[Qemu-devel] [PATCH v17 00/21] Deterministic replay core

[Qemu-devel] [PATCH v17 00/21] Deterministic replay core

[Pavel Dovgalyuk]

This set of patches is related to the reverse execution and deterministic 
replay of qemu execution. This implementation of deterministic replay can 
be used for deterministic debugging of guest code through gdb remote
interface.

Core set of patches does not include support for reverse debugging commands
of gdb, block devices' operations, USB replay support.

These patches include only core function of the replay,
excluding the support for replaying serial, audio, network, and USB devices'
operations. Reverse debugging and monitor commands were also excluded to
be submitted later as separate patches.

Execution recording writes non-deterministic events log, which can be later 
used for replaying the execution anywhere and for unlimited number of times. 
It also supports checkpointing for faster rewinding during reverse debugging. 
Execution replaying reads the log and replays all non-deterministic events 
including external input, hardware clocks, and interrupts.

Full version of deterministic replay has the following features:
 * Deterministically replays whole system execution and all contents of the memory,
   state of the hadrware devices, clocks, and screen of the VM.
 * Writes execution log into the file for latter replaying for multiple times 
   on different machines.
 * Supports i386, x86_64, ARM, PowerPC, and MIPS hardware platforms.
 * Performs deterministic replay of all operations with keyboard and mouse
   input devices.
 * Supports auto-checkpointing for convenient reverse debugging.

Usage of the record/replay core:
 * First, record the execution, by adding the following string to the command line:
   '-icount shift=7,rr=record,rrfile=replay.bin -net none'. 
   Block devices' images are not actually changed in the recording mode, 
   because all of the changes are written to the temporary overlay file.
 * Then you can replay it for the multiple times by using another command
   line option: '-icount shift=7,rr=replay,rrfile=replay.bin -net none'
 * '-net none' option should also be specified if network replay patches
   are not applied.
 * Do not add any disk images to VM, because they are not supported by
   the core patches.

Papers with description of deterministic replay implementation:
http://www.computer.org/csdl/proceedings/csmr/2012/4666/00/4666a553-abs.html
http://dl.acm.org/citation.cfm?id=2786805.2803179

Public repository with current version of the patches:
https://github.com/Dovgalyuk/qemu/tree/rr-17

Modifications of qemu include:
 * wrappers for clock and time functions to save their return values in the log
 * saving different asynchronous events (e.g. system shutdown) into the log
 * synchronization of the threads from thread pool
 * recording/replaying user input (mouse and keyboard)
 * adding internal events for cpu and io synchronization

v17 changes:
 * Removed useless stub functions (as suggested by Paolo Bonzini)
 * Refined checkpoint-related code (as suggested by Paolo Bonzini)
 * Improved icount processing (as suggested by Paolo Bonzini)
 * Added checkpoint for suspend event (as suggested by Paolo Bonzini)
 * Fixed linux-user configurations build
 * Minor fixes

v16 changes:
 * Several warnings were fixed

v15 changes:
 * Tested record/replay with MIPS and PowerPC guests
 * Published the patches on github
 * Fixed replay mutex operation in icount mode
 * Fixed timers processing in record/replay mode

v14 changes:
 * Minor fixes

v13 changes:
 * Introduced "ptimer trigger" event (as suggested by Paolo Bonzini)

v12 changes:
 * Removed block patches from the core patches set.

v11 changes:
 * Fixed instructions event processing.
 * Added some mutex protection calls for replay.
 * Fixed replaying read operations for qcow2.
 * Fixed rtc reads on initializations stage.
 * Eliminated some warnings in replay module.
 * Fixed misprints in documentation for replay (as suggested by Eric Blake)

v10 changes:
 * Fixed queue processing for bottom halves (as suggested by Paolo Bonzini)
 * Rewritten several replay functions (as suggested by Paolo Bonzini)
 * Some minor fixes.

v9 changes:
 * Replaced fwrite/fread with putc/getc (as suggested by Paolo Bonzini)
 * Stopping virtual machine in case of replay file end (as suggested by Paolo Bonzini)
 * Removed one of the replay mutexes (as suggested by Paolo Bonzini)
 * Fixed RCU queue for bottom halves (as suggested by Paolo Bonzini)
 * Updated command line options' names (as suggested by Paolo Bonzini)
 * Added design document for record/replay (as suggested by Paolo Bonzini)
 * Simplified checkpoints for the timers
 * Added cloning InputEvent objects for replay (as suggested by Paolo Bonzini)
 * Added replay blockers instead of checking the command line (as suggested by Paolo Bonzini)
 * Some functions renaming and extracting.

v8 changes:
 * Simplified processing of the shutdown event (as suggested by Paolo Bonzini)
 * Replaced stack of bottom halves in AIO context with QSIMPLEQ (as suggested by Paolo Bonzini)
 * Moved replay_submode out of the series (as suggested by Paolo Bonzini)
 * Moved suffix option out of the series
 * Converted some of the defines into enums (as suggested by Paolo Bonzini)
 * Encapsulated save_tm/read_tm calls into the single function (as suggested by Paolo Bonzini)
 * Moved record/replay options to icount group (as suggested by Paolo Bonzini)
 * Updated mutex protection for the events queue (as suggested by Paolo Bonzini)
 * Added mutex to protect replay log file (as suggested by Paolo Bonzini)
 * Minor cleanups

v7 changes:
 * Removed patches that were applied to upstream.

v6 changes:
 * Fixed replay stub return value (as suggested by Eric Blake)
 * Fixed icount warping.
 * Virtual rt clock now uses cpu_get_clock() (as suggested by Paolo Bonzini)
 * Replated get_clock_realtime and get_clock calls with qemu clock requests (as suggested by Paolo Bonzini)
 * Modified can_do_io logic to allow requesting icount from cpu_exec function (as suggested by Paolo Bonzini)
 * Removed applied patches.

v5 changes:
 * Minor changes.
 * Used fixed-width integer types for read/write functions (as suggested by Alex Bennee)
 * Moved savevm-related code out of the core.
 * Added new traced clock for deterministic virtual clock warping (as suggested by Paolo Bonzini)
 * Fixed exception_index reset for user mode (as suggested by Paolo Bonzini)
 * Adopted Paolo's icount patches
 * Fixed hardware interrupts replaying

v4 changes:
 * Updated block drivers to support new bdrv_open interface.
 * Moved migration patches into separate series (as suggested by Paolo Bonzini)
 * Fixed a bug in replay_break operation.
 * Fixed rtl8139 migration for replay.
 * Fixed 'period' parameter processing for record mode.
 * Fixed bug in 'reverse-stepi' implementation.
 * Fixed replay without making any snapshots (even the starting one).
 * Moved core replay patches into the separate series.
 * Fixed reverse step and reverse continue support.
 * Fixed several bugs in icount subsystem.
 * Reusing native qemu icount for replay instructions counting.
 * Separated core patches into their own series.

v3 changes:
 * Fixed bug with replay of the aio write operations.
 * Added virtual clock based on replay icount.
 * Removed duplicated saving of interrupt_request CPU field.
 * Fixed some coding style issues.
 * Renamed QMP commands for controlling reverse execution (as suggested by Eric Blake)
 * Replay mode and submode implemented as QAPI enumerations (as suggested by Eric Blake)
 * Added description and example for replay-info command (as suggested by Eric Blake)
 * Added information about the current breakpoint to the output of replay-info (as suggested by Eric Blake)
 * Updated version id for HPET vmstate (as suggested by Paolo Bonzini)
 * Removed static fields from parallel vmstate (as suggested by Paolo Bonzini)
 * New vmstate fields for mc146818rtc, pckbd, kvmapic, serial, fdc, rtl8139 moved to subsection (as suggested by Paolo Bonzini)
 * Disabled textmode cursor blinking, when virtual machine is stopped (as suggested by Paolo Bonzini)
 * Extracted saving of exception_index to separate patch (as suggested by Paolo Bonzini)

v2 changes:
 * Patches are split to be reviewable and bisectable (as suggested by Kirill Batuzov)
 * Added QMP versions of replay commands (as suggested by Eric Blake)
 * Removed some optional features of replay to make patches cleaner
 * Minor changes and code cleanup were made

---

Pavel Dovgalyuk (21):
      i386: partial revert of interrupt poll fix
      replay: global variables and function stubs
      replay: internal functions for replay log
      replay: introduce mutex to protect the replay log
      replay: introduce icount event
      cpu-exec: allow temporary disabling icount
      cpu: replay instructions sequence
      i386: interrupt poll processing
      replay: interrupts and exceptions
      replay: asynchronous events infrastructure
      replay: recording and replaying clock ticks
      replay: shutdown event
      icount: improve counting for record/replay
      replay: checkpoints
      bottom halves: introduce bh call function
      replay: ptimer
      typedef: add typedef for QemuOpts
      replay: initialization and deinitialization
      replay: replay blockers for devices
      replay: command line options
      replay: recording of the user input


 Makefile.target           |    1 
 async.c                   |    7 +
 cpu-exec.c                |   63 +++++++-
 cpus.c                    |   57 +++++---
 docs/replay.txt           |  168 ++++++++++++++++++++++
 exec.c                    |    2 
 hw/bt/hci.c               |    7 +
 hw/core/ptimer.c          |    7 +
 include/block/aio.h       |    5 +
 include/exec/exec-all.h   |    1 
 include/qapi/qmp/qerror.h |    3 
 include/qemu/option.h     |    5 -
 include/qemu/typedefs.h   |    3 
 include/ui/input.h        |    2 
 main-loop.c               |    3 
 qapi-schema.json          |   18 ++
 qemu-options.hx           |    8 +
 qemu-timer.c              |   48 +++++-
 replay/Makefile.objs      |    6 +
 replay/replay-events.c    |  274 ++++++++++++++++++++++++++++++++++++
 replay/replay-input.c     |  160 +++++++++++++++++++++
 replay/replay-internal.c  |  206 +++++++++++++++++++++++++++
 replay/replay-internal.h  |  140 ++++++++++++++++++
 replay/replay-time.c      |   64 ++++++++
 replay/replay.c           |  342 +++++++++++++++++++++++++++++++++++++++++++++
 replay/replay.h           |  120 ++++++++++++++++
 stubs/Makefile.objs       |    1 
 stubs/replay.c            |   32 ++++
 target-i386/cpu.c         |   10 -
 target-i386/seg_helper.c  |    3 
 translate-all.c           |    2 
 ui/input.c                |   27 +++-
 vl.c                      |   53 ++++++-
 33 files changed, 1779 insertions(+), 69 deletions(-)
 create mode 100755 docs/replay.txt
 create mode 100755 replay/Makefile.objs
 create mode 100755 replay/replay-events.c
 create mode 100755 replay/replay-input.c
 create mode 100755 replay/replay-internal.c
 create mode 100755 replay/replay-internal.h
 create mode 100755 replay/replay-time.c
 create mode 100755 replay/replay.c
 create mode 100755 replay/replay.h
 create mode 100755 stubs/replay.c

-- 
Pavel Dovgalyuk

[Qemu-devel] [PATCH v17 01/21] i386: partial revert of interrupt poll fix

[Pavel Dovgalyuk]

Processing CPU_INTERRUPT_POLL requests in cpu_has_work functions
break the determinism of cpu_exec. This patch is required to make
interrupts processing deterministic.

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

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpu-exec.c        |    9 +++++++++
 target-i386/cpu.c |   10 ++--------
 2 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/cpu-exec.c b/cpu-exec.c
index 6b6942d..1f8046f 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -28,6 +28,9 @@
 #include "exec/memory-internal.h"
 #include "qemu/rcu.h"
 #include "exec/tb-hash.h"
+#if defined(TARGET_I386) && !defined(CONFIG_USER_ONLY)
+#include "hw/i386/apic.h"
+#endif
 
 /* -icount align implementation. */
 
@@ -369,6 +372,12 @@ int cpu_exec(CPUState *cpu)
     volatile bool have_tb_lock = false;
 
     if (cpu->halted) {
+#if defined(TARGET_I386) && !defined(CONFIG_USER_ONLY)
+        if (cpu->interrupt_request & CPU_INTERRUPT_POLL) {
+            apic_poll_irq(x86_cpu->apic_state);
+            cpu_reset_interrupt(cpu, CPU_INTERRUPT_POLL);
+        }
+#endif
         if (!cpu_has_work(cpu)) {
             return EXCP_HALTED;
         }
diff --git a/target-i386/cpu.c b/target-i386/cpu.c
index cfb8aa7..3d2e2f0 100644
--- a/target-i386/cpu.c
+++ b/target-i386/cpu.c
@@ -3098,14 +3098,8 @@ static bool x86_cpu_has_work(CPUState *cs)
     X86CPU *cpu = X86_CPU(cs);
     CPUX86State *env = &cpu->env;
 
-#if !defined(CONFIG_USER_ONLY)
-    if (cs->interrupt_request & CPU_INTERRUPT_POLL) {
-        apic_poll_irq(cpu->apic_state);
-        cpu_reset_interrupt(cs, CPU_INTERRUPT_POLL);
-    }
-#endif
-
-    return ((cs->interrupt_request & CPU_INTERRUPT_HARD) &&
+    return ((cs->interrupt_request & (CPU_INTERRUPT_HARD |
+                                      CPU_INTERRUPT_POLL)) &&
             (env->eflags & IF_MASK)) ||
            (cs->interrupt_request & (CPU_INTERRUPT_NMI |
                                      CPU_INTERRUPT_INIT |

[Qemu-devel] [PATCH v17 02/21] replay: global variables and function stubs

[Pavel Dovgalyuk]

This patch adds global variables, defines, function declarations,
and function stubs for deterministic VM replay used by external modules.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 Makefile.target      |    1 
 docs/replay.txt      |  168 ++++++++++++++++++++++++++++++++++++++++++++++++++
 qapi-schema.json     |   18 +++++
 replay/Makefile.objs |    2 +
 replay/replay.c      |   14 ++++
 replay/replay.h      |   19 ++++++
 stubs/Makefile.objs  |    1 
 stubs/replay.c       |    3 +
 8 files changed, 226 insertions(+), 0 deletions(-)
 create mode 100755 docs/replay.txt
 create mode 100755 replay/Makefile.objs
 create mode 100755 replay/replay.c
 create mode 100755 replay/replay.h
 create mode 100755 stubs/replay.c

diff --git a/Makefile.target b/Makefile.target
index 3e7aafd..e149ec9 100644
--- a/Makefile.target
+++ b/Makefile.target
@@ -86,6 +86,7 @@ all: $(PROGS) stap
 # cpu emulator library
 obj-y = exec.o translate-all.o cpu-exec.o
 obj-y += tcg/tcg.o tcg/tcg-op.o tcg/optimize.o
+obj-y += replay/
 obj-$(CONFIG_TCG_INTERPRETER) += tci.o
 obj-$(CONFIG_TCG_INTERPRETER) += disas/tci.o
 obj-y += fpu/softfloat.o
diff --git a/docs/replay.txt b/docs/replay.txt
new file mode 100755
index 0000000..645d462
--- /dev/null
+++ b/docs/replay.txt
@@ -0,0 +1,168 @@
+Copyright (c) 2010-2015 Institute for System Programming
+                        of the Russian Academy of Sciences.
+
+This work is licensed under the terms of the GNU GPL, version 2 or later.
+See the COPYING file in the top-level directory.
+
+Record/replay
+-------------
+
+Record/replay functions are used for the reverse execution and deterministic
+replay of qemu execution. This implementation of deterministic replay can
+be used for deterministic debugging of guest code through a gdb remote
+interface.
+
+Execution recording writes a non-deterministic events log, which can be later
+used for replaying the execution anywhere and for unlimited number of times.
+It also supports checkpointing for faster rewinding during reverse debugging.
+Execution replaying reads the log and replays all non-deterministic events
+including external input, hardware clocks, and interrupts.
+
+Deterministic replay has the following features:
+ * Deterministically replays whole system execution and all contents of
+   the memory, state of the hardware devices, clocks, and screen of the VM.
+ * Writes execution log into the file for later replaying for multiple times
+   on different machines.
+ * Supports i386, x86_64, and ARM hardware platforms.
+ * Performs deterministic replay of all operations with keyboard and mouse
+   input devices.
+
+Usage of the record/replay:
+ * First, record the execution, by adding the following arguments to the command line:
+   '-icount shift=7,rr=record,rrfile=replay.bin -net none'. 
+   Block devices' images are not actually changed in the recording mode,
+   because all of the changes are written to the temporary overlay file.
+ * Then you can replay it by using another command
+   line option: '-icount shift=7,rr=replay,rrfile=replay.bin -net none'
+ * '-net none' option should also be specified if network replay patches
+   are not applied.
+
+Papers with description of deterministic replay implementation:
+http://www.computer.org/csdl/proceedings/csmr/2012/4666/00/4666a553-abs.html
+http://dl.acm.org/citation.cfm?id=2786805.2803179
+
+Modifications of qemu include:
+ * wrappers for clock and time functions to save their return values in the log
+ * saving different asynchronous events (e.g. system shutdown) into the log
+ * synchronization of the bottom halves execution
+ * synchronization of the threads from thread pool
+ * recording/replaying user input (mouse and keyboard)
+ * adding internal checkpoints for cpu and io synchronization
+
+Non-deterministic events
+------------------------
+
+Our record/replay system is based on saving and replaying non-deterministic
+events (e.g. keyboard input) and simulating deterministic ones (e.g. reading
+from HDD or memory of the VM). Saving only non-deterministic events makes
+log file smaller, simulation faster, and allows using reverse debugging even
+for realtime applications.
+
+The following non-deterministic data from peripheral devices is saved into
+the log: mouse and keyboard input, network packets, audio controller input,
+USB packets, serial port input, and hardware clocks (they are non-deterministic
+too, because their values are taken from the host machine). Inputs from
+simulated hardware, memory of VM, software interrupts, and execution of
+instructions are not saved into the log, because they are deterministic and
+can be replayed by simulating the behavior of virtual machine starting from
+initial state.
+
+We had to solve three tasks to implement deterministic replay: recording
+non-deterministic events, replaying non-deterministic events, and checking
+that there is no divergence between record and replay modes.
+
+We changed several parts of QEMU to make event log recording and replaying.
+Devices' models that have non-deterministic input from external devices were
+changed to write every external event into the execution log immediately.
+E.g. network packets are written into the log when they arrive into the virtual
+network adapter.
+
+All non-deterministic events are coming from these devices. But to
+replay them we need to know at which moments they occur. We specify
+these moments by counting the number of instructions executed between
+every pair of consecutive events.
+
+Instruction counting
+--------------------
+
+QEMU should work in icount mode to use record/replay feature. icount was
+designed to allow deterministic execution in absence of external inputs
+of the virtual machine. We also use icount to control the occurrence of the
+non-deterministic events. The number of instructions elapsed from the last event
+is written to the log while recording the execution. In replay mode we
+can predict when to inject that event using the instruction counter.
+
+Timers
+------
+
+Timers are used to execute callbacks from different subsystems of QEMU
+at the specified moments of time. There are several kinds of timers:
+ * Real time clock. Based on host time and used only for callbacks that
+   do not change the virtual machine state. For this reason real time
+   clock and timers does not affect deterministic replay at all.
+ * Virtual clock. These timers run only during the emulation. In icount
+   mode virtual clock value is calculated using executed instructions counter.
+   That is why it is completely deterministic and does not have to be recorded.
+ * Host clock. This clock is used by device models that simulate real time
+   sources (e.g. real time clock chip). Host clock is the one of the sources
+   of non-determinism. Host clock read operations should be logged to
+   make the execution deterministic.
+ * Real time clock for icount. This clock is similar to real time clock but
+   it is used only for increasing virtual clock while virtual machine is
+   sleeping. Due to its nature it is also non-deterministic as the host clock
+   and has to be logged too.
+
+Checkpoints
+-----------
+
+Replaying of the execution of virtual machine is bound by sources of
+non-determinism. These are inputs from clock and peripheral devices,
+and QEMU thread scheduling. Thread scheduling affect on processing events
+from timers, asynchronous input-output, and bottom halves.
+
+Invocations of timers are coupled with clock reads and changing the state
+of the virtual machine. Reads produce non-deterministic data taken from
+host clock. And VM state changes should preserve their order. Their relative
+order in replay mode must replicate the order of callbacks in record mode.
+To preserve this order we use checkpoints. When a specific clock is processed
+in record mode we save to the log special "checkpoint" event.
+Checkpoints here do not refer to virtual machine snapshots. They are just
+record/replay events used for synchronization.
+
+QEMU in replay mode will try to invoke timers processing in random moment
+of time. That's why we do not process a group of timers until the checkpoint
+event will be read from the log. Such an event allows synchronizing CPU
+execution and timer events.
+
+Another checkpoints application in record/replay is instruction counting
+while the virtual machine is idle. This function (qemu_clock_warp) is called
+from the wait loop. It changes virtual machine state and must be deterministic
+then. That is why we added checkpoint to this function to prevent its
+operation in replay mode when it does not correspond to record mode.
+
+Bottom halves
+-------------
+
+Disk I/O events are completely deterministic in our model, because
+in both record and replay modes we start virtual machine from the same
+disk state. But callbacks that virtual disk controller uses for reading and
+writing the disk may occur at different moments of time in record and replay
+modes.
+
+Reading and writing requests are created by CPU thread of QEMU. Later these
+requests proceed to block layer which creates "bottom halves". Bottom
+halves consist of callback and its parameters. They are processed when
+main loop locks the global mutex. These locks are not synchronized with
+replaying process because main loop also processes the events that do not
+affect the virtual machine state (like user interaction with monitor).
+
+That is why we had to implement saving and replaying bottom halves callbacks
+synchronously to the CPU execution. When the callback is about to execute
+it is added to the queue in the replay module. This queue is written to the
+log when its callbacks are executed. In replay mode callbacks are not processed
+until the corresponding event is read from the events log file.
+
+Sometimes the block layer uses asynchronous callbacks for its internal purposes
+(like reading or writing VM snapshots or disk image cluster tables). In this
+case bottom halves are not marked as "replayable" and do not saved
+into the log.
diff --git a/qapi-schema.json b/qapi-schema.json
index 4342a08..563321c 100644
--- a/qapi-schema.json
+++ b/qapi-schema.json
@@ -3794,3 +3794,21 @@
 
 # Rocker ethernet network switch
 { 'include': 'qapi/rocker.json' }
+
+##
+# ReplayMode:
+#
+# Mode of the replay subsystem.
+#
+# @none: normal execution mode. Replay or record are not enabled.
+#
+# @record: record mode. All non-deterministic data is written into the
+#          replay log.
+#
+# @play: replay mode. Non-deterministic data required for system execution
+#        is read from the log.
+#
+# Since: 2.5
+##
+{ 'enum': 'ReplayMode',
+  'data': [ 'none', 'record', 'play' ] }
diff --git a/replay/Makefile.objs b/replay/Makefile.objs
new file mode 100755
index 0000000..0b9cb99
--- /dev/null
+++ b/replay/Makefile.objs
@@ -0,0 +1,2 @@
+obj-$(CONFIG_SOFTMMU) += replay.o
+obj-$(CONFIG_USER_ONLY) += replay-user.o
diff --git a/replay/replay.c b/replay/replay.c
new file mode 100755
index 0000000..5ce066f
--- /dev/null
+++ b/replay/replay.c
@@ -0,0 +1,14 @@
+/*
+ * replay.c
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "replay.h"
+
+ReplayMode replay_mode = REPLAY_MODE_NONE;
diff --git a/replay/replay.h b/replay/replay.h
new file mode 100755
index 0000000..d6b73c3
--- /dev/null
+++ b/replay/replay.h
@@ -0,0 +1,19 @@
+#ifndef REPLAY_H
+#define REPLAY_H
+
+/*
+ * replay.h
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qapi-types.h"
+
+extern ReplayMode replay_mode;
+
+#endif
diff --git a/stubs/Makefile.objs b/stubs/Makefile.objs
index 9937a12..200adc9 100644
--- a/stubs/Makefile.objs
+++ b/stubs/Makefile.objs
@@ -25,6 +25,7 @@ stub-obj-y += monitor-init.o
 stub-obj-y += notify-event.o
 stub-obj-$(CONFIG_SPICE) += qemu-chr-open-spice.o
 stub-obj-y += qtest.o
+stub-obj-y += replay.o
 stub-obj-y += reset.o
 stub-obj-y += runstate-check.o
 stub-obj-y += set-fd-handler.o
diff --git a/stubs/replay.c b/stubs/replay.c
new file mode 100755
index 0000000..563c777
--- /dev/null
+++ b/stubs/replay.c
@@ -0,0 +1,3 @@
+#include "replay/replay.h"
+
+ReplayMode replay_mode;

[Qemu-devel] [PATCH v17 03/21] replay: internal functions for replay log

[Pavel Dovgalyuk]

This patch adds functions to perform read and write operations
with replay log.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 replay/Makefile.objs     |    1 
 replay/replay-internal.c |  155 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay-internal.h |   46 ++++++++++++++
 3 files changed, 202 insertions(+), 0 deletions(-)
 create mode 100755 replay/replay-internal.c
 create mode 100755 replay/replay-internal.h

diff --git a/replay/Makefile.objs b/replay/Makefile.objs
index 0b9cb99..9711e14 100755
--- a/replay/Makefile.objs
+++ b/replay/Makefile.objs
@@ -1,2 +1,3 @@
 obj-$(CONFIG_SOFTMMU) += replay.o
+obj-$(CONFIG_SOFTMMU) += replay-internal.o
 obj-$(CONFIG_USER_ONLY) += replay-user.o
diff --git a/replay/replay-internal.c b/replay/replay-internal.c
new file mode 100755
index 0000000..04efeee
--- /dev/null
+++ b/replay/replay-internal.c
@@ -0,0 +1,155 @@
+/*
+ * replay-internal.c
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "replay-internal.h"
+#include "qemu/error-report.h"
+#include "sysemu/sysemu.h"
+
+unsigned int replay_data_kind = -1;
+static unsigned int replay_has_unread_data;
+
+/* File for replay writing */
+FILE *replay_file;
+
+void replay_put_byte(uint8_t byte)
+{
+    if (replay_file) {
+        putc(byte, replay_file);
+    }
+}
+
+void replay_put_event(uint8_t event)
+{
+    replay_put_byte(event);
+}
+
+
+void replay_put_word(uint16_t word)
+{
+    replay_put_byte(word >> 8);
+    replay_put_byte(word);
+}
+
+void replay_put_dword(uint32_t dword)
+{
+    replay_put_word(dword >> 16);
+    replay_put_word(dword);
+}
+
+void replay_put_qword(int64_t qword)
+{
+    replay_put_dword(qword >> 32);
+    replay_put_dword(qword);
+}
+
+void replay_put_array(const uint8_t *buf, size_t size)
+{
+    if (replay_file) {
+        replay_put_dword(size);
+        fwrite(buf, 1, size, replay_file);
+    }
+}
+
+uint8_t replay_get_byte(void)
+{
+    uint8_t byte = 0;
+    if (replay_file) {
+        byte = getc(replay_file);
+    }
+    return byte;
+}
+
+uint16_t replay_get_word(void)
+{
+    uint16_t word = 0;
+    if (replay_file) {
+        word = replay_get_byte();
+        word = (word << 8) + replay_get_byte();
+    }
+
+    return word;
+}
+
+uint32_t replay_get_dword(void)
+{
+    uint32_t dword = 0;
+    if (replay_file) {
+        dword = replay_get_word();
+        dword = (dword << 16) + replay_get_word();
+    }
+
+    return dword;
+}
+
+int64_t replay_get_qword(void)
+{
+    int64_t qword = 0;
+    if (replay_file) {
+        qword = replay_get_dword();
+        qword = (qword << 32) + replay_get_dword();
+    }
+
+    return qword;
+}
+
+void replay_get_array(uint8_t *buf, size_t *size)
+{
+    if (replay_file) {
+        *size = replay_get_dword();
+        if (fread(buf, 1, *size, replay_file) != *size) {
+            error_report("replay read error");
+        }
+    }
+}
+
+void replay_get_array_alloc(uint8_t **buf, size_t *size)
+{
+    if (replay_file) {
+        *size = replay_get_dword();
+        *buf = g_malloc(*size);
+        if (fread(*buf, 1, *size, replay_file) != *size) {
+            error_report("replay read error");
+        }
+    }
+}
+
+void replay_check_error(void)
+{
+    if (replay_file) {
+        if (feof(replay_file)) {
+            error_report("replay file is over");
+            qemu_system_vmstop_request_prepare();
+            qemu_system_vmstop_request(RUN_STATE_PAUSED);
+        } else if (ferror(replay_file)) {
+            error_report("replay file is over or something goes wrong");
+            qemu_system_vmstop_request_prepare();
+            qemu_system_vmstop_request(RUN_STATE_INTERNAL_ERROR);
+        }
+    }
+}
+
+void replay_fetch_data_kind(void)
+{
+    if (replay_file) {
+        if (!replay_has_unread_data) {
+            replay_data_kind = replay_get_byte();
+            replay_check_error();
+            replay_has_unread_data = 1;
+        }
+    }
+}
+
+void replay_finish_event(void)
+{
+    replay_has_unread_data = 0;
+    replay_fetch_data_kind();
+}
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
new file mode 100755
index 0000000..17600de
--- /dev/null
+++ b/replay/replay-internal.h
@@ -0,0 +1,46 @@
+#ifndef REPLAY_INTERNAL_H
+#define REPLAY_INTERNAL_H
+
+/*
+ * replay-internal.h
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include <stdio.h>
+
+extern unsigned int replay_data_kind;
+
+/* File for replay writing */
+extern FILE *replay_file;
+
+void replay_put_byte(uint8_t byte);
+void replay_put_event(uint8_t event);
+void replay_put_word(uint16_t word);
+void replay_put_dword(uint32_t dword);
+void replay_put_qword(int64_t qword);
+void replay_put_array(const uint8_t *buf, size_t size);
+
+uint8_t replay_get_byte(void);
+uint16_t replay_get_word(void);
+uint32_t replay_get_dword(void);
+int64_t replay_get_qword(void);
+void replay_get_array(uint8_t *buf, size_t *size);
+void replay_get_array_alloc(uint8_t **buf, size_t *size);
+
+/*! Checks error status of the file. */
+void replay_check_error(void);
+
+/*! Finishes processing of the replayed event and fetches
+    the next event from the log. */
+void replay_finish_event(void);
+/*! Reads data type from the file and stores it in the
+    replay_data_kind variable. */
+void replay_fetch_data_kind(void);
+
+#endif

[Qemu-devel] [PATCH v17 04/21] replay: introduce mutex to protect the replay log

[Pavel Dovgalyuk]

This mutex will protect read/write operations for replay log.
Using mutex is necessary because most of the events consist of
several fields stored in the log. The mutex will help to avoid races.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 replay/replay-internal.c |   27 +++++++++++++++++++++++++++
 replay/replay-internal.h |    7 +++++++
 2 files changed, 34 insertions(+), 0 deletions(-)

diff --git a/replay/replay-internal.c b/replay/replay-internal.c
index 04efeee..efae672 100755
--- a/replay/replay-internal.c
+++ b/replay/replay-internal.c
@@ -17,6 +17,13 @@
 unsigned int replay_data_kind = -1;
 static unsigned int replay_has_unread_data;
 
+/* Mutex to protect reading and writing events to the log.
+   replay_data_kind and replay_has_unread_data are also protected
+   by this mutex.
+   It also protects replay events queue which stores events to be
+   written or read to the log. */
+static QemuMutex lock;
+
 /* File for replay writing */
 FILE *replay_file;
 
@@ -153,3 +160,23 @@ void replay_finish_event(void)
     replay_has_unread_data = 0;
     replay_fetch_data_kind();
 }
+
+void replay_mutex_init(void)
+{
+    qemu_mutex_init(&lock);
+}
+
+void replay_mutex_destroy(void)
+{
+    qemu_mutex_destroy(&lock);
+}
+
+void replay_mutex_lock(void)
+{
+    qemu_mutex_lock(&lock);
+}
+
+void replay_mutex_unlock(void)
+{
+    qemu_mutex_unlock(&lock);
+}
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 17600de..8a0de0d 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -33,6 +33,13 @@ int64_t replay_get_qword(void);
 void replay_get_array(uint8_t *buf, size_t *size);
 void replay_get_array_alloc(uint8_t **buf, size_t *size);
 
+/* Mutex functions for protecting replay log file */
+
+void replay_mutex_init(void);
+void replay_mutex_destroy(void);
+void replay_mutex_lock(void);
+void replay_mutex_unlock(void);
+
 /*! Checks error status of the file. */
 void replay_check_error(void);
 

[Qemu-devel] [PATCH v17 05/21] replay: introduce icount event

[Pavel Dovgalyuk]

This patch adds icount event to the replay subsystem. This event corresponds
to execution of several instructions and used to synchronize input events
in the replay phase.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 replay/replay-internal.c |   24 ++++++++++++++++++++++++
 replay/replay-internal.h |   21 +++++++++++++++++++++
 replay/replay.c          |   33 +++++++++++++++++++++++++++++++++
 replay/replay.h          |    7 +++++++
 4 files changed, 85 insertions(+), 0 deletions(-)

diff --git a/replay/replay-internal.c b/replay/replay-internal.c
index efae672..69fe49f 100755
--- a/replay/replay-internal.c
+++ b/replay/replay-internal.c
@@ -10,6 +10,7 @@
  */
 
 #include "qemu-common.h"
+#include "replay.h"
 #include "replay-internal.h"
 #include "qemu/error-report.h"
 #include "sysemu/sysemu.h"
@@ -36,6 +37,7 @@ void replay_put_byte(uint8_t byte)
 
 void replay_put_event(uint8_t event)
 {
+    assert(event < EVENT_COUNT);
     replay_put_byte(event);
 }
 
@@ -149,8 +151,15 @@ void replay_fetch_data_kind(void)
     if (replay_file) {
         if (!replay_has_unread_data) {
             replay_data_kind = replay_get_byte();
+            if (replay_data_kind == EVENT_INSTRUCTION) {
+                replay_state.instructions_count = replay_get_dword();
+            }
             replay_check_error();
             replay_has_unread_data = 1;
+            if (replay_data_kind >= EVENT_COUNT) {
+                error_report("Replay: unknown event kind %d", replay_data_kind);
+                exit(1);
+            }
         }
     }
 }
@@ -180,3 +189,18 @@ void replay_mutex_unlock(void)
 {
     qemu_mutex_unlock(&lock);
 }
+
+/*! Saves cached instructions. */
+void replay_save_instructions(void)
+{
+    if (replay_file && replay_mode == REPLAY_MODE_RECORD) {
+        replay_mutex_lock();
+        int diff = (int)(replay_get_current_step() - replay_state.current_step);
+        if (first_cpu != NULL && diff > 0) {
+            replay_put_event(EVENT_INSTRUCTION);
+            replay_put_dword(diff);
+            replay_state.current_step += diff;
+        }
+        replay_mutex_unlock();
+    }
+}
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 8a0de0d..ff4fabc 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -14,6 +14,20 @@
 
 #include <stdio.h>
 
+enum ReplayEvents {
+    /* for instruction event */
+    EVENT_INSTRUCTION,
+    EVENT_COUNT
+};
+
+typedef struct ReplayState {
+    /*! Current step - number of processed instructions and timer events. */
+    uint64_t current_step;
+    /*! Number of instructions to be executed before other events happen. */
+    int instructions_count;
+} ReplayState;
+extern ReplayState replay_state;
+
 extern unsigned int replay_data_kind;
 
 /* File for replay writing */
@@ -50,4 +64,11 @@ void replay_finish_event(void);
     replay_data_kind variable. */
 void replay_fetch_data_kind(void);
 
+/*! Saves queued events (like instructions and sound). */
+void replay_save_instructions(void);
+
+/*! Skips async events until some sync event will be found.
+    \return true, if event was found */
+bool replay_next_event_is(int event);
+
 #endif
diff --git a/replay/replay.c b/replay/replay.c
index 5ce066f..7c5af4c 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -9,6 +9,39 @@
  *
  */
 
+#include "qemu-common.h"
 #include "replay.h"
+#include "replay-internal.h"
+#include "qemu/timer.h"
 
 ReplayMode replay_mode = REPLAY_MODE_NONE;
+
+ReplayState replay_state;
+
+bool replay_next_event_is(int event)
+{
+    bool res = false;
+
+    /* nothing to skip - not all instructions used */
+    if (replay_state.instructions_count != 0) {
+        assert(replay_data_kind == EVENT_INSTRUCTION);
+        return event == EVENT_INSTRUCTION;
+    }
+
+    while (true) {
+        if (event == replay_data_kind) {
+            res = true;
+        }
+        switch (replay_data_kind) {
+        default:
+            /* clock, time_t, checkpoint and other events */
+            return res;
+        }
+    }
+    return res;
+}
+
+uint64_t replay_get_current_step(void)
+{
+    return cpu_get_icount_raw();
+}
diff --git a/replay/replay.h b/replay/replay.h
index d6b73c3..a03c748 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -12,8 +12,15 @@
  *
  */
 
+#include <stdbool.h>
+#include <stdint.h>
 #include "qapi-types.h"
 
 extern ReplayMode replay_mode;
 
+/* Processing the instructions */
+
+/*! Returns number of executed instructions. */
+uint64_t replay_get_current_step(void);
+
 #endif

[Qemu-devel] [PATCH v17 06/21] cpu-exec: allow temporary disabling icount

[Pavel Dovgalyuk]

This patch is required for deterministic replay to generate an exception
by trying executing an instruction without changing icount.
It adds new flag to TB for disabling icount while translating it.

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

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpu-exec.c              |    7 ++++---
 include/exec/exec-all.h |    1 +
 translate-all.c         |    2 +-
 3 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/cpu-exec.c b/cpu-exec.c
index 1f8046f..4bcd7a6 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -240,7 +240,7 @@ static inline tcg_target_ulong cpu_tb_exec(CPUState *cpu, uint8_t *tb_ptr)
 /* Execute the code without caching the generated code. An interpreter
    could be used if available. */
 static void cpu_exec_nocache(CPUState *cpu, int max_cycles,
-                             TranslationBlock *orig_tb)
+                             TranslationBlock *orig_tb, bool ignore_icount)
 {
     TranslationBlock *tb;
 
@@ -250,7 +250,8 @@ static void cpu_exec_nocache(CPUState *cpu, int max_cycles,
         max_cycles = CF_COUNT_MASK;
 
     tb = tb_gen_code(cpu, orig_tb->pc, orig_tb->cs_base, orig_tb->flags,
-                     max_cycles | CF_NOCACHE);
+                     max_cycles | CF_NOCACHE
+                         | (ignore_icount ? CF_IGNORE_ICOUNT : 0));
     tb->orig_tb = tcg_ctx.tb_ctx.tb_invalidated_flag ? NULL : orig_tb;
     cpu->current_tb = tb;
     /* execute the generated code */
@@ -555,7 +556,7 @@ int cpu_exec(CPUState *cpu)
                             if (insns_left > 0) {
                                 /* Execute remaining instructions.  */
                                 tb = (TranslationBlock *)(next_tb & ~TB_EXIT_MASK);
-                                cpu_exec_nocache(cpu, insns_left, tb);
+                                cpu_exec_nocache(cpu, insns_left, tb, false);
                                 align_clocks(&sc, cpu);
                             }
                             cpu->exception_index = EXCP_INTERRUPT;
diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index f500dff..878a16e 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -199,6 +199,7 @@ struct TranslationBlock {
 #define CF_LAST_IO     0x8000 /* Last insn may be an IO access.  */
 #define CF_NOCACHE     0x10000 /* To be freed after execution */
 #define CF_USE_ICOUNT  0x20000
+#define CF_IGNORE_ICOUNT 0x40000 /* Do not generate icount code */
 
     void *tc_ptr;    /* pointer to the translated code */
     /* next matching tb for physical address. */
diff --git a/translate-all.c b/translate-all.c
index 2a40530..c2ba423 100644
--- a/translate-all.c
+++ b/translate-all.c
@@ -1007,7 +1007,7 @@ TranslationBlock *tb_gen_code(CPUState *cpu,
     int code_gen_size;
 
     phys_pc = get_page_addr_code(env, pc);
-    if (use_icount) {
+    if (use_icount && !(cflags & CF_IGNORE_ICOUNT)) {
         cflags |= CF_USE_ICOUNT;
     }
     tb = tb_alloc(pc);

[Qemu-devel] [PATCH v17 07/21] cpu: replay instructions sequence

[Pavel Dovgalyuk]

This patch adds calls to replay functions into the icount setup block.
In record mode number of executed instructions is written to the log.
In replay mode number of istructions to execute is taken from the replay log.
When replayed instructions counter is expired qemu_notify_event()
function is called to wake up the iothread.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpus.c          |   38 +++++++++++++++++++++++++-------------
 replay/replay.c |   34 ++++++++++++++++++++++++++++++++++
 replay/replay.h |    4 ++++
 3 files changed, 63 insertions(+), 13 deletions(-)

diff --git a/cpus.c b/cpus.c
index c1e74d9..ec3f21d 100644
--- a/cpus.c
+++ b/cpus.c
@@ -42,6 +42,7 @@
 #include "qemu/seqlock.h"
 #include "qapi-event.h"
 #include "hw/nmi.h"
+#include "replay/replay.h"
 
 #ifndef _WIN32
 #include "qemu/compatfd.h"
@@ -1363,6 +1364,28 @@ int vm_stop_force_state(RunState state)
     }
 }
 
+static int64_t tcg_get_icount_limit(void)
+{
+    int64_t deadline;
+
+    if (replay_mode != REPLAY_MODE_PLAY) {
+        deadline = qemu_clock_deadline_ns_all(QEMU_CLOCK_VIRTUAL);
+
+        /* Maintain prior (possibly buggy) behaviour where if no deadline
+         * was set (as there is no QEMU_CLOCK_VIRTUAL timer) or it is more than
+         * INT32_MAX nanoseconds ahead, we still use INT32_MAX
+         * nanoseconds.
+         */
+        if ((deadline < 0) || (deadline > INT32_MAX)) {
+            deadline = INT32_MAX;
+        }
+
+        return qemu_icount_round(deadline);
+    } else {
+        return replay_get_instructions();
+    }
+}
+
 static int tcg_cpu_exec(CPUState *cpu)
 {
     int ret;
@@ -1375,24 +1398,12 @@ static int tcg_cpu_exec(CPUState *cpu)
 #endif
     if (use_icount) {
         int64_t count;
-        int64_t deadline;
         int decr;
         timers_state.qemu_icount -= (cpu->icount_decr.u16.low
                                     + cpu->icount_extra);
         cpu->icount_decr.u16.low = 0;
         cpu->icount_extra = 0;
-        deadline = qemu_clock_deadline_ns_all(QEMU_CLOCK_VIRTUAL);
-
-        /* Maintain prior (possibly buggy) behaviour where if no deadline
-         * was set (as there is no QEMU_CLOCK_VIRTUAL timer) or it is more than
-         * INT32_MAX nanoseconds ahead, we still use INT32_MAX
-         * nanoseconds.
-         */
-        if ((deadline < 0) || (deadline > INT32_MAX)) {
-            deadline = INT32_MAX;
-        }
-
-        count = qemu_icount_round(deadline);
+        count = tcg_get_icount_limit();
         timers_state.qemu_icount += count;
         decr = (count > 0xffff) ? 0xffff : count;
         count -= decr;
@@ -1410,6 +1421,7 @@ static int tcg_cpu_exec(CPUState *cpu)
                         + cpu->icount_extra);
         cpu->icount_decr.u32 = 0;
         cpu->icount_extra = 0;
+        replay_account_executed_instructions();
     }
     return ret;
 }
diff --git a/replay/replay.c b/replay/replay.c
index 7c5af4c..731eec3 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -13,6 +13,7 @@
 #include "replay.h"
 #include "replay-internal.h"
 #include "qemu/timer.h"
+#include "qemu/main-loop.h"
 
 ReplayMode replay_mode = REPLAY_MODE_NONE;
 
@@ -45,3 +46,36 @@ uint64_t replay_get_current_step(void)
 {
     return cpu_get_icount_raw();
 }
+
+int replay_get_instructions(void)
+{
+    int res = 0;
+    replay_mutex_lock();
+    if (replay_next_event_is(EVENT_INSTRUCTION)) {
+        res = replay_state.instructions_count;
+    }
+    replay_mutex_unlock();
+    return res;
+}
+
+void replay_account_executed_instructions(void)
+{
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        replay_mutex_lock();
+        if (replay_state.instructions_count > 0) {
+            int count = (int)(replay_get_current_step()
+                              - replay_state.current_step);
+            replay_state.instructions_count -= count;
+            replay_state.current_step += count;
+            if (replay_state.instructions_count == 0) {
+                assert(replay_data_kind == EVENT_INSTRUCTION);
+                replay_finish_event();
+                /* Wake up iothread. This is required because
+                   timers will not expire until clock counters
+                   will be read from the log. */
+                qemu_notify_event();
+            }
+        }
+        replay_mutex_unlock();
+    }
+}
diff --git a/replay/replay.h b/replay/replay.h
index a03c748..d19715f 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -22,5 +22,9 @@ extern ReplayMode replay_mode;
 
 /*! Returns number of executed instructions. */
 uint64_t replay_get_current_step(void);
+/*! Returns number of instructions to execute in replay mode. */
+int replay_get_instructions(void);
+/*! Updates instructions counter in replay mode. */
+void replay_account_executed_instructions(void);
 
 #endif

[Qemu-devel] [PATCH v17 08/21] i386: interrupt poll processing

[Pavel Dovgalyuk]

This patch updates x86_cpu_exec_interrupt function.
It can process two interrupt request at a time (poll and another one).
This makes its execution non-deterministic. Determinism is requred
for recorded icount execution.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 target-i386/seg_helper.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/target-i386/seg_helper.c b/target-i386/seg_helper.c
index 1a3a2e7..1cbe559 100644
--- a/target-i386/seg_helper.c
+++ b/target-i386/seg_helper.c
@@ -1321,6 +1321,9 @@ bool x86_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
     if (interrupt_request & CPU_INTERRUPT_POLL) {
         cs->interrupt_request &= ~CPU_INTERRUPT_POLL;
         apic_poll_irq(cpu->apic_state);
+        /* Don't process multiple interrupt requests in a single call.
+           This is required to make icount-driven execution deterministic. */
+        return true;
     }
 #endif
     if (interrupt_request & CPU_INTERRUPT_SIPI) {

[Qemu-devel] [PATCH v17 09/21] replay: interrupts and exceptions

[Pavel Dovgalyuk]

This patch includes modifications of common cpu files. All interrupts and
exceptions occured during recording are written into the replay log.
These events allow correct replaying the execution by kicking cpu thread
when one of these events is found in the log.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpu-exec.c               |   49 ++++++++++++++++++++++++++--------
 replay/replay-internal.h |    4 +++
 replay/replay.c          |   67 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay.h          |   17 ++++++++++++
 4 files changed, 126 insertions(+), 11 deletions(-)

diff --git a/cpu-exec.c b/cpu-exec.c
index 4bcd7a6..aff6d83 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -31,6 +31,7 @@
 #if defined(TARGET_I386) && !defined(CONFIG_USER_ONLY)
 #include "hw/i386/apic.h"
 #endif
+#include "replay/replay.h"
 
 /* -icount align implementation. */
 
@@ -372,22 +373,25 @@ int cpu_exec(CPUState *cpu)
     /* This must be volatile so it is not trashed by longjmp() */
     volatile bool have_tb_lock = false;
 
+    /* replay_interrupt may need current_cpu */
+    current_cpu = cpu;
+
     if (cpu->halted) {
 #if defined(TARGET_I386) && !defined(CONFIG_USER_ONLY)
-        if (cpu->interrupt_request & CPU_INTERRUPT_POLL) {
+        if ((cpu->interrupt_request & CPU_INTERRUPT_POLL)
+            && replay_interrupt()) {
             apic_poll_irq(x86_cpu->apic_state);
             cpu_reset_interrupt(cpu, CPU_INTERRUPT_POLL);
         }
 #endif
         if (!cpu_has_work(cpu)) {
+            current_cpu = NULL;
             return EXCP_HALTED;
         }
 
         cpu->halted = 0;
     }
 
-    current_cpu = cpu;
-
     /* As long as current_cpu is null, up to the assignment just above,
      * requests by other threads to exit the execution loop are expected to
      * be issued using the exit_request global. We must make sure that our
@@ -436,10 +440,22 @@ int cpu_exec(CPUState *cpu)
                     cpu->exception_index = -1;
                     break;
 #else
-                    cc->do_interrupt(cpu);
-                    cpu->exception_index = -1;
+                    if (replay_exception()) {
+                        cc->do_interrupt(cpu);
+                        cpu->exception_index = -1;
+                    } else if (!replay_has_interrupt()) {
+                        /* give a chance to iothread in replay mode */
+                        ret = EXCP_INTERRUPT;
+                        break;
+                    }
 #endif
                 }
+            } else if (replay_has_exception()
+                       && cpu->icount_decr.u16.low + cpu->icount_extra == 0) {
+                /* try to cause an exception pending in the log */
+                cpu_exec_nocache(cpu, 1, tb_find_fast(cpu), true);
+                ret = -1;
+                break;
             }
 
             next_tb = 0; /* force lookup of first TB */
@@ -455,30 +471,40 @@ int cpu_exec(CPUState *cpu)
                         cpu->exception_index = EXCP_DEBUG;
                         cpu_loop_exit(cpu);
                     }
-                    if (interrupt_request & CPU_INTERRUPT_HALT) {
+                    if (replay_mode == REPLAY_MODE_PLAY
+                        && !replay_has_interrupt()) {
+                        /* Do nothing */
+                    } else if (interrupt_request & CPU_INTERRUPT_HALT) {
+                        replay_interrupt();
                         cpu->interrupt_request &= ~CPU_INTERRUPT_HALT;
                         cpu->halted = 1;
                         cpu->exception_index = EXCP_HLT;
                         cpu_loop_exit(cpu);
                     }
 #if defined(TARGET_I386)
-                    if (interrupt_request & CPU_INTERRUPT_INIT) {
+                    else if (interrupt_request & CPU_INTERRUPT_INIT) {
+                        replay_interrupt();
                         cpu_svm_check_intercept_param(env, SVM_EXIT_INIT, 0);
                         do_cpu_init(x86_cpu);
                         cpu->exception_index = EXCP_HALTED;
                         cpu_loop_exit(cpu);
                     }
 #else
-                    if (interrupt_request & CPU_INTERRUPT_RESET) {
+                    else if (interrupt_request & CPU_INTERRUPT_RESET) {
+                        replay_interrupt();
                         cpu_reset(cpu);
+                        cpu_loop_exit(cpu);
                     }
 #endif
                     /* The target hook has 3 exit conditions:
                        False when the interrupt isn't processed,
                        True when it is, and we should restart on a new TB,
                        and via longjmp via cpu_loop_exit.  */
-                    if (cc->cpu_exec_interrupt(cpu, interrupt_request)) {
-                        next_tb = 0;
+                    else {
+                        replay_interrupt();
+                        if (cc->cpu_exec_interrupt(cpu, interrupt_request)) {
+                            next_tb = 0;
+                        }
                     }
                     /* Don't use the cached interrupt_request value,
                        do_interrupt may have updated the EXITTB flag. */
@@ -489,7 +515,8 @@ int cpu_exec(CPUState *cpu)
                         next_tb = 0;
                     }
                 }
-                if (unlikely(cpu->exit_request)) {
+                if (unlikely(cpu->exit_request
+                             || replay_has_interrupt())) {
                     cpu->exit_request = 0;
                     cpu->exception_index = EXCP_INTERRUPT;
                     cpu_loop_exit(cpu);
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index ff4fabc..5ff1c14 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -17,6 +17,10 @@
 enum ReplayEvents {
     /* for instruction event */
     EVENT_INSTRUCTION,
+    /* for software interrupt */
+    EVENT_INTERRUPT,
+    /* for emulated exceptions */
+    EVENT_EXCEPTION,
     EVENT_COUNT
 };
 
diff --git a/replay/replay.c b/replay/replay.c
index 731eec3..d072ef7 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -79,3 +79,70 @@ void replay_account_executed_instructions(void)
         replay_mutex_unlock();
     }
 }
+
+bool replay_exception(void)
+{
+    if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_save_instructions();
+        replay_mutex_lock();
+        replay_put_event(EVENT_EXCEPTION);
+        replay_mutex_unlock();
+        return true;
+    } else if (replay_mode == REPLAY_MODE_PLAY) {
+        bool res = replay_has_exception();
+        if (res) {
+            replay_mutex_lock();
+            replay_finish_event();
+            replay_mutex_unlock();
+        }
+        return res;
+    }
+
+    return true;
+}
+
+bool replay_has_exception(void)
+{
+    bool res = false;
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        replay_account_executed_instructions();
+        replay_mutex_lock();
+        res = replay_next_event_is(EVENT_EXCEPTION);
+        replay_mutex_unlock();
+    }
+
+    return res;
+}
+
+bool replay_interrupt(void)
+{
+    if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_save_instructions();
+        replay_mutex_lock();
+        replay_put_event(EVENT_INTERRUPT);
+        replay_mutex_unlock();
+        return true;
+    } else if (replay_mode == REPLAY_MODE_PLAY) {
+        bool res = replay_has_interrupt();
+        if (res) {
+            replay_mutex_lock();
+            replay_finish_event();
+            replay_mutex_unlock();
+        }
+        return res;
+    }
+
+    return true;
+}
+
+bool replay_has_interrupt(void)
+{
+    bool res = false;
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        replay_account_executed_instructions();
+        replay_mutex_lock();
+        res = replay_next_event_is(EVENT_INTERRUPT);
+        replay_mutex_unlock();
+    }
+    return res;
+}
diff --git a/replay/replay.h b/replay/replay.h
index d19715f..8915523 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -27,4 +27,21 @@ int replay_get_instructions(void);
 /*! Updates instructions counter in replay mode. */
 void replay_account_executed_instructions(void);
 
+/* Interrupts and exceptions */
+
+/*! Called by exception handler to write or read
+    exception processing events. */
+bool replay_exception(void);
+/*! Used to determine that exception is pending.
+    Does not proceed to the next event in the log. */
+bool replay_has_exception(void);
+/*! Called by interrupt handlers to write or read
+    interrupt processing events.
+    \return true if interrupt should be processed */
+bool replay_interrupt(void);
+/*! Tries to read interrupt event from the file.
+    Returns true, when interrupt request is pending */
+bool replay_has_interrupt(void);
+
+
 #endif

[Qemu-devel] [PATCH v17 10/21] replay: asynchronous events infrastructure

[Pavel Dovgalyuk]

This patch adds module for saving and replaying asynchronous events.
These events include network packets, keyboard and mouse input,
USB packets, thread pool and bottom halves callbacks.
All events are stored in the queue to be processed at synchronization points
such as beginning of TB execution, or checkpoint in the iothread.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 replay/Makefile.objs     |    1 
 replay/replay-events.c   |  225 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay-internal.h |   29 ++++++
 replay/replay.h          |    6 +
 4 files changed, 261 insertions(+), 0 deletions(-)
 create mode 100755 replay/replay-events.c

diff --git a/replay/Makefile.objs b/replay/Makefile.objs
index 9711e14..1ce61ec 100755
--- a/replay/Makefile.objs
+++ b/replay/Makefile.objs
@@ -1,3 +1,4 @@
 obj-$(CONFIG_SOFTMMU) += replay.o
 obj-$(CONFIG_SOFTMMU) += replay-internal.o
+obj-$(CONFIG_SOFTMMU) += replay-events.o
 obj-$(CONFIG_USER_ONLY) += replay-user.o
diff --git a/replay/replay-events.c b/replay/replay-events.c
new file mode 100755
index 0000000..f973cb0
--- /dev/null
+++ b/replay/replay-events.c
@@ -0,0 +1,225 @@
+/*
+ * replay-events.c
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "qemu/error-report.h"
+#include "replay.h"
+#include "replay-internal.h"
+
+typedef struct Event {
+    ReplayAsyncEventKind event_kind;
+    void *opaque;
+    void *opaque2;
+    uint64_t id;
+
+    QTAILQ_ENTRY(Event) events;
+} Event;
+
+static QTAILQ_HEAD(, Event) events_list = QTAILQ_HEAD_INITIALIZER(events_list);
+static unsigned int read_event_kind = -1;
+static uint64_t read_id = -1;
+static int read_checkpoint = -1;
+
+static bool events_enabled;
+
+/* Functions */
+
+static void replay_run_event(Event *event)
+{
+    switch (event->event_kind) {
+    default:
+        error_report("Replay: invalid async event ID (%d) in the queue",
+                    event->event_kind);
+        exit(1);
+        break;
+    }
+}
+
+void replay_enable_events(void)
+{
+    events_enabled = true;
+}
+
+bool replay_has_events(void)
+{
+    return !QTAILQ_EMPTY(&events_list);
+}
+
+void replay_flush_events(void)
+{
+    replay_mutex_lock();
+    while (!QTAILQ_EMPTY(&events_list)) {
+        Event *event = QTAILQ_FIRST(&events_list);
+        replay_mutex_unlock();
+        replay_run_event(event);
+        replay_mutex_lock();
+        QTAILQ_REMOVE(&events_list, event, events);
+        g_free(event);
+    }
+    replay_mutex_unlock();
+}
+
+void replay_disable_events(void)
+{
+    if (replay_mode != REPLAY_MODE_NONE) {
+        events_enabled = false;
+        /* Flush events queue before waiting of completion */
+        replay_flush_events();
+    }
+}
+
+void replay_clear_events(void)
+{
+    replay_mutex_lock();
+    while (!QTAILQ_EMPTY(&events_list)) {
+        Event *event = QTAILQ_FIRST(&events_list);
+        QTAILQ_REMOVE(&events_list, event, events);
+
+        g_free(event);
+    }
+    replay_mutex_unlock();
+}
+
+/*! Adds specified async event to the queue */
+static void replay_add_event(ReplayAsyncEventKind event_kind,
+                             void *opaque,
+                             void *opaque2, uint64_t id)
+{
+    assert(event_kind < REPLAY_ASYNC_COUNT);
+
+    if (!replay_file || replay_mode == REPLAY_MODE_NONE
+        || !events_enabled) {
+        Event e;
+        e.event_kind = event_kind;
+        e.opaque = opaque;
+        e.opaque2 = opaque2;
+        e.id = id;
+        replay_run_event(&e);
+        return;
+    }
+
+    Event *event = g_malloc0(sizeof(Event));
+    event->event_kind = event_kind;
+    event->opaque = opaque;
+    event->opaque2 = opaque2;
+    event->id = id;
+
+    replay_mutex_lock();
+    QTAILQ_INSERT_TAIL(&events_list, event, events);
+    replay_mutex_unlock();
+}
+
+static void replay_save_event(Event *event, int checkpoint)
+{
+    if (replay_mode != REPLAY_MODE_PLAY) {
+        /* put the event into the file */
+        replay_put_event(EVENT_ASYNC);
+        replay_put_byte(checkpoint);
+        replay_put_byte(event->event_kind);
+
+        /* save event-specific data */
+        switch (event->event_kind) {
+        default:
+            error_report("Unknown ID %d of replay event", read_event_kind);
+            exit(1);
+            break;
+        }
+    }
+}
+
+/* Called with replay mutex locked */
+void replay_save_events(int checkpoint)
+{
+    while (!QTAILQ_EMPTY(&events_list)) {
+        Event *event = QTAILQ_FIRST(&events_list);
+        replay_save_event(event, checkpoint);
+
+        replay_mutex_unlock();
+        replay_run_event(event);
+        replay_mutex_lock();
+        QTAILQ_REMOVE(&events_list, event, events);
+        g_free(event);
+    }
+}
+
+static Event *replay_read_event(int checkpoint)
+{
+    Event *event;
+    if (read_event_kind == -1) {
+        read_checkpoint = replay_get_byte();
+        read_event_kind = replay_get_byte();
+        read_id = -1;
+        replay_check_error();
+    }
+
+    if (checkpoint != read_checkpoint) {
+        return NULL;
+    }
+
+    /* Events that has not to be in the queue */
+    switch (read_event_kind) {
+    default:
+        error_report("Unknown ID %d of replay event", read_event_kind);
+        exit(1);
+        break;
+    }
+
+    QTAILQ_FOREACH(event, &events_list, events) {
+        if (event->event_kind == read_event_kind
+            && (read_id == -1 || read_id == event->id)) {
+            break;
+        }
+    }
+
+    if (event) {
+        QTAILQ_REMOVE(&events_list, event, events);
+    } else {
+        return NULL;
+    }
+
+    /* Read event-specific data */
+
+    return event;
+}
+
+/* Called with replay mutex locked */
+void replay_read_events(int checkpoint)
+{
+    while (replay_data_kind == EVENT_ASYNC) {
+        Event *event = replay_read_event(checkpoint);
+        if (!event) {
+            break;
+        }
+        replay_mutex_unlock();
+        replay_run_event(event);
+        replay_mutex_lock();
+
+        g_free(event);
+        replay_finish_event();
+        read_event_kind = -1;
+    }
+}
+
+void replay_init_events(void)
+{
+    read_event_kind = -1;
+}
+
+void replay_finish_events(void)
+{
+    events_enabled = false;
+    replay_clear_events();
+}
+
+bool replay_events_enabled(void)
+{
+    return events_enabled;
+}
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 5ff1c14..23807ca 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -21,9 +21,19 @@ enum ReplayEvents {
     EVENT_INTERRUPT,
     /* for emulated exceptions */
     EVENT_EXCEPTION,
+    /* for async events */
+    EVENT_ASYNC,
     EVENT_COUNT
 };
 
+/* Asynchronous events IDs */
+
+enum ReplayAsyncEventKind {
+    REPLAY_ASYNC_COUNT
+};
+
+typedef enum ReplayAsyncEventKind ReplayAsyncEventKind;
+
 typedef struct ReplayState {
     /*! Current step - number of processed instructions and timer events. */
     uint64_t current_step;
@@ -75,4 +85,23 @@ void replay_save_instructions(void);
     \return true, if event was found */
 bool replay_next_event_is(int event);
 
+/* Asynchronous events queue */
+
+/*! Initializes events' processing internals */
+void replay_init_events(void);
+/*! Clears internal data structures for events handling */
+void replay_finish_events(void);
+/*! Enables storing events in the queue */
+void replay_enable_events(void);
+/*! Flushes events queue */
+void replay_flush_events(void);
+/*! Clears events list before loading new VM state */
+void replay_clear_events(void);
+/*! Returns true if there are any unsaved events in the queue */
+bool replay_has_events(void);
+/*! Saves events from queue into the file */
+void replay_save_events(int checkpoint);
+/*! Read events from the file into the input queue */
+void replay_read_events(int checkpoint);
+
 #endif
diff --git a/replay/replay.h b/replay/replay.h
index 8915523..c2a7651 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -43,5 +43,11 @@ bool replay_interrupt(void);
     Returns true, when interrupt request is pending */
 bool replay_has_interrupt(void);
 
+/* Asynchronous events queue */
+
+/*! Disables storing events in the queue */
+void replay_disable_events(void);
+/*! Returns true when saving events is enabled */
+bool replay_events_enabled(void);
 
 #endif

[Qemu-devel] [PATCH v17 11/21] replay: recording and replaying clock ticks

[Pavel Dovgalyuk]

Clock ticks are considered as the sources of non-deterministic data for
virtual machine. This patch implements saving the clock values when they
are acquired (virtual, host clock).
When replaying the execution corresponding values are read from log and
transfered to the module, which wants to read the values.
Such a design required the clock polling to be synchronized. Sometimes
it is not true - e.g. when timeouts for timer lists are checked. In this case
we use a cached value of the clock, passing it to the client code.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpus.c                   |    3 +-
 qemu-timer.c             |    8 ++++--
 replay/Makefile.objs     |    1 +
 replay/replay-internal.h |   11 ++++++++
 replay/replay-time.c     |   64 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay.h          |   23 +++++++++++++++++
 stubs/replay.c           |   13 +++++++++
 7 files changed, 119 insertions(+), 4 deletions(-)
 create mode 100755 replay/replay-time.c

diff --git a/cpus.c b/cpus.c
index ec3f21d..821ebbe 100644
--- a/cpus.c
+++ b/cpus.c
@@ -338,7 +338,8 @@ static void icount_warp_rt(void *opaque)
 
     seqlock_write_lock(&timers_state.vm_clock_seqlock);
     if (runstate_is_running()) {
-        int64_t clock = cpu_get_clock_locked();
+        int64_t clock = REPLAY_CLOCK(REPLAY_CLOCK_VIRTUAL_RT,
+                                     cpu_get_clock_locked());
         int64_t warp_delta;
 
         warp_delta = clock - vm_clock_warp_start;
diff --git a/qemu-timer.c b/qemu-timer.c
index 2463fe6..c7bd643 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -24,6 +24,7 @@
 
 #include "qemu/main-loop.h"
 #include "qemu/timer.h"
+#include "replay/replay.h"
 
 #ifdef CONFIG_POSIX
 #include <pthread.h>
@@ -570,15 +571,16 @@ int64_t qemu_clock_get_ns(QEMUClockType type)
             return cpu_get_clock();
         }
     case QEMU_CLOCK_HOST:
-        now = get_clock_realtime();
+        now = REPLAY_CLOCK(REPLAY_CLOCK_HOST, get_clock_realtime());
         last = clock->last;
         clock->last = now;
-        if (now < last || now > (last + get_max_clock_jump())) {
+        if ((now < last || now > (last + get_max_clock_jump()))
+            && replay_mode == REPLAY_MODE_NONE) {
             notifier_list_notify(&clock->reset_notifiers, &now);
         }
         return now;
     case QEMU_CLOCK_VIRTUAL_RT:
-        return cpu_get_clock();
+        return REPLAY_CLOCK(REPLAY_CLOCK_VIRTUAL_RT, cpu_get_clock());
     }
 }
 
diff --git a/replay/Makefile.objs b/replay/Makefile.objs
index 1ce61ec..4aa527e 100755
--- a/replay/Makefile.objs
+++ b/replay/Makefile.objs
@@ -1,4 +1,5 @@
 obj-$(CONFIG_SOFTMMU) += replay.o
 obj-$(CONFIG_SOFTMMU) += replay-internal.o
 obj-$(CONFIG_SOFTMMU) += replay-events.o
+obj-$(CONFIG_SOFTMMU) += replay-time.o
 obj-$(CONFIG_USER_ONLY) += replay-user.o
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 23807ca..f042c46 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -23,6 +23,10 @@ enum ReplayEvents {
     EVENT_EXCEPTION,
     /* for async events */
     EVENT_ASYNC,
+    /* for clock read/writes */
+    /* some of greater codes are reserved for clocks */
+    EVENT_CLOCK,
+    EVENT_CLOCK_LAST = EVENT_CLOCK + REPLAY_CLOCK_COUNT - 1,
     EVENT_COUNT
 };
 
@@ -35,6 +39,8 @@ enum ReplayAsyncEventKind {
 typedef enum ReplayAsyncEventKind ReplayAsyncEventKind;
 
 typedef struct ReplayState {
+    /*! Cached clock values. */
+    int64_t cached_clock[REPLAY_CLOCK_COUNT];
     /*! Current step - number of processed instructions and timer events. */
     uint64_t current_step;
     /*! Number of instructions to be executed before other events happen. */
@@ -85,6 +91,11 @@ void replay_save_instructions(void);
     \return true, if event was found */
 bool replay_next_event_is(int event);
 
+/*! Reads next clock value from the file.
+    If clock kind read from the file is different from the parameter,
+    the value is not used. */
+void replay_read_next_clock(unsigned int kind);
+
 /* Asynchronous events queue */
 
 /*! Initializes events' processing internals */
diff --git a/replay/replay-time.c b/replay/replay-time.c
new file mode 100755
index 0000000..a05bfd9
--- /dev/null
+++ b/replay/replay-time.c
@@ -0,0 +1,64 @@
+/*
+ * replay-time.c
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "replay.h"
+#include "replay-internal.h"
+#include "qemu/error-report.h"
+
+int64_t replay_save_clock(ReplayClockKind kind, int64_t clock)
+{
+    replay_save_instructions();
+
+    if (replay_file) {
+        replay_mutex_lock();
+        replay_put_event(EVENT_CLOCK + kind);
+        replay_put_qword(clock);
+        replay_mutex_unlock();
+    }
+
+    return clock;
+}
+
+void replay_read_next_clock(ReplayClockKind kind)
+{
+    unsigned int read_kind = replay_data_kind - EVENT_CLOCK;
+
+    assert(read_kind == kind);
+
+    int64_t clock = replay_get_qword();
+
+    replay_check_error();
+    replay_finish_event();
+
+    replay_state.cached_clock[read_kind] = clock;
+}
+
+/*! Reads next clock event from the input. */
+int64_t replay_read_clock(ReplayClockKind kind)
+{
+    replay_account_executed_instructions();
+
+    if (replay_file) {
+        int64_t ret;
+        replay_mutex_lock();
+        if (replay_next_event_is(EVENT_CLOCK + kind)) {
+            replay_read_next_clock(kind);
+        }
+        ret = replay_state.cached_clock[kind];
+        replay_mutex_unlock();
+
+        return ret;
+    }
+
+    error_report("REPLAY INTERNAL ERROR %d", __LINE__);
+    exit(1);
+}
diff --git a/replay/replay.h b/replay/replay.h
index c2a7651..2398509 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -16,6 +16,16 @@
 #include <stdint.h>
 #include "qapi-types.h"
 
+/* replay clock kinds */
+enum ReplayClockKind {
+    /* host_clock */
+    REPLAY_CLOCK_HOST,
+    /* virtual_rt_clock */
+    REPLAY_CLOCK_VIRTUAL_RT,
+    REPLAY_CLOCK_COUNT
+};
+typedef enum ReplayClockKind ReplayClockKind;
+
 extern ReplayMode replay_mode;
 
 /* Processing the instructions */
@@ -43,6 +53,19 @@ bool replay_interrupt(void);
     Returns true, when interrupt request is pending */
 bool replay_has_interrupt(void);
 
+/* Processing clocks and other time sources */
+
+/*! Save the specified clock */
+int64_t replay_save_clock(ReplayClockKind kind, int64_t clock);
+/*! Read the specified clock from the log or return cached data */
+int64_t replay_read_clock(ReplayClockKind kind);
+/*! Saves or reads the clock depending on the current replay mode. */
+#define REPLAY_CLOCK(clock, value)                                      \
+    (replay_mode == REPLAY_MODE_PLAY ? replay_read_clock((clock))       \
+        : replay_mode == REPLAY_MODE_RECORD                             \
+            ? replay_save_clock((clock), (value))                       \
+        : (value))
+
 /* Asynchronous events queue */
 
 /*! Disables storing events in the queue */
diff --git a/stubs/replay.c b/stubs/replay.c
index 563c777..3e7c3fe 100755
--- a/stubs/replay.c
+++ b/stubs/replay.c
@@ -1,3 +1,16 @@
 #include "replay/replay.h"
+#include <stdlib.h>
 
 ReplayMode replay_mode;
+
+int64_t replay_save_clock(unsigned int kind, int64_t clock)
+{
+    abort();
+    return 0;
+}
+
+int64_t replay_read_clock(unsigned int kind)
+{
+    abort();
+    return 0;
+}

[Qemu-devel] [PATCH v17 12/21] replay: shutdown event

[Pavel Dovgalyuk]

This patch records and replays simulator shutdown event.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 replay/replay-internal.h |    2 ++
 replay/replay.c          |   14 ++++++++++++++
 replay/replay.h          |    5 +++++
 vl.c                     |    1 +
 4 files changed, 22 insertions(+), 0 deletions(-)

diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index f042c46..4414695 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -23,6 +23,8 @@ enum ReplayEvents {
     EVENT_EXCEPTION,
     /* for async events */
     EVENT_ASYNC,
+    /* for shutdown request */
+    EVENT_SHUTDOWN,
     /* for clock read/writes */
     /* some of greater codes are reserved for clocks */
     EVENT_CLOCK,
diff --git a/replay/replay.c b/replay/replay.c
index d072ef7..83b0f8c 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -14,6 +14,7 @@
 #include "replay-internal.h"
 #include "qemu/timer.h"
 #include "qemu/main-loop.h"
+#include "sysemu/sysemu.h"
 
 ReplayMode replay_mode = REPLAY_MODE_NONE;
 
@@ -34,6 +35,10 @@ bool replay_next_event_is(int event)
             res = true;
         }
         switch (replay_data_kind) {
+        case EVENT_SHUTDOWN:
+            replay_finish_event();
+            qemu_system_shutdown_request();
+            break;
         default:
             /* clock, time_t, checkpoint and other events */
             return res;
@@ -146,3 +151,12 @@ bool replay_has_interrupt(void)
     }
     return res;
 }
+
+void replay_shutdown_request(void)
+{
+    if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_mutex_lock();
+        replay_put_event(EVENT_SHUTDOWN);
+        replay_mutex_unlock();
+    }
+}
diff --git a/replay/replay.h b/replay/replay.h
index 2398509..fcc93d1 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -66,6 +66,11 @@ int64_t replay_read_clock(ReplayClockKind kind);
             ? replay_save_clock((clock), (value))                       \
         : (value))
 
+/* Events */
+
+/*! Called when qemu shutdown is requested. */
+void replay_shutdown_request(void);
+
 /* Asynchronous events queue */
 
 /*! Disables storing events in the queue */
diff --git a/vl.c b/vl.c
index 584ca88..9ff801d 100644
--- a/vl.c
+++ b/vl.c
@@ -1814,6 +1814,7 @@ void qemu_system_killed(int signal, pid_t pid)
 void qemu_system_shutdown_request(void)
 {
     trace_qemu_system_shutdown_request();
+    replay_shutdown_request();
     shutdown_requested = 1;
     qemu_notify_event();
 }

[Qemu-devel] [PATCH v17 13/21] icount: improve counting for record/replay

[Pavel Dovgalyuk]

icount_warp_rt function is called by qemu_clock_warp and as
callback of icount_warp timer. This patch adds call to qemu_clock_warp
into main_loop_wait function, because icount warp may be missed
in record/replay mode, when CPU is sleeping.
This patch also disables of calling this function by timer, because
it is not needed after making modifications of main_loop_wait.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpus.c      |   11 ++++++++---
 main-loop.c |    3 +++
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/cpus.c b/cpus.c
index 821ebbe..00d50f1 100644
--- a/cpus.c
+++ b/cpus.c
@@ -327,7 +327,7 @@ static int64_t qemu_icount_round(int64_t count)
     return (count + (1 << icount_time_shift) - 1) >> icount_time_shift;
 }
 
-static void icount_warp_rt(void *opaque)
+static void icount_warp_rt(void)
 {
     /* The icount_warp_timer is rescheduled soon after vm_clock_warp_start
      * changes from -1 to another value, so the race here is okay.
@@ -362,6 +362,11 @@ static void icount_warp_rt(void *opaque)
     }
 }
 
+static void icount_dummy_timer(void *opaque)
+{
+    (void)opaque;
+}
+
 void qtest_clock_warp(int64_t dest)
 {
     int64_t clock = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL);
@@ -406,7 +411,7 @@ void qemu_clock_warp(QEMUClockType type)
          * the CPU starts running, in case the CPU is woken by an event other
          * than the earliest QEMU_CLOCK_VIRTUAL timer.
          */
-        icount_warp_rt(NULL);
+        icount_warp_rt();
         timer_del(icount_warp_timer);
     }
     if (!all_cpu_threads_idle()) {
@@ -529,7 +534,7 @@ void configure_icount(QemuOpts *opts, Error **errp)
     icount_sleep = qemu_opt_get_bool(opts, "sleep", true);
     if (icount_sleep) {
         icount_warp_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL_RT,
-                                         icount_warp_rt, NULL);
+                                         icount_dummy_timer, NULL);
     }
 
     icount_align_option = qemu_opt_get_bool(opts, "align", false);
diff --git a/main-loop.c b/main-loop.c
index 3997043..cd6b4cf 100644
--- a/main-loop.c
+++ b/main-loop.c
@@ -505,6 +505,9 @@ int main_loop_wait(int nonblocking)
     slirp_pollfds_poll(gpollfds, (ret < 0));
 #endif
 
+    /* CPU thread can infinitely wait for event after
+       missing the warp */
+    qemu_clock_warp(QEMU_CLOCK_VIRTUAL);
     qemu_clock_run_all_timers();
 
     return ret;

[Qemu-devel] [PATCH v17 14/21] replay: checkpoints

[Pavel Dovgalyuk]

This patch introduces checkpoints that synchronize cpu thread and iothread.
When checkpoint is met in the code all asynchronous events from the queue
are executed.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 cpus.c                   |    5 +++++
 qemu-timer.c             |   40 ++++++++++++++++++++++++++++++++++++----
 replay/replay-internal.h |    4 ++++
 replay/replay.c          |   34 ++++++++++++++++++++++++++++++++++
 replay/replay.h          |   20 ++++++++++++++++++++
 stubs/replay.c           |   11 +++++++++++
 vl.c                     |   23 +++++++++++++++++++----
 7 files changed, 129 insertions(+), 8 deletions(-)

diff --git a/cpus.c b/cpus.c
index 00d50f1..4f0bcce 100644
--- a/cpus.c
+++ b/cpus.c
@@ -402,6 +402,11 @@ void qemu_clock_warp(QEMUClockType type)
         return;
     }
 
+    /* warp clock deterministically in record/replay mode */
+    if (!replay_checkpoint(CHECKPOINT_CLOCK_WARP)) {
+        return;
+    }
+
     if (icount_sleep) {
         /*
          * If the CPUs have been sleeping, advance QEMU_CLOCK_VIRTUAL timer now.
diff --git a/qemu-timer.c b/qemu-timer.c
index c7bd643..e7a5c96 100644
--- a/qemu-timer.c
+++ b/qemu-timer.c
@@ -25,6 +25,7 @@
 #include "qemu/main-loop.h"
 #include "qemu/timer.h"
 #include "replay/replay.h"
+#include "sysemu/sysemu.h"
 
 #ifdef CONFIG_POSIX
 #include <pthread.h>
@@ -478,10 +479,34 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
     void *opaque;
 
     qemu_event_reset(&timer_list->timers_done_ev);
-    if (!timer_list->clock->enabled) {
+    if (!timer_list->clock->enabled || !timer_list->active_timers) {
         goto out;
     }
 
+    switch (timer_list->clock->type) {
+    case QEMU_CLOCK_REALTIME:
+        break;
+    default:
+    case QEMU_CLOCK_VIRTUAL:
+        if ((replay_mode != REPLAY_MODE_NONE && !runstate_is_running())
+            || !replay_checkpoint(CHECKPOINT_CLOCK_VIRTUAL)) {
+            goto out;
+        }
+        break;
+    case QEMU_CLOCK_HOST:
+        if ((replay_mode != REPLAY_MODE_NONE && !runstate_is_running())
+            || !replay_checkpoint(CHECKPOINT_CLOCK_HOST)) {
+            goto out;
+        }
+        break;
+    case QEMU_CLOCK_VIRTUAL_RT:
+        if ((replay_mode != REPLAY_MODE_NONE && !runstate_is_running())
+            || !replay_checkpoint(CHECKPOINT_CLOCK_VIRTUAL_RT)) {
+            goto out;
+        }
+        break;
+    }
+
     current_time = qemu_clock_get_ns(timer_list->clock->type);
     for(;;) {
         qemu_mutex_lock(&timer_list->active_timers_lock);
@@ -545,11 +570,18 @@ int64_t timerlistgroup_deadline_ns(QEMUTimerListGroup *tlg)
 {
     int64_t deadline = -1;
     QEMUClockType type;
+    bool play = replay_mode == REPLAY_MODE_PLAY;
     for (type = 0; type < QEMU_CLOCK_MAX; type++) {
         if (qemu_clock_use_for_deadline(tlg->tl[type]->clock->type)) {
-            deadline = qemu_soonest_timeout(deadline,
-                                            timerlist_deadline_ns(
-                                                tlg->tl[type]));
+            if (!play || tlg->tl[type]->clock->type == QEMU_CLOCK_REALTIME) {
+                deadline = qemu_soonest_timeout(deadline,
+                                                timerlist_deadline_ns(
+                                                    tlg->tl[type]));
+            } else {
+                /* Read clock from the replay file and
+                   do not calculate the deadline, based on virtual clock. */
+                qemu_clock_get_ns(tlg->tl[type]->clock->type);
+            }
         }
     }
     return deadline;
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 4414695..bf64be5 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -29,6 +29,10 @@ enum ReplayEvents {
     /* some of greater codes are reserved for clocks */
     EVENT_CLOCK,
     EVENT_CLOCK_LAST = EVENT_CLOCK + REPLAY_CLOCK_COUNT - 1,
+    /* for checkpoint event */
+    /* some of greater codes are reserved for checkpoints */
+    EVENT_CHECKPOINT,
+    EVENT_CHECKPOINT_LAST = EVENT_CHECKPOINT + CHECKPOINT_COUNT - 1,
     EVENT_COUNT
 };
 
diff --git a/replay/replay.c b/replay/replay.c
index 83b0f8c..ca9996e 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -160,3 +160,37 @@ void replay_shutdown_request(void)
         replay_mutex_unlock();
     }
 }
+
+bool replay_checkpoint(ReplayCheckpoint checkpoint)
+{
+    bool res = false;
+    assert(EVENT_CHECKPOINT + checkpoint <= EVENT_CHECKPOINT_LAST);
+    replay_save_instructions();
+
+    if (!replay_file) {
+        return true;
+    }
+
+    replay_mutex_lock();
+
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        if (replay_next_event_is(EVENT_CHECKPOINT + checkpoint)) {
+            replay_finish_event();
+        } else if (replay_data_kind != EVENT_ASYNC) {
+            res = false;
+            goto out;
+        }
+        replay_read_events(checkpoint);
+        /* replay_read_events may leave some unread events.
+           Return false if not all of the events associated with
+           checkpoint were processed */
+        res = replay_data_kind != EVENT_ASYNC;
+    } else if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_put_event(EVENT_CHECKPOINT + checkpoint);
+        replay_save_events(checkpoint);
+        res = true;
+    }
+out:
+    replay_mutex_unlock();
+    return res;
+}
diff --git a/replay/replay.h b/replay/replay.h
index fcc93d1..e2696fe 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -26,6 +26,20 @@ enum ReplayClockKind {
 };
 typedef enum ReplayClockKind ReplayClockKind;
 
+/* IDs of the checkpoints */
+enum ReplayCheckpoint {
+    CHECKPOINT_CLOCK_WARP,
+    CHECKPOINT_RESET_REQUESTED,
+    CHECKPOINT_SUSPEND_REQUESTED,
+    CHECKPOINT_CLOCK_VIRTUAL,
+    CHECKPOINT_CLOCK_HOST,
+    CHECKPOINT_CLOCK_VIRTUAL_RT,
+    CHECKPOINT_INIT,
+    CHECKPOINT_RESET,
+    CHECKPOINT_COUNT
+};
+typedef enum ReplayCheckpoint ReplayCheckpoint;
+
 extern ReplayMode replay_mode;
 
 /* Processing the instructions */
@@ -70,6 +84,12 @@ int64_t replay_read_clock(ReplayClockKind kind);
 
 /*! Called when qemu shutdown is requested. */
 void replay_shutdown_request(void);
+/*! Should be called at check points in the execution.
+    These check points are skipped, if they were not met.
+    Saves checkpoint in the SAVE mode and validates in the PLAY mode.
+    Returns 0 in PLAY mode if checkpoint was not found.
+    Returns 1 in all other cases. */
+bool replay_checkpoint(ReplayCheckpoint checkpoint);
 
 /* Asynchronous events queue */
 
diff --git a/stubs/replay.c b/stubs/replay.c
index 3e7c3fe..e737dad 100755
--- a/stubs/replay.c
+++ b/stubs/replay.c
@@ -1,5 +1,6 @@
 #include "replay/replay.h"
 #include <stdlib.h>
+#include "sysemu/sysemu.h"
 
 ReplayMode replay_mode;
 
@@ -14,3 +15,13 @@ int64_t replay_read_clock(unsigned int kind)
     abort();
     return 0;
 }
+
+bool replay_checkpoint(ReplayCheckpoint checkpoint)
+{
+    return 0;
+}
+
+int runstate_is_running(void)
+{
+    return 0;
+}
diff --git a/vl.c b/vl.c
index 9ff801d..c461a98 100644
--- a/vl.c
+++ b/vl.c
@@ -122,6 +122,7 @@ int main(int argc, char **argv)
 #include "qapi-event.h"
 #include "exec/semihost.h"
 #include "crypto/init.h"
+#include "replay/replay.h"
 
 #define MAX_VIRTIO_CONSOLES 1
 #define MAX_SCLP_CONSOLES 1
@@ -1661,15 +1662,21 @@ static void qemu_kill_report(void)
 static int qemu_reset_requested(void)
 {
     int r = reset_requested;
-    reset_requested = 0;
-    return r;
+    if (r && replay_checkpoint(CHECKPOINT_RESET_REQUESTED)) {
+        reset_requested = 0;
+        return r;
+    }
+    return false;
 }
 
 static int qemu_suspend_requested(void)
 {
     int r = suspend_requested;
-    suspend_requested = 0;
-    return r;
+    if (r && replay_checkpoint(CHECKPOINT_SUSPEND_REQUESTED)) {
+        suspend_requested = 0;
+        return r;
+    }
+    return false;
 }
 
 static WakeupReason qemu_wakeup_requested(void)
@@ -4500,6 +4507,10 @@ int main(int argc, char **argv, char **envp)
     }
     qemu_add_globals();
 
+    /* This checkpoint is required by replay to separate prior clock
+       reading from the other reads, because timer polling functions query
+       clock values from the log. */
+    replay_checkpoint(CHECKPOINT_INIT);
     qdev_machine_init();
 
     current_machine->ram_size = ram_size;
@@ -4615,6 +4626,10 @@ int main(int argc, char **argv, char **envp)
         exit(1);
     }
 
+    /* This checkpoint is required by replay to separate prior clock
+       reading from the other reads, because timer polling functions query
+       clock values from the log. */
+    replay_checkpoint(CHECKPOINT_RESET);
     qemu_system_reset(VMRESET_SILENT);
     register_global_state();
     if (loadvm) {

[Qemu-devel] [PATCH v17 15/21] bottom halves: introduce bh call function

[Pavel Dovgalyuk]

This patch introduces aio_bh_call function. It is used to execute
bottom halves as callbacks without adding them to the queue.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 async.c             |    7 ++++++-
 include/block/aio.h |    5 +++++
 2 files changed, 11 insertions(+), 1 deletions(-)

diff --git a/async.c b/async.c
index efce14b..e1db2dc 100644
--- a/async.c
+++ b/async.c
@@ -59,6 +59,11 @@ QEMUBH *aio_bh_new(AioContext *ctx, QEMUBHFunc *cb, void *opaque)
     return bh;
 }
 
+void aio_bh_call(QEMUBH *bh)
+{
+    bh->cb(bh->opaque);
+}
+
 /* Multiple occurrences of aio_bh_poll cannot be called concurrently */
 int aio_bh_poll(AioContext *ctx)
 {
@@ -84,7 +89,7 @@ int aio_bh_poll(AioContext *ctx)
                 ret = 1;
             }
             bh->idle = 0;
-            bh->cb(bh->opaque);
+            aio_bh_call(bh);
         }
     }
 
diff --git a/include/block/aio.h b/include/block/aio.h
index 400b1b0..7c31b19 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -207,6 +207,11 @@ void aio_notify(AioContext *ctx);
 void aio_notify_accept(AioContext *ctx);
 
 /**
+ * aio_bh_call: Executes callback function of the specified BH.
+ */
+void aio_bh_call(QEMUBH *bh);
+
+/**
  * aio_bh_poll: Poll bottom halves for an AioContext.
  *
  * These are internal functions used by the QEMU main loop.

[Qemu-devel] [PATCH v17 16/21] replay: ptimer

[Pavel Dovgalyuk]

This patch adds deterministic replay for hardware periodic countdown timers.
ptimer uses bottom halves layer to execute such an asynchronous callback.
We put this callback into the replay queue instead of bottom halves one.
When checkpoint is met by main loop thread, the replay queue is processed
and callback is executed. Binding callback moment to one of the checkpoints
makes it deterministic.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 hw/core/ptimer.c         |    7 ++++++-
 replay/replay-events.c   |   18 +++++++++++++++++-
 replay/replay-internal.h |    1 +
 replay/replay.h          |    2 ++
 4 files changed, 26 insertions(+), 2 deletions(-)

diff --git a/hw/core/ptimer.c b/hw/core/ptimer.c
index 8437bd6..c56078d 100644
--- a/hw/core/ptimer.c
+++ b/hw/core/ptimer.c
@@ -9,6 +9,7 @@
 #include "qemu/timer.h"
 #include "hw/ptimer.h"
 #include "qemu/host-utils.h"
+#include "replay/replay.h"
 
 struct ptimer_state
 {
@@ -27,7 +28,11 @@ struct ptimer_state
 static void ptimer_trigger(ptimer_state *s)
 {
     if (s->bh) {
-        qemu_bh_schedule(s->bh);
+        if (replay_mode != REPLAY_MODE_NONE) {
+            replay_add_ptimer_event(s->bh, replay_get_current_step());
+        } else {
+            qemu_bh_schedule(s->bh);
+        }
     }
 }
 
diff --git a/replay/replay-events.c b/replay/replay-events.c
index f973cb0..f356493 100755
--- a/replay/replay-events.c
+++ b/replay/replay-events.c
@@ -13,6 +13,7 @@
 #include "qemu/error-report.h"
 #include "replay.h"
 #include "replay-internal.h"
+#include "block/aio.h"
 
 typedef struct Event {
     ReplayAsyncEventKind event_kind;
@@ -35,6 +36,9 @@ static bool events_enabled;
 static void replay_run_event(Event *event)
 {
     switch (event->event_kind) {
+    case REPLAY_ASYNC_EVENT_PTIMER:
+        aio_bh_call(event->opaque);
+        break;
     default:
         error_report("Replay: invalid async event ID (%d) in the queue",
                     event->event_kind);
@@ -117,6 +121,11 @@ static void replay_add_event(ReplayAsyncEventKind event_kind,
     replay_mutex_unlock();
 }
 
+void replay_add_ptimer_event(void *bh, uint64_t id)
+{
+    replay_add_event(REPLAY_ASYNC_EVENT_PTIMER, bh, NULL, id);
+}
+
 static void replay_save_event(Event *event, int checkpoint)
 {
     if (replay_mode != REPLAY_MODE_PLAY) {
@@ -127,10 +136,12 @@ static void replay_save_event(Event *event, int checkpoint)
 
         /* save event-specific data */
         switch (event->event_kind) {
+        case REPLAY_ASYNC_EVENT_PTIMER:
+            replay_put_qword(event->id);
+            break;
         default:
             error_report("Unknown ID %d of replay event", read_event_kind);
             exit(1);
-            break;
         }
     }
 }
@@ -166,6 +177,11 @@ static Event *replay_read_event(int checkpoint)
 
     /* Events that has not to be in the queue */
     switch (read_event_kind) {
+    case REPLAY_ASYNC_EVENT_PTIMER:
+        if (read_id == -1) {
+            read_id = replay_get_qword();
+        }
+        break;
     default:
         error_report("Unknown ID %d of replay event", read_event_kind);
         exit(1);
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index bf64be5..89c1dc7 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -39,6 +39,7 @@ enum ReplayEvents {
 /* Asynchronous events IDs */
 
 enum ReplayAsyncEventKind {
+    REPLAY_ASYNC_EVENT_PTIMER,
     REPLAY_ASYNC_COUNT
 };
 
diff --git a/replay/replay.h b/replay/replay.h
index e2696fe..c7de964 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -97,5 +97,7 @@ bool replay_checkpoint(ReplayCheckpoint checkpoint);
 void replay_disable_events(void);
 /*! Returns true when saving events is enabled */
 bool replay_events_enabled(void);
+/*! Adds ptimer event to the queue */
+void replay_add_ptimer_event(void *bh, uint64_t id);
 
 #endif

[Qemu-devel] [PATCH v17 17/21] typedef: add typedef for QemuOpts

[Pavel Dovgalyuk]

This patch moves typedefs for QemuOpts and related types
to qemu/typedefs.h file.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 include/qemu/option.h   |    5 +----
 include/qemu/typedefs.h |    3 +++
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/qemu/option.h b/include/qemu/option.h
index 57e51c9..71f5f27 100644
--- a/include/qemu/option.h
+++ b/include/qemu/option.h
@@ -30,6 +30,7 @@
 #include "qemu/queue.h"
 #include "qapi/error.h"
 #include "qapi/qmp/qdict.h"
+#include "qemu/typedefs.h"
 
 const char *get_opt_name(char *buf, int buf_size, const char *p, char delim);
 const char *get_opt_value(char *buf, int buf_size, const char *p);
@@ -44,10 +45,6 @@ void parse_option_size(const char *name, const char *value,
 bool has_help_option(const char *param);
 bool is_valid_option_list(const char *param);
 
-typedef struct QemuOpt QemuOpt;
-typedef struct QemuOpts QemuOpts;
-typedef struct QemuOptsList QemuOptsList;
-
 enum QemuOptType {
     QEMU_OPT_STRING = 0,  /* no parsing (use string as-is)                        */
     QEMU_OPT_BOOL,        /* on/off                                               */
diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h
index f8a9dd6..d154116 100644
--- a/include/qemu/typedefs.h
+++ b/include/qemu/typedefs.h
@@ -68,6 +68,9 @@ typedef struct QEMUBH QEMUBH;
 typedef struct QemuConsole QemuConsole;
 typedef struct QEMUFile QEMUFile;
 typedef struct QEMUMachine QEMUMachine;
+typedef struct QemuOpt QemuOpt;
+typedef struct QemuOpts QemuOpts;
+typedef struct QemuOptsList QemuOptsList;
 typedef struct QEMUSGList QEMUSGList;
 typedef struct QEMUSizedBuffer QEMUSizedBuffer;
 typedef struct QEMUTimerListGroup QEMUTimerListGroup;

[Qemu-devel] [PATCH v17 18/21] replay: initialization and deinitialization

[Pavel Dovgalyuk]

This patch introduces the functions for enabling the record/replay and for
freeing the resources when simulator closes.

Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 exec.c                   |    2 +
 replay/replay-internal.h |    2 +
 replay/replay.c          |  129 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay.h          |   10 ++++
 stubs/replay.c           |    5 ++
 vl.c                     |    4 +
 6 files changed, 152 insertions(+), 0 deletions(-)

diff --git a/exec.c b/exec.c
index 54cd70a..395bcc4 100644
--- a/exec.c
+++ b/exec.c
@@ -51,6 +51,7 @@
 #include "qemu/main-loop.h"
 #include "exec/cputlb.h"
 #include "translate-all.h"
+#include "replay/replay.h"
 
 #include "exec/memory-internal.h"
 #include "exec/ram_addr.h"
@@ -841,6 +842,7 @@ void cpu_abort(CPUState *cpu, const char *fmt, ...)
     }
     va_end(ap2);
     va_end(ap);
+    replay_finish();
 #if defined(CONFIG_USER_ONLY)
     {
         struct sigaction act;
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 89c1dc7..820134f 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -33,6 +33,8 @@ enum ReplayEvents {
     /* some of greater codes are reserved for checkpoints */
     EVENT_CHECKPOINT,
     EVENT_CHECKPOINT_LAST = EVENT_CHECKPOINT + CHECKPOINT_COUNT - 1,
+    /* end of log event */
+    EVENT_END,
     EVENT_COUNT
 };
 
diff --git a/replay/replay.c b/replay/replay.c
index ca9996e..e22e595 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -16,8 +16,16 @@
 #include "qemu/main-loop.h"
 #include "sysemu/sysemu.h"
 
+/* Current version of the replay mechanism.
+   Increase it when file format changes. */
+#define REPLAY_VERSION              0xe02002
+/* Size of replay log header */
+#define HEADER_SIZE                 (sizeof(uint32_t) + sizeof(uint64_t))
+
 ReplayMode replay_mode = REPLAY_MODE_NONE;
 
+/* Name of replay file  */
+static char *replay_filename;
 ReplayState replay_state;
 
 bool replay_next_event_is(int event)
@@ -194,3 +202,124 @@ out:
     replay_mutex_unlock();
     return res;
 }
+
+static void replay_enable(const char *fname, int mode)
+{
+    const char *fmode = NULL;
+    assert(!replay_file);
+
+    switch (mode) {
+    case REPLAY_MODE_RECORD:
+        fmode = "wb";
+        break;
+    case REPLAY_MODE_PLAY:
+        fmode = "rb";
+        break;
+    default:
+        fprintf(stderr, "Replay: internal error: invalid replay mode\n");
+        exit(1);
+    }
+
+    atexit(replay_finish);
+
+    replay_mutex_init();
+
+    replay_file = fopen(fname, fmode);
+    if (replay_file == NULL) {
+        fprintf(stderr, "Replay: open %s: %s\n", fname, strerror(errno));
+        exit(1);
+    }
+
+    replay_filename = g_strdup(fname);
+
+    replay_mode = mode;
+    replay_data_kind = -1;
+    replay_state.instructions_count = 0;
+    replay_state.current_step = 0;
+
+    /* skip file header for RECORD and check it for PLAY */
+    if (replay_mode == REPLAY_MODE_RECORD) {
+        fseek(replay_file, HEADER_SIZE, SEEK_SET);
+    } else if (replay_mode == REPLAY_MODE_PLAY) {
+        unsigned int version = replay_get_dword();
+        if (version != REPLAY_VERSION) {
+            fprintf(stderr, "Replay: invalid input log file version\n");
+            exit(1);
+        }
+        /* go to the beginning */
+        fseek(replay_file, HEADER_SIZE, SEEK_SET);
+        replay_fetch_data_kind();
+    }
+
+    replay_init_events();
+}
+
+void replay_configure(QemuOpts *opts)
+{
+    const char *fname;
+    const char *rr;
+    ReplayMode mode = REPLAY_MODE_NONE;
+
+    rr = qemu_opt_get(opts, "rr");
+    if (!rr) {
+        /* Just enabling icount */
+        return;
+    } else if (!strcmp(rr, "record")) {
+        mode = REPLAY_MODE_RECORD;
+    } else if (!strcmp(rr, "replay")) {
+        mode = REPLAY_MODE_PLAY;
+    } else {
+        error_report("Invalid icount rr option: %s", rr);
+        exit(1);
+    }
+
+    fname = qemu_opt_get(opts, "rrfile");
+    if (!fname) {
+        error_report("File name not specified for replay");
+        exit(1);
+    }
+
+    replay_enable(fname, mode);
+}
+
+void replay_start(void)
+{
+    if (replay_mode == REPLAY_MODE_NONE) {
+        return;
+    }
+
+    /* Timer for snapshotting will be set up here. */
+
+    replay_enable_events();
+}
+
+void replay_finish(void)
+{
+    if (replay_mode == REPLAY_MODE_NONE) {
+        return;
+    }
+
+    replay_save_instructions();
+
+    /* finalize the file */
+    if (replay_file) {
+        if (replay_mode == REPLAY_MODE_RECORD) {
+            /* write end event */
+            replay_put_event(EVENT_END);
+
+            /* write header */
+            fseek(replay_file, 0, SEEK_SET);
+            replay_put_dword(REPLAY_VERSION);
+        }
+
+        fclose(replay_file);
+        replay_file = NULL;
+    }
+    if (replay_filename) {
+        g_free(replay_filename);
+        replay_filename = NULL;
+    }
+
+    replay_finish_events();
+    replay_mutex_destroy();
+}
diff --git a/replay/replay.h b/replay/replay.h
index c7de964..736555d 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -15,6 +15,7 @@
 #include <stdbool.h>
 #include <stdint.h>
 #include "qapi-types.h"
+#include "qemu/typedefs.h"
 
 /* replay clock kinds */
 enum ReplayClockKind {
@@ -42,6 +43,15 @@ typedef enum ReplayCheckpoint ReplayCheckpoint;
 
 extern ReplayMode replay_mode;
 
+/* Replay process control functions */
+
+/*! Enables recording or saving event log with specified parameters */
+void replay_configure(struct QemuOpts *opts);
+/*! Initializes timers used for snapshotting and enables events recording */
+void replay_start(void);
+/*! Closes replay log file and frees other resources. */
+void replay_finish(void);
+
 /* Processing the instructions */
 
 /*! Returns number of executed instructions. */
diff --git a/stubs/replay.c b/stubs/replay.c
index e737dad..5cf587b 100755
--- a/stubs/replay.c
+++ b/stubs/replay.c
@@ -25,3 +25,8 @@ int runstate_is_running(void)
 {
     return 0;
 }
+
+bool replay_events_enabled(void)
+{
+    return false;
+}
diff --git a/vl.c b/vl.c
index c461a98..cfc8d26 100644
--- a/vl.c
+++ b/vl.c
@@ -4626,6 +4626,8 @@ int main(int argc, char **argv, char **envp)
         exit(1);
     }
 
+    replay_start();
+
     /* This checkpoint is required by replay to separate prior clock
        reading from the other reads, because timer polling functions query
        clock values from the log. */
@@ -4667,6 +4669,8 @@ int main(int argc, char **argv, char **envp)
     }
 
     main_loop();
+    replay_disable_events();
+
     bdrv_close_all();
     pause_all_vcpus();
     res_free();

[Qemu-devel] [PATCH v17 19/21] replay: replay blockers for devices

[Pavel Dovgalyuk]

Some devices are not supported by record/replay subsystem.
This patch introduces replay blocker which denies starting record/replay
if such devices are included into the configuration.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 hw/bt/hci.c               |    7 +++++++
 include/qapi/qmp/qerror.h |    3 +++
 replay/replay.c           |   13 +++++++++++++
 replay/replay.h           |    3 +++
 vl.c                      |   10 ++++++++++
 5 files changed, 36 insertions(+), 0 deletions(-)

diff --git a/hw/bt/hci.c b/hw/bt/hci.c
index 7ea3dc6..9be30e1 100644
--- a/hw/bt/hci.c
+++ b/hw/bt/hci.c
@@ -23,6 +23,8 @@
 #include "hw/usb.h"
 #include "sysemu/bt.h"
 #include "hw/bt.h"
+#include "qapi/qmp/qerror.h"
+#include "replay/replay.h"
 
 struct bt_hci_s {
     uint8_t *(*evt_packet)(void *opaque);
@@ -72,6 +74,8 @@ struct bt_hci_s {
 
     struct HCIInfo info;
     struct bt_device_s device;
+
+    Error *replay_blocker;
 };
 
 #define DEFAULT_RSSI_DBM	20
@@ -2191,6 +2195,9 @@ struct HCIInfo *bt_new_hci(struct bt_scatternet_s *net)
 
     s->device.handle_destroy = bt_hci_destroy;
 
+    error_set(&s->replay_blocker, QERR_REPLAY_NOT_SUPPORTED, "bt hci");
+    replay_add_blocker(s->replay_blocker);
+
     return &s->info;
 }
 
diff --git a/include/qapi/qmp/qerror.h b/include/qapi/qmp/qerror.h
index 842b27a..0781a7f 100644
--- a/include/qapi/qmp/qerror.h
+++ b/include/qapi/qmp/qerror.h
@@ -106,4 +106,7 @@
 #define QERR_UNSUPPORTED \
     "this feature or command is not currently supported"
 
+#define QERR_REPLAY_NOT_SUPPORTED \
+    ERROR_CLASS_GENERIC_ERROR, "Record/replay feature is not supported for '%s'"
+
 #endif /* QERROR_H */
diff --git a/replay/replay.c b/replay/replay.c
index e22e595..671b7b3 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -15,6 +15,7 @@
 #include "qemu/timer.h"
 #include "qemu/main-loop.h"
 #include "sysemu/sysemu.h"
+#include "qemu/error-report.h"
 
 /* Current version of the replay mechanism.
    Increase it when file format changes. */
@@ -27,6 +28,7 @@ ReplayMode replay_mode = REPLAY_MODE_NONE;
 /* Name of replay file  */
 static char *replay_filename;
 ReplayState replay_state;
+static GSList *replay_blockers;
 
 bool replay_next_event_is(int event)
 {
@@ -288,6 +290,12 @@ void replay_start(void)
         return;
     }
 
+    if (replay_blockers) {
+        error_report("Record/replay: %s",
+                     error_get_pretty(replay_blockers->data));
+        exit(1);
+    }
+
     /* Timer for snapshotting will be set up here. */
 
     replay_enable_events();
@@ -323,3 +331,8 @@ void replay_finish(void)
     replay_finish_events();
     replay_mutex_destroy();
 }
+
+void replay_add_blocker(Error *reason)
+{
+    replay_blockers = g_slist_prepend(replay_blockers, reason);
+}
diff --git a/replay/replay.h b/replay/replay.h
index 736555d..3539d31 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -15,6 +15,7 @@
 #include <stdbool.h>
 #include <stdint.h>
 #include "qapi-types.h"
+#include "qapi/error.h"
 #include "qemu/typedefs.h"
 
 /* replay clock kinds */
@@ -51,6 +52,8 @@ void replay_configure(struct QemuOpts *opts);
 void replay_start(void);
 /*! Closes replay log file and frees other resources. */
 void replay_finish(void);
+/*! Adds replay blocker with the specified error description */
+void replay_add_blocker(Error *reason);
 
 /* Processing the instructions */
 
diff --git a/vl.c b/vl.c
index cfc8d26..7ceec83 100644
--- a/vl.c
+++ b/vl.c
@@ -123,6 +123,7 @@ int main(int argc, char **argv)
 #include "exec/semihost.h"
 #include "crypto/init.h"
 #include "replay/replay.h"
+#include "qapi/qmp/qerror.h"
 
 #define MAX_VIRTIO_CONSOLES 1
 #define MAX_SCLP_CONSOLES 1
@@ -843,7 +844,11 @@ static void configure_rtc(QemuOpts *opts)
         if (!strcmp(value, "utc")) {
             rtc_utc = 1;
         } else if (!strcmp(value, "localtime")) {
+            Error *blocker = NULL;
             rtc_utc = 0;
+            error_set(&blocker, QERR_REPLAY_NOT_SUPPORTED,
+                      "-rtc base=localtime");
+            replay_add_blocker(blocker);
         } else {
             configure_rtc_date_offset(value, 0);
         }
@@ -1246,6 +1251,11 @@ static void smp_parse(QemuOpts *opts)
         exit(1);
     }
 
+    if (smp_cpus > 1 || smp_cores > 1 || smp_threads > 1) {
+        Error *blocker = NULL;
+        error_set(&blocker, QERR_REPLAY_NOT_SUPPORTED, "smp");
+        replay_add_blocker(blocker);
+    }
 }
 
 static void realtime_init(void)

[Qemu-devel] [PATCH v17 20/21] replay: command line options

[Pavel Dovgalyuk]

This patch introduces command line options for enabling recording or replaying
virtual machine behavior. These options are added to icount command line
parameter. They include 'rr' which switches between record and replay
and 'rrfile' for specifying the filename for replay log.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 qemu-options.hx |    8 ++++++--
 replay/replay.c |    4 ++++
 vl.c            |   15 ++++++++++++---
 3 files changed, 22 insertions(+), 5 deletions(-)

diff --git a/qemu-options.hx b/qemu-options.hx
index 77f5853..1d1ef0b 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -3123,12 +3123,12 @@ re-inject them.
 ETEXI
 
 DEF("icount", HAS_ARG, QEMU_OPTION_icount, \
-    "-icount [shift=N|auto][,align=on|off][,sleep=no]\n" \
+    "-icount [shift=N|auto][,align=on|off][,sleep=no,rr=record|replay,rrfile=<filename>]\n" \
     "                enable virtual instruction counter with 2^N clock ticks per\n" \
     "                instruction, enable aligning the host and virtual clocks\n" \
     "                or disable real time cpu sleeping\n", QEMU_ARCH_ALL)
 STEXI
-@item -icount [shift=@var{N}|auto]
+@item -icount [shift=@var{N}|auto][,rr=record|replay,rrfile=@var{filename}]
 @findex -icount
 Enable virtual instruction counter.  The virtual cpu will execute one
 instruction every 2^@var{N} ns of virtual time.  If @code{auto} is specified
@@ -3157,6 +3157,10 @@ Currently this option does not work when @option{shift} is @code{auto}.
 Note: The sync algorithm will work for those shift values for which
 the guest clock runs ahead of the host clock. Typically this happens
 when the shift value is high (how high depends on the host machine).
+
+When @option{rr} option is specified deterministic record/replay is enabled.
+Replay log is written into @var{filename} file in record mode and
+read from this file in replay mode.
 ETEXI
 
 DEF("watchdog", HAS_ARG, QEMU_OPTION_watchdog, \
diff --git a/replay/replay.c b/replay/replay.c
index 671b7b3..dfa3d6f 100755
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -295,6 +295,10 @@ void replay_start(void)
                      error_get_pretty(replay_blockers->data));
         exit(1);
     }
+    if (!use_icount) {
+        error_report("Please enable icount to use record/replay");
+        exit(1);
+    }
 
     /* Timer for snapshotting will be set up here. */
 
diff --git a/vl.c b/vl.c
index 7ceec83..8df818e 100644
--- a/vl.c
+++ b/vl.c
@@ -476,6 +476,12 @@ static QemuOptsList qemu_icount_opts = {
         }, {
             .name = "sleep",
             .type = QEMU_OPT_BOOL,
+        }, {
+            .name = "rr",
+            .type = QEMU_OPT_STRING,
+        }, {
+            .name = "rrfile",
+            .type = QEMU_OPT_STRING,
         },
         { /* end of list */ }
     },
@@ -4027,6 +4033,8 @@ int main(int argc, char **argv, char **envp)
         }
     }
 
+    replay_configure(icount_opts);
+
     opts = qemu_get_machine_opts();
     optarg = qemu_opt_get(opts, "type");
     if (optarg) {
@@ -4460,9 +4468,10 @@ int main(int argc, char **argv, char **envp)
     }
 
     /* open the virtual block devices */
-    if (snapshot)
-        qemu_opts_foreach(qemu_find_opts("drive"),
-                          drive_enable_snapshot, NULL, NULL);
+    if (snapshot || replay_mode != REPLAY_MODE_NONE) {
+        qemu_opts_foreach(qemu_find_opts("drive"), drive_enable_snapshot,
+                          NULL, NULL);
+    }
     if (qemu_opts_foreach(qemu_find_opts("drive"), drive_init_func,
                           &machine_class->block_default_type, NULL)) {
         exit(1);

[Qemu-devel] [PATCH v17 21/21] replay: recording of the user input

[Pavel Dovgalyuk]

This records user input (keyboard and mouse events) in record mode and replays
these input events in replay mode.

Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
---
 include/ui/input.h       |    2 +
 replay/Makefile.objs     |    1 
 replay/replay-events.c   |   33 +++++++++
 replay/replay-input.c    |  160 ++++++++++++++++++++++++++++++++++++++++++++++
 replay/replay-internal.h |   13 ++++
 replay/replay.h          |    4 +
 ui/input.c               |   27 +++++---
 7 files changed, 232 insertions(+), 8 deletions(-)
 create mode 100755 replay/replay-input.c

diff --git a/include/ui/input.h b/include/ui/input.h
index 5d5ac00..d06a12d 100644
--- a/include/ui/input.h
+++ b/include/ui/input.h
@@ -33,7 +33,9 @@ void qemu_input_handler_bind(QemuInputHandlerState *s,
                              const char *device_id, int head,
                              Error **errp);
 void qemu_input_event_send(QemuConsole *src, InputEvent *evt);
+void qemu_input_event_send_impl(QemuConsole *src, InputEvent *evt);
 void qemu_input_event_sync(void);
+void qemu_input_event_sync_impl(void);
 
 InputEvent *qemu_input_event_new_key(KeyValue *key, bool down);
 void qemu_input_event_send_key(QemuConsole *src, KeyValue *key, bool down);
diff --git a/replay/Makefile.objs b/replay/Makefile.objs
index 4aa527e..689b1e9 100755
--- a/replay/Makefile.objs
+++ b/replay/Makefile.objs
@@ -2,4 +2,5 @@ obj-$(CONFIG_SOFTMMU) += replay.o
 obj-$(CONFIG_SOFTMMU) += replay-internal.o
 obj-$(CONFIG_SOFTMMU) += replay-events.o
 obj-$(CONFIG_SOFTMMU) += replay-time.o
+obj-$(CONFIG_SOFTMMU) += replay-input.o
 obj-$(CONFIG_USER_ONLY) += replay-user.o
diff --git a/replay/replay-events.c b/replay/replay-events.c
index f356493..23f3b12 100755
--- a/replay/replay-events.c
+++ b/replay/replay-events.c
@@ -14,6 +14,7 @@
 #include "replay.h"
 #include "replay-internal.h"
 #include "block/aio.h"
+#include "ui/input.h"
 
 typedef struct Event {
     ReplayAsyncEventKind event_kind;
@@ -39,6 +40,13 @@ static void replay_run_event(Event *event)
     case REPLAY_ASYNC_EVENT_PTIMER:
         aio_bh_call(event->opaque);
         break;
+    case REPLAY_ASYNC_EVENT_INPUT:
+        qemu_input_event_send_impl(NULL, (InputEvent *)event->opaque);
+        qapi_free_InputEvent((InputEvent *)event->opaque);
+        break;
+    case REPLAY_ASYNC_EVENT_INPUT_SYNC:
+        qemu_input_event_sync_impl();
+        break;
     default:
         error_report("Replay: invalid async event ID (%d) in the queue",
                     event->event_kind);
@@ -126,6 +134,16 @@ void replay_add_ptimer_event(void *bh, uint64_t id)
     replay_add_event(REPLAY_ASYNC_EVENT_PTIMER, bh, NULL, id);
 }
 
+void replay_add_input_event(struct InputEvent *event)
+{
+    replay_add_event(REPLAY_ASYNC_EVENT_INPUT, event, NULL, 0);
+}
+
+void replay_add_input_sync_event(void)
+{
+    replay_add_event(REPLAY_ASYNC_EVENT_INPUT_SYNC, NULL, NULL, 0);
+}
+
 static void replay_save_event(Event *event, int checkpoint)
 {
     if (replay_mode != REPLAY_MODE_PLAY) {
@@ -139,6 +157,11 @@ static void replay_save_event(Event *event, int checkpoint)
         case REPLAY_ASYNC_EVENT_PTIMER:
             replay_put_qword(event->id);
             break;
+        case REPLAY_ASYNC_EVENT_INPUT:
+            replay_save_input_event(event->opaque);
+            break;
+        case REPLAY_ASYNC_EVENT_INPUT_SYNC:
+            break;
         default:
             error_report("Unknown ID %d of replay event", read_event_kind);
             exit(1);
@@ -182,6 +205,16 @@ static Event *replay_read_event(int checkpoint)
             read_id = replay_get_qword();
         }
         break;
+    case REPLAY_ASYNC_EVENT_INPUT:
+        event = g_malloc0(sizeof(Event));
+        event->event_kind = read_event_kind;
+        event->opaque = replay_read_input_event();
+        return event;
+    case REPLAY_ASYNC_EVENT_INPUT_SYNC:
+        event = g_malloc0(sizeof(Event));
+        event->event_kind = read_event_kind;
+        event->opaque = 0;
+        return event;
     default:
         error_report("Unknown ID %d of replay event", read_event_kind);
         exit(1);
diff --git a/replay/replay-input.c b/replay/replay-input.c
new file mode 100755
index 0000000..bdb4db1
--- /dev/null
+++ b/replay/replay-input.c
@@ -0,0 +1,160 @@
+/*
+ * replay-input.c
+ *
+ * Copyright (c) 2010-2015 Institute for System Programming
+ *                         of the Russian Academy of Sciences.
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
+ * See the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "replay.h"
+#include "replay-internal.h"
+#include "qemu/notify.h"
+#include "ui/input.h"
+#include "qapi/qmp-output-visitor.h"
+#include "qapi/qmp-input-visitor.h"
+#include "qapi-visit.h"
+
+static InputEvent *qapi_clone_InputEvent(InputEvent *src)
+{
+    QmpOutputVisitor *qov;
+    QmpInputVisitor *qiv;
+    Visitor *ov, *iv;
+    QObject *obj;
+    InputEvent *dst = NULL;
+
+    qov = qmp_output_visitor_new();
+    ov = qmp_output_get_visitor(qov);
+    visit_type_InputEvent(ov, &src, NULL, &error_abort);
+    obj = qmp_output_get_qobject(qov);
+    qmp_output_visitor_cleanup(qov);
+    if (!obj) {
+        return NULL;
+    }
+
+    qiv = qmp_input_visitor_new(obj);
+    iv = qmp_input_get_visitor(qiv);
+    visit_type_InputEvent(iv, &dst, NULL, &error_abort);
+    qmp_input_visitor_cleanup(qiv);
+    qobject_decref(obj);
+
+    return dst;
+}
+
+void replay_save_input_event(InputEvent *evt)
+{
+    replay_put_dword(evt->kind);
+
+    switch (evt->kind) {
+    case INPUT_EVENT_KIND_KEY:
+        replay_put_dword(evt->key->key->kind);
+
+        switch (evt->key->key->kind) {
+        case KEY_VALUE_KIND_NUMBER:
+            replay_put_qword(evt->key->key->number);
+            replay_put_byte(evt->key->down);
+            break;
+        case KEY_VALUE_KIND_QCODE:
+            replay_put_dword(evt->key->key->qcode);
+            replay_put_byte(evt->key->down);
+            break;
+        case KEY_VALUE_KIND_MAX:
+            /* keep gcc happy */
+            break;
+        }
+        break;
+    case INPUT_EVENT_KIND_BTN:
+        replay_put_dword(evt->btn->button);
+        replay_put_byte(evt->btn->down);
+        break;
+    case INPUT_EVENT_KIND_REL:
+        replay_put_dword(evt->rel->axis);
+        replay_put_qword(evt->rel->value);
+        break;
+    case INPUT_EVENT_KIND_ABS:
+        replay_put_dword(evt->abs->axis);
+        replay_put_qword(evt->abs->value);
+        break;
+    case INPUT_EVENT_KIND_MAX:
+        /* keep gcc happy */
+        break;
+    }
+}
+
+InputEvent *replay_read_input_event(void)
+{
+    InputEvent evt;
+    KeyValue keyValue;
+    InputKeyEvent key;
+    key.key = &keyValue;
+    InputBtnEvent btn;
+    InputMoveEvent rel;
+    InputMoveEvent abs;
+
+    evt.kind = replay_get_dword();
+    switch (evt.kind) {
+    case INPUT_EVENT_KIND_KEY:
+        evt.key = &key;
+        evt.key->key->kind = replay_get_dword();
+
+        switch (evt.key->key->kind) {
+        case KEY_VALUE_KIND_NUMBER:
+            evt.key->key->number = replay_get_qword();
+            evt.key->down = replay_get_byte();
+            break;
+        case KEY_VALUE_KIND_QCODE:
+            evt.key->key->qcode = (QKeyCode)replay_get_dword();
+            evt.key->down = replay_get_byte();
+            break;
+        case KEY_VALUE_KIND_MAX:
+            /* keep gcc happy */
+            break;
+        }
+        break;
+    case INPUT_EVENT_KIND_BTN:
+        evt.btn = &btn;
+        evt.btn->button = (InputButton)replay_get_dword();
+        evt.btn->down = replay_get_byte();
+        break;
+    case INPUT_EVENT_KIND_REL:
+        evt.rel = &rel;
+        evt.rel->axis = (InputAxis)replay_get_dword();
+        evt.rel->value = replay_get_qword();
+        break;
+    case INPUT_EVENT_KIND_ABS:
+        evt.abs = &abs;
+        evt.abs->axis = (InputAxis)replay_get_dword();
+        evt.abs->value = replay_get_qword();
+        break;
+    case INPUT_EVENT_KIND_MAX:
+        /* keep gcc happy */
+        break;
+    }
+
+    return qapi_clone_InputEvent(&evt);
+}
+
+void replay_input_event(QemuConsole *src, InputEvent *evt)
+{
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        /* Nothing */
+    } else if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_add_input_event(qapi_clone_InputEvent(evt));
+    } else {
+        qemu_input_event_send_impl(src, evt);
+    }
+}
+
+void replay_input_sync_event(void)
+{
+    if (replay_mode == REPLAY_MODE_PLAY) {
+        /* Nothing */
+    } else if (replay_mode == REPLAY_MODE_RECORD) {
+        replay_add_input_sync_event();
+    } else {
+        qemu_input_event_sync_impl();
+    }
+}
diff --git a/replay/replay-internal.h b/replay/replay-internal.h
index 820134f..04d2e1b 100755
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -42,6 +42,8 @@ enum ReplayEvents {
 
 enum ReplayAsyncEventKind {
     REPLAY_ASYNC_EVENT_PTIMER,
+    REPLAY_ASYNC_EVENT_INPUT,
+    REPLAY_ASYNC_EVENT_INPUT_SYNC,
     REPLAY_ASYNC_COUNT
 };
 
@@ -124,4 +126,15 @@ void replay_save_events(int checkpoint);
 /*! Read events from the file into the input queue */
 void replay_read_events(int checkpoint);
 
+/* Input events */
+
+/*! Saves input event to the log */
+void replay_save_input_event(InputEvent *evt);
+/*! Reads input event from the log */
+InputEvent *replay_read_input_event(void);
+/*! Adds input event to the queue */
+void replay_add_input_event(struct InputEvent *event);
+/*! Adds input sync event to the queue */
+void replay_add_input_sync_event(void);
+
 #endif
diff --git a/replay/replay.h b/replay/replay.h
index 3539d31..cbb4e11 100755
--- a/replay/replay.h
+++ b/replay/replay.h
@@ -112,5 +112,9 @@ void replay_disable_events(void);
 bool replay_events_enabled(void);
 /*! Adds ptimer event to the queue */
 void replay_add_ptimer_event(void *bh, uint64_t id);
+/*! Adds input event to the queue */
+void replay_input_event(QemuConsole *src, InputEvent *evt);
+/*! Adds input sync event to the queue */
+void replay_input_sync_event(void);
 
 #endif
diff --git a/ui/input.c b/ui/input.c
index 1a552d1..9939722 100644
--- a/ui/input.c
+++ b/ui/input.c
@@ -6,6 +6,7 @@
 #include "trace.h"
 #include "ui/input.h"
 #include "ui/console.h"
+#include "replay/replay.h"
 
 struct QemuInputHandlerState {
     DeviceState       *dev;
@@ -300,14 +301,10 @@ static void qemu_input_queue_sync(struct QemuInputEventQueueHead *queue)
     QTAILQ_INSERT_TAIL(queue, item, node);
 }
 
-void qemu_input_event_send(QemuConsole *src, InputEvent *evt)
+void qemu_input_event_send_impl(QemuConsole *src, InputEvent *evt)
 {
     QemuInputHandlerState *s;
 
-    if (!runstate_is_running() && !runstate_check(RUN_STATE_SUSPENDED)) {
-        return;
-    }
-
     qemu_input_event_trace(src, evt);
 
     /* pre processing */
@@ -324,14 +321,19 @@ void qemu_input_event_send(QemuConsole *src, InputEvent *evt)
     s->events++;
 }
 
-void qemu_input_event_sync(void)
+void qemu_input_event_send(QemuConsole *src, InputEvent *evt)
 {
-    QemuInputHandlerState *s;
-
     if (!runstate_is_running() && !runstate_check(RUN_STATE_SUSPENDED)) {
         return;
     }
 
+    replay_input_event(src, evt);
+}
+
+void qemu_input_event_sync_impl(void)
+{
+    QemuInputHandlerState *s;
+
     trace_input_event_sync();
 
     QTAILQ_FOREACH(s, &handlers, node) {
@@ -345,6 +347,15 @@ void qemu_input_event_sync(void)
     }
 }
 
+void qemu_input_event_sync(void)
+{
+    if (!runstate_is_running() && !runstate_check(RUN_STATE_SUSPENDED)) {
+        return;
+    }
+
+    replay_input_sync_event();
+}
+
 InputEvent *qemu_input_event_new_key(KeyValue *key, bool down)
 {
     InputEvent *evt = g_new0(InputEvent, 1);

Re: [Qemu-devel] [PATCH v17 00/21] Deterministic replay core

[Pavel Dovgaluk]

Paolo,

Are these patches good enough?

Pavel Dovgalyuk

> -----Original Message-----
> From: Pavel Dovgalyuk [mailto:Pavel.Dovgaluk@ispras.ru]
> Sent: Monday, September 07, 2015 11:40 AM
> To: qemu-devel@nongnu.org
> Cc: edgar.iglesias@xilinx.com; peter.maydell@linaro.org; igor.rubinov@gmail.com;
> eblake@redhat.com; mark.burton@greensocs.com; real@ispras.ru; batuzovk@ispras.ru;
> maria.klimushenkova@ispras.ru; pavel.dovgaluk@ispras.ru; pbonzini@redhat.com; hines@cert.org;
> alex.bennee@linaro.org; fred.konrad@greensocs.com
> Subject: [PATCH v17 00/21] Deterministic replay core
> 
> This set of patches is related to the reverse execution and deterministic
> replay of qemu execution. This implementation of deterministic replay can
> be used for deterministic debugging of guest code through gdb remote
> interface.
> 
> Core set of patches does not include support for reverse debugging commands
> of gdb, block devices' operations, USB replay support.
> 
> These patches include only core function of the replay,
> excluding the support for replaying serial, audio, network, and USB devices'
> operations. Reverse debugging and monitor commands were also excluded to
> be submitted later as separate patches.
> 
> Execution recording writes non-deterministic events log, which can be later
> used for replaying the execution anywhere and for unlimited number of times.
> It also supports checkpointing for faster rewinding during reverse debugging.
> Execution replaying reads the log and replays all non-deterministic events
> including external input, hardware clocks, and interrupts.
> 
> Full version of deterministic replay has the following features:
>  * Deterministically replays whole system execution and all contents of the memory,
>    state of the hadrware devices, clocks, and screen of the VM.
>  * Writes execution log into the file for latter replaying for multiple times
>    on different machines.
>  * Supports i386, x86_64, ARM, PowerPC, and MIPS hardware platforms.
>  * Performs deterministic replay of all operations with keyboard and mouse
>    input devices.
>  * Supports auto-checkpointing for convenient reverse debugging.
> 
> Usage of the record/replay core:
>  * First, record the execution, by adding the following string to the command line:
>    '-icount shift=7,rr=record,rrfile=replay.bin -net none'.
>    Block devices' images are not actually changed in the recording mode,
>    because all of the changes are written to the temporary overlay file.
>  * Then you can replay it for the multiple times by using another command
>    line option: '-icount shift=7,rr=replay,rrfile=replay.bin -net none'
>  * '-net none' option should also be specified if network replay patches
>    are not applied.
>  * Do not add any disk images to VM, because they are not supported by
>    the core patches.
> 
> Papers with description of deterministic replay implementation:
> http://www.computer.org/csdl/proceedings/csmr/2012/4666/00/4666a553-abs.html
> http://dl.acm.org/citation.cfm?id=2786805.2803179
> 
> Public repository with current version of the patches:
> https://github.com/Dovgalyuk/qemu/tree/rr-17
> 
> Modifications of qemu include:
>  * wrappers for clock and time functions to save their return values in the log
>  * saving different asynchronous events (e.g. system shutdown) into the log
>  * synchronization of the threads from thread pool
>  * recording/replaying user input (mouse and keyboard)
>  * adding internal events for cpu and io synchronization
> 
> v17 changes:
>  * Removed useless stub functions (as suggested by Paolo Bonzini)
>  * Refined checkpoint-related code (as suggested by Paolo Bonzini)
>  * Improved icount processing (as suggested by Paolo Bonzini)
>  * Added checkpoint for suspend event (as suggested by Paolo Bonzini)
>  * Fixed linux-user configurations build
>  * Minor fixes
> 
> v16 changes:
>  * Several warnings were fixed
> 
> v15 changes:
>  * Tested record/replay with MIPS and PowerPC guests
>  * Published the patches on github
>  * Fixed replay mutex operation in icount mode
>  * Fixed timers processing in record/replay mode
> 
> v14 changes:
>  * Minor fixes
> 
> v13 changes:
>  * Introduced "ptimer trigger" event (as suggested by Paolo Bonzini)
> 
> v12 changes:
>  * Removed block patches from the core patches set.
> 
> v11 changes:
>  * Fixed instructions event processing.
>  * Added some mutex protection calls for replay.
>  * Fixed replaying read operations for qcow2.
>  * Fixed rtc reads on initializations stage.
>  * Eliminated some warnings in replay module.
>  * Fixed misprints in documentation for replay (as suggested by Eric Blake)
> 
> v10 changes:
>  * Fixed queue processing for bottom halves (as suggested by Paolo Bonzini)
>  * Rewritten several replay functions (as suggested by Paolo Bonzini)
>  * Some minor fixes.
> 
> v9 changes:
>  * Replaced fwrite/fread with putc/getc (as suggested by Paolo Bonzini)
>  * Stopping virtual machine in case of replay file end (as suggested by Paolo Bonzini)
>  * Removed one of the replay mutexes (as suggested by Paolo Bonzini)
>  * Fixed RCU queue for bottom halves (as suggested by Paolo Bonzini)
>  * Updated command line options' names (as suggested by Paolo Bonzini)
>  * Added design document for record/replay (as suggested by Paolo Bonzini)
>  * Simplified checkpoints for the timers
>  * Added cloning InputEvent objects for replay (as suggested by Paolo Bonzini)
>  * Added replay blockers instead of checking the command line (as suggested by Paolo Bonzini)
>  * Some functions renaming and extracting.
> 
> v8 changes:
>  * Simplified processing of the shutdown event (as suggested by Paolo Bonzini)
>  * Replaced stack of bottom halves in AIO context with QSIMPLEQ (as suggested by Paolo
> Bonzini)
>  * Moved replay_submode out of the series (as suggested by Paolo Bonzini)
>  * Moved suffix option out of the series
>  * Converted some of the defines into enums (as suggested by Paolo Bonzini)
>  * Encapsulated save_tm/read_tm calls into the single function (as suggested by Paolo Bonzini)
>  * Moved record/replay options to icount group (as suggested by Paolo Bonzini)
>  * Updated mutex protection for the events queue (as suggested by Paolo Bonzini)
>  * Added mutex to protect replay log file (as suggested by Paolo Bonzini)
>  * Minor cleanups
> 
> v7 changes:
>  * Removed patches that were applied to upstream.
> 
> v6 changes:
>  * Fixed replay stub return value (as suggested by Eric Blake)
>  * Fixed icount warping.
>  * Virtual rt clock now uses cpu_get_clock() (as suggested by Paolo Bonzini)
>  * Replated get_clock_realtime and get_clock calls with qemu clock requests (as suggested by
> Paolo Bonzini)
>  * Modified can_do_io logic to allow requesting icount from cpu_exec function (as suggested by
> Paolo Bonzini)
>  * Removed applied patches.
> 
> v5 changes:
>  * Minor changes.
>  * Used fixed-width integer types for read/write functions (as suggested by Alex Bennee)
>  * Moved savevm-related code out of the core.
>  * Added new traced clock for deterministic virtual clock warping (as suggested by Paolo
> Bonzini)
>  * Fixed exception_index reset for user mode (as suggested by Paolo Bonzini)
>  * Adopted Paolo's icount patches
>  * Fixed hardware interrupts replaying
> 
> v4 changes:
>  * Updated block drivers to support new bdrv_open interface.
>  * Moved migration patches into separate series (as suggested by Paolo Bonzini)
>  * Fixed a bug in replay_break operation.
>  * Fixed rtl8139 migration for replay.
>  * Fixed 'period' parameter processing for record mode.
>  * Fixed bug in 'reverse-stepi' implementation.
>  * Fixed replay without making any snapshots (even the starting one).
>  * Moved core replay patches into the separate series.
>  * Fixed reverse step and reverse continue support.
>  * Fixed several bugs in icount subsystem.
>  * Reusing native qemu icount for replay instructions counting.
>  * Separated core patches into their own series.
> 
> v3 changes:
>  * Fixed bug with replay of the aio write operations.
>  * Added virtual clock based on replay icount.
>  * Removed duplicated saving of interrupt_request CPU field.
>  * Fixed some coding style issues.
>  * Renamed QMP commands for controlling reverse execution (as suggested by Eric Blake)
>  * Replay mode and submode implemented as QAPI enumerations (as suggested by Eric Blake)
>  * Added description and example for replay-info command (as suggested by Eric Blake)
>  * Added information about the current breakpoint to the output of replay-info (as suggested
> by Eric Blake)
>  * Updated version id for HPET vmstate (as suggested by Paolo Bonzini)
>  * Removed static fields from parallel vmstate (as suggested by Paolo Bonzini)
>  * New vmstate fields for mc146818rtc, pckbd, kvmapic, serial, fdc, rtl8139 moved to
> subsection (as suggested by Paolo Bonzini)
>  * Disabled textmode cursor blinking, when virtual machine is stopped (as suggested by Paolo
> Bonzini)
>  * Extracted saving of exception_index to separate patch (as suggested by Paolo Bonzini)
> 
> v2 changes:
>  * Patches are split to be reviewable and bisectable (as suggested by Kirill Batuzov)
>  * Added QMP versions of replay commands (as suggested by Eric Blake)
>  * Removed some optional features of replay to make patches cleaner
>  * Minor changes and code cleanup were made
> 
> ---
> 
> Pavel Dovgalyuk (21):
>       i386: partial revert of interrupt poll fix
>       replay: global variables and function stubs
>       replay: internal functions for replay log
>       replay: introduce mutex to protect the replay log
>       replay: introduce icount event
>       cpu-exec: allow temporary disabling icount
>       cpu: replay instructions sequence
>       i386: interrupt poll processing
>       replay: interrupts and exceptions
>       replay: asynchronous events infrastructure
>       replay: recording and replaying clock ticks
>       replay: shutdown event
>       icount: improve counting for record/replay
>       replay: checkpoints
>       bottom halves: introduce bh call function
>       replay: ptimer
>       typedef: add typedef for QemuOpts
>       replay: initialization and deinitialization
>       replay: replay blockers for devices
>       replay: command line options
>       replay: recording of the user input
> 
> 
>  Makefile.target           |    1
>  async.c                   |    7 +
>  cpu-exec.c                |   63 +++++++-
>  cpus.c                    |   57 +++++---
>  docs/replay.txt           |  168 ++++++++++++++++++++++
>  exec.c                    |    2
>  hw/bt/hci.c               |    7 +
>  hw/core/ptimer.c          |    7 +
>  include/block/aio.h       |    5 +
>  include/exec/exec-all.h   |    1
>  include/qapi/qmp/qerror.h |    3
>  include/qemu/option.h     |    5 -
>  include/qemu/typedefs.h   |    3
>  include/ui/input.h        |    2
>  main-loop.c               |    3
>  qapi-schema.json          |   18 ++
>  qemu-options.hx           |    8 +
>  qemu-timer.c              |   48 +++++-
>  replay/Makefile.objs      |    6 +
>  replay/replay-events.c    |  274 ++++++++++++++++++++++++++++++++++++
>  replay/replay-input.c     |  160 +++++++++++++++++++++
>  replay/replay-internal.c  |  206 +++++++++++++++++++++++++++
>  replay/replay-internal.h  |  140 ++++++++++++++++++
>  replay/replay-time.c      |   64 ++++++++
>  replay/replay.c           |  342 +++++++++++++++++++++++++++++++++++++++++++++
>  replay/replay.h           |  120 ++++++++++++++++
>  stubs/Makefile.objs       |    1
>  stubs/replay.c            |   32 ++++
>  target-i386/cpu.c         |   10 -
>  target-i386/seg_helper.c  |    3
>  translate-all.c           |    2
>  ui/input.c                |   27 +++-
>  vl.c                      |   53 ++++++-
>  33 files changed, 1779 insertions(+), 69 deletions(-)
>  create mode 100755 docs/replay.txt
>  create mode 100755 replay/Makefile.objs
>  create mode 100755 replay/replay-events.c
>  create mode 100755 replay/replay-input.c
>  create mode 100755 replay/replay-internal.c
>  create mode 100755 replay/replay-internal.h
>  create mode 100755 replay/replay-time.c
>  create mode 100755 replay/replay.c
>  create mode 100755 replay/replay.h
>  create mode 100755 stubs/replay.c
> 
> --
> Pavel Dovgalyuk

Re: [Qemu-devel] [PATCH v17 00/21] Deterministic replay core

[Paolo Bonzini]

On 11/09/2015 07:52, Pavel Dovgaluk wrote:
> Paolo,
> 
> Are these patches good enough?

Haven't reviewed them as I already have a 50 patch queue.  I will look
at them next week.

I can already tell you that I'd like some comments in front of clock
warp calls, but that can be added as a follow up.

Paolo