From: "Alex Bennée" <alex.bennee@linaro.org>
To: pbonzini@redhat.com, boost.lists@gmail.com, pavel.dovgaluk@ispras.ru
Cc: cota@braap.org, qemu-devel@nongnu.org,
"Alex Bennée" <alex.bennee@linaro.org>
Subject: [Qemu-devel] [RFC PATCH v1 5/9] replay: make locking visible outside replay code
Date: Fri, 5 May 2017 11:38:18 +0100 [thread overview]
Message-ID: <20170505103822.20641-6-alex.bennee@linaro.org> (raw)
In-Reply-To: <20170505103822.20641-1-alex.bennee@linaro.org>
The replay_mutex_lock/unlock/locked functions are now going to be used
for ensuring lock-step behaviour between the two threads. Make them
public API functions and also provide stubs for non-QEMU builds on
common paths.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
include/sysemu/replay.h | 14 ++++++++++++++
replay/replay-internal.c | 5 ++---
replay/replay-internal.h | 5 ++---
stubs/replay.c | 15 +++++++++++++++
4 files changed, 33 insertions(+), 6 deletions(-)
diff --git a/include/sysemu/replay.h b/include/sysemu/replay.h
index f1c0712795..08bffa2dc2 100644
--- a/include/sysemu/replay.h
+++ b/include/sysemu/replay.h
@@ -46,6 +46,20 @@ extern ReplayMode replay_mode;
/* Name of the initial VM snapshot */
extern char *replay_snapshot;
+/* Replay locking
+ *
+ * The locks are needed to protect the shared structures and log file
+ * when doing record/replay. They also are the main sync-point between
+ * the main-loop thread and the vCPU thread. This was a role
+ * previously filled by the BQL which has been busy trying to reduce
+ * its impact across the code. This ensures blocks of events stay
+ * sequential and reproducible.
+ */
+
+void replay_mutex_lock(void);
+void replay_mutex_unlock(void);
+bool replay_mutex_locked(void);
+
/* Replay process control functions */
/*! Enables recording or saving event log with specified parameters */
diff --git a/replay/replay-internal.c b/replay/replay-internal.c
index 157c863e25..e6b2fdb6c1 100644
--- a/replay/replay-internal.c
+++ b/replay/replay-internal.c
@@ -181,7 +181,7 @@ void replay_mutex_destroy(void)
static __thread bool replay_locked;
-static bool replay_mutex_locked(void)
+bool replay_mutex_locked(void)
{
return replay_locked;
}
@@ -204,7 +204,7 @@ void replay_mutex_unlock(void)
void replay_save_instructions(void)
{
if (replay_file && replay_mode == REPLAY_MODE_RECORD) {
- replay_mutex_lock();
+ g_assert(replay_mutex_locked());
int diff = (int)(replay_get_current_step() - replay_state.current_step);
/* Time can only go forward */
@@ -215,6 +215,5 @@ void replay_save_instructions(void)
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 ed66ed803c..b0cf78f16f 100644
--- a/replay/replay-internal.h
+++ b/replay/replay-internal.h
@@ -97,12 +97,11 @@ 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 */
+/* Mutex functions for protecting replay log file and ensuring
+ * synchronisation between vCPU and main-loop threads. */
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);
diff --git a/stubs/replay.c b/stubs/replay.c
index 9c8aa48c9c..39c5746043 100644
--- a/stubs/replay.c
+++ b/stubs/replay.c
@@ -72,3 +72,18 @@ uint64_t blkreplay_next_id(void)
{
return 0;
}
+
+void replay_mutex_lock(void)
+{
+ abort();
+}
+
+void replay_mutex_unlock(void)
+{
+ abort();
+}
+
+bool replay_mutex_locked(void)
+{
+ return false;
+}
--
2.11.0
next prev parent reply other threads:[~2017-05-05 10:37 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-05-05 10:38 [Qemu-devel] [RFC PATCH v1 0/9] BQL and Replay Lock changes Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 1/9] target/arm/arm-powertctl: drop BQL assertions Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 2/9] cpus: push BQL lock to qemu_*_wait_io_event Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 3/9] cpus: only take BQL for sleeping threads Alex Bennée
2017-05-05 15:15 ` Paolo Bonzini
2017-05-05 15:28 ` Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 4/9] replay/replay-internal.c: track holding of replay_lock Alex Bennée
2017-05-05 10:38 ` Alex Bennée [this message]
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 6/9] replay: push replay_mutex_lock up the call tree Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 7/9] scripts/qemu-gdb: add simple tcg lock status helper Alex Bennée
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 8/9] util/qemu-thread-*: add qemu_lock, locked and unlock trace events Alex Bennée
2017-05-05 15:17 ` Paolo Bonzini
2017-05-05 15:59 ` Alex Bennée
2017-05-06 8:19 ` Paolo Bonzini
2017-05-08 17:52 ` Stefan Hajnoczi
2017-05-09 13:50 ` Alex Bennée
2017-05-09 13:55 ` Paolo Bonzini
2017-05-05 10:38 ` [Qemu-devel] [RFC PATCH v1 9/9] scripts/analyse-locks-simpletrace.py: script to analyse lock times Alex Bennée
2017-05-05 15:38 ` [Qemu-devel] [RFC PATCH v1 0/9] BQL and Replay Lock changes no-reply
2017-05-10 13:51 ` Pavel Dovgalyuk
2017-05-10 15:24 ` Alex Bennée
2017-05-11 11:23 ` Pavel Dovgalyuk
2017-06-05 10:52 ` Pavel Dovgalyuk
2017-06-06 9:34 ` Alex Bennée
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=20170505103822.20641-6-alex.bennee@linaro.org \
--to=alex.bennee@linaro.org \
--cc=boost.lists@gmail.com \
--cc=cota@braap.org \
--cc=pavel.dovgaluk@ispras.ru \
--cc=pbonzini@redhat.com \
--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).