Documentation: x86: rework IOMMU documentation
PATCH 1/1 Documentation: x86: rework IOMMU documentation
Add preliminary documentation for AMD IOMMU and combine with the existing Intel IOMMU documentation and clean up and modernize some of the existing documentation to align with the current state of the kernel.
---
V2: Incorporate feedback from Robin to clarify IOMMU vs DMA engine (e.g.,
a device) and document proper DMA API. Also correct the fact that the AMD IOMMU is not limited to managing PCI devices.
v3: Fix spelling and rework text as suggested by Vasant
v4: Combine Intel and AMD documents into a single document as suggested
by Dave Hansen
v5: Clarify that keywords are related to ACPI, grammatical fixes
v6: Make more stuff common based on feedback from Robin
Documentation/x86/index.rst | 2 +- Documentation/x86/intel-iommu.rst | 115 ------------------------ Documentation/x86/iommu.rst | 143 ++++++++++++++++++++++++++++++ 3 files changed, 144 insertions(+), 116 deletions(-) delete mode 100644 Documentation/x86/intel-iommu.rst create mode 100644 Documentation/x86/iommu.rst
Documentation/x86/index.rst +1 -1
@@ -21,7 +21,7 @@ x86-specific Documentation tlb mtrr pat - intel-iommu + iommu intel_txt amd-memory-encryption pti
Documentation/x86/intel-iommu.rst deleted +0 -115
115 lines not on this page · raw message
Documentation/x86/iommu.rst new file +143 -0
143 lines not on this page · raw message
Signature · 6 lines
2.35.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Quoted 4 lines · Alex Deucher
Add preliminary documentation for AMD IOMMU and combine with the existing Intel IOMMU documentation and clean up and modernize some of the existing documentation to align with the current state of the kernel.
FWIW,
Quoted 292 lines
Signed-off-by Alex Deucher <[email protected]>---
V2: Incorporate feedback from Robin to clarify IOMMU vs DMA engine (e.g.,
a device) and document proper DMA API. Also correct the fact that the AMD IOMMU is not limited to managing PCI devices.
v3: Fix spelling and rework text as suggested by Vasant
v4: Combine Intel and AMD documents into a single document as suggested
by Dave Hansen
v5: Clarify that keywords are related to ACPI, grammatical fixes
v6: Make more stuff common based on feedback from RobinDocumentation/x86/index.rst | 2 +- Documentation/x86/intel-iommu.rst | 115 ------------------------ Documentation/x86/iommu.rst | 143 ++++++++++++++++++++++++++++++ 3 files changed, 144 insertions(+), 116 deletions(-)delete mode 100644
Documentation/x86/intel-iommu.rstcreate mode 100644Documentation/x86/iommu.rstDocumentation/x86/index.rst +1
-1@@ -21,7 +21,7 @@ x86-specific Documentation tlb mtrr pat - intel-iommu + iommu intel_txt amd-memory-encryption ptiDocumentation/x86/intel-iommu.rst deleted +0
-115@@ -1,115 +0,0 @@ -=================== -Linux IOMMU Support -=================== - -The architecture spec can be obtained from the below location. - -http://www.intel.com/content/dam/www/public/us/en/documents/product-specifications/vt-directed-io-spec.pdf - -This guide gives a quick cheat sheet for some basic understanding. - -Some Keywords - -- DMAR - DMA remapping -- DRHD - DMA Remapping Hardware Unit Definition -- RMRR - Reserved memory Region Reporting Structure -- ZLR - Zero length reads from PCI devices -- IOVA - IO Virtual address. - -Basic stuff ------------ - -ACPI enumerates and lists the different DMA engines in the platform, and -device scope relationships between PCI devices and which DMA engine controls -them. - -What is RMRR? -------------- - -There are some devices the BIOS controls, for e.g USB devices to perform -PS2 emulation. The regions of memory used for these devices are marked -reserved in the e820 map. When we turn on DMA translation, DMA to those -regions will fail. Hence BIOS uses RMRR to specify these regions along with -devices that need to access these regions. OS is expected to setup -unity mappings for these regions for these devices to access these regions. - -How is IOVA generated? ----------------------- - -Well behaved drivers call pci_map_*() calls before sending command to device -that needs to perform DMA. Once DMA is completed and mapping is no longer -required, device performs a pci_unmap_*() calls to unmap the region. - -The Intel IOMMU driver allocates a virtual address per domain. Each PCIE -device has its own domain (hence protection). Devices under p2p bridges -share the virtual address with all devices under the p2p bridge due to -transaction id aliasing for p2p bridges. - -IOVA generation is pretty generic. We used the same technique as vmalloc() -but these are not global address spaces, but separate for each domain. -Different DMA engines may support different number of domains. - -We also allocate guard pages with each mapping, so we can attempt to catch -any overflow that might happen. - - -Graphics Problems? ------------------- -If you encounter issues with graphics devices, you can try adding -option intel_iommu=igfx_off to turn off the integrated graphics engine. -If this fixes anything, please ensure you file a bug reporting the problem. - -Some exceptions to IOVA ------------------------ -Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). -The same is true for peer to peer transactions. Hence we reserve the -address from PCI MMIO ranges so they are not allocated for IOVA addresses. - - -Fault reporting ---------------- -When errors are reported, the DMA engine signals via an interrupt. The fault -reason and device that caused it with fault reason is printed on console. - -See below for sample. - - -Boot Message Sample -------------------- - -Something like this gets printed indicating presence of DMAR tables -in ACPI. - -ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @ 0x000000007f5b5ef0 - -When DMAR is being processed and initialized by ACPI, prints DMAR locations -and any RMRR's processed:: - - ACPI DMAR:Host address width 36 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 - ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 - ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff - ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff - -When DMAR is enabled for use, you will notice.. - -PCI-DMA: Using DMAR IOMMU -------------------------- - -Fault reporting -^^^^^^^^^^^^^^^ - -:: - - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - -TBD ----- - -- For compatibility testing, could use unity map domain for all devices, just - provide a 1-1 for all useful memory under a single domain for all devices. -- API for paravirt ops for abstracting functionality for VMM folks.Documentation/x86/iommu.rst new file +133
-0@@ -0,0 +1,143 @@ +================= +x86 IOMMU Support +================= + +The architecture specs can be obtained from the below locations. + +- Intel: http://www.intel.com/content/dam/www/public/us/en/documents/product-specifications/vt-directed-io-spec.pdf +- AMD: https://www.amd.com/system/files/TechDocs/48882_IOMMU.pdf + +This guide gives a quick cheat sheet for some basic understanding. + +Basic stuff +----------- + +ACPI enumerates and lists the different IOMMUs on the platform, and +device scope relationships between devices and which IOMMU controls +them. + +Some ACPI Keywords: + +- DMAR - Intel DMA Remapping table +- DRHD - Intel DMA Remapping Hardware Unit Definition +- RMRR - Intel Reserved Memory Region Reporting Structure +- IVRS - AMD I/O Virtualization Reporting Structure +- IVDB - AMD I/O Virtualization Definition Block +- IVHD - AMD I/O Virtualization Hardware Definition + +What is Intel RMRR? +^^^^^^^^^^^^^^^^^^^ + +There are some devices the BIOS controls, for e.g USB devices to perform +PS2 emulation. The regions of memory used for these devices are marked +reserved in the e820 map. When we turn on DMA translation, DMA to those +regions will fail. Hence BIOS uses RMRR to specify these regions along with +devices that need to access these regions. OS is expected to setup +unity mappings for these regions for these devices to access these regions. + +What is AMD IVRS? +^^^^^^^^^^^^^^^^^ + +The architecture defines an ACPI-compatible data structure called an I/O +Virtualization Reporting Structure (IVRS) that is used to convey information +related to I/O virtualization to system software. The IVRS describes the +configuration and capabilities of the IOMMUs contained in the platform as +well as information about the devices that each IOMMU virtualizes. + +The IVRS provides information about the following: + +- IOMMUs present in the platform including their capabilities and proper configuration +- System I/O topology relevant to each IOMMU +- Peripheral devices that cannot be otherwise enumerated +- Memory regions used by SMI/SMM, platform firmware, and platform hardware. These are generally exclusion ranges to be configured by system software. + +How is an I/O Virtual Address (IOVA) generated? +----------------------------------------------- + +Well behaved drivers call dma_map_*() calls before sending command to device +that needs to perform DMA. Once DMA is completed and mapping is no longer +required, driver performs dma_unmap_*() calls to unmap the region. + +Some exceptions to IOVAs +------------------------ + +Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). +The same is true for peer to peer transactions. Hence we reserve the +address from PCI MMIO ranges so they are not allocated for IOVA addresses. + +Graphics Problems? +------------------ + +If you encounter issues with integrated graphics devices, you can try +adding the option "iommu.passthrough=1", or the equivalent "iommu=pt", +to the kernel command line to use a 1:1 mapping for the IOMMU in +general. On Intel you can also try "intel_iommu=igfx_off" to turn off +translation specifically for the integrated graphics engine only. If +this fixes anything, please ensure you file a bug reporting the problem. + +Fault reporting +--------------- +When errors are reported, the IOMMU signals via an interrupt. The fault +reason and device that caused it is printed on the console. + + +Kernel Log Samples +------------------ + +Intel Boot Messages +^^^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of DMAR tables +in ACPI: + +:: + + ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @ 0x000000007f5b5ef0 + +When DMAR is being processed and initialized by ACPI, prints DMAR locations +and any RMRR's processed: + +:: + + ACPI DMAR:Host address width 36 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 + ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 + ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff + ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff + +When DMAR is enabled for use, you will notice: + +:: + + PCI-DMA: Using DMAR IOMMU + +Intel Fault reporting +^^^^^^^^^^^^^^^^^^^^^ + +:: + + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + +AMD Boot Messages +^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of the IOMMU: + +:: + + AMD-Vi: Found IOMMU cap 0x40 + AMD-Vi: Extended features (0x4f77ef22294ada): PPR NX GT IA GA PC GA_vAPIC
Documentation/x86/iommu.rst line 143
Quoted 10 lines
Documentation/x86/iommu.rst new file +10
-0+ AMD-Vi: Interrupt remapping enabled + AMD-Vi: Virtual APIC enabled + +AMD Fault reporting +^^^^^^^^^^^^^^^^^^^ + +:: + + AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0007 address=0xffffc02000 flags=0x0000] + AMD-Vi: Event logged [IO_PAGE_FAULT device=07:00.0 domain=0x0007 address=0xffffc02000 flags=0x0000]
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Hi Alex,
Thanks for doing this, really helps to catch up the current state. Please see my comments inline.
Quoted 51 lines · Alex Deucher
Add preliminary documentation for AMD IOMMU and combine with the existing Intel IOMMU documentation and clean up and modernize some of the existing documentation to align with the current state of the kernel.
Signed-off-by Alex Deucher <[email protected]>---
V2: Incorporate feedback from Robin to clarify IOMMU vs DMA engine (e.g.,
a device) and document proper DMA API. Also correct the fact that the AMD IOMMU is not limited to managing PCI devices.
v3: Fix spelling and rework text as suggested by Vasant
v4: Combine Intel and AMD documents into a single document as suggested
by Dave Hansen
v5: Clarify that keywords are related to ACPI, grammatical fixes
v6: Make more stuff common based on feedback from RobinDocumentation/x86/index.rst | 2 +- Documentation/x86/intel-iommu.rst | 115 ------------------------ Documentation/x86/iommu.rst | 143 ++++++++++++++++++++++++++++++ 3 files changed, 144 insertions(+), 116 deletions(-) delete mode 100644 Documentation/x86/intel-iommu.rst create mode 100644 Documentation/x86/iommu.rstDocumentation/x86/index.rst +1
-1@@ -21,7 +21,7 @@ x86-specific Documentation tlb mtrr pat - intel-iommu + iommu intel_txt amd-memory-encryption pti(quoted diff) +0
-0
b/Documentation/x86/intel-iommu.rstdeleted file mode 100644 index 099f13d51d5f..000000000000Documentation/x86/intel-iommu.rst deleted +0
-8@@ -1,115 +0,0 @@ -=================== -Linux IOMMU Support -=================== - -The architecture spec can be obtained from the below location. - -http://www.intel.com/content/dam/www/public/us/en/documents/product-specifications/vt-directed-io-spec.pdf -
Documentation/x86/intel-iommu.rst line 18
Quoted 10 lines
Documentation/x86/intel-iommu.rst deleted +0
-10-This guide gives a quick cheat sheet for some basic understanding. - -Some Keywords - -- DMAR - DMA remapping -- DRHD - DMA Remapping Hardware Unit Definition -- RMRR - Reserved memory Region Reporting Structure -- ZLR - Zero length reads from PCI devices -- IOVA - IO Virtual address. -
I feel this combined document only focus on IOVA and DMA APIs, it is considered as legacy DMA after scalable mode is introduced by Intel to support DMA with PASID, shared virtual addressing (SVA). Perhaps, we can also combine ./Documentation/x86/sva.rst
With scalable mode, it affects boot messages, fault reporting, etc. I am not saying no to this document, just suggesting. I don't know where AMD is at in terms of PASID support but there are lots of things in common between VT-d and ARM's SMMU in terms of PASID/SVA. Should we broaden the purpose of this document even further?
Quoted 249 lines
Documentation/x86/intel-iommu.rst deleted +0
-5-Basic stuff ------------ - -ACPI enumerates and lists the different DMA engines in the platform, and -device scope relationships between PCI devices and which DMA enginecontrols -them.
Documentation/x86/intel-iommu.rst deleted +0
-8- -What is RMRR? -------------- - -There are some devices the BIOS controls, for e.g USB devices to perform -PS2 emulation. The regions of memory used for these devices are marked -reserved in the e820 map. When we turn on DMA translation, DMA to those -regions will fail. Hence BIOS uses RMRR to specify these regions alongwith -devices that need to access these regions. OS is expected to setup -unity mappings for these regions for these devices to access these regions. -
Documentation/x86/intel-iommu.rst deleted +0
-4-How is IOVA generated? ----------------------- - -Well behaved drivers call pci_map_*() calls before sending command todevice -that needs to perform DMA. Once DMA is completed and mapping is no longer -required, device performs a pci_unmap_*() calls to unmap the region. -
Documentation/x86/intel-iommu.rst deleted +0
-6-The Intel IOMMU driver allocates a virtual address per domain. Each PCIE -device has its own domain (hence protection). Devices under p2p bridges -share the virtual address with all devices under the p2p bridge due to -transaction id aliasing for p2p bridges. - -IOVA generation is pretty generic. We used the same technique as
vmalloc()-but these are not global address spaces, but separate for each domain. -Different DMA engines may support different number of domains.Documentation/x86/intel-iommu.rst deleted +0
-2- -We also allocate guard pages with each mapping, so we can attempt tocatch -any overflow that might happen.
Documentation/x86/intel-iommu.rst deleted +0
-7- - -Graphics Problems? ------------------- -If you encounter issues with graphics devices, you can try adding -option intel_iommu=igfx_off to turn off the integrated graphics engine. -If this fixes anything, please ensure you file a bug reporting theproblem. -
Documentation/x86/intel-iommu.rst deleted +0
-5-Some exceptions to IOVA ------------------------ -Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). -The same is true for peer to peer transactions. Hence we reserve the -address from PCI MMIO ranges so they are not allocated for IOVAaddresses. -
Documentation/x86/intel-iommu.rst deleted +0
-4- -Fault reporting ---------------- -When errors are reported, the DMA engine signals via an interrupt. Thefault -reason and device that caused it with fault reason is printed on console. -
Documentation/x86/intel-iommu.rst deleted +0
-10-See below for sample. - - -Boot Message Sample -------------------- - -Something like this gets printed indicating presence of DMAR tables -in ACPI. - -ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @
0x000000007f5b5ef0-
-When DMAR is being processed and initialized by ACPI, prints DMAR locations -and any RMRR's processed::Documentation/x86/intel-iommu.rst deleted +0
-27- - ACPI DMAR:Host address width 36 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 - ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 - ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff - ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff - -When DMAR is enabled for use, you will notice.. - -PCI-DMA: Using DMAR IOMMU -------------------------- - -Fault reporting -^^^^^^^^^^^^^^^ - -:: - - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - -TBD ----- - -- For compatibility testing, could use unity map domain for all devices,just
- provide a 1-1 for all useful memory under a single domain for all devices. -- API for paravirt ops for abstracting functionality for VMM folks. diff --gita/Documentation/x86/iommu.rstb/Documentation/x86/iommu.rstnew file mode 100644
index 000000000000..ed87d76a38d5Documentation/x86/iommu.rst new file +7
-0@@ -0,0 +1,143 @@ +================= +x86 IOMMU Support +================= + +The architecture specs can be obtained from the below locations. + +- Intel:http://www.intel.com/content/dam/www/public/us/en/documents/product-specifications/vt-directed-io-spec.pdf
+- AMD: https://www.amd.com/system/files/TechDocs/48882_IOMMU.pdf +Documentation/x86/iommu.rst new file +25
-0+This guide gives a quick cheat sheet for some basic understanding. + +Basic stuff +----------- + +ACPI enumerates and lists the different IOMMUs on the platform, and +device scope relationships between devices and which IOMMU controls +them. + +Some ACPI Keywords: + +- DMAR - Intel DMA Remapping table +- DRHD - Intel DMA Remapping Hardware Unit Definition +- RMRR - Intel Reserved Memory Region Reporting Structure +- IVRS - AMD I/O Virtualization Reporting Structure +- IVDB - AMD I/O Virtualization Definition Block +- IVHD - AMD I/O Virtualization Hardware Definition + +What is Intel RMRR? +^^^^^^^^^^^^^^^^^^^ + +There are some devices the BIOS controls, for e.g USB devices to perform +PS2 emulation. The regions of memory used for these devices are marked +reserved in the e820 map. When we turn on DMA translation, DMA to those +regions will fail. Hence BIOS uses RMRR to specify these regions alongwith +devices that need to access these regions. OS is expected to setup +unity mappings for these regions for these devices to access these regions. +
Documentation/x86/iommu.rst new file +5
-0+What is AMD IVRS? +^^^^^^^^^^^^^^^^^ + +The architecture defines an ACPI-compatible data structure called an I/O +Virtualization Reporting Structure (IVRS) that is used to conveyinformation +related to I/O virtualization to system software. The IVRS describes the +configuration and capabilities of the IOMMUs contained in the platform as +well as information about the devices that each IOMMU virtualizes. +
Documentation/x86/iommu.rst new file +3
-0+The IVRS provides information about the following: + +- IOMMUs present in the platform including their capabilities and properconfiguration +- System I/O topology relevant to each IOMMU
+- Peripheral devices that cannot be otherwise enumerated +- Memory regions used by SMI/SMM, platform firmware, and platformhardware. These are generally exclusion ranges to be configured by system software. + +How is an I/O Virtual Address (IOVA) generated?
+-----------------------------------------------Documentation/x86/iommu.rst new file +2
-0+ +Well behaved drivers call dma_map_*() calls before sending command todevice +that needs to perform DMA. Once DMA is completed and mapping is no longer +required, driver performs dma_unmap_*() calls to unmap the region. +
Documentation/x86/iommu.rst new file +6
-0+Some exceptions to IOVAs +------------------------ + +Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). +The same is true for peer to peer transactions. Hence we reserve the +address from PCI MMIO ranges so they are not allocated for IOVAaddresses. +
Documentation/x86/iommu.rst new file +28
-0+Graphics Problems? +------------------ + +If you encounter issues with integrated graphics devices, you can try +adding the option "iommu.passthrough=1", or the equivalent "iommu=pt", +to the kernel command line to use a 1:1 mapping for the IOMMU in +general. On Intel you can also try "intel_iommu=igfx_off" to turn off +translation specifically for the integrated graphics engine only. If +this fixes anything, please ensure you file a bug reporting the problem. + +Fault reporting +--------------- +When errors are reported, the IOMMU signals via an interrupt. The fault +reason and device that caused it is printed on the console. + + +Kernel Log Samples +------------------ + +Intel Boot Messages +^^^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of DMAR tables +in ACPI: + +:: + + ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @
0x000000007f5b5ef0+
+When DMAR is being processed and initialized by ACPI, prints DMAR locations +and any RMRR's processed:Documentation/x86/iommu.rst new file +35
-0+ +:: + + ACPI DMAR:Host address width 36 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 + ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 + ACPI DMAR:RMRR base: 0x00000000000ed000 end: 0x00000000000effff + ACPI DMAR:RMRR base: 0x000000007f600000 end: 0x000000007fffffff + +When DMAR is enabled for use, you will notice: + +:: + + PCI-DMA: Using DMAR IOMMU + +Intel Fault reporting +^^^^^^^^^^^^^^^^^^^^^ + +:: + + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + +AMD Boot Messages +^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of the IOMMU: + +:: + + AMD-Vi: Found IOMMU cap 0x40 + AMD-Vi: Extended features (0x4f77ef22294ada): PPR NX GT IA GA PC
GA_vAPICDocumentation/x86/iommu.rst new file +9
-0+ AMD-Vi: Interrupt remapping enabled + AMD-Vi: Virtual APIC enabled + +AMD Fault reporting +^^^^^^^^^^^^^^^^^^^ + +:: + + AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0007address=0xffffc02000 flags=0x0000]
Documentation/x86/iommu.rst new file +1
-0+ AMD-Vi: Event logged [IO_PAGE_FAULT device=07:00.0 domain=0x0007
Quoted 1 line
address=0xffffc02000 flags=0x0000]
Thanks,
Jacob
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
[Public]
Quoted 91 lines
-----Original Message-----
From: Jacob Pan <[email protected]>
Sent: Tuesday, April 26, 2022 12:45 PM
To: Deucher, Alexander <[email protected]>
Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Suthikulpanit, Suravee <[email protected]>; [email protected]; [email protected] foundation.org; [email protected]; Hegde, Vasant <[email protected]>; [email protected]; Lu, Baolu <[email protected]>
Subject: Re: [PATCH] Documentation: x86: rework IOMMU documentationHi Alex,
Thanks for doing this, really helps to catch up the current state. Please see my comments inline.
Quoted 17 lines · Alex Deucher
Add preliminary documentation for AMD IOMMU and combine with the existing Intel IOMMU documentation and clean up and modernize some of the existing documentation to align with the current state of the kernel.
Signed-off-by Alex Deucher <[email protected]>---
V2: Incorporate feedback from Robin to clarify IOMMU vs DMA engine (e.g.,
a device) and document proper DMA API. Also correct the fact that the AMD IOMMU is not limited to managing PCI devices.
v3: Fix spelling and rework text as suggested by Vasant
v4: Combine Intel and AMD documents into a single document as suggested
by Dave Hansen
v5: Clarify that keywords are related to ACPI, grammatical fixes
v6: Make more stuff common based on feedback from RobinQuoted 3 lines
Documentation/x86/index.rst | 2 +- Documentation/x86/intel-iommu.rst | 115 ------------------------ Documentation/x86/iommu.rst | 143++++++++++++++++++++++++++++++
Quoted 23 lines
3 files changed, 144 insertions(+), 116 deletions(-) delete mode100644
Documentation/x86/intel-iommu.rstcreate mode 100644Documentation/x86/iommu.rstDocumentation/x86/index.rst +1
-1@@ -21,7 +21,7 @@ x86-specific Documentation tlb mtrr pat - intel-iommu + iommu intel_txt amd-memory-encryption pti(quoted diff) +0
-0
b/Documentation/x86/intel-iommu.rstdeleted file mode 100644 index 099f13d51d5f..000000000000Documentation/x86/intel-iommu.rst line 7
Quoted 7 lines
Documentation/x86/intel-iommu.rst deleted +0
-7@@ -1,115 +0,0 @@ -=================== -Linux IOMMU Support -=================== - -The architecture spec can be obtained from the below location. - -https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.
Quoted 1 line
intel.com%2Fcontent%2Fdam%2Fwww%2Fpublic%2Fus%2Fen%2Fdocuments %2Fprodu
Quoted 1 line
ct-specifications%2Fvt-directed-io-
spec.pdf&data=05%7C01%7Calexand
Quoted 1 line
er.deucher%40amd.com%7C929847a4b2524432d1a608da27a3c9b0%7C3dd
8961fe488Quoted 1 line
4e608e11a82d994e183d%7C0%7C0%7C637865881851295857%7CUnknow n%7CTWFpbGZs
Quoted 1 line
b3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0 %3D
Quoted 1 line
%7C3000%7C%7C%7C&sdata=KaPkvBSHWbn1cKBRzyk9H%2BuNDll%2Bq 3wPfR3SFVA
Quoted 16 lines
Documentation/x86/intel-iommu.rst line 18
Quoted 12 lines
LwkU%3D&reserved=0
Documentation/x86/intel-iommu.rst deleted +0
-11- -This guide gives a quick cheat sheet for some basic understanding. - -Some Keywords - -- DMAR - DMA remapping -- DRHD - DMA Remapping Hardware Unit Definition -- RMRR - Reserved memory Region Reporting Structure -- ZLR - Zero length reads from PCI devices -- IOVA - IO Virtual address. -I feel this combined document only focus on IOVA and DMA APIs, it is considered as legacy DMA after scalable mode is introduced by Intel to support DMA with PASID, shared virtual addressing (SVA). Perhaps, we can also combine
./Documentation/x86/sva.rst
I think it would make sense to take that up in a separate patch set.
Quoted 6 lines
With scalable mode, it affects boot messages, fault reporting, etc. I am not saying no to this document, just suggesting. I don't know where AMD is at in terms of PASID support but there are lots of things in common between VT-d and ARM's SMMU in terms of PASID/SVA. Should we broaden the purpose of this document even further?
I think that would make sense for a future clean up. I'd like to land the current clean up first.
AMD's IOMMU driver has supported PASID for probably 8-10 years. When we originally added it no other vendors were interested in supporting it so we made it a private API which was used by other AMD drivers that needed it at the time. Suravee can probably comment on the status of our support for the cross vendor API.
Alex
Quoted 289 lines
Quoted 12 lines
Documentation/x86/intel-iommu.rst deleted +0
-4-Basic stuff ------------ - -ACPI enumerates and lists the different DMA engines in the platform,and -device scope relationships between PCI devices and which DMA engine controls -them.
Documentation/x86/intel-iommu.rst deleted +0
-5- -What is RMRR? -------------- - -There are some devices the BIOS controls, for e.g USB devices toQuoted 3 lines
perform
-PS2 emulation. The regions of memory used for these devices are marked -reserved in the e820 map. When we turn on DMA translation,DMA
Quoted 15 lines
to those -regions will fail. Hence BIOS uses RMRR to specify these regions along with -devices that need to access these regions. OS is expected to setup -unity mappings for these regions for these devices to access these regions. - -How is IOVA generated?
-----------------------Documentation/x86/intel-iommu.rst deleted +0
-2- -Well behaved drivers call pci_map_*() calls before sending command todevice -that needs to perform DMA. Once DMA is completed and mapping is no longer -required, device performs a pci_unmap_*() calls to unmap the region. - -The Intel IOMMU driver allocates a virtual address per domain. Each PCIE -device has its own domain (hence protection). Devices under p2p bridges -share the virtual address with all devices under the p2p bridge due to -transaction id aliasing for p2p bridges.
Documentation/x86/intel-iommu.rst deleted +0
-2- -IOVA generation is pretty generic. We used the same technique asQuoted 2 lines
vmalloc()-but these are not global address spaces, but separate for each domain. -Different DMA engines may support different number ofdomains.
Quoted 34 lines
Documentation/x86/intel-iommu.rst deleted +0
-2- -We also allocate guard pages with each mapping, so we can attempt tocatch -any overflow that might happen.
Documentation/x86/intel-iommu.rst deleted +0
-7- - -Graphics Problems? ------------------- -If you encounter issues with graphics devices, you can try adding -option intel_iommu=igfx_off to turn off the integrated graphics engine. -If this fixes anything, please ensure you file a bug reporting theproblem. - -Some exceptions to IOVA
------------------------Documentation/x86/intel-iommu.rst deleted +0
-3-Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). -The same is true for peer to peer transactions. Hence we reserve the -address from PCI MMIO ranges so they are not allocated for IOVAaddresses. -
Documentation/x86/intel-iommu.rst deleted +0
-4- -Fault reporting ---------------- -When errors are reported, the DMA engine signals via an interrupt.The fault -reason and device that caused it with fault reason is printed on console. - -See below for sample.
Documentation/x86/intel-iommu.rst deleted +0
-9- - -Boot Message Sample -------------------- - -Something like this gets printed indicating presence of DMAR tables -in ACPI. - -ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT 0x00000097) @
0x000000007f5b5ef0-
-When DMAR is being processed and initialized by ACPI, prints DMAR locations -and any RMRR's processed::Documentation/x86/intel-iommu.rst
Quoted 6 lines
Documentation/x86/intel-iommu.rst deleted +0
-6- - ACPI DMAR:Host address width 36 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 - ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 - ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 - ACPI DMAR:RMRR base: 0x00000000000ed000 end:
0x00000000000effffDocumentation/x86/intel-iommu.rst
Quoted 1 line
Documentation/x86/intel-iommu.rst deleted +0
-1- ACPI DMAR:RMRR base: 0x000000007f600000 end:
0x000000007fffffffQuoted 30 lines
Documentation/x86/intel-iommu.rst deleted +0
-20- -When DMAR is enabled for use, you will notice.. - -PCI-DMA: Using DMAR IOMMU -------------------------- - -Fault reporting -^^^^^^^^^^^^^^^ - -:: - - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 - DMAR:[fault reason 05] PTE Write access is not set - -TBD ----- - -- For compatibility testing, could use unity map domain for alldevices, just
- provide a 1-1 for all useful memory under a single domain for all devices. -- API for paravirt ops for abstracting functionality for VMM folks. diff --gita/Documentation/x86/iommu.rstb/Documentation/x86/iommu.rstnew file mode 100644 index 000000000000..ed87d76a38d5Documentation/x86/iommu.rst line 7
Quoted 7 lines
Documentation/x86/iommu.rst new file +7
-0@@ -0,0 +1,143 @@ +================= +x86 IOMMU Support +================= + +The architecture specs can be obtained from the below locations. + +- Intel:https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.i
Quoted 1 line
ntel.com%2Fcontent%2Fdam%2Fwww%2Fpublic%2Fus%2Fen%2Fdocuments% 2Fproduc
Quoted 1 line
t-specifications%2Fvt-directed-io-
spec.pdf&data=05%7C01%7Calexande
Quoted 1 line
r.deucher%40amd.com%7C929847a4b2524432d1a608da27a3c9b0%7C3dd8
961fe4884Quoted 1 line
e608e11a82d994e183d%7C0%7C0%7C637865881851295857%7CUnknown %7CTWFpbGZsb
Quoted 1 line
3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0 %3D%
Quoted 1 line
7C3000%7C%7C%7C&sdata=KaPkvBSHWbn1cKBRzyk9H%2BuNDll%2Bq3 wPfR3SFVAL
Quoted 3 lines
wkU%3D&reserved=0
+- AMD:+https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww .amd.com%2Fsystem%2Ffiles%2FTechDocs%2F48882_IOMMU.pdf&data =05%7C01%7Calexander.deucher%40amd.com%7C929847a4b2524432d1a6 08da27a3c9b0%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C63 7865881851295857%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMD AiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C& amp;sdata=QFq8HBY6cm2XaZ50u5LQA3RX3NX%2BwVsMqfJFDH9bFDs%3D&a mp;reserved=0 + This guide gives a quick cheat sheet for some basic understanding.
Quoted 14 lines
Documentation/x86/iommu.rst new file +14
-0+ +Basic stuff +----------- + +ACPI enumerates and lists the different IOMMUs on the platform, and +device scope relationships between devices and which IOMMU controls +them. + +Some ACPI Keywords: + +- DMAR - Intel DMA Remapping table +- DRHD - Intel DMA Remapping Hardware Unit Definition +- RMRR - Intel Reserved Memory Region Reporting Structure +- IVRS - AMD I/O Virtualization Reporting StructureDocumentation/x86/iommu.rst
Quoted 10 lines
Documentation/x86/iommu.rst new file +10
-0+- IVDB - AMD I/O Virtualization Definition Block +- IVHD - AMD I/O Virtualization Hardware Definition + +What is Intel RMRR? +^^^^^^^^^^^^^^^^^^^ + +There are some devices the BIOS controls, for e.g USB devices to +perform +PS2 emulation. The regions of memory used for these devices are +marked reserved in the e820 map. When we turn on DMA translation,DMA
Quoted 37 lines
Documentation/x86/iommu.rst new file +2
-0+to those regions will fail. Hence BIOS uses RMRR to specify these +regions alongwith +devices that need to access these regions. OS is expected to setup
+unity mappings for these regions for these devices to access these regions. +Documentation/x86/iommu.rst new file +5
-0+What is AMD IVRS? +^^^^^^^^^^^^^^^^^ + +The architecture defines an ACPI-compatible data structure called an +I/O Virtualization Reporting Structure (IVRS) that is used to conveyinformation +related to I/O virtualization to system software. The IVRS describes the +configuration and capabilities of the IOMMUs contained in the platform as +well as information about the devices that each IOMMU virtualizes. +
Documentation/x86/iommu.rst new file +4
-0+The IVRS provides information about the following: + +- IOMMUs present in the platform including their capabilities and +properconfiguration +- System I/O topology relevant to each IOMMU
+- Peripheral devices that cannot be otherwise enumerated +- Memory regions used by SMI/SMM, platform firmware, and platformhardware. These are generally exclusion ranges to be configured by system software. + +How is an I/O Virtual Address (IOVA) generated?
+-----------------------------------------------Documentation/x86/iommu.rst new file +2
-0+ +Well behaved drivers call dma_map_*() calls before sending command todevice +that needs to perform DMA. Once DMA is completed and mapping is no longer +required, driver performs dma_unmap_*() calls to unmap the region. +
Documentation/x86/iommu.rst new file +6
-0+Some exceptions to IOVAs +------------------------ + +Interrupt ranges are not address translated, (0xfee00000 - 0xfeefffff). +The same is true for peer to peer transactions. Hence we reserve the +address from PCI MMIO ranges so they are not allocated for IOVAaddresses. +
Documentation/x86/iommu.rst
Quoted 9 lines
Documentation/x86/iommu.rst new file +9
-0+Graphics Problems? +------------------ + +If you encounter issues with integrated graphics devices, you can try +adding the option "iommu.passthrough=1", or the equivalent +"iommu=pt", to the kernel command line to use a 1:1 mapping for the +IOMMU in general. On Intel you can also try "intel_iommu=igfx_off" +to turn off translation specifically for the integrated graphics +engine only. If this fixes anything, please ensure you file a bug reportingthe problem.
Quoted 9 lines
Documentation/x86/iommu.rst new file +9
-0+ +Fault reporting +--------------- +When errors are reported, the IOMMU signals via an interrupt. The +fault reason and device that caused it is printed on the console. + + +Kernel Log Samples +------------------Documentation/x86/iommu.rst
Quoted 10 lines
Documentation/x86/iommu.rst new file +10
-0+ +Intel Boot Messages +^^^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of DMAR tables +in ACPI: + +:: + + ACPI: DMAR (v001 A M I OEMDMAR 0x00000001 MSFT
0x00000097) @Documentation/x86/iommu.rst
Quoted 11 lines
0x000000007f5b5ef0+
+When DMAR is being processed and initialized by ACPI, prints DMAR locations +and any RMRR's processed:Documentation/x86/iommu.rst new file +8
-0+ +:: + + ACPI DMAR:Host address width 36 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed90000 + ACPI DMAR:DRHD (flags: 0x00000000)base: 0x00000000fed91000 + ACPI DMAR:DRHD (flags: 0x00000001)base: 0x00000000fed93000 + ACPI DMAR:RMRR base: 0x00000000000ed000 end:
0x00000000000effffDocumentation/x86/iommu.rst
Quoted 1 line
Documentation/x86/iommu.rst new file +1
-0+ ACPI DMAR:RMRR base: 0x000000007f600000 end:
0x000000007fffffffQuoted 37 lines
Documentation/x86/iommu.rst new file +26
-0+ +When DMAR is enabled for use, you will notice: + +:: + + PCI-DMA: Using DMAR IOMMU + +Intel Fault reporting +^^^^^^^^^^^^^^^^^^^^^ + +:: + + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + DMAR:[DMA Write] Request device [00:02.0] fault addr 6df084000 + DMAR:[fault reason 05] PTE Write access is not set + +AMD Boot Messages +^^^^^^^^^^^^^^^^^ + +Something like this gets printed indicating presence of the IOMMU: + +:: + + AMD-Vi: Found IOMMU cap 0x40 + AMD-Vi: Extended features (0x4f77ef22294ada): PPR NX GT IA GA PC
GA_vAPICDocumentation/x86/iommu.rst new file +9
-0+ AMD-Vi: Interrupt remapping enabled + AMD-Vi: Virtual APIC enabled + +AMD Fault reporting +^^^^^^^^^^^^^^^^^^^ + +:: + + AMD-Vi: Event logged [IO_PAGE_FAULT domain=0x0007address=0xffffc02000 flags=0x0000]
Documentation/x86/iommu.rst
Quoted 1 line
Documentation/x86/iommu.rst new file +1
-0+ AMD-Vi: Event logged [IO_PAGE_FAULT device=07:00.0domain=0x0007
Quoted 6 lines
Quoted 1 line
address=0xffffc02000 flags=0x0000]
Thanks,
Jacob
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Quoted 17 lines · Alex Deucher
Add preliminary documentation for AMD IOMMU and combine with the existing Intel IOMMU documentation and clean up and modernize some of the existing documentation to align with the current state of the kernel.
Signed-off-by Alex Deucher <[email protected]>---
V2: Incorporate feedback from Robin to clarify IOMMU vs DMA engine (e.g.,
a device) and document proper DMA API. Also correct the fact that the AMD IOMMU is not limited to managing PCI devices.
v3: Fix spelling and rework text as suggested by Vasant
v4: Combine Intel and AMD documents into a single document as suggested
by Dave Hansen
v5: Clarify that keywords are related to ACPI, grammatical fixes
v6: Make more stuff common based on feedback from Robin
Quoted 6 lines
Documentation/x86/index.rst | 2 +- Documentation/x86/intel-iommu.rst | 115 ------------------------ Documentation/x86/iommu.rst | 143 ++++++++++++++++++++++++++++++ 3 files changed, 144 insertions(+), 116 deletions(-) delete mode 100644 Documentation/x86/intel-iommu.rst create mode 100644 Documentation/x86/iommu.rst
Jonathan, will you merge that through the documentation tree?
Jonathan Corbet
maintainerQuoted 3 lines · Joerg Roedel
Acked-by Joerg Roedel <[email protected]>Jonathan, will you merge that through the documentation tree?
Done.
Thanks,
jon