virtual address below 32-bit forcing dual address cycle on pci bus for cards supporting greater than 32-bit addressing. The default is to look for translation below 32-bit and if not available then look in the higher range. strict [Default Off] With this option on every unmap_single operation will result in a hardware IOTLB flush operation as opposed to batching them for performance. sp_off [Default Off] By default, super page will be supported if Intel IOMMU has the capability. With this option, super page will not be supported. ecs_off [Default Off] By default, extended context tables will be supported if the hardware advertises that it has support both for the extended tables themselves, and also PASID support. With this option set, extended tables will not be used even on hardware which claims to support them. tboot_noforce [Default Off] Do not force the Intel IOMMU enabled under tboot. By default, tboot will force Intel IOMMU on, which could harm performance of some high-throughput devices like 40GBit network cards, even if identity mapping is enabled. Note that using this option lowers the security provided by tboot because it makes the system vulnerable to DMA attacks.
forcedac might work.
Did you try "iommu=soft" (NOT "intel_iommu")
Tried both, neither worked. Perhaps it's time to report a kernel bug or something? how do I do that?
Offline
https://www.kernel.org/doc/html/latest/ … -bugs.html please do not report it on the arch bug tracker as it is an upstream issue.
Offline
Pages: 1