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>
A container can be used instead of an alias to allow switching between
multiple subregions. In this case we cannot directly share the
subregions (since they only belong to a single parent), but if the
subregions are aliases we can in turn walk those.
This is not enough to remove all source of quadratic FlatView creation,
but it enables sharing of the PCI bus master FlatViews (and their
AddressSpaceDispatch structures) across all PCI devices. For 112
virtio-net-pci devices, boot time is reduced from 25 to 10 seconds and
memory consumption from 1.4 to 1 G.
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 creates a new AS object without any FlatView as
memory_region_transaction_commit() may want to reuse the empty FV.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-14-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>
So it is called (twice) from the same function. This is to make the next
patches a bit simpler.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-12-aik@ozlabs.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is to make next patches simpler.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-11-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>
This renames some helpers to reflect better what they do.
This should cause no behavioural change.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Message-Id: <20170921085110.25598-9-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>
It's possible for address_space_get_flatview() as it currently stands
to cause a use-after-free for the returned FlatView, if the reference
count is incremented after the FlatView has been replaced by a writer:
thread 1 thread 2 RCU thread
-------------------------------------------------------------
rcu_read_lock
read as->current_map
set as->current_map
flatview_unref
'--> call_rcu
flatview_ref
[ref=1]
rcu_read_unlock
flatview_destroy
<badness>
Since FlatViews are not updated very often, we can just detect the
situation using a new atomic op atomic_fetch_inc_nonzero, similar to
Linux's atomic_inc_not_zero, which performs the refcount increment only if
it hasn't already hit zero. This is similar to Linux commit de09a9771a53
("CRED: Fix get_task_cred() and task_state() to not resurrect dead
credentials", 2010-07-29).
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This avoids a name clash with the access macro on windows 64:
make
CHK version_gen.h
CC aarch64-softmmu/memory.o
/home/konrad/qemu/memory.c: In function 'access_with_adjusted_size':
/home/konrad/qemu/memory.c:591:73: error: macro "access" passed 7 arguments, \
but takes just 2
(size - access_size - i) * 8, access_mask, attrs);
^
Signed-off-by: KONRAD Frederic <frederic.konrad@adacore.com>
Message-Id: <1505988260-8483-1-git-send-email-frederic.konrad@adacore.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SunOS declares struct queue in <netinet/in.h>.
This fixes build on SmartOS (Joyent).
Patch cherry-picked from pkgsrc by jperkin (Joyent).
Signed-off-by: Kamil Rytarowski <n54@gmx.com>
Message-Id: <20170903163304.17919-1-n54@gmx.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Qemu_savevm_state_cleanup takes about 300ms in my ram migration tests
with a 8U24G vm(20G is really occupied), the main cost comes from
KVM_SET_USER_MEMORY_REGION ioctl when mem.memory_size = 0 in
kvm_set_user_memory_region. In kmod, the main cost is
kvm_zap_obsolete_pages, which traverses the active_mmu_pages list to
zap the unsync sptes.
It can be optimized by delaying memory_global_dirty_log_stop to the next
vm_start.
Changes v2->v3:
- NULL VMChangeStateHandler if it is deleted and protect the scenario
of nested invocations of memory_global_dirty_log_start/stop [Paolo]
Changes v1->v2:
- create a VMChangeStateHandler in memory.c to reduce the coupling [Paolo]
Signed-off-by: Jay Zhou <jianjay.zhou@huawei.com>
Message-Id: <1501237733-2736-1-git-send-email-jianjay.zhou@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add new utility functions which both initialize a RAM
MemoryRegion and arrange for its contents to be migrated;
we give thes the memory_region_init_ram(), memory_region_init_rom()
and memory_region_init_rom_device() names that we just freed up
by renaming the old implementations to _nomigrate().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1499438577-7674-6-git-send-email-peter.maydell@linaro.org
Rename memory_region_init_rom() to memory_region_init_rom_nomigrate()
and memory_region_init_rom_device() to
memory_region_init_rom_device_nomigrate().
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1499438577-7674-5-git-send-email-peter.maydell@linaro.org
Rename memory_region_init_ram() to memory_region_init_ram_nomigrate().
This leaves the way clear for us to provide a memory_region_init_ram()
which does handle migration.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 1499438577-7674-4-git-send-email-peter.maydell@linaro.org
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>
This introduces a special callback which allows to run code from some MMIO
devices.
SysBusDevice with a MemoryRegion which implements the request_ptr callback will
be notified when the guest try to execute code from their offset. Then it will
be able to eg: pre-load some code from an SPI device or ask a pointer from an
external simulator, etc..
When the pointer or the data in it are no longer valid the device has to
invalidate it.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Add a new function to initialize a RAM memory region with a file
descriptor to be mmap-ed.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20170602141229.15326-5-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We were always passing in that one as "false" to assume that's an read
operation, and we also assume that IOMMU translation would always have
that read permission. A better permission would be IOMMU_NONE since the
replay is after all not a real read operation, but just a page table
rebuilding process.
CC: David Gibson <david@gibson.dropbear.id.au>
CC: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
This patch converts the old "is_write" bool into IOMMUAccessFlags. The
difference is that "is_write" can only express either read/write, but
sometimes what we really want is "none" here (neither read nor write).
Replay is an good example - during replay, we should not check any RW
permission bits since thats not an actual IO at all.
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: David Gibson <david@gibson.dropbear.id.au>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jason Wang <jasowang@redhat.com>
This patch adds support for getting and using a local copy of the dirty
bitmap.
memory_region_snapshot_and_clear_dirty() will create a snapshot of the
dirty bitmap for the specified range, clear the dirty bitmap and return
the copy. The returned bitmap can be a bit larger than requested, the
range is expanded so the code can copy unsigned longs from the bitmap
and avoid atomic bit update operations.
memory_region_snapshot_get_dirty() will return the dirty status of
pages, pretty much like memory_region_get_dirty(), but using the copy
returned by memory_region_copy_and_clear_dirty().
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20170421091632.30900-3-kraxel@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Originally we have one memory_region_iommu_replay() function, which is
the default behavior to replay the translations of the whole IOMMU
region. However, on some platform like x86, we may want our own replay
logic for IOMMU regions. This patch adds one more hook for IOMMUOps for
the callback, and it'll override the default if set.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: \"Michael S. Tsirkin\" <mst@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-6-git-send-email-peterx@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Generalizing the notify logic in memory_region_notify_iommu() into a
single function. This can be further used in customized replay()
functions for IOMMUs.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: \"Michael S. Tsirkin\" <mst@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-5-git-send-email-peterx@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
This is an "global" version of existing memory_region_iommu_replay() -
we announce the translations to all the registered notifiers, instead of
a specific one.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: \"Michael S. Tsirkin\" <mst@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-4-git-send-email-peterx@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
A new macro is provided to iterate all the IOMMU notifiers hooked
under specific IOMMU memory region.
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: \"Michael S. Tsirkin\" <mst@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-3-git-send-email-peterx@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
In this patch, IOMMUNotifier.{start|end} are introduced to store section
information for a specific notifier. When notification occurs, we not
only check the notification type (MAP|UNMAP), but also check whether the
notified iova range overlaps with the range of specific IOMMU notifier,
and skip those notifiers if not in the listened range.
When removing an region, we need to make sure we removed the correct
VFIOGuestIOMMU by checking the IOMMUNotifier.start address as well.
This patch is solving the problem that vfio-pci devices receive
duplicated UNMAP notification on x86 platform when vIOMMU is there. The
issue is that x86 IOMMU has a (0, 2^64-1) IOMMU region, which is
splitted by the (0xfee00000, 0xfeefffff) IRQ region. AFAIK
this (splitted IOMMU region) is only happening on x86.
This patch also helps vhost to leverage the new interface as well, so
that vhost won't get duplicated cache flushes. In that sense, it's an
slight performance improvement.
Suggested-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1491562755-23867-2-git-send-email-peterx@redhat.com>
[ehabkost: included extra vhost_iommu_region_del() change from Peter Xu]
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
clear pending status before calling memory commit.
Otherwise when memory_region_finalize is called,
memory_region_transaction_depth is 0 and
memory_region_update_pending is true.
That's wrong.
Signed-off -by: Anthony Xu <anthony.xu@intel.com>
Message-Id: <4712D8F4B26E034E80552F30A67BE0B1A2E3D5@ORSMSX112.amr.corp.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The address of memory regions might overflow when something wrong
happened, like reported in:
https://lists.gnu.org/archive/html/qemu-devel/2017-03/msg02043.html
For easier debugging, let's try to detect it.
Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1489496187-624-1-git-send-email-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
At the moment ram device's memory regions are DEVICE_NATIVE_ENDIAN. It's
incorrect. This memory region is backed by a MMIO area in host, so the
uint64_t data that MemoryRegionOps read from/write to this area should be
host-endian rather than target-endian. Hence, current code does not work
when target and host endianness are different which is the most common case
on PPC64. To fix it, this introduces DEVICE_HOST_ENDIAN for the ram device.
This has been tested on PPC64 BE/LE host/guest in all possible combinations
including TCG.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Yongji Xie <xyjxie@linux.vnet.ibm.com>
Message-Id: <1488171164-28319-1-git-send-email-xyjxie@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This finally allows TCG to benefit from the iothread introduction: Drop
the global mutex while running pure TCG CPU code. Reacquire the lock
when entering MMIO or PIO emulation, or when leaving the TCG loop.
We have to revert a few optimization for the current TCG threading
model, namely kicking the TCG thread in qemu_mutex_lock_iothread and not
kicking it in qemu_cpu_kick. We also need to disable RAM block
reordering until we have a more efficient locking mechanism at hand.
Still, a Linux x86 UP guest and my Musicpal ARM model boot fine here.
These numbers demonstrate where we gain something:
20338 jan 20 0 331m 75m 6904 R 99 0.9 0:50.95 qemu-system-arm
20337 jan 20 0 331m 75m 6904 S 20 0.9 0:26.50 qemu-system-arm
The guest CPU was fully loaded, but the iothread could still run mostly
independent on a second core. Without the patch we don't get beyond
32206 jan 20 0 330m 73m 7036 R 82 0.9 1:06.00 qemu-system-arm
32204 jan 20 0 330m 73m 7036 S 21 0.9 0:17.03 qemu-system-arm
We don't benefit significantly, though, when the guest is not fully
loading a host CPU.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Message-Id: <1439220437-23957-10-git-send-email-fred.konrad@greensocs.com>
[FK: Rebase, fix qemu_devices_reset deadlock, rm address_space_* mutex]
Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com>
[EGC: fixed iothread lock for cpu-exec IRQ handling]
Signed-off-by: Emilio G. Cota <cota@braap.org>
[AJB: -smp single-threaded fix, clean commit msg, BQL fixes]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Pranith Kumar <bobby.prani@gmail.com>
[PM: target-arm changes]
Acked-by: Peter Maydell <peter.maydell@linaro.org>
Make it easy to unregister a MemoryListener without tracking whether it
had been registered before.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Introduce rules in the top level Makefile that are able to generate
trace.[ch] files in every subdirectory which has a trace-events file.
The top level directory is handled specially, so instead of creating
trace.h, it creates trace-root.h. This allows sub-directories to
include the top level trace-root.h file, without ambiguity wrt to
the trace.g file in the current sub-dir.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-id: 20170125161417.31949-7-berrange@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Adding one more option "-f" for "info mtree" to dump the flat views of
all the address spaces.
This will be useful to debug the memory rendering logic, also it'll be
much easier with it to know what memory region is handling what address
range.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1484556005-29701-3-git-send-email-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We were dumping RW bits for each memory region, that might be confusing.
It'll make more sense to dump the memory region type directly rather
than the RW bits since that's how the bits are derived.
Meanwhile, with some slight cleanup in the function.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1484556005-29701-2-git-send-email-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
With a vfio assigned device we lay down a base MemoryRegion registered
as an IO region, giving us read & write accessors. If the region
supports mmap, we lay down a higher priority sub-region MemoryRegion
on top of the base layer initialized as a RAM device pointer to the
mmap. Finally, if we have any quirks for the device (ie. address
ranges that need additional virtualization support), we put another IO
sub-region on top of the mmap MemoryRegion. When this is flattened,
we now potentially have sub-page mmap MemoryRegions exposed which
cannot be directly mapped through KVM.
This is as expected, but a subtle detail of this is that we end up
with two different access mechanisms through QEMU. If we disable the
mmap MemoryRegion, we make use of the IO MemoryRegion and service
accesses using pread and pwrite to the vfio device file descriptor.
If the mmap MemoryRegion is enabled and results in one of these
sub-page gaps, QEMU handles the access as RAM, using memcpy to the
mmap. Using either pread/pwrite or the mmap directly should be
correct, but using memcpy causes us problems. I expect that not only
does memcpy not necessarily honor the original width and alignment in
performing a copy, but it potentially also uses processor instructions
not intended for MMIO spaces. It turns out that this has been a
problem for Realtek NIC assignment, which has such a quirk that
creates a sub-page mmap MemoryRegion access.
To resolve this, we disable memory_access_is_direct() for ram_device
regions since QEMU assumes that it can use memcpy for those regions.
Instead we access through MemoryRegionOps, which replaces the memcpy
with simple de-references of standard sizes to the host memory.
With this patch we attempt to provide unrestricted access to the RAM
device, allowing byte through qword access as well as unaligned
access. The assumption here is that accesses initiated by the VM are
driven by a device specific driver, which knows the device
capabilities. If unaligned accesses are not supported by the device,
we don't want them to work in a VM by performing multiple aligned
accesses to compose the unaligned access. A down-side of this
philosophy is that the xp command from the monitor attempts to use
the largest available access weidth, unaware of the underlying
device. Using memcpy had this same restriction, but at least now an
operator can dump individual registers, even if blocks of device
memory may result in access widths beyond the capabilities of a
given device (RTL NICs only support up to dword).
Reported-by: Thorsten Kohfeldt <thorsten.kohfeldt@gmx.de>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Setting skip_dump on a MemoryRegion allows us to modify one specific
code path, but the restriction we're trying to address encompasses
more than that. If we have a RAM MemoryRegion backed by a physical
device, it not only restricts our ability to dump that region, but
also affects how we should manipulate it. Here we recognize that
MemoryRegions do not change to sometimes allow dumps and other times
not, so we replace setting the skip_dump flag with a new initializer
so that we know exactly the type of region to which we're applying
this behavior.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Avoid walking the FlatView of all address spaces. Most of the
address spaces will have no log_sync callback on their listeners.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
