qemu_rearm_alarm_timer partially duplicates the code in
qemu_next_alarm_deadline to figure out if it needs to rearm the timer.
If it calls qemu_next_alarm_deadline, it always rearms the timer even if
the next deadline is INT64_MAX.

This patch simplifies the behavior of qemu_rearm_alarm_timer and removes
the duplicated code, always calling qemu_next_alarm_deadline and only
rearming the timer if the deadline is less than INT64_MAX.
Signed-off-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: NStefan Weil <sw@weilnetz.de>
Tested-by: NAndreas Färber <andreas.faerber@web.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

For command line options which permit '?' meaning 'please list the
permitted values', add support for 'help' as a synonym, by abstracting
the check out into a helper function.

This change means that in some cases where we were being lazy in
our string parsing, "?junk" will now be rejected as an invalid option
rather than being (undocumentedly) treated the same way as "?".

Update the documentation to use 'help' rather than '?', since '?'
is a shell metacharacter and thus prone to fail confusingly if there
is a single character filename in the current working directory and
the '?' has not been escaped. It's therefore better to steer users
towards 'help', though '?' is retained for backwards compatibility.

We do not, however, update the output of the system emulator's -help
(or any documentation autogenerated from the qemu-options.hx which
is the source of the -help text) because libvirt parses our -help
output and will break. At a later date when QEMU provides a better
interface so libvirt can avoid having to do this, we can update the
-help text too.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Function timeSetEvent returns 0 when it fails, but it does not set
an error code which can be retrieved by GetLastError.

Therefore calling GetLastError is useless.
Signed-off-by: NStefan Weil <sw@weilnetz.de>

sys/param.h is needed for __FreeBSD_version.
Pointed out by Juergen, thanks.
Signed-off-by: NAndreas Faerber <andreas.faerber@web.de>
Cc: Juergen Lock <nox@jelal.kn-bremen.de>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

timeSetEvent only accepts delays in the range which is returned by
timeGetDevCaps.

The lower limit is typically 1 (= 1 ms), so the constant value of 1
in the old code usually worked.

The upper limit can be as low as 10000 ms, so the latest changes in
QEMU's timer handling which introduced timeout values above that limit
could result in failures of timeSetEvent when the timer was re-armed.
Signed-off-by: NStefan Weil <sw@weilnetz.de>

Remove all holes which were found by pahole on Linux x86_64
(and replace "struct QEMUTimer" by "QEMUTimer").
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <sw@weilnetz.de>

Some time ago, the last time which did not have dynticks was removed,
so now all timers have dynticks.

I also removed a misleading error message for the dynticks timer.
If timer_create fails, there is already an error message, and
QEMU will use the unix timer which also provides dynamic ticks,
therefore dynamic ticks are not disabled.

v2:
Remove two if statements because they were always true
(thanks to Paolo Bonzini for this correction).
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <sw@weilnetz.de>

This avoids conversions between int and bool / char.

It also makes the code more readable.
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <sw@weilnetz.de>

The last user of this function was removed by commit
12d4536f.
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <sw@weilnetz.de>

qemu-timer.h includes qemu-common.h which already includes time.h,
sys/time.h, windows.h, unistd.h, fcntl.h, errno.h and signal.h.

Therefore those include statements are redundant and can be removed.
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <sw@weilnetz.de>

- remove qemu_calculate_timeout;

- explicitly size timeout to uint32_t;

- introduce slirp_update_timeout;

- pass NULL as timeout argument to select in case timeout is the maximum
value;
Signed-off-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: NPaul Brook <paul@codesourcery.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Also delta in qemu_next_alarm_deadline is a 64 bit value so set the
default to INT64_MAX instead of INT32_MAX.
Signed-off-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Fix win32_rearm_timer and mm_rearm_timer: they should be able to handle
INT64_MAX as a delta parameter without overflowing.
Also, the next deadline in ms should be calculated rounding down rather
than up (see unix_rearm_timer and dynticks_rearm_timer).

Finally ChangeTimerQueueTimer takes an unsigned long and timeSetEvent
takes an unsigned int as delta, so cast the ms delta to the appropriate
unsigned integer.
Signed-off-by: NStefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Basically, the main wait loop calls qemu_run_all_timers() unconditionally. The
first thing this routine used to do is to see if a timer had been serviced,
and then reset the loop timeout to the next deadline.

However, the new deadlines had not been calculated at that point, as
qemu_run_timers() had not been called yet for each of the clocks. So
qemu_rearm_alarm_timer() would end up with a negative or zero deadline, and
default to setting a 250us timeout for the loop.

As qemu_run_timers() is called for each clock, the real deadlines would be put
in place, but because a loop timeout was already set, the loop timeout would
not be changed.

Once that 250us timeout fired, the real deadline would be used for the
subsequent timeout.

For idle VMs, this effectively doubles the number of times through the loop,
doubling the number of select() system calls, timer calls, etc. putting added
scheduling pressure on the kernel. And under cgroups, this really causes a big
problem because the cgroup code does not scale well.

By simply running the timers before trying to rearm the timer, we always rearm
with a non-zero deadline, effectively halving the number of system calls.
Signed-off-by: NPeter Portante <pportant@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch combines qtest and -icount together to turn the vm_clock
into a source that can be fully managed by the client.  To this end new
commands clock_step and clock_set are added.  Hooking them with libqtest
is left as an exercise to the reader.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Notifiers do not need to access both ends of the list, and using
a QLIST also simplifies the API.
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: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>

The non-dynticks timer variations are broken, so they can be
removed.
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: NPaolo Bonzini <pbonzini@redhat.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

None of this is needed by tools, and most of it can even be made static
inside cpus.c.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

These will be used when moving icount accounting to cpus.c.
Reviewed-by: NAnthony Liguori <aliguori@us.ibm.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

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

Embed the list in the QEMUClock instead.
Reviewed-by: NAnthony Liguori <aliguori@us.ibm.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Use runstate_is_running() instead, which is introduced by this commit.
Signed-off-by: NLuiz Capitulino <lcapitulino@redhat.com>

Today, when notifying a VM state change with vm_state_notify(),
we pass a VMSTOP macro as the 'reason' argument. This is not ideal
because the VMSTOP macros tell why qemu stopped and not exactly
what the current VM state is.

One example to demonstrate this problem is that vm_start() calls
vm_state_notify() with reason=0, which turns out to be VMSTOP_USER.

This commit fixes that by replacing the VMSTOP macros with a proper
state type called RunState.
Signed-off-by: NLuiz Capitulino <lcapitulino@redhat.com>

Enabling the I/O thread by default seems like an important part of declaring
1.0.  Besides allowing true SMP support with KVM, the I/O thread means that the
TCG VCPU doesn't have to multiplex itself with the I/O dispatch routines which
currently requires a (racey) signal based alarm system.

I know there have been concerns about performance.  I think so far the ones that
have come up (virtio-net) are most likely due to secondary reasons like
decreased batching.

I think we ought to force enabling I/O thread early in 1.0 development and
commit to resolving any lingering issues.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

qemu_malloc/qemu_free no longer exist after this commit.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Derived from kvm-tool patch
http://thread.gmane.org/gmane.comp.emulators.kvm.devel/74309

Ingo Molnar pointed out that sending the timer signal to the whole
process, just blocking it everywhere, is suboptimal with an increasing
number of threads. QEMU is also using this pattern so far.

Linux provides a (non-portable) way to restrict the signal to a single
thread: We can use SIGEV_THREAD_ID unless we are forced to emulate
signalfd via an additional thread. That case could theoretically be
optimized as well, but it doesn't look worth bothering.
Reviewed-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

QEMU_CLOCK_HOST is based on the system time which may jump backward in
case the admin or NTP adjusts it. RTC emulations and other device models
can suffer in this case as timers will stall for the period the clock
was tuned back.

This adds a detection mechanism that checks on every host clock readout
if the new time is before the last result. If that is the case a
notifier list is informed. Device models interested in this event can
register a notifier with the clock.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

A timer that wakes up every millisecond puts a lot of stress on the
iothread.  The large amount of IPIs causes very high context switch
activity, making emulation slow and the UI unusable.  This is by the
way the same reason why the Windows timers were switched to dynticks.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Tested-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

dynticks will provide equally good timer granularity on all modern Linux
systems.  This is more or less dead code these days.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Commit 68c23e55 removed the
multimedia timer, but this timer is needed for certain
Linux kernels. Otherwise Linux boot stops with this error:

    MP-BIOS bug: 8254 timer not connected to IO-APIC

So the multimedia timer is added again here.

Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NStefan Weil <weil@mail.berlios.de>

The type casts are no longer needed after some small changes
in struct qemu_alarm_timer. This also improves readability
of the code.
Signed-off-by: NStefan Weil <weil@mail.berlios.de>

This simply moves code which is used three times
into a new function thus improving readability.
Signed-off-by: NStefan Weil <weil@mail.berlios.de>

It is purely for icount-based virtual timers.  And now that we got the
code right, rename the function to clarify the intended scope.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Tested-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>

This reverts commits 225d02cd and c9f7383c.  While some parts of
the latter could be saved, I preferred a smooth, complete revert.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Tested-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>

The previous patch however is not enough, because if the virtual CPU
goes to sleep waiting for a future timer interrupt to wake it up, qemu
deadlocks.  The timer interrupt never comes because time is driven by
icount, but the vCPU doesn't run any insns.

You could say that VCPUs should never go to sleep in icount
mode if there is a pending vm_clock timer; rather time should
just warp to the next vm_clock event with no sleep ever taking place.
Even better, you can sleep for some time related to the
time left until the next event, to avoid that the warps are too visible
externally; for example, you could be sending network packets continously
instead of every 100ms.

This is what this patch implements.  qemu_clock_warp is called: 1)
whenever a vm_clock timer is adjusted, to ensure the warp_timer is
synchronized; 2) at strategic points in the CPU thread, to make sure
the insn counter is synchronized before the CPU starts running.
In any case, the warp_timer is disabled while the CPU is running,
because the insn counter will then be making progress on its own.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Tested-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>