When the TCG thread is started but not yet the machine, we wait in
qemu_tcg_cpu_thread_fn on tcg_halt_cond. To allow run_on_cpu already at
this time, we need to process pending request in that loop.

CC: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

CPUState::next_cpu is already CPUState *.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

As we have thread-local cpu_single_env now and KVM uses exactly one
thread per VCPU, we can drop the cpu_single_env updates from the loop
and initialize this variable only once during setup.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Call to kvm_cpu_synchronize_state() is missing.
kvm_arch_stop_on_emulation_error may look at outdated registers here.
Signed-off-by: NGleb Natapov <gleb@redhat.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>
Reviewed-by: NJan Kiszka <jan.kiszka@siemens.com>

* qemu-kvm/uq/master:
  apic: Fix legacy vmstate loading for KVM
  kvm: Implement kvm_irqchip_in_kernel like kvm_enabled
  kvm: Allow to set shadow MMU size

* kraxel/vnc.2:
  vnc: lift modifier keys on client disconnect.
  vnc: implement shared flag handling.
  vnc: fix ctrl key in vnc terminal emulation
  Fix vnc memory corruption with width = 1400

* kiszka/queues/slirp:
  slirp: Prevent sending ICMP error replies to source-only addresses
  slirp: Remove unused variable and unused code

* kraxel/usb.38: (28 commits)
  xhci: handle USB_RET_NAK
  xhci: remote wakeup support
  xhci: kill port arg from xhci_setup_packet
  xhci: stop on errors
  xhci: add trb type name lookup support.
  xhci: signal low- and fullspeed support
  usb: add USBBusOps->wakeup_endpoint
  usb: pass USBEndpoint to usb_wakeup
  usb: maintain async packet list per endpoint
  usb: Set USBEndpoint in usb_packet_setup().
  usb: add USBEndpoint->{nr,pid}
  usb: USBPacket: add status, rename owner -> ep
  usb: fold usb_generic_handle_packet into usb_handle_packet
  usb: kill handle_packet callback
  usb-xhci: switch to usb_find_device()
  usb-musb: switch to usb_find_device()
  usb-ohci: switch to usb_find_device()
  usb-ehci: switch to usb_find_device()
  usb-uhci: switch to usb_find_device()
  usb: handle dev == NULL in usb_handle_packet()
  ...

* kwolf/for-anthony:
  AHCI: Masking of IRQs actually masks them
  sheepdog: fix co_recv coroutine context
  AHCI: Fix port reset race
  rewrite QEMU_BUILD_BUG_ON
  qcow2: Keep unknown header extension when rewriting header
  qcow2: Update whole header at once
  vpc: Round up image size during fixed image creation
  vpc: Add support for Fixed Disk type
  iSCSI: add configuration variables for iSCSI
  qemu-io: add write -z option for bdrv_co_write_zeroes
  qed: add .bdrv_co_write_zeroes() support
  qed: replace is_write with flags field
  block: perform zero-detection during copy-on-read
  block: add .bdrv_co_write_zeroes() interface
  cutils: extract buffer_is_zero() from qemu-img.c

Otherwise we end up with a dangling reference which causes qdev_free() to fail.
Reported-by: NMichael Tsirkin <mst@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Replace device_init() with generalized type_init().

While at it, unify naming convention: type_init([$prefix_]register_types)
Also, type_init() is a function, so add preceding blank line where
necessary and don't put a semicolon after the closing brace.
Signed-off-by: NAndreas Färber <afaerber@suse.de>
Cc: Anthony Liguori <anthony@codemonkey.ws>
Cc: malc <av1474@comtv.ru>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

qdev_prop_set_* functions are always called by machine init functions
that should know what they're doing, so they abort on error.  Still,
an assert(!errp) does not aid debugging.  Print an error before aborting.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

SPARC and PPC set properties to NULL.  This can be done with an
empty string value.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Fix this error:
/src/qemu/target-ppc/helper.c: In function 'booke206_tlb_to_page_size':
/src/qemu/target-ppc/helper.c:1296:14: error: variable 'tlbncfg' set but not used [-Werror=unused-but-set-variable]
Tested-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

The conversion to memory regions broke lazy ROMD switching by forgetting
to update the rom_mode state variable.
Signed-off-by: NJan Kiszka <jan.kiszka@web.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Helpful to understand guest configurations of things like the i440FX's
PAM or the state of ROM devices.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

This fixes the regression introduced by cd7a45c9: We lost the or'ing
with the full_update flag.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Fix confusion in length calculation in commit
cd7a45c9.
Reported-by: NJan Kiszka <jan.kiszka@web.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Add a field to XHCITransfer to correctly keep track of NAK'ed usb
packets.  Retry transfers when the endpoint is kicked again.  Implement
wakeup_endpoint bus op so we can kick the endpoint when needed.

With this patch applied the emulated hid devices are working correctly
when hooked up to xhci.  usb-tabled without polling, yay!
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Unused argument, remove it.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

When some error happened we'll have to stop processing the endpoint.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

When logging TRBs add a the type name for more readable debug output.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add usb bus op which is called whenever a usb endpoint becomes ready,
so the host adapter emulation can react on that event.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Devices must specify which endpoint has data to transfer now.
The plan is to use the usb_wakeup() not only for remove wakeup support,
but for "data ready" signaling in general, so we can move away from
constant polling to event driven usb device emulation.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Maintain a list of async packets per endpoint.  With the current code
the list will never receive more than a single item.  I think you can
guess what the future plan is though ;)
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

With the separation of the device lookup (via usb_find_device) and
packet processing we can lookup device and endpoint before setting up
the usb packet.  So we can initialize USBPacket->ep early and keep it
valid for the whole lifecycle of the USBPacket.  Also the devaddr and
devep fields are not needed any more.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add a "nr" and "pid" fields to USBEndpoint so you can easily figure the
endpoint number and direction of any given endpoint.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add enum to track the status of USBPackets, use that instead of the
owner pointer to figure whenever a usb packet is currently in flight
or not.  Add some more packet status sanity checks.  Also rename the
USBEndpoint pointer from "owner" to "ep".
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

There is no reason to have a separate usb_generic_handle_packet function
any more, fold it into usb_handle_packet().  Also call the do_token_*
functions which handle control transfer emulation for control pipe
packets only.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

All drivers except usb-hub use usb_generic_handle_packet.  The only
reason the usb hub has its own function is that it used to be called
with packets which are intended for downstream devices.  With the new,
separate device lookup step this doesn't happen any more, so the need
for a different handle_packet callback is gone.

So we can kill the handle_packet callback and just call
usb_generic_handle_packet directly.  The special hub handling in
usb_handle_packet() can go away for the same reason.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Switch over xHCI to use the new usb_find_device()
function for device lookup.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Switch over musb to use the new usb_find_device()
function for device lookup.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Switch over OHCI to use the new usb_find_device()
function for device lookup.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Switch over EHCI to use the new usb_find_device()
function for device lookup.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Switch over UHCI to use the new usb_find_device()
function for device lookup.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Allow passing in a NULL pointer, return USB_RET_NODEV in that case.
Removes the burden to to a NULL pointer check from the callers.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Implement the find_device callback for the usb hub.  It'll loop over all
ports, calling usb_find_device for all enabled ports until it finds a
matching device.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add usb_find_device().  This function will check whenever a device with
a specific address is connected to the specified port.  Usually this
will just check state and address of the device hooked up to the port,
but in case of a hub it will ask the hub to check all hub ports for a
matching device.

This patch doesn't put the code into use yet, see the following patches
for details.

The master plan is to separate device lookup and packet processing.
Right now the usb code simply walks all devices, calls
usb_handle_packet() on each until one accepts the packet (by returning
something different that USB_RET_NODEV).  I want to have a device lookup
first, then call usb_handle_packet() once, for the device which actually
processes the packet.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>