drivers/pci/intel-iommu.c: In function `__iommu_calculate_agaw':
drivers/pci/intel-iommu.c:437: sorry, unimplemented: inlining failed in call to 'width_to_agaw': function body not available
drivers/pci/intel-iommu.c:445: sorry, unimplemented: called from here
Move the offending function (and its siblings) to top-of-file, remove the
forward declaration.
Addresses https://bugzilla.kernel.org/show_bug.cgi?id=17441
Reported-by: Martin Mokrejs <mmokrejs@ribosome.natur.cuni.cz>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 9eecabcb9a ("intel-iommu: Abort
IOMMU setup for igfx if BIOS gave no shadow GTT space") uses a bunch of
magic numbers. Provide #defines for those to make it look slightly saner.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
drivers/pci/intel-iommu.c: In function 'dma_pte_addr':
drivers/pci/intel-iommu.c:239: warning: passing argument 1 of '__cmpxchg64' from incompatible pointer type
It seems that __cmpxchg64() now cares about the type of its pointer argument,
so give it a (uint64_t *) instead of a pointer to a structure which contains
only that.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
On some platforms (MacPro3,1) the BIOS assigns the ioatdma device to the
incorrect iommu causing faults when the driver initializes. Add a quirk
to catch this misconfiguration and try falling back to untranslated
operation (which works in the MacPro3,1 case).
Assuming there are other platforms with misconfigured iommus teach the
ioatdma driver to treat initialization failures as non-fatal (just fail
the driver load and emit a warning instead of triggering a BUG_ON).
This can be classified as a boot regression since 2.6.32 on affected
platforms since the ioatdma module did not autoload prior to that
kernel.
Cc: <stable@kernel.org>
Acked-by: David Woodhouse <David.Woodhouse@intel.com>
Reported-by: Chris Li <lkml@chrisli.org>
Tested-by: Chris Li <lkml@chrisli.org>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This patch allows IOMMU users to determine whether the
hardware and software support safe, isolated interrupt
remapping. Not all Intel IOMMUs have the hardware, and the
software for AMD is not there yet.
Signed-off-by: Tom Lyon <pugs@cisco.com>
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Certain revisions of this chipset appear to be broken. There is a shadow
GTT which mirrors the real GTT but contains pre-translated physical
addresses, for performance reasons. When a GTT update happens, the
translations are done once and the resulting physical addresses written
back to the shadow GTT.
Except sometimes, the physical address is actually written back to the
_real_ GTT, not the shadow GTT. Thus we start to see faults when that
physical address is fed through translation again.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
stanse found the following double lock.
In get_domain_for_dev:
spin_lock_irqsave(&device_domain_lock, flags);
domain_exit(domain);
domain_remove_dev_info(domain);
spin_lock_irqsave(&device_domain_lock, flags);
spin_unlock_irqrestore(&device_domain_lock, flags);
spin_unlock_irqrestore(&device_domain_lock, flags);
This happens when the domain is created by another CPU at the same time
as this function is creating one, and the other CPU wins the race to
attach it to the device in question, so we have to destroy our own
newly-created one.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Commit a99c47a2 "intel-iommu: errors with smaller iommu widths" replace the
dmar_domain->pgd with the first entry of page table when iommu's supported
width is smaller than dmar_domain's. But it use physical address directly
for new dmar_domain->pgd...
This result in KVM oops with VT-d on some machines.
Reported-by: Allen Kay <allen.m.kay@intel.com>
Cc: Tom Lyon <pugs@cisco.com>
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
intel_iommu_map_range() doesn't allow allocation at the very end of the
address space; that code has been simplified and corrected.
Signed-off-by: Tom Lyon <pugs@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
When using iommu_domain_alloc with the Intel iommu, the domain address
width is always initialized to 48 bits (agaw 2). This domain->agaw value
is then used by pfn_to_dma_pte to (always) build a 4 level page table.
However, not all systems support iommu width of 48 or 4 level page tables.
In particular, the Core i5-660 and i5-670 support an address width of 36
bits (not 39!), an agaw of only 1, and only 3 level page tables.
This version of the patch simply lops off extra levels of the page tables
if the agaw value of the iommu is less than what is currently allocated
for the domain (in intel_iommu_attach_device). If there were already
allocated addresses above what the new iommu can handle, EFAULT is
returned.
Signed-off-by: Tom Lyon <pugs@cisco.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
While it may be efficient on real hardware, emulation of global
invalidations is very expensive as all shadow entries must be examined.
This patch changes the behaviour when caching mode is enabled (which is
the case when IOMMU emulation takes place). In this case, page specific
invalidation is used instead.
Signed-off-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
In caching-mode mappings of pages (changes from non-present to present)
require invalidation.
Currently, this IOTLB flush is performed with domain ID of zero.
This is not according to the VT-d spec and causes big problems for
emulating software.
This patch uses the correct domain ID in IOTLB flushes.
Device IOTLB invalidation is performed only on present to non-present
changes. This decision is now based on explicit parameter instead of
zero domain-ID.
Signed-off-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
intel_map_sg used offset_pfn which was set to zero when invalidating the IOTLB.
intel_map_sg now uses size variable for this matter.
Signed-off-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Replace open-coded loop with for_each_set_bit().
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This patch changes the iommu-api functions for mapping and
unmapping page ranges to use the new page-size based
interface. This allows to remove the range based functions
later.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The new function pointer names match better with the
top-level functions of the iommu-api which are using them.
Main intention of this change is to make the ->{un}map
pointer names free for two new mapping functions.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
commit eb3fa7cb51 said Intel IOMMU
Intel IOMMU driver needs memory during DMA map calls to setup its
internal page tables and for other data structures. As we all know
that these DMA map calls are mostly called in the interrupt context
or with the spinlock held by the upper level drivers(network/storage
drivers), so in order to avoid any memory allocation failure due to
low memory issues, this patch makes memory allocation by temporarily
setting PF_MEMALLOC flags for the current task before making memory
allocation calls.
We evaluated mempools as a backup when kmem_cache_alloc() fails
and found that mempools are really not useful here because
1) We don't know for sure how much to reserve in advance
2) And mempools are not useful for GFP_ATOMIC case (as we call
memory alloc functions with GFP_ATOMIC)
(akpm: point 2 is wrong...)
The above description doesn't justify to waste system emergency memory
at all. Non MM subsystem must not use PF_MEMALLOC. Memory reclaim need
few memory, anyone must not prevent it. Otherwise the system cause
mysterious hang-up and/or OOM Killer invokation.
Plus, akpm already pointed out what we should do.
Then, this patch revert it.
Cc: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
We are seeing a bug when booting w/ iommu=pt with current upstream
(bisect blames 19943b0e30 "intel-iommu:
Unify hardware and software passthrough support).
The issue is specific to this loop during identity map initialization
of each device:
domain_context_mapping_one(si_domain, ..., CONTEXT_TT_PASS_THROUGH)
...
/* Skip top levels of page tables for
* iommu which has less agaw than default.
*/
for (agaw = domain->agaw; agaw != iommu->agaw; agaw--) {
pgd = phys_to_virt(dma_pte_addr(pgd));
if (!dma_pte_present(pgd)) { <------ failing here
spin_unlock_irqrestore(&iommu->lock, flags);
return -ENOMEM;
}
This box has 2 iommu's in it. The catchall iommu has MGAW == 48, and
SAGAW == 4. The other iommu has MGAW == 39, SAGAW == 2.
The device that's failing the above pgd test is the only device connected
to the non-catchall iommu, which has a smaller address width than the
domain default. This test is not necessary since the context is in PT
mode and the ASR is ignored.
Thanks to Don Dutile for discovering and debugging this one.
Cc: stable@kernel.org
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The hotplug notifier will call find_domain() to see if the device in
question has been assigned an IOMMU domain. However, this should never
be called for devices with a "dummy" domain, such as graphics devices
when intel_iommu=igfx_off is set and the corresponding IOMMU isn't even
initialised. If you do that, it'll oops as it dereferences the (-1)
pointer.
The notifier function should check iommu_no_mapping() for the
device before doing anything else.
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Some HP BIOSes report an RMRR region (a region which needs a 1:1 mapping
in the IOMMU for a given device) which has an end address lower than its
start address. Detect that and warn, rather than triggering the
BUG() in dma_pte_clear_range().
Cc: stable@kernel.org
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Change for PCI core to use pci_is_pcie() instead of checking
pci_dev->is_pcie.
Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
To support PCIe hot plug in IOMMU, we register a notifier to respond to device
change action.
When the notifier gets BUS_NOTIFY_UNBOUND_DRIVER, it removes the device
from its DMAR domain.
A hot added device will be added into an IOMMU domain when it first does IOMMU
op. So there is no need to add more code for hot add.
Without the patch, after a hot-remove, a hot-added device on the same
slot will not work.
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Tested-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The model for IOMMU passthrough is that decent devices that can cope
with DMA to all of memory get passthrough; crappy devices with a limited
dma_mask don't -- they get to use the IOMMU anyway.
This is done on the basis that IOMMU passthrough is usually wanted for
performance reasons, and it's only the decent PCI devices that you
really care about performance for, while the crappy 32-bit ones like
your USB controller can just use the IOMMU and you won't really care.
Unfortunately, the check for this was only looking at dev->dma_mask, not
at dev->coherent_dma_mask. And some devices have a 32-bit
coherent_dma_mask even though they have a full 64-bit dma_mask.
Even more unfortunately, fixing that simple oversight would upset
certain broken HP devices. Not only do they have a 32-bit
coherent_dma_mask, but they also have a tendency to do stray DMA to
unmapped addresses. And then they die when they take the DMA fault they
so richly deserve.
So if we do the 'correct' fix, it'll mean that affected users have to
disable IOMMU support completely on "a large percentage of servers from
a major vendor."
Personally, I have little sympathy -- given that this is the _same_
'major vendor' who is shipping machines which claim to have IOMMU
support but have obviously never _once_ booted a VT-d capable OS to do
any form of QA. But strictly speaking, it _would_ be a regression even
though it only ever worked by fluke.
For 2.6.33, we'll come up with a quirk which gives swiotlb support
for this particular device, and other devices with an inadequate
coherent_dma_mask will just get normal IOMMU mapping.
The simplest fix for 2.6.32, though, is just to jump through some hoops
to try to allocate coherent DMA memory for such devices in a place that
they can reach. We'd use dma_generic_alloc_coherent() for this if it
existed on IA64.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
If HW IOMMU initialization fails (Intel VT-d often does this,
typically due to BIOS bugs), we fall back to nommu. It doesn't
work for the majority since nowadays we have more than 4GB
memory so we must use swiotlb instead of nommu.
The problem is that it's too late to initialize swiotlb when HW
IOMMU initialization fails. We need to allocate swiotlb memory
earlier from bootmem allocator. Chris explained the issue in
detail:
http://marc.info/?l=linux-kernel&m=125657444317079&w=2
The current x86 IOMMU initialization sequence is too complicated
and handling the above issue makes it more hacky.
This patch changes x86 IOMMU initialization sequence to handle
the above issue cleanly.
The new x86 IOMMU initialization sequence are:
1. we initialize the swiotlb (and setting swiotlb to 1) in the case
of (max_pfn > MAX_DMA32_PFN && !no_iommu). dma_ops is set to
swiotlb_dma_ops or nommu_dma_ops. if swiotlb usage is forced by
the boot option, we finish here.
2. we call the detection functions of all the IOMMUs
3. the detection function sets x86_init.iommu.iommu_init to the
IOMMU initialization function (so we can avoid calling the
initialization functions of all the IOMMUs needlessly).
4. if the IOMMU initialization function doesn't need to swiotlb
then sets swiotlb to zero (e.g. the initialization is
sucessful).
5. if we find that swiotlb is set to zero, we free swiotlb
resource.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: chrisw@sous-sol.org
Cc: dwmw2@infradead.org
Cc: joerg.roedel@amd.com
Cc: muli@il.ibm.com
LKML-Reference: <1257849980-22640-10-git-send-email-fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Allocate dma-remapping structures using numa locality info. On platforms
having remapping hardware units span different nodes, this enables
optimized dma-remapping transalation structures access by remapping hardware.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Asus decided to ship a BIOS which configures sound DMA to go via the
dedicated IOMMU unit, but assigns precisely zero TLB entries to that
unit. Which causes the whole thing to deadlock, including the DMA
traffic on the _other_ IOMMU units. Nice one.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The following 64 bit promotions are necessary to handle memory above the
4GiB boundary correctly.
[dwmw2: Fix the second part not to need 64-bit arithmetic at all]
Signed-off-by: Benjamin LaHaise <ben.lahaise@neterion.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
If end_pfn is equal to (unsigned long)-1, then the loop will never end.
Seen on 32-bit kernel, but could have happened on 64-bit too once we get
hardware that supports 64-bit guest addresses.
Change both functions to a 'do {} while' loop with the test at the end,
and check for the PFN having wrapper round to zero.
Reported-by: Benjamin LaHaise <ben.lahaise@neterion.com>
Tested-by: Benjamin LaHaise <ben.lahaise@neterion.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This means we're limited to 44-bit addresses on 32-bit kernels, and
makes it sane for us to use 'unsigned long' for PFNs throughout.
Which is just as well, really, since we already do that.
Reported-by: Benjamin LaHaise <ben.lahaise@neterion.com>
Tested-by: Benjamin LaHaise <ben.lahaise@neterion.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Move tboot.h from asm to linux to fix the build errors of intel_txt
patch on non-X86 platforms. Remove the tboot code from generic code
init/main.c and kernel/cpu.c.
Signed-off-by: Shane Wang <shane.wang@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
This file needs to include linux/dmi.h directly rather than relying on
it being pulled in from elsewhere.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The kcalloc() failure path in iommu_init_domains() calls
free_dmar_iommu(), which assumes that ->domains, ->domain_ids,
and ->lock have been properly initialized.
Add checks in free_[dmar]_iommu to not use ->domains,->domain_ids
if not alloced. Move the lock init to prior to the kcalloc()'s,
so it is valid in free_context_table() when free_dmar_iommu() invokes
it at the end.
Patch based on iommu-2.6,
commit 132032274a
Signed-off-by: Donald Dutile <ddutile@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Mark si_domain_init and iommu_prepare_static_identity_mapping with
__init, to eliminate the following warnings:
WARNING: drivers/pci/built-in.o(.text+0xf1f4): Section mismatch in reference from the function si_domain_init() to the function .init.text:si_domain_work_fn()
The function si_domain_init() references
the function __init si_domain_work_fn().
This is often because si_domain_init lacks a __init
annotation or the annotation of si_domain_work_fn is wrong.
WARNING: drivers/pci/built-in.o(.text+0xe340): Section mismatch in reference from the function iommu_prepare_static_identity_mapping() to the function .init.text:si_domain_init()
The function iommu_prepare_static_identity_mapping() references
the function __init si_domain_init().
This is often because iommu_prepare_static_identity_mapping lacks a __init
annotation or the annotation of si_domain_init is wrong.
Signed-off-by: Matt Kraai <kraai@ftbfs.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
All callers of the former were also calling the latter, in one order or
the other, and failing to correctly clean up if the second returned
failure.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Two defects work together result in KVM device passthrough randomly can't
work:
1. iommu_snooping is not initialized to zero when vm_iommu_init() called.
So it is possible to get a random value.
2. One line added by commit 2c2e2c38("IOMMU Identity Mapping Support")
change the code path, let it bypass domain_update_iommu_cap(), as well as
missing the increment of domain iommu reference count.
The latter is also likely to cause a leak of domains on repeated VMM
assignment and deassignment.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The physical address passed to domain_pfn_mapping() should be rounded
down to the start of the MM page, not the VT-d page.
This issue causes kernel panic on PAGE_SIZE>VTD_PAGE_SIZE platforms e.g. ia64
platforms.
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
In domain_sg_mapping(), use aligned_nrpages() instead of hand-coded
rounding code for calculating the size of each sg elem. This means that
on IA64 we correctly round up to the MM page size, not just to the VT-d
page size.
Also remove the incorrect mm_to_dma_pfn() when intel_map_sg() calls
domain_sg_mapping() -- the 'size' variable is in VT-d pages already.
Signed-off-by: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
This makes the hardware passthrough mode work a lot more like the
software version, so that the behaviour of a kernel with 'iommu=pt'
is the same whether the hardware supports passthrough or not.
In particular:
- We use a single si_domain for the pass-through devices.
- 32-bit devices can be taken out of the pass-through domain so that
they don't have to use swiotlb.
- Devices will work again after being removed from a KVM guest.
- A potential oops on OOM (in init_context_pass_through()) is fixed.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
The tboot module will DMA protect all of memory in order to ensure the that
kernel will be able to initialize without compromise (from DMA). Consequently,
the kernel must enable Intel Virtualization Technology for Directed I/O
(VT-d or Intel IOMMU) in order to replace this broad protection with the
appropriate page-granular protection. Otherwise DMA devices will be unable
to read or write from memory and the kernel will eventually panic.
Because runtime IOMMU support is configurable by command line options, this
patch will force it to be enabled regardless of the options specified, and will
log a message if it was required to force it on.
dmar.c | 7 +++++++
intel-iommu.c | 17 +++++++++++++++--
2 files changed, 22 insertions(+), 2 deletions(-)
Signed-off-by: Joseph Cihula <joseph.cihula@intel.com>
Signed-off-by: Shane Wang <shane.wang@intel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
g_iommus is freed after we "goto error;".
Found by smatch (http://repo.or.cz/w/smatch.git).
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
We only ever obtain this lock immediately before the iova_rbtree_lock,
and release it immediately after the iova_rbtree_lock. So ditch it and
just use iova_rbtree_lock.
[v2: Remove the lockdep bits this time too]
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
After some API change, intel_iommu_unmap_range() introduced a assumption that
parameter size != 0, otherwise the dma_pte_clean_range() would have a
overflowed argument. But the user like KVM don't have this assumption before,
then some BUG() triggered.
Fix it by ignoring size = 0.
Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We did before, in the end -- but it was at the bottom of a long stack of
functions. Add an inline wrapper get_valid_domain_for_dev() which will
use the cached one _first_ and only make the out-of-line call if it's
not already set.
This takes the average time taken for a 1-page intel_map_sg() from 5961
cycles to 4812 cycles on my Lenovo x200s test box -- a modest 20%.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Andrew Morton