Right now, downscript is not invoked reliably.  If you execute 'quit' from the
monitor, it won't be invoked.

This fixes that by converting tap to use an exit_notifier to execute the
downscript.  In this case, allowing an exit notifier to include state is
critically important for the conversion.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

All of these users have global state so we really don't see a benefit from
exit_notifier.  However, using exit_notifier means that there's one less
justification for having global state in the first place.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Like atexit() but with state
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Today we poll the mouse mode whenever there is a mouse movement.  There is a
subtle usability problem with this though.

If we're in relative mode and grab is enabled, when we change to absolute mode,
we break grab.  This gives a user a seamless transition when the new pointer
is enabled.

But because we poll for mouse change, this grab break won't occur until the user
attempts to move the mouse.  By using notifiers, the grab break happens as soon
as possible.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

When we switch to absolute mode, we send out a notification (if the client
supports it).  Today, we only send this notification when the client sends us
a mouse event and we're in the wrong mode.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

For QMP, we just add an attribute which is backwards compatible.  For the human
monitor, we add (absolute) to the end of the line.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Right now, DisplayState clients rely on polling the mouse mode to determine
when the device is changed to an absolute device.  Use a notification list to
add an explicit notification.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

kbd_mouse_is_absolute tells us whether the current mouse handler is an absolute
device.  kbd_mouse_has_absolute tells us whether we have any device that is
capable of absolute input.

This lets us tell a user that they have configured an absolute device but that
the guest is not currently using it.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

And convert usb-hid to use it (to avoid regression with bisection)

Right now, when we do info mice and we've added a usb tablet, we don't see it
until the guest starts using the tablet.  We implement this behavior in order
to provide a means to delay registration of a mouse handler since we treat
the last registered handler as the current handler.

This is a usability problem though as we would like to give the user feedback
that they've either 1) not added an absolute device 2) there is an absolute
device but the guest isn't using it 3) we have an absolute device and it's
active.

By using QTAILQ and having an explicit activation function that moves the
handler to the front of the queue, we can implement the same semantics as
before with respect to automatically switching to usb tablet while providing
the user with a whole lot more information.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Notifiers are data-less callbacks and a notifier list is a list of registered
notifiers that all are interested in a particular event.

We'll use this in a few patches to implement mouse change notification.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>
---
v1 -> v2
 - Do not do memory allocations by placing list nodes in notifier

This reverts commit 3c9c706c.

This breaks build (gcc 4.3.2):
    CC    usb-linux.o
 cc1: warnings being treated as errors
 /src/qemu/usb-linux.c: In function 'usb_linux_update_endp_table':
 /src/qemu/usb-linux.c:759: error: 'type' may be used uninitialized in
 this function
Reported-by: NBlue Swirl <blauwirbel@gmail.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

It allways returned true, that is the equivalent of not having the
callback.
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

It was always only used for reads
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

values are already pointers, no need to cast them to void *
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NAvi Kivity <avi@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

qemu-option.o(.text+0x20f8): In function `qemu_opts_from_qdict_1':
/src/qemu/qemu-option.c:813: warning: strcpy() is almost always misused, please use strlcpy()
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

When building with -DNDEBUG, assert(0) will not stop execution
so it must not be used for abnormal termination.

Use cpu_abort() when in CPU context, abort() otherwise.
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Split long unbreakable lines to smaller sections.

Spotted by Debian Lintian tool:
http://lintian.debian.org/full/pkg-qemu-devel@lists.alioth.debian.org.html#qemuSigned-off-by: NBlue Swirl <blauwirbel@gmail.com>

The usb-msd device emulation needs some small tweaks in the requests
emulations. For instance, the reset/maxlun requests are class/interface
specific so requests for them with the type class and recipient interface
bits sets have to be handled.
Signed-off-by: NArnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

In case s->version is shorter than 4 bytes we overflow the memcpy src
buffer.  Fix it by clearing the target buffer, then copy only the
amount of bytes we actually have.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Add an option to disable the heuristics which try to keep
capslock and numlock state for guest and host in sync.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This was the only incoming migration without autostart check
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Not clearing the fd and closing the file makes qemu spin using 100%CPU
after incoming migration error.

See for instance bug:
https://bugzilla.redhat.com/show_bug.cgi?id=518032Signed-off-by: NJuan Quintela <quintela@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Don't rely on CDROM hint for read_only attribute
Signed-off-by: NNaphtali Sprei <nsprei@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Really use read-only flags for opening the file when asked for read-only
Signed-off-by: NNaphtali Sprei <nsprei@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Just tell main_loop_wait whether to be blocking or nonblocking, so that
there is no need to call qemu_cpus_have_work from the timer subsystem.
Instead, tcg_cpu_exec can say "we want the main loop not to block because
we have stuff to do".
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Tweaking the rounding in qemu_next_deadline ensures that there's
no change whatsoever.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

A simple patch to place together all handling of -icount.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

By adding the possibility to turn on/off a clock, yet another
incestuous relationship between timers and CPUs can be disentangled.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Make the timer subsystem register its own callback instead.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Instead of testing specially next_cpu in host_alarm_handler, just do
that in qemu_notify_event.  The idea is, if we are not running (or
not yet running) target CPU code, prepare things so that the execution
loop is exited asap; just make that clear.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>