A while ago we discussed the concept of the crashdump kernel dealing with the
legacy DMA from the (old) panic'd kernel by allowing the (new) crashdump
kernel: to accept the iommu hardware in an active state, to leave the current
translations in-place so that legacy DMA will continue using its current
buffers until the device drivers in the crashdump kernel initialize, and to use
different portions of the iova address ranges for the device drivers in the
crashdump kernel.
As time permitted I have created prototype code to explore this concept.
The "crashdump accepting active iommu" prototype/proof-of-concept code has
successfully completed its first crashdump and rebooted the machine. The
latest version of the 'crash' utility reads the vmcore dump file that was
created.
At a high-level, here are the changes I made:
The code is entirely within intel-iommu.c
A key concept is that the active translation tables are mapped into kernel
virtual addresses in the new kernel. This allows most of the existing iommu
code to operate without change.
Hardware initialization:
------------------------
In intel_iommu_init(void)
* If (This is the crash kernel)
. Set flag: crashdump_accepting_active_iommu (all changes below check this)
. Skip disabling the iommu hardware translations
In init_dmars()
* Copy the intel iommu translation tables from the old kernel into the new
kernel
. The root-entry table, all context-entry tables, and all
page-translation-entry tables
. The copied tables contain updated physical addresses to link them together.
. The copied tables are also mapped into kernel virtual addresses in the new
kernel
which allows most of the existing iommu code to operate without change.
. Do some minimal sanity-checks during the copy
. Place the address of the new root-entry structure into "struct intel_iommu"
* Skip setting-up new domains for 'si', 'rmrr', 'isa'
. Translations for 'rmrr' and 'isa' ranges have been copied from the old
kernel
. This prototype does not yet handle pass-through
* Existing (unchanged) code near the end of dmar_init:
. Loads the address of the (now new) root-entry structure from "struct
intel_iommu"
into the iommu hardware and does the proper iommu hardware flushes. This
changes the
active translation tables from the ones in the old kernel to the copies in
the new kernel.
. This is legal because the translations in the two sets of tables are
currently identical:
- Intel(r) Virtualization Technology for Directed I/O. Architecture
Specification,
February 2011, Rev. 1.3 (section 11.2, paragraph 2)
In iommu_init_domains()
* Mark as in-use all domain-id's from the old kernel
. In case the new kernel contains a device that was not in the old kernel
and a new, unused domain-id is actually needed, the bitmap will give us one.
When a new domain is created for a device:
------------------------------------------
* If (this device has a context in the old kernel)
. Get domain-id, address-width, and IOVA ranges from the old kernel context;
. Get adr(page-entry-tables) from the copy in the new kernel;
. And apply all of the above values to the new domain structure.
* Else
. Create a new domain as normal
Most of the code to implement the new functionality was added at the end of
intel-iommu.c
Bill
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
