qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Pavel Dovgalyuk" <Pavel.Dovgaluk@gmail.com>,
	"Alex Bennée" <alex.bennee@linaro.org>,
	"Pavel Dovgalyuk" <Pavel.Dovgaluk@ispras.ru>
Subject: [PULL 07/31] replay: synchronize on every virtual timer callback
Date: Wed, 24 Jun 2020 06:50:24 -0400	[thread overview]
Message-ID: <20200624105048.375353-7-pbonzini@redhat.com> (raw)
In-Reply-To: <20200624104917.375143-1-pbonzini@redhat.com>

From: Pavel Dovgalyuk <Pavel.Dovgaluk@gmail.com>

Sometimes virtual timer callbacks depend on order
of virtual timer processing and warping of virtual clock.
Therefore every callback should be logged to make replay deterministic.
This patch creates a checkpoint before every virtual timer callback.
With these checkpoints virtual timers processing and clock warping
events order is completely deterministic.

Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Acked-by: Alex Bennée <alex.bennee@linaro.org>

--

v2:
  - remove mutex lock/unlock for virtual clock checkpoint since it is
    not process any asynchronous events (commit ca9759c2a92f528f256fef0e3922416f7bb47bf9)
  - bump record/replay log file version
Message-Id: <159012932716.27256.8854065545365559921.stgit@pasha-ThinkPad-X280>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 replay/replay.c   |  2 +-
 util/qemu-timer.c | 32 +++++++++-----------------------
 2 files changed, 10 insertions(+), 24 deletions(-)

diff --git a/replay/replay.c b/replay/replay.c
index 7d93746c73..83ed9e0e24 100644
--- a/replay/replay.c
+++ b/replay/replay.c
@@ -22,7 +22,7 @@
 
 /* Current version of the replay mechanism.
    Increase it when file format changes. */
-#define REPLAY_VERSION              0xe02009
+#define REPLAY_VERSION              0xe0200a
 /* Size of replay log header */
 #define HEADER_SIZE                 (sizeof(uint32_t) + sizeof(uint64_t))
 
diff --git a/util/qemu-timer.c b/util/qemu-timer.c
index b6575a2cd5..f62b4feecd 100644
--- a/util/qemu-timer.c
+++ b/util/qemu-timer.c
@@ -501,7 +501,6 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
     bool progress = false;
     QEMUTimerCB *cb;
     void *opaque;
-    bool need_replay_checkpoint = false;
 
     if (!atomic_read(&timer_list->active_timers)) {
         return false;
@@ -517,16 +516,6 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
         break;
     default:
     case QEMU_CLOCK_VIRTUAL:
-        if (replay_mode != REPLAY_MODE_NONE) {
-            /* Checkpoint for virtual clock is redundant in cases where
-             * it's being triggered with only non-EXTERNAL timers, because
-             * these timers don't change guest state directly.
-             * Since it has conditional dependence on specific timers, it is
-             * subject to race conditions and requires special handling.
-             * See below.
-             */
-            need_replay_checkpoint = true;
-        }
         break;
     case QEMU_CLOCK_HOST:
         if (!replay_checkpoint(CHECKPOINT_CLOCK_HOST)) {
@@ -559,19 +548,16 @@ bool timerlist_run_timers(QEMUTimerList *timer_list)
              */
             break;
         }
-        if (need_replay_checkpoint
-                && !(ts->attributes & QEMU_TIMER_ATTR_EXTERNAL)) {
-            /* once we got here, checkpoint clock only once */
-            need_replay_checkpoint = false;
+        /* Checkpoint for virtual clock is redundant in cases where
+         * it's being triggered with only non-EXTERNAL timers, because
+         * these timers don't change guest state directly.
+         */
+        if (replay_mode != REPLAY_MODE_NONE
+            && timer_list->clock->type == QEMU_CLOCK_VIRTUAL
+            && !(ts->attributes & QEMU_TIMER_ATTR_EXTERNAL)
+            && !replay_checkpoint(CHECKPOINT_CLOCK_VIRTUAL)) {
             qemu_mutex_unlock(&timer_list->active_timers_lock);
-            if (!replay_checkpoint(CHECKPOINT_CLOCK_VIRTUAL)) {
-                goto out;
-            }
-            qemu_mutex_lock(&timer_list->active_timers_lock);
-            /* The lock was released; start over again in case the list was
-             * modified.
-             */
-            continue;
+            goto out;
         }
 
         /* remove timer from the list before calling the callback */
-- 
2.26.2




  parent reply	other threads:[~2020-06-24 11:05 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-24 10:48 [PULL 00/31] Misc patches for 2020-06-24 Paolo Bonzini
2020-06-24 10:50 ` [PULL 01/31] kvm: support to get/set dirty log initial-all-set capability Paolo Bonzini
2020-06-24 10:50 ` [PULL 02/31] util/getauxval: Porting to FreeBSD getauxval feature Paolo Bonzini
2020-06-24 10:50 ` [PULL 03/31] libqos: usb-hcd-ehci: use 32-bit write for config register Paolo Bonzini
2020-06-24 10:50 ` [PULL 04/31] libqos: pci-pc: use 32-bit write for EJ register Paolo Bonzini
2020-06-24 10:50 ` [PULL 05/31] memory: Revert "memory: accept mismatching sizes in memory_region_access_valid" Paolo Bonzini
2020-06-24 10:50 ` [PULL 06/31] replay: notify the main loop when there are no instructions Paolo Bonzini
2020-06-24 10:50 ` Paolo Bonzini [this message]
2020-06-24 10:50 ` [PULL 08/31] configure: add libdaxctl support Paolo Bonzini
2020-06-24 10:50 ` [PULL 09/31] exec: fetch the alignment of Linux devdax pmem character device nodes Paolo Bonzini
2020-06-24 10:50 ` [PULL 10/31] docs/nvdimm: add description of alignment requirement of device dax Paolo Bonzini
2020-06-24 10:50 ` [PULL 11/31] hw/scsi/megasas: Fix possible out-of-bounds array access in tracepoints Paolo Bonzini
2020-06-24 10:50 ` [PULL 12/31] Makefile: Install qemu-[qmp/ga]-ref.* into the directory "interop" Paolo Bonzini
2020-06-24 10:50 ` [PULL 13/31] xen: Actually fix build without passthrough Paolo Bonzini
2020-06-24 10:50 ` [PULL 14/31] target/i386: reimplement f2xm1 using floatx80 operations Paolo Bonzini
2020-07-14 14:09   ` Laszlo Ersek
2020-06-24 10:50 ` [PULL 15/31] softfloat: merge floatx80_mod and floatx80_rem Paolo Bonzini
2020-06-24 10:50 ` [PULL 16/31] softfloat: fix floatx80 remainder pseudo-denormal check for zero Paolo Bonzini
2020-06-24 10:50 ` [PULL 17/31] softfloat: do not return pseudo-denormal from floatx80 remainder Paolo Bonzini
2020-06-24 10:50 ` [PULL 18/31] softfloat: do not set denominator high bit for " Paolo Bonzini
2020-06-24 10:50 ` [PULL 19/31] softfloat: return low bits of quotient from floatx80_modrem Paolo Bonzini
2020-06-24 10:50 ` [PULL 20/31] target/i386: reimplement fprem, fprem1 using floatx80 operations Paolo Bonzini
2020-06-24 10:50 ` [PULL 21/31] target/i386: reimplement fyl2xp1 " Paolo Bonzini
2020-06-24 10:50 ` [PULL 22/31] target/i386: reimplement fyl2x " Paolo Bonzini
2020-06-24 10:50 ` [PULL 23/31] target/i386: reimplement fpatan " Paolo Bonzini
2020-06-24 10:50 ` [PULL 24/31] target/i386: Add notes for versioned CPU models Paolo Bonzini
2020-06-24 10:50 ` [PULL 25/31] osdep: Make MIN/MAX evaluate arguments only once Paolo Bonzini
2020-06-24 12:13   ` Eric Blake
2020-06-24 12:21     ` Daniel P. Berrangé
2020-06-24 13:19     ` Philippe Mathieu-Daudé
2020-06-24 10:50 ` [PULL 26/31] numa: forbid '-numa node, mem' for 5.1 and newer machine types Paolo Bonzini
2020-06-24 10:50 ` [PULL 27/31] kvm: i386: allow TSC to differ by NTP correction bounds without TSC scaling Paolo Bonzini
2020-06-24 10:50 ` [PULL 28/31] hyperv: vmbus: Remove the 2nd IRQ Paolo Bonzini
2020-06-24 10:50 ` [PULL 29/31] vmport: move compat properties to hw_compat_5_0 Paolo Bonzini
2020-06-24 10:50 ` [PULL 30/31] ibex_uart: fix XOR-as-pow Paolo Bonzini
2020-06-24 10:50 ` [PULL 31/31] i386: Mask SVM features if nested SVM is disabled Paolo Bonzini
2020-06-24 11:29 ` [PULL 00/31] Misc patches for 2020-06-24 no-reply
2020-06-25 15:50 ` Peter Maydell
2020-06-25 16:33   ` Eric Blake

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200624105048.375353-7-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=Pavel.Dovgaluk@gmail.com \
    --cc=Pavel.Dovgaluk@ispras.ru \
    --cc=alex.bennee@linaro.org \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).