At the moment the PPC64/pseries guest only supports 4K/64K/16M IOMMU
pages and POWER8 CPU supports the exact same set of page size so
so far things worked fine.
However POWER9 supports different set of sizes - 4K/64K/2M/1G and
the last two - 2M and 1G - are not even allowed in the paravirt interface
(RTAS DDW) so we always end up using 64K IOMMU pages, although we could
back guest's 16MB IOMMU pages with 2MB pages on the host.
This stores the supported host IOMMU page sizes in VFIOContainer and uses
this later when creating a new DMA window. This uses the system page size
(64k normally, 2M/16M/1G if hugepages used) as the upper limit of
the IOMMU pagesize.
This changes the type of @pagesize to uint64_t as this is what
memory_region_iommu_get_min_page_size() returns and clz64() takes.
There should be no behavioral changes on platforms other than pseries.
The guest will keep using the IOMMU page size selected by the PHB pagesize
property as this only changes the underlying hardware TCE table
granularity.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This includes gcc8.1 fixes and the image is compiled using gcc 8.1 as well.
The full list of changes is:
> Fix bad assembler statements for compiling with gcc 8.1 / as 2.30
> libelf: Add REL32 to the list of ignored relocations
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This adds owners/parents (which are the same, just occasionally
owner==NULL) printing for memory regions; a new '-o' flag
enabled new output.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20180604032511.6980-1-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The changes are:
1. fixed broken_sc1;
2. added switching between boot consoles;
3. added PXE boot.
The full list is:
> lib/libnet/pxelinux: Fix two off-by-one bugs in the pxelinux.cfg parser
> lib/libnet/pxelinux: Make the size handling for pxelinux_load_cfg more logical
> libc: Add a simple implementation of an assert() function
> libnet: Support UUID-based pxelinux.cfg file names
> slof: Add a helper function to get the contents of a property in C code
> libnet: Add support for DHCPv4 options 209 and 210
> libnet: Wire up pxelinux.cfg network booting
> libnet: Add functions for downloading and parsing pxelinux.cfg files
> libnet: Put code for determing TFTP error strings into a separate function
> libc: Add the snprintf() function
> libnet: Pass ip_version via struct filename_ip
> resolve ihandle and xt handle in the input command (like for the output)
> Fix output word
> obp-tftp: Make sure to not overwrite paflof in memory
> libnet: Get rid of unused huge_load and block_size parameters
> libc: Check for NULL pointers in free()
> libc: Implement strrchr()
> libnet: Get rid of unnecessary (char *) casts
> broken_sc1: check for H_PRIVILEGE
> OF: Use new property "stdout-path" for boot console
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The recently introduced qom-list-properties QMP command raised
a question what properties it (and its cousin - device-list-properties)
can possibly print - only those defined by DeviceClass::props
or dynamically created in TypeInfo::instance_init() so properties created
elsewhere won't show up and this behaviour might confuse the user.
For example, PIIX4 does that from piix4_pm_realize() via
piix4_pm_add_propeties():
object_property_add_uint8_ptr(OBJECT(s), ACPI_PM_PROP_ACPI_ENABLE_CMD,
&acpi_enable_cmd, NULL);
This adds a note to the command descriptions about the limitation.
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20180530071129.9013-1-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
At the moment the device tree produced by the H_CAS handler has no
reserved map initialized at all which is not correct as at least one
empty record is required to be present as a marker of the end.
This does not cause problems now as the only consumer is SLOF which
does not look at the reserved map area.
However when DTC's "Improve libfdt's memory safety" changeset hits
the QEMU upstream, there will be errors reported and crashes observed.
This fixes the problem by adding an empty entry to the reserved map,
just like create_device_tree() does already.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
A "powernv" machine type defines an ISA bus but it does not add any DMA
controller to it so it is possible to hit assert(fdctrl->dma) by
adding "-machine powernv -device isa-fdc".
This replaces assert() with an error message.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[thuth: Slightly adjusted error message and updated scripts/device-crash-test]
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1521193892-15552-3-git-send-email-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This adds a possibility for the platform to tell VFIO not to emulate MSIX
so MMIO memory regions do not get split into chunks in flatview and
the entire page can be registered as a KVM memory slot and make direct
MMIO access possible for the guest.
This enables the entire MSIX BAR mapping to the guest for the pseries
platform in order to achieve the maximum MMIO preformance for certain
devices.
Tested on:
LSI Logic / Symbios Logic SAS3008 PCI-Express Fusion-MPT SAS-3 (rev 02)
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment we unconditionally avoid mapping MSIX data of a BAR and
emulate MSIX table in QEMU. However it is 1) not always necessary as
a platform may provide a paravirt interface for MSIX configuration;
2) can affect the speed of MMIO access by emulating them in QEMU when
frequently accessed registers share same system page with MSIX data,
this is particularly a problem for systems with the page size bigger
than 4KB.
A new capability - VFIO_REGION_INFO_CAP_MSIX_MAPPABLE - has been added
to the kernel [1] which tells the userspace that mapping of the MSIX data
is possible now. This makes use of it so from now on QEMU tries mapping
the entire BAR as a whole and emulate MSIX on top of that.
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=a32295c612c57990d17fb0f41e7134394b2f35f6
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
At the moment if vfio_memory_listener is registered in the system memory
address space, it maps/unmaps every RAM memory region for DMA.
It expects system page size aligned memory sections so vfio_dma_map
would not fail and so far this has been the case. A mapping failure
would be fatal. A side effect of such behavior is that some MMIO pages
would not be mapped silently.
However we are going to change MSIX BAR handling so we will end having
non-aligned sections in vfio_memory_listener (more details is in
the next patch) and vfio_dma_map will exit QEMU.
In order to avoid fatal failures on what previously was not a failure and
was just silently ignored, this checks the section alignment to
the smallest supported IOMMU page size and prints an error if not aligned;
it also prints an error if vfio_dma_map failed despite the page size check.
Both errors are not fatal; only MMIO RAM regions are checked
(aka "RAM device" regions).
If the amount of errors printed is overwhelming, the MSIX relocation
could be used to avoid excessive error output.
This is unlikely to cause any behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[aw: Fix Int128 bit ops]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
There is already 'device-list-properties' which does most of the job,
however it does not handle everything returned by qom-list-types such
as machines as they inherit directly from TYPE_OBJECT and not TYPE_DEVICE.
It does not handle abstract classes either.
This adds a new qom-list-properties command which prints properties
of a specific class and its instance. It is pretty much a simplified copy
of the device-list-properties handler.
Since it creates an object instance, device properties should appear
in the output as they are copied to QOM properties at the instance_init
hook.
This adds a object_class_property_iter_init() helper to allow class
properties enumeration uses it in the new QMP command to allow properties
listing for abstract classes.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20180301130939.15875-3-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
ObjectPropertyInfo is more generic and only missing @description.
This adds a description to ObjectPropertyInfo and removes
DevicePropertyInfo so the resulting ObjectPropertyInfo can be used
elsewhere.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20180301130939.15875-2-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Even with --disable-git-update, ./configure tries updating the capstone
submodule instead of marking it "no"; this disables capstone submodule
if git update is disabled.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Reviewed-by: Thomas Huth <thuth@redhat.com>
The "-machine xxx,help" prints kernel-irqchip possible values as
"OnOffSplit", this adds separators to the printed line.
Also, since only lower case letters are specified in qapi/common.json,
this changes the letter cases too.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
There is already @hostwin in vfio_listener_region_add() so there is no
point in having the other one.
Fixes: 2e4109de8e ("vfio/spapr: Create DMA window dynamically (SPAPR IOMMU v2)")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In order to enable TCE operations support in KVM, we have to inform
the KVM about VFIO groups being attached to specific LIOBNs;
the necessary bits are implemented already by IOMMU MR and VFIO.
This defines get_attr() for the SPAPR TCE IOMMU MR which makes VFIO
call the KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE ioctl and establish
LIOBN-to-IOMMU link.
This changes spapr_tce_set_need_vfio() to avoid TCE table reallocation
if the kernel supports the TCE acceleration.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
[aw - remove unnecessary sys/ioctl.h include]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
In order to enable TCE operations support in KVM, we have to inform
the KVM about VFIO groups being attached to specific LIOBNs. The KVM
already knows about VFIO groups, the only bit missing is which
in-kernel TCE table (the one with user visible TCEs) should update
the attached broups. There is an KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE
attribute of the VFIO KVM device which receives a groupfd/tablefd couple.
This uses a new memory_region_iommu_get_attr() helper to get the IOMMU fd
and calls KVM to establish the link.
As get_attr() is not implemented yet, this should cause no behavioural
change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This adds get_attr() to IOMMUMemoryRegionClass, like
iommu_ops::domain_get_attr in the Linux kernel.
This defines the first attribute - IOMMU_ATTR_SPAPR_TCE_FD - which
will be used between the pSeries machine and VFIO-PCI.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This adds a tracepoint to trace the KVM_SET_USER_MEMORY_REGION ioctl
parameters which is quite useful for debugging VFIO memory regions
being actually registered with KVM.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20171215052326.21386-1-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
As stated in the 1ad9f0a464 commit log, the returned entries are not
a whole PTEG. It was not a problem before 1ad9f0a464 as it would read
a single record assuming it contains a whole PTEG but now the code tries
reading the entire PTEG and "if ((n - i) < invalid)" produces negative
values which then are converted to size_t for memset() and that throws
seg fault.
This fixes the math.
While here, fix the last @i increment as well.
Fixes: 1ad9f0a464 "target/ppc: Fix KVM-HV HPTE accessors"
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The main changes are:
- able to handle more devices with specified bootindex;
- implements flatten device tree rendering, for both QEMU and guest kernel.
The full list is:
> boot: use a temporary bootdev-buf
> boot: do not concatenate bootdev
> libvirtio: Mark struct virtio_scsi_req_cmd as packed
> fdt: Implement "fdt-fetch" method for client interface
> rtas: Store RTAS address and entry in the device tree
> board-qemu: Fix slof-build-id length
> fdt: Pass the resulting device tree to QEMU
> fdt: Fix version and add a word for FDT header size
> tree: Rework set-chosen-cpu and store /chosen ihandle and phandle
> node: Add some documentation
> Revert various SLOF-to-QEMU private hypercalls
> Use input-device and output-device
> netboot: Create bootp-response when bootp is used
> libnet/ipv6: assign times_asked value directly
> usb-xhci: Reset ERSTSZ together with ERSTBA
> virtio-net: rework the driver to support multiple open
> board-qemu: add private hcall to inform host on "phandle" update
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
When support for multiple mappings per a region were added, this was
left behind, let's finish and remove unused bits.
Fixes: db0da029a1 ("vfio: Generalize region support")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
The vfio_iommu_spapr_tce driver advertises kernel's support for
v1 and v2 IOMMU support, however it is not always possible to use
the requested IOMMU type. For example, a pseries host platform does not
support dynamic DMA windows so v2 cannot initialize and QEMU fails to
start.
This adds a fallback to the v1 IOMMU if v2 cannot be used.
Fixes: 318f67ce13 ("vfio: spapr: Add DMA memory preregistering (SPAPR IOMMU v2)")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This moves pci_dev->name initialization earlier so
pci_dev->bus_master_as could get a name instead of an empty string.
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
The modern bar is accessed now via yet another address space created just
for that purpose and it does not really need FlatView and dispatch tree
as it has a single memory region so it is just a waste of memory. Things
get even worse when there are dozens or hundreds of virtio-pci devices -
since these address spaces are global, changing any of them triggers
rebuilding all address spaces.
This replaces indirect accesses to the modern BAR with a simple lookup
and direct calls to memory_region_dispatch_read/write.
This is expected to save lots of memory at boot time after applying:
[Qemu-devel] [PULL 00/32] Misc changes for 2017-09-22
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
This shares an cached empty FlatView among address spaces. The empty
FV is used every time when a root MR renders into a FV without memory
sections which happens when MR or its children are not enabled or
zero-sized. The empty_view is not NULL to keep the rest of memory
API intact; it also has a dispatch tree for the same reason.
On POWER8 with 255 CPUs, 255 virtio-net, 40 PCI bridges guest this halves
the amount of FlatView's in use (557 -> 260) and dispatch tables
(~800000 -> ~370000). In an unrelated experiment with 112 non-virtio
devices on x86 ("-M pc"), only 4 FlatViews are alive, and about ~2000
are created at startup.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-16-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This avoids usual memory_region_transaction_commit() which rebuilds
all FVs.
On POWER8 with 255 CPUs, 255 virtio-net, 40 PCI bridges guest this brings
down the boot time from 25s to 20s and reduces the amount of temporary FVs
allocated during machine constructon (~800000 -> ~640000) and amount of
temporary dispatch trees (~370000 -> ~300000), the total memory footprint
goes down (18G -> 17G).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-18-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since FlatViews are shared now and ASes not, this gets rid of
address_space_init_shareable().
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-17-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This adds a new "-d" switch to "info mtree" to print dispatch tree
internals.
This changes the way "-f" is handled - it prints now flat views and
associated address spaces.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-15-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This allows sharing flat views between address spaces (AS) when
the same root memory region is used when creating a new address space.
This is done by walking through all ASes and caching one FlatView per
a physical root MR (i.e. not aliased).
This removes search for duplicates from address_space_init_shareable() as
FlatViews are shared elsewhere and keeping as::ref_count correct seems
an unnecessary and useless complication.
This should cause no change and memory use or boot time yet.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-13-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Address spaces get to keep a root MR (alias or not) but FlatView stores
the actual MR as this is going to be used later on to decide whether to
share a particular FlatView or not.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-10-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We store AddressSpaceDispatch* in FlatView anyway so there is no need
to carry it from mem_add() to register_subpage/register_multipage.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-8-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
AS in ASD is only used to pass AS from mem_begin() to register_subpage()
to store it in MemoryRegionSection, we can do this directly now.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-6-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
As we are going to share FlatView's between AddressSpace's,
and AddressSpaceDispatch is a structure to perform quick lookup
in FlatView, this moves ASD to FlatView.
After previosly open coded ASD rendering, we can also remove
as->next_dispatch as the new FlatView pointer is stored
on a stack and set to an AS atomically.
flatview_destroy() is executed under RCU instead of
address_space_dispatch_free() now.
This makes mem_begin/mem_commit to work with ASD and mem_add with FV
as later on mem_add will be taking FV as an argument anyway.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-5-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This moves a FlatView allocation and initialization to a helper.
While we are nere, replace g_new with g_new0 to not to bother if we add
new fields in the future.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-4-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We are going to share FlatView's between AddressSpace's and per-AS
memory listeners won't suit the purpose anymore so open code
the dispatch tree rendering.
Since there is a good chance that dispatch_listener was the only
listener, this avoids address_space_update_topology_pass() if there is
no registered listeners; this should improve starting time.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-3-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This adds an AS** parameter to address_space_do_translate()
to make it easier for the next patch to share FlatViews.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-2-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The existing tries to round up the number of pages but @pages is always
calculated as the rounded up value minus one which makes ctz64() always
return 0 and have create.levels always set 1.
This removes wrong "-1" and allows having more than 1 levels. This becomes
handy for >128GB guests with standard 64K pages as this requires blocks
with zone order 9 and the popular limit of CONFIG_FORCE_MAX_ZONEORDER=9
means that only blocks up to order 8 are allowed.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The existing XHCI code reads the Event Ring Segment Table Base Address
Register (ERSTBA) every time when it is changed. However zero is its
default state so one would think that zero there means it is not in use.
This adds a check for ERSTBA in addition to the existing check for
the Event Ring Segment Table Size Register (ERSTSZ).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-id: 20170911065606.40600-1-aik@ozlabs.ru
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
This replaces g_malloc() with spapr_tce_alloc_table() as this is
the standard way of allocating tables and this allows moving the table
back to KVM when unplugging a VFIO PCI device and VFIO TCE acceleration
support is not present in the KVM.
Although spapr_tce_alloc_table() is expected to fail with EBUSY
if called when previous fd is not closed yet, in practice we will not
see it because cap_spapr_vfio is false at the moment.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This reverts c8e1158cf6 "elf-loader: warn about invalid endianness"
as it produces a useless message every time an LE kernel image is
passed via -kernel on a ppc64-pseries machine. The pseries machine
already checks for ELF_LOAD_WRONG_ENDIAN and tries with big_endian=0.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Yongbok Kim <yongbok.kim@imgtec.com>
The main changes are:
- fixes in PCI bridges code;
- LUN>255 are allowed not in virtio-scsi.
The full list is:
> pci-scan: Fix pci-bridge-set-mem-base and pci-bridge-set-mem-limit
> pci: Avoid 32-bit prefetchable memory area if possible
> Remove unused functions ishexdigit and $cat-comma
> pci: Translate PCI addresses to host addresses at the end of map-in
> Define 'open' and 'close' words of the /aliases nodes right from the start
> virtio-scsi: Allow LUNs bigger than 255
> paflof: Silence gcc's -Warray-bounds warning for stack pointers
> board_qemu: move code out of fdt-fix-node-phandle
> board_qemu: drop unused values early in fdt-fix-node-phandle
> pci: Improve the pci-var-out debug function
> libhvcall: drop unused KVMPPC_H_REPORT_MC_ERR and KVMPPC_H_NMI_MCE defines
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
f1c2dc7c86 "spapr-pci: rework MSI/MSIX" (07/2013) changed MSIX encoding
but forgot to change the comment so this changes it.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At the moment VFIO PCI device initialization works as follows:
vfio_realize
vfio_get_group
vfio_connect_container
register memory listeners (1)
update QEMU groups lists
vfio_kvm_device_add_group
Then (example for pseries) the machine reset hook triggers region_add()
for all regions where listeners from (1) are listening:
ppc_spapr_reset
spapr_phb_reset
spapr_tce_table_enable
memory_region_add_subregion
vfio_listener_region_add
vfio_spapr_create_window
This scheme works fine until we need to handle VFIO PCI device hotplug
and we want to enable PPC64/sPAPR in-kernel TCE acceleration on,
i.e. after PCI hotplug we need a place to call
ioctl(vfio_kvm_device_fd, KVM_DEV_VFIO_GROUP_SET_SPAPR_TCE).
Since the ioctl needs a LIOBN fd (from sPAPRTCETable) and a IOMMU group fd
(from VFIOGroup), vfio_listener_region_add() seems to be the only place
for this ioctl().
However this only works during boot time because the machine reset
happens strictly after all devices are finalized. When hotplug happens,
vfio_listener_region_add() is called when a memory listener is registered
but when this happens:
1. new group is not added to the container->group_list yet;
2. VFIO KVM device is unaware of the new IOMMU group.
This moves bits around to have all necessary VFIO infrastructure
in place for both initial startup and hotplug cases.
[aw: ie, register vfio groups with kvm prior to memory listener
registration such that kvm-vfio pseudo device ioctls are available
during the region_add callback]
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
e6f7e110ee "ppc/xics: remove the XICSState classes" got rid of
XICSState, this is just an leftover.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Cédric Le Goater <clg@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This finishes QOM'fication of IOMMUMemoryRegion by introducing
a IOMMUMemoryRegionClass. This also provides a fastpath analog for
IOMMU_MEMORY_REGION_GET_CLASS().
This makes IOMMUMemoryRegion an abstract class.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170711035620.4232-3-aik@ozlabs.ru>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This defines new QOM object - IOMMUMemoryRegion - with MemoryRegion
as a parent.
This moves IOMMU-related fields from MR to IOMMU MR. However to avoid
dymanic QOM casting in fast path (address_space_translate, etc),
this adds an @is_iommu boolean flag to MR and provides new helper to
do simple cast to IOMMU MR - memory_region_get_iommu. The flag
is set in the instance init callback. This defines
memory_region_is_iommu as memory_region_get_iommu()!=NULL.
This switches MemoryRegion to IOMMUMemoryRegion in most places except
the ones where MemoryRegion may be an alias.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <20170711035620.4232-2-aik@ozlabs.ru>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If a page size used by QEMU is not enabled in the PHB IOMMU page mask,
in-kernel acceleration of TCE handling won't be enabled and performance
might be slower than expected.
This prints a warning if system page size is not enabled. This should
print a warning if huge pages are enabled but sphb.pgsz still uses
the default value of 4K|64K.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This enables in-kernel handling of H_PUT_TCE_INDIRECT and
H_STUFF_TCE hypercalls. The host kernel support is there since v4.6,
in particular d3695aa4f452
("KVM: PPC: Add support for multiple-TCE hcalls").
H_PUT_TCE is already accelerated and does not need any special enablement.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
KVM_CAP_SPAPR_TCE capability allows creating TCE tables in KVM which
allows having in-kernel acceleration for H_PUT_TCE_xxx hypercalls.
However it only supports 32bit DMA windows at zero bus offset.
There is a new KVM_CAP_SPAPR_TCE_64 capability which supports 64bit
window size, variable page size and bus offset.
This makes use of the new capability. The kernel headers are already
updated as the kernel support went in to v4.6.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Normally pci_init_bus_master() would be called either via
bus->machine_done.notify or directly from do_pci_register_device().
However if a device's realize() failed, pci_init_bus_master() is not
called, and do_pci_unregister_device() fails on
memory_region_del_subregion() as it was not mapped.
This adds a check that subregion was mapped before unmapping it.
Fixes: c53598ed18 ("pci: Add missing drop of bus master AS reference")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Tested-by: John Snow <jsnow@redhat.com>
The recent introduction of a bus master container added
memory_region_add_subregion() into the PCI device registering path but
missed memory_region_del_subregion() in the unregistering path leaving
a reference to the root memory region of the new container.
This adds missing memory_region_del_subregion().
Fixes: 3716d5902d ("pci: introduce a bus master container")
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Various fixes in this update, the full list is:
> qemu-bootlist: Take the "-boot strict=off" setting properly into account
> virtio-scsi: initialize vring avail queue buffers
> virtio: Remove global variables in block and 9p driver
> Remove superfluous checkpoints in tree.fs
> Provide "write" function in the disk-label package
> virtio: Implement block write support
> scsi: Add SCSI block write support
> deblocker: Add a 'write' function
> virtio-scsi: Fix descriptor order for SCSI WRITE commands
> board-qemu: Add a possibility to use hvterm input instead of USB keyboard
> Do not try to use virtio-gpu in VGA mode
> virtio: Fix stack comment of virtio-blk-read
> envvar: Do not read default values for /options from the NVRAM anymore
> envvar: Set properties in /options during "(set-defaults)"
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
getrampagesize() returns the largest supported page size and mainly
used to know if huge pages are enabled.
However is implemented in target-ppc/kvm.c and not available
in TCG or other architectures.
This renames and moves gethugepagesize() to mmap-alloc.c where
fd-based analog of it is already implemented. This renames and moves
getrampagesize() to exec.c as it seems to be the common place for
helpers like this.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
When configured to compile out of tree, the configure script
copies BIOS blobs to the build directory. However since the PPC64 powernv
machine ROM has .lid extension, it is ignored and "make check" fails
when trying the powernv machine.
This adds *.lid to the list of copied blobs.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The main changes are:
* virtio-serial
* booting speed imrovement
* better PCI bridge support
The complete changelog is:
> virtio-serial: Fix compile error
> scsi: Remove debug functions from scsi-loader.fs
> scsi: Remove unused read-6 command
> obp-tftp: Remove the ciregs-buffer
> libnet: Simplify the net-load arguments passing
> libnet: Simplify the Forth-to-C wrapper of ping()
> Do not link libnet to net-snk anymore, and remove net-snk from board-qemu
> Add a Forth-to-C wrapper for the ping command, too
> Link libnet code to Paflof and add a wrapper for netboot()
> Remember execution tokens of "write" and "read" for socket operations
> Add virtio-serial device support
> Generalize output banner write routine
> Improve indentation in OF.fs
> scsi: implement READ (16) command
> rtas: Improve rtas-do-config-@ and rtas-do-config-! a little bit
> libnet: Make netapps.h includable from .code files
> libnet: Remove unused prototypes from netapps.h
> libnet: Fix the printout of the ping command
> libnet: Make sure to close sockets when we're done
> scsi: implement read-capacity-16
> pci: Fix secondary and subordinate PCI bus enumeration with board-qemu
> pci-phb: Fix stack underflow in phb-pci-walk-bridge
> paflof: Add a read() function to read keyboard input
> paflof: Add socket(), send() and recv() functions to paflof
> paflof: Provide get_timer() and set_timer() helper functions
> paflof: Add a write_mm_log helper function
> paflof: Copy sbrk code from net-snk
> paflof: Use CFLAGS from make.rules instead of completely redefining them
> Do not include the FCode evaluator by default anymore
> Source code beautification of board-qemu/slof/pci-interrupts.fs
> Allow PCI devices in PCI bridge slots greater than 4
> Fix bad interrupt pin numbering in interrupt-map property of PCI bridges
> Improve SLOF_alloc_mem_aligned()
> instance: Fix set-my-args for empty arguments
> Fix remaining compiler warnings in sloffs.c
> Remove misleading padding fields from ROM header definition
> Improve indentation in calculatecrc.h
> Do not include calculatecrc.h from assembler files
> Remove unused defines in calculatecrc.h
> libnet: Re-initialize global variables at the beginning of tftp()
> Remove dependency on cpu/@0 for booting
> usb: Set XHCI slot speed according to port status
> usb: Build correct route string for USB3 devices behind a hub
> usb: Initialize USB3 devices on a hub and keep track of hub topology
> usb: Increase amount of maximum slot IDs and add a sanity check
> usb: Move XHCI port state arrays from header to .c file
> tools: add copy functionality
> tools: added support to sloffs to read from /dev/slof_flash
> tools: added file append functionality
> tools: use crc checking code from romfs/tools
> tools: added initial version of sloffs
> romfs: factored out crc code, to make it usable from other locations
> tools: remove unused parts from the Makefile
> usb-hid: Fix non-working comma key
> fat-files: Fix access to FAT32 dir/files when cluster > 16-bits
> virtio-net: fix ring handling in receive
> net: Remove remainders of the MTFTP code
> net: Move also files from clients/net-snk/app/netapps/ to lib/libnet/
> net: Move files from clients/net-snk/app/netlib/ to lib/libnet/
> net-snk: Get rid of netlib and netapps prefixes in include statements
> usb-xhci: assign field4 before conditional
> Improve F12 key handling in boot menu
> Fix stack underflow that occurs with duplicated ESC in input
> rtas-nvram: optimize erase
> ipv6: Replace magic number 1500 with ETH_MTU_SIZE (i.e. 1518)
> ipv6: Fix NULL pointer dereference in ip6addr_add()
> ipv6: Fix memory leak in set_ipv6_address() / ip6_create_ll_address()
> ipv6: Clear memory after malloc if necessary
> ipv6: Fix possible NULL-pointer dereference in send_ipv6()
> ping: use gateway address for routing
> ping: add netmask in the ping argument
> xhci: fix missing keys from keyboard
> xhci: add memory barrier after filling the trb
> loaders: Remove netflash command
> boot: Remove legacy Forth words for network loading
> base: Move cnt-bits and bcd-to-bin to board-js2x folder
> base: Move huge-tftp-load variable to obp-tftp package
> base: Remove unused IP address conversion functions
> virtio: White space cleanup in virtio-9p.c
> virtio: Add modern version 1.0 support to 9p driver
> virtio: Set a proper name for virtio-9p device tree nodes
> pci: Fix mistype in "unkown-bridge"
> ipv6: Indent code with tabs, not with spaces
> ipv6: send_ipv6() has to return after doing NDP
> ipv6: Do not use unitialized MAC address array
> ipv6: Add support for sending packets through a router
> Remove unused sms code.
> virtio-net: initialize to populate mac address
> libbootmsg: Do not use '\b' characters when printing checkpoints
> dev-null: The "read" function has to return 0 if nothing has been read
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The tap backend is already using qemu-bridge-helper to attach tap
interface to a bridge but (unlike the bridge backend) it always uses
the default bridge name - br0.
This adds a "br" property support to the tap backend.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Greg Kurz <groug@kaod.org>
Tested-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Jason Wang <jasowang@redhat.com>
This adds a numa id property to a PHB to allow linking passed PCI device
to CPU/memory. It is up to the management stack to do CPU/memory pinning
to the node with the actual PCI device.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[dwg: Renamed property from "node" to "numa_node" to match the similar
one in the pxb device]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
A static analysis tool called BEAM detected possible side effect from
assert() calling a helper which may change an XHCI ring after every call.
This moves xhci_ring_fetch() out of assert() so it will be called
with and without enabled debug.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-id: 1468812548-31868-1-git-send-email-aik@ozlabs.ru
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
This adds support for Dynamic DMA Windows (DDW) option defined by
the SPAPR specification which allows to have additional DMA window(s)
The "ddw" property is enabled by default on a PHB but for compatibility
the pseries-2.6 machine and older disable it.
This also creates a single DMA window for the older machines to
maintain backward migration.
This implements DDW for PHB with emulated and VFIO devices. The host
kernel support is required. The advertised IOMMU page sizes are 4K and
64K; 16M pages are supported but not advertised by default, in order to
enable them, the user has to specify "pgsz" property for PHB and
enable huge pages for RAM.
The existing linux guests try creating one additional huge DMA window
with 64K or 16MB pages and map the entire guest RAM to. If succeeded,
the guest switches to dma_direct_ops and never calls TCE hypercalls
(H_PUT_TCE,...) again. This enables VFIO devices to use the entire RAM
and not waste time on map/unmap later. This adds a "dma64_win_addr"
property which is a bus address for the 64bit window and by default
set to 0x800.0000.0000.0000 as this is what the modern POWER8 hardware
uses and this allows having emulated and VFIO devices on the same bus.
This adds 4 RTAS handlers:
* ibm,query-pe-dma-window
* ibm,create-pe-dma-window
* ibm,remove-pe-dma-window
* ibm,reset-pe-dma-window
These are registered from type_init() callback.
These RTAS handlers are implemented in a separate file to avoid polluting
spapr_iommu.c with PCI.
This changes sPAPRPHBState::dma_liobn to an array to allow 2 LIOBNs
and updates all references to dma_liobn. However this does not add
64bit LIOBN to the migration stream as in fact even 32bit LIOBN is
rather pointless there (as it is a PHB property and the management
software can/should pass LIOBNs via CLI) but we keep it for the backward
migration support.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
New VFIO_SPAPR_TCE_v2_IOMMU type supports dynamic DMA window management.
This adds ability to VFIO common code to dynamically allocate/remove
DMA windows in the host kernel when new VFIO container is added/removed.
This adds a helper to vfio_listener_region_add which makes
VFIO_IOMMU_SPAPR_TCE_CREATE ioctl and adds just created IOMMU into
the host IOMMU list; the opposite action is taken in
vfio_listener_region_del.
When creating a new window, this uses heuristic to decide on the TCE table
levels number.
This should cause no guest visible change in behavior.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
[dwg: Added some casts to prevent printf() warnings on certain targets
where the kernel headers' __u64 doesn't match uint64_t or PRIx64]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
There are going to be multiple IOMMUs per a container. This moves
the single host IOMMU parameter set to a list of VFIOHostDMAWindow.
This should cause no behavioral change and will be used later by
the SPAPR TCE IOMMU v2 which will also add a vfio_host_win_del() helper.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This makes use of the new "memory registering" feature. The idea is
to provide the userspace ability to notify the host kernel about pages
which are going to be used for DMA. Having this information, the host
kernel can pin them all once per user process, do locked pages
accounting (once) and not spent time on doing that in real time with
possible failures which cannot be handled nicely in some cases.
This adds a prereg memory listener which listens on address_space_memory
and notifies a VFIO container about memory which needs to be
pinned/unpinned. VFIO MMIO regions (i.e. "skip dump" regions) are skipped.
The feature is only enabled for SPAPR IOMMU v2. The host kernel changes
are required. Since v2 does not need/support VFIO_IOMMU_ENABLE, this does
not call it when v2 is detected and enabled.
This enforces guest RAM blocks to be host page size aligned; however
this is not new as KVM already requires memory slots to be host page
size aligned.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[dwg: Fix compile error on 32-bit host]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The sPAPR TCE tables manage 2 copies when VFIO is using an IOMMU -
a guest view of the table and a hardware TCE table. If there is no VFIO
presense in the address space, then just the guest view is used, if
this is the case, it is allocated in the KVM. However since there is no
support yet for VFIO in KVM TCE hypercalls, when we start using VFIO,
we need to move the guest view from KVM to the userspace; and we need
to do this for every IOMMU on a bus with VFIO devices.
This implements the callbacks for the sPAPR IOMMU - notify_started()
reallocated the guest view to the user space, notify_stopped() does
the opposite.
This removes explicit spapr_tce_set_need_vfio() call from PCI hotplug
path as the new callbacks do this better - they notify IOMMU at
the exact moment when the configuration is changed, and this also
includes the case of PCI hot unplug.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The IOMMU driver may change behavior depending on whether a notifier
client is present. In the case of POWER, this represents a change in
the visibility of the IOTLB, for other drivers such as intel-iommu and
future AMD-Vi emulation, notifier support is not yet enabled and this
provides the opportunity to flag that incompatibility.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Peter Xu <peterx@redhat.com>
Tested-by: Peter Xu <peterx@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
[new log & extracted from [PATCH qemu v17 12/12] spapr_iommu, vfio, memory: Notify IOMMU about starting/stopping listening]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Every IOMMU has some granularity which MemoryRegionIOMMUOps::translate
uses when translating, however this information is not available outside
the translate context for various checks.
This adds a get_min_page_size callback to MemoryRegionIOMMUOps and
a wrapper for it so IOMMU users (such as VFIO) can know the minimum
actual page size supported by an IOMMU.
As IOMMU MR represents a guest IOMMU, this uses TARGET_PAGE_SIZE
as fallback.
This removes vfio_container_granularity() and uses new helper in
memory_region_iommu_replay() when replaying IOMMU mappings on added
IOMMU memory region.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
[dwg: Removed an unnecessary calculation]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This will be later used by the "ibm,reset-pe-dma-window" RTAS handler
which resets the DMA configuration to the defaults.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
LoPAPR dictates that during system reset all DMA windows must be removed
and the default DMA32 window must be created so does the patch.
At the moment there is just one window supported so no change in
behaviour is expected.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
We are going to have multiple DMA windows at different offsets on
a PCI bus. For the sake of migration, we will have as many TCE table
objects pre-created as many windows supported.
So we need a way to map windows dynamically onto a PCI bus
when migration of a table is completed but at this stage a TCE table
object does not have access to a PHB to ask it to map a DMA window
backed by just migrated TCE table.
This adds a "root" memory region (UINT64_MAX long) to the TCE object.
This new region is mapped on a PCI bus with enabled overlapping as
there will be one root MR per TCE table, each of them mapped at 0.
The actual IOMMU memory region is a subregion of the root region and
a TCE table enables/disables this subregion and maps it at
the specific offset inside the root MR which is 1:1 mapping of
a PCI address space.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The source guest could have reallocated the default TCE table and
migrate bigger/smaller table. This adds reallocation in post_load()
if the default table size is different on source and destination.
This adds @bus_offset, @page_shift to the migration stream as
a subsection so when DDW is added, migration to older machines will
still be possible. As @bus_offset and @page_shift are not used yet,
this makes no change in behavior.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Currently TCE tables are created once at start and their sizes never
change. We are going to change that by introducing a Dynamic DMA windows
support where DMA configuration may change during the guest execution.
This changes spapr_tce_new_table() to create an empty zero-size IOMMU
memory region (IOMMU MR). Only LIOBN is assigned by the time of creation.
It still will be called once at the owner object (VIO or PHB) creation.
This introduces an "enabled" state for TCE table objects, some
helper functions are added:
- spapr_tce_table_enable() receives TCE table parameters, stores in
sPAPRTCETable and allocates a guest view of the TCE table
(in the user space or KVM) and sets the correct size on the IOMMU MR;
- spapr_tce_table_disable() disposes the table and resets the IOMMU MR
size; it is made public as the following DDW code will be using it.
This changes the PHB reset handler to do the default DMA initialization
instead of spapr_phb_realize(). This does not make differenct now but
later with more than just one DMA window, we will have to remove them all
and create the default one on a system reset.
No visible change in behaviour is expected except the actual table
will be reallocated every reset. We might optimize this later.
The other way to implement this would be dynamically create/remove
the TCE table QOM objects but this would make migration impossible
as the migration code expects all QOM objects to exist at the receiver
so we have to have TCE table objects created when migration begins.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This allows dynamic allocation for migrating arrays.
Already existing VMSTATE_VARRAY_UINT32 requires an array to be
pre-allocated, however there are cases when the size is not known in
advance and there is no real need to enforce it.
This defines another variant of VMSTATE_VARRAY_UINT32 with WMS_ALLOC
flag which tells the receiving side to allocate memory for the array
before receiving the data.
The first user of it is a dynamic DMA window which existence and size
are totally dynamic.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At the moment presence of vfio-pci devices on a bus affect the way
the guest view table is allocated. If there is no vfio-pci on a PHB
and the host kernel supports KVM acceleration of H_PUT_TCE, a table
is allocated in KVM. However, if there is vfio-pci and we do yet not
KVM acceleration for these, the table has to be allocated by
the userspace. At the moment the table is allocated once at boot time
but next patches will reallocate it.
This moves kvmppc_create_spapr_tce/g_malloc0 and their counterparts
to helpers.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The user could have picked LIOBN via the CLI but the device tree
rendering code would still use the value derived from the PHB index
(which is the default fallback if LIOBN is not set in the CLI).
This replaces SPAPR_PCI_LIOBN() with the actual DMA LIOBN value.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At the moment IOMMU MR only translate to the system memory.
However if some new code changes this, we will need clear indication why
it is not working so here is the check.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Since a788f227 "memory: Allow replay of IOMMU mapping notifications"
when new VFIO listener is added, all existing IOMMU mappings are
replayed. However there is a problem that the base address of
an IOMMU memory region (IOMMU MR) is ignored which is not a problem
for the existing user (which is pseries) with its default 32bit DMA
window starting at 0 but it is if there is another DMA window.
This stores the IOMMU's offset_within_address_space and adjusts
the IOVA before calling vfio_dma_map/vfio_dma_unmap.
As the IOMMU notifier expects IOVA offset rather than the absolute
address, this also adjusts IOVA in sPAPR H_PUT_TCE handler before
calling notifier(s).
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
7532d3cbf "vfio: Fix 128 bit handling" added support for 64bit IOMMU
memory regions when those are added to VFIO address space; however
removing code cannot cope with these as int128_get64() will fail on
1<<64.
This copies 128bit handling from region_add() to region_del().
Since the only machine type which is actually going to use 64bit IOMMU
is pseries and it never really removes them (instead it will dynamically
add/remove subregions), this should cause no behavioral change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
ePAPR defines "hcall-instructions" device-tree property which contains
code to call hypercalls in ePAPR paravirtualized guests. In general
pseries guests won't use this property, instead using the PAPR defined
hypercall interface.
However, this property has been re-used to implement a hack to allow
PR KVM to run (slightly modified) guests in some situations where it
otherwise wouldn't be able to (because the system's L0 hypervisor
doesn't forward the PAPR hypercalls to the PR KVM kernel).
Hence, this property is always present in the device tree for pseries
guests. All KVM guests use it at least to read features via the
KVM_HC_FEATURES hypercall.
The property is populated by the code returned from the KVM's
KVM_PPC_GET_PVINFO ioctl; if not implemented in the KVM, QEMU supplies
code which will fail all hypercall attempts. If QEMU does not create
the property, and the guest kernel is compiled with
CONFIG_EPAPR_PARAVIRT (which is normally the case), there is exactly
the same stub at @epapr_hypercall_start already.
Rather than maintaining this fairly useless stub implementation, it
makes more sense not to create the property in the device tree in the
first place if the host kernel does not implement it.
This changes kvmppc_get_hypercall() to return 1 if the host kernel
does not implement KVM_CAP_PPC_GET_PVINFO. The caller can use it to decide
on whether to create the property or not.
This changes the pseries machine to not create the property if KVM does
not implement KVM_PPC_GET_PVINFO. In practice this means that from now
on the property will not be created if either HV KVM or TCG is used.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
[reworded commit message for clarity --dwg]
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
This adds a new POWER8+NVLink CPU PVR which core is identical to POWER8
but has a different PVR. The only available machine now has PVR
pvr 004c 0100 so this defines "POWER8NVL" alias as v1.0.
The corresponding kernel commit is
https://github.com/torvalds/linux/commit/ddee09c099c3
"powerpc: Add PVR for POWER8NVL processor"
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The main change is virtio 1.0 support.
The complete changelog is:
> dhcp: fix warning messages when calling strtoip()
> virtio-scsi: enable virtio 1.0
> virtio-scsi: use virtio_fill desc api
> virtio-scsi: use idx during initialization
> virtio-net: enable virtio 1.0
> virtio-blk: enable virtio 1.0
> virtio: 1.0 helper to read 16/32/64 bit value
> virtio: add and enable 1.0 device setup
> virtio: 1.0 guest features negotiation
> virtio: update features set/get register accessor
> virtio: make all virtio apis 1.0 aware
> virtio: add 64-bit virtio helpers for 1.0
> virtio: add virtio 1.0 related struct and defines
> virtio: get rid of type variable in virtio_device
> virtio-net: move setup-mac to the open routine
> virtio-net: make net_hdr_size a variable
> virtio-net: replace vq array with vq_{tx,rx}
> virtio-net: use virtio_fill_desc
> virtio-{net,blk,scsi,9p}: use status variable
> virtio-blk: add helpers for filling descriptors
> virtio-{blk,9p}: enable resetting the device
> virtio: introduce helper for initializing virt queue
> virtio: fix code style/design issues.
> fix code style in byteorder.h
> pci: add byte read/write helper routines
> virtio-net: fix gcc warnings (-Wextra)
> virtio-blk: fix gcc warnings (-Wextra)
> readme: Add a note about coding style
> dhcp: Remove duplicated strtoip()
> ethernet: Fix gcc warnings
> net-snk: Fix gcc warnings
> net-snk: Fix coding style
> net-snk: Fix memory leak in dhcp6_process_options()
> net-snk: Fix memory leak in ip6_to_multicast_mac() / send_ipv6()
> net-snk: Remove bad NEIGHBOUR_SOLICITATION code in send_ipv6()
> Fix dma-alloc and dma-map-in functions on board-js2x
> net-snk: Allow stateless autoconfig IPv6 addresses with IP_INIT_IPV6_MANUAL
> net-snk: Simplify the ip6_is_multicast() function
> net-snk: Move global variable definition out of the header file
> net-snk: Prefer non-link-local unicast IPv6 addresses if possible
> net-snk: Fix the check for link-local addresses when receiving RAs
> net-snk: Remove junk at the end of IPv6 TFTP ACK and error packets
> Fix format strings in usb-ohci.c
> net-snk: Get rid of junk at the end of sent DHCPv6 packets
> net-snk: Use transaction IDs in DHCPv4, too
> net-snk: Make use of DHCPv6 transaction IDs
> net-snk: Seed the pseudo-random number generator
> libc: Add srand() call
> libc: Fix the rand() function to return non-zero values
> net-snk: Improve printed text when booting via network
> Increase temporary buffer size of ibm,client-architecture-support call
> Move archsupport.fs into board-qemu directory
> boot: stop booting when we encounter HALT
> fat-files: Fix bug with root-entries = 0 on certain FAT32 file systems
> usb: print unhandled descriptor in debug mode
> Improve stack usage with libnvram get_partition function
> Improve stack usage in libnvram environment variable code
> libc: Port vsnprintf back from skiboot
> Move the code for rfill into a separate function
> Rework wrapper for new_nvram_partition() and fix possible bug in there
> Stack optimization in libusb: split up setup_new_device()
> Check for stack overflow in paflof engine
> Clean up pending packet variable in ipv4 code
> Fix tracking of pending outgoing packets when handling ARP replies
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Section B.6.2.1 Root Node Properties of PAPR specification defines
a set of properties which shall be present in the device tree root,
one of these properties is "system-id" which "should be unique across
all systems and all manufacturers". Since UUID is meant to be unique,
it makes sense to use it as "system-id".
This adds "system-id" property to the device tree root when not empty.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
At the moment get_monitor_def() returns only registers from statically
defined monitor_defs array. However there is a lot of BOOK3S SPRs
which are not in the list and cannot be printed from the monitor.
This adds a new target platform hook - target_get_monitor_def().
The hook is called if a register was not found in the static
array returned by the target_monitor_defs() hook.
The hook is only defined for POWERPC, it returns registered
SPRs and fails on unregistered ones providing the user with information
on what is actually supported on the running CPU. The register value is
saved as uint64_t as it is the biggest supported register size;
target_ulong cannot be used because of the stub - it is in a "common"
code and cannot include "cpu.h", etc; this is also why the hook prototype
is redefined in the stub instead of being included from some header.
This replaces static descriptors for GPRs, FPRs, SRs with a helper which
looks for a value in a corresponding array in the CPUPPCState.
The immediate effect is that all 32 SRs can be printed now (instead of 16);
later this can be reused for VSX or TM registers.
This replaces callbacks for MSR and XER with static descriptors in
monitor_defs as they are stored in CPUPPCState.
While we are here, this adds "cr" as a synonym of "ccr".
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The changes are:
1. supports recent binutils;
2. 64bit BARs behind PCI bridges supported;
3. Many fixes for USB keyboard support - keys, XHCI;
4. virtio-vga support.
This image was built with:
gcc version 4.8.3 20140911 (Red Hat 4.8.3-7) (GCC)
GNU ld version 2.23.2
The full changelog is:
> version: update to 20151103
> documentation: Add a clause about signing off
> qemu/js2x/client: Support binutils >= 2.25.1
> Fix special keys on USB
> Fix function keys on USB
> pci-scan: program 64-bit mem bar range in pci-bridge bar
> Allow to build SLOF on Little Endian host
> usb-xhci: add keyboard support
> usb-xhci: ready the link trb early
> usb-xhci: scan usb high speed ports
> usb-xhci: bulk improve event handling loop
> usb-xhci: return on allocation failure
> usb-xhci: add delay in shutdown path
> usb-xhci: event trbs does not need link trb
> usb-hid: refactor usb key reading
> takeover: Fix header includes
> board-js2x: Add missing file dma-function.fs
> vga: Add support for virtio-vga
> qemu-vga: Use MMIO BAR instead of legacy IO ports
> slof: Change call_c() function to a proper assembler function
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
As the comment in kvm_set_phys_mem() says, KVM works in page size chunks.
However it uses hardcoded TARGET_PAGE_SIZE which is 4K on most platforms
while actual host may use different page size, for example, PPC64 hosts
use 64K system pages.
This replaces static TARGET_PAGE_SIZE with run-time calculated
qemu_real_host_page_size.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <1444102257-17405-1-git-send-email-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
sPAPR uses hard coded limit of maximum 255 supported CPUs which is
exactly the same as QEMU-wide limit which is MAX_CPUMASK_BITS and also
defined as 255.
This makes use of a global CPU number limit for the "pseries" machine.
In order to anticipate future increase of the MAX_CPUMASK_BITS
(or to help debugging large systems), this also bumps the FDT_MAX_SIZE
limit from 256K to 1M assuming that 1 CPU core needs roughly 512 bytes
in the device tree so the new limit can cover up to 2048 CPU cores.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
For setting debug watchpoints, sPAPR guests use H_SET_MODE hypercall.
The existing QEMU H_SET_MODE handler does not support this but
the KVM handler in HV KVM does. However it is not enabled.
This enables the in-kernel H_SET_MODE handler which handles:
- Completed Instruction Address Breakpoint Register
- Watch point 0 registers.
The rest is still handled in QEMU.
Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The changes are:
1. GPT support;
2. Much faster VGA support.
The full changelog is:
> Add missing half word access case to _FASTRMOVE and _FASTMOVE
> Remove unused RMOVE64 stub
> fbuffer: Implement RFILL as an accelerated primitive
> fbuffer: Implement MRMOVE as an accelerated primitive
> fbuffer: Precalculate line length in bytes
> terminal: Disable the terminal-write trace by default
> boot: remove trailing ":" in the bootpath
> ci: implement boot client interface
> boot: bootpath should be complete device path
> fbuffer: Use a smaller cursor
> fbuffer: Improve invert-region helper
> usb-hid: Caps is not always shift
> cas: Increase FDT buffer size to accomodate larger ibm, cas node properties
> README: Update with patch submittion note
> disk-label: add support for booting from GPT FAT partition
> disk-label: introduce helper to check fat filesystem
> introduce 8-byte LE helpers
> disk-label: simplify gpt-prep-partition? routine
> fbuffer: introduce the invert-region-x helper
> fbuffer: introduce the invert-region helper
> fbuffer: simplify address computations in fb8-toggle-cursor
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
The changelog is:
> version: update to 20150429
> pci: Use QEMU created PCI device nodes
> usb: support 64-bit pci bars
> pci: Support 64-bit address translation
> pci: program correct bridge limit registers during probe
> scsi: handle report-luns failure
> Fix "key?" Forth word when using USB keyboards
> Remove bulk.fs package
> Include make.rules in the library Makefiles
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Alexander Graf <agraf@suse.de>
On systems with guest visible IOMMU, adding a new memory region onto
PCI bus calls vfio_listener_region_add() for every DMA window. This
installs a notifier for IOMMU memory regions. The notifier is supposed
to be removed vfio_listener_region_del(), however in the case of mixed
PHB (emulated + VFIO devices) when last VFIO device is unplugged and
container gets destroyed, all existing DMA windows stay alive altogether
with the notifiers which are on the linked list which head was in
the destroyed container.
This unregisters IOMMU memory region notifier when a container is
destroyed.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
This updates linux-headers against master 4.2-rc1 (commit
d770e558e21961ad6cfdf0ff7df0eb5d7d4f0754). This is the result of
./scripts/update-linux-headers.sh work.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
