Currently the only use of the periodic timer tick in kvmtool is to
handle reading from stdin. Though functional, this periodic tick can be
problematic on slow (eg FPGA) platforms and can cause low interactivity or
even stop the execution from progressing at all.
This patch removes the periodic tick in favour of a dedicated thread blocked
waiting for input from the console. In order to reflect the new behaviour,
the old 'kvm__arch_periodic_tick' function is renamed to 'kvm__arch_read_term'.
In making this change it is necessary to actively flush the emulated serial
console's output buffer after the guest writes to it, as otherwise flushing
only happens with terminal input. Similarly, it is no longer necessary to
flush the buffer when we process input.
Signed-off-by: Jonathan Austin <jonathan.austin@arm.com>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
A recent fix to virtio MMIO (72a7541ce305 ["kvm tools: virtio-mmio:
init_ioeventfd should use MMIO for ioeventfd__add_event()"]) highlighted
the confusing parameters expected by ioeventfd__add_event.
As per Pekka's suggestion, replace the bool parameters to this function
with a single `flags' argument instead.
Cc: Ying-Shiuan Pan <yingshiuan.pan@gmail.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
When fa7226f (kvm tools: init network devices only when the virtio
driver is ready to go) was introduced, a tiny detail was overlooked:
- Initialization of the uip layer is now coming in very late (only
when the guest driver says it is ready).
- In parallel, the rx thread is created quite early (as soon as the
queues are allocated).
This cause the rx thread to call uip_rx, which calls uip_buf_get_used,
which starts to use buf_lock mutex/the buf_used_cond, which haven't
been initialized yet. Tears and devastation follow, not to mention a
certain lack of network connectivity for the unsuspecting guest.
The (not so pretty) fix is to split uip_init:
- uip_static_init: initialize the lists, mutexes and conditions,
called from virtio_net__init_one.
- uip_init: perform the dynamic memory allocations, called from
notify_status.
This allows the network to be safely initialized.
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Asias writes:
In include/kvm/virtio-scsi.h, we have
/* TODO: Remove this when tcm_vhost goes upstream */
We are redefining VHOST_SCSI_GET_ABI_VERSION and friends because
tcm_vhost was not upstream at that time. Pekka, I guess we need to
upgrade kernel version in lkvm repo and drop our own tcm_vhost
defines.
Ingo writes:
I can confirm that the patch below fixes the build problem.
Analyzed-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Commit "kvm tools: virtio: remove hardcoded assumptions
about guest page size" has introduced a bug that prevented
guests with more than 4gb of ram from booting.
The issue is that 'pfn' is a 32bit integer, so when multiplying
it by page size to get the actual page will cause an overflow if
the pfn referred to a memory area above 4gb.
Acked-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Support mergable rx buffers for virtio-net. This helps reduce the amount
of memory the guest kernel has to allocate per rx vq.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
They're hidden inside net/core/iovec.c. It'd be nice to just link to that
but they're not too generic and come with tons of net/ specific code we
don't want. So we just copy over the relevant parts.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Some devices want to know their status, use this hook to allow them to
get that notification.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
We want to make the size of the virtio net header opaque to
uip.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Devices in the ioport space may have fdt bindings (e.g. 8250, rtc) and
on architectures where a fixed ioport is not defined, it is necessary
to generate fdt nodes for them.
This patch creates a new virtual bus (tree of device_headers) for ioport
devices, allowing architecture code to callback to the emulation code
when generating fdt nodes, as we do for virtio-mmio devices.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
If an architecture other than x86 wants to make use of ioport devices, the
interrupt lines will likely need remapping from their fixed values.
This patch allows an architecture callback, ioport__map_irq, to map
interrupts as appropriate.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
This patch makes the maximum amount of vqs configurable. To use it pass a 'mq'
option to network device configuration. For example:
vm run -n mode=tap,mq=4
Will allow up to 4 queue pairs for that network device.
Note that not specifiying mq, or setting mq=0 will disable virtio-net
multiqueuing.
Acked-by: Will Deacon <will.deacon@arm.com>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
It's barely usable but it isn't getting any better sitting alone in a
private git branch.
You can start a new VM with the GTK UI like this:
./vm run --gtk
It's rough around the edges:
- Red and blue color channels are inverted.
- Some keys do not work.
- Mouse does not work.
- GTK assertion failure pops up on shutdown.
but I'm sure there's someone out there that's just dying to improve the
user experience.
Cc: Asias He <asias.hejun@gmail.com>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Commit 4d789d4a2050 ("kvm tools: Increase amount of possible interrupts
per PCI device") increased the maximum amount of virtio queues for the
PCI transport, but neglected to do the same for MMIO.
This patch makes the same change for virtio-mmio.
Cc: Sasha Levin <sasha.levin@oracle.com>
Reported-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
We've limited the amount of interrupts per PCI device to be the maximum amount of VQs.
Now that the maximum amount of VQs has increased, time to increase this as well.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
1. print mem debug info into debugfd instead guest console
2. always print page table info
Signed-off-by: Asias He <asias.hejun@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
virtio-based PCI devices deal only with 4k memory granules, making
direct use of the VIRTIO_PCI_VRING_ALIGN and VIRTIO_PCI_QUEUE_ADDR_SHIFT
constants when initialising the virtqueues for a device.
For MMIO-based devices, the guest page size is arbitrary and may differ
from that of the host (this is the case on AArch64, where both 4k and
64k pages are supported).
This patch fixes the virtio drivers to honour the guest page size passed
when configuring the virtio device and align the virtqueues accordingly.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Since nothing depends on the max_high field values anymore, we can just
remove the field and the code that was used to maintain it.
Signed-off-by: Michel Lespinasse <walken@google.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Currently, only x86 has architecture command-line options (for setting
the BIOS video mode) however this is likely to become more common in the
future.
This patch adds some simple macros and a struct definition to allow
architectures to augment the command-line options with private
definitions. The BIOS video mode option (--vidmode) is also migrated to
the new framework.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
When generating a device tree for a guest, it is useful to have a helper
for converting host addresses to guest addresses in order to populate
the device nodes correctly.
This patch adds such a helper, following a similar implementation to the
reverse translation function that already exists.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Running a guest with multiple banks of memory based above 0 causes the
guest_flat_to_host address conversion to fail, as it is assumed that
guest memory addresses are offset linearly from 0.
This patch changes the translation function so that the kvm_mem_bank
structures registered by kvm__register_mem are used to translate guest
addresses, rather than use an offset from the start of host memory.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
When registering memory banks for a guest, it is useful to keep the
range information around for translating between guest and host address
spaces.
This patch adds a list of kvm_mem_bank structures to struct kvm, which
is updated when a new bank is registered.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
PCI devices are currently registered into the pci_devices array via the
pci__register function, which can then be indexed later by architecture
code to construct device tree nodes. For MMIO devices, there is no such
utility.
Rather than invent a similar mechanism for MMIO, this patch creates a
global device registration mechanism, which allows the device type to be
specified when registered or indexing a device. Current users of the pci
registration code are migrated to the new infrastructure and virtio MMIO
devices are registered at init time.
As part of the device registration, allocation of the device number is
moved out of irq__register_device and performed when adding the device
header to the relevant bus tree, allowing us to maintain separate device
numberspaces for each bus.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
linux/compiler.h will never give us a definition for __compiler_offsetof
because __KERNEL__ isn't defined, so just use the simple definition that
we have already.
This patch removes the redundant code.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
The _FDT macro is useful when generating device trees for a guest, so
make it available to other architectures.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
I've noticed message on kvm exit:
Warning: serial8250__exit failed.
kvm tool is not able to remove ioport range which was added previously.
The issue is caused by bug in rbtree-interval. Search algorithm in
rb_int_search_single() expects correct value of max_high. But the tree
can contain leaf nodes, which never were updated by propagate_callback().
For this kind of nodes high_max will be 0 and we will not be able to
find and remove them.
Let's initialize max_high on RB_INT_INIT() time.
Fixing this bug makes other bug visible: propagate_callback() can be
called for empty tree: node == NULL. The callback is not ready for empty
tree. Let's fix that as well.
Tested-by: William Dauchy <william@gandi.net>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
We already have something to wrap pthread with mutex_[init,lock,unlock]
calls. This patch creates a new struct mutex abstraction and moves
everything to work with it.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Give threads a meaningful name. This makes debugging much easier, and
everything else much prettier.
Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
[ penberg@kernel.org: specify vcpu names ]
Signed-off-by: Pekka Enberg <penberg@kernel.org>
There were a number of changes that caused problems:
- UAPI conversion
- the old version of the rbtree-augmented API got removed, I
updated the implementation to the new API
- lib/rbtree.c wants a 'true' definition
Lightly tested: it boots a v3.7-rc1 defconfig+kvmconfig kernel.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
If SDL isn't compiled in we shouldn't die unless we actually try using it.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
We used wrong prototypes for sdl init/exit when libsdl wasn't installed when
building. This would cause build errors.
Reported-by: Kashyap Chamarthy <kashyap.cv@gmail.com>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Switch to using init/exit calls instead of the repeating call blocks in builtin-run.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
This adds a method to call init/exit functions similar to the kernel's init functions.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Sort out init/exit calls, move parser into the 9p code and make sure
rootfs config is initialized before virtio-9p (or any other init func)
is called.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Sort out the config initialization order so that configuration is fully initialized
before init functions start running, and move the firmware initialization code into
kvm.c.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Move the vesa initialization logic into sdl__init() and vnc__init(), builtin-run
shouldn't have to know about the conditions for initializing vesa on it's own.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Check if i8042 is supported only within the initialization call itself, so that
builtin-run won't need to know which archs are supported by it.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Move all the kvm-ipc specific code into the relevant file, and modify
the ipc callback to pass a ptr to struct kvm as well.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init/exit of virtio-net self-contained, so the global init code
won't need to check if it was selected or not.
This also moves the bulk of the net-specific initialization code, including
the parser, into virtio-net itself.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init/exit of pci-shmem self-contained, so the global init code
won't need to check if it was selected or not.
Also move the parser out of builtin-run into the pci-shmem code.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init/exit of virtio-balloon self-contained, so the global init code
won't need to check if it was selected or not.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init/exit of virtio-console self-contained, so the global init code
won't need to check if it was selected or not.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the timer init/exit follow the rest of the code, and move it out of
builtin-run.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Add an exit function for the threadpool which will stop all running threads in the
pool. Also clean up the init code a bit.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init and exit functions of the term code similar to the rest
of the code.
Also move in the pty parser into the term code out of builtin-run.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Make the init and exit functions of the framebuffer similar to the rest
of the code.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
There's no reason the array of guest specific vcpus is global. Move it into
struct kvm.
Also split up arch specific vcpu init from the generic code and call it from
the kvm_cpu initializer.
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Pekka Enberg <penberg@kernel.org>
Jonathan Austin