All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Alex Bennée" <alex.bennee@linaro.org>
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Peter Crosthwaite" <crosthwaite.peter@gmail.com>,
	"Richard Henderson" <rth@twiddle.net>,
	"open list:Overall" <qemu-devel@nongnu.org>
Subject: [Qemu-devel] [PATCH v8 06/25] tcg: add kick timer for single-threaded vCPU emulation
Date: Fri, 27 Jan 2017 10:34:46 +0000	[thread overview]
Message-ID: <20170127103505.18606-7-alex.bennee@linaro.org> (raw)
In-Reply-To: <20170127103505.18606-1-alex.bennee@linaro.org>

Currently we rely on the side effect of the main loop grabbing the
iothread_mutex to give any long running basic block chains a kick to
ensure the next vCPU is scheduled. As this code is being re-factored and
rationalised we now do it explicitly here.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
---
v2
  - re-base fixes
  - get_ticks_per_sec() -> NANOSECONDS_PER_SEC
v3
  - add define for TCG_KICK_FREQ
  - fix checkpatch warning
v4
  - wrap next calc in inline qemu_tcg_next_kick() instead of macro
v5
  - move all kick code into own section
  - use global for timer
  - add helper functions to start/stop timer
  - stop timer when all cores paused
v7
  - checkpatch > 80 char fix
---
 cpus.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 61 insertions(+)

diff --git a/cpus.c b/cpus.c
index 76b6e04332..a98925105c 100644
--- a/cpus.c
+++ b/cpus.c
@@ -767,6 +767,53 @@ void configure_icount(QemuOpts *opts, Error **errp)
 }
 
 /***********************************************************/
+/* TCG vCPU kick timer
+ *
+ * The kick timer is responsible for moving single threaded vCPU
+ * emulation on to the next vCPU. If more than one vCPU is running a
+ * timer event with force a cpu->exit so the next vCPU can get
+ * scheduled.
+ *
+ * The timer is removed if all vCPUs are idle and restarted again once
+ * idleness is complete.
+ */
+
+static QEMUTimer *tcg_kick_vcpu_timer;
+
+static void qemu_cpu_kick_no_halt(void);
+
+#define TCG_KICK_PERIOD (NANOSECONDS_PER_SECOND / 10)
+
+static inline int64_t qemu_tcg_next_kick(void)
+{
+    return qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + TCG_KICK_PERIOD;
+}
+
+static void kick_tcg_thread(void *opaque)
+{
+    timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick());
+    qemu_cpu_kick_no_halt();
+}
+
+static void start_tcg_kick_timer(void)
+{
+    if (!tcg_kick_vcpu_timer && CPU_NEXT(first_cpu)) {
+        tcg_kick_vcpu_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL,
+                                           kick_tcg_thread, NULL);
+        timer_mod(tcg_kick_vcpu_timer, qemu_tcg_next_kick());
+    }
+}
+
+static void stop_tcg_kick_timer(void)
+{
+    if (tcg_kick_vcpu_timer) {
+        timer_del(tcg_kick_vcpu_timer);
+        tcg_kick_vcpu_timer = NULL;
+    }
+}
+
+
+/***********************************************************/
 void hw_error(const char *fmt, ...)
 {
     va_list ap;
@@ -1020,9 +1067,12 @@ static void qemu_wait_io_event_common(CPUState *cpu)
 static void qemu_tcg_wait_io_event(CPUState *cpu)
 {
     while (all_cpu_threads_idle()) {
+        stop_tcg_kick_timer();
         qemu_cond_wait(cpu->halt_cond, &qemu_global_mutex);
     }
 
+    start_tcg_kick_timer();
+
     while (iothread_requesting_mutex) {
         qemu_cond_wait(&qemu_io_proceeded_cond, &qemu_global_mutex);
     }
@@ -1222,6 +1272,15 @@ static void deal_with_unplugged_cpus(void)
     }
 }
 
+/* Single-threaded TCG
+ *
+ * In the single-threaded case each vCPU is simulated in turn. If
+ * there is more than a single vCPU we create a simple timer to kick
+ * the vCPU and ensure we don't get stuck in a tight loop in one vCPU.
+ * This is done explicitly rather than relying on side-effects
+ * elsewhere.
+ */
+
 static void *qemu_tcg_cpu_thread_fn(void *arg)
 {
     CPUState *cpu = arg;
@@ -1248,6 +1307,8 @@ static void *qemu_tcg_cpu_thread_fn(void *arg)
         }
     }
 
+    start_tcg_kick_timer();
+
     /* process any pending work */
     atomic_mb_set(&exit_request, 1);
 
-- 
2.11.0

  parent reply	other threads:[~2017-01-27 10:35 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20170127103505.18606-1-alex.bennee@linaro.org>
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 01/25] docs: new design document multi-thread-tcg.txt Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 02/25] mttcg: translate-all: Enable locking debug in a debug build Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 03/25] mttcg: Add missing tb_lock/unlock() in cpu_exec_step() Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 04/25] tcg: move TCG_MO/BAR types into own file Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 05/25] tcg: add options for enabling MTTCG Alex Bennée
2017-01-27 10:34 ` Alex Bennée [this message]
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 07/25] tcg: rename tcg_current_cpu to tcg_current_rr_cpu Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 08/25] tcg: drop global lock during TCG code execution Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 09/25] tcg: remove global exit_request Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 10/25] tcg: enable tb_lock() for SoftMMU Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 11/25] tcg: enable thread-per-vCPU Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 12/25] tcg: handle EXCP_ATOMIC exception for system emulation Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 13/25] cputlb: add assert_cpu_is_self checks Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 14/25] cputlb: tweak qemu_ram_addr_from_host_nofail reporting Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 15/25] cputlb: introduce tlb_flush_* async work Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 16/25] cputlb and arm/sparc targets: convert mmuidx flushes from varg to bitmap Alex Bennée
2017-01-27 13:03   ` Artyom Tarasenko
2017-01-27 13:27   ` Peter Maydell
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 17/25] cputlb: add tlb_flush_by_mmuidx async routines Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 18/25] cputlb: atomically update tlb fields used by tlb_reset_dirty Alex Bennée
2017-01-27 10:34 ` [Qemu-devel] [PATCH v8 19/25] cputlb: introduce tlb_flush_*_all_cpus[_synced] Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 20/25] target-arm/powerctl: defer cpu reset work to CPU context Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 21/25] target-arm: don't generate WFE/YIELD calls for MTTCG Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 22/25] target-arm/cpu.h: make ARM_CP defined consistent Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 23/25] target-arm: introduce ARM_CP_EXIT_PC Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 24/25] target-arm: ensure all cross vCPUs TLB flushes complete Alex Bennée
2017-01-27 10:35 ` [Qemu-devel] [PATCH v8 25/25] tcg: enable MTTCG by default for ARM on x86 hosts Alex Bennée
2017-01-27 16:49   ` [Qemu-devel] [Qemu-arm] " Pranith Kumar
2017-01-27 17:07     ` Alex Bennée
2017-01-27 10:38 [Qemu-devel] [PATCH v8 00/25] Remaining MTTCG Base patches and ARM enablement Alex Bennée
2017-01-27 10:39 ` [Qemu-devel] [PATCH v8 06/25] tcg: add kick timer for single-threaded vCPU emulation Alex Bennée
2017-01-29 21:00   ` Pranith Kumar

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=20170127103505.18606-7-alex.bennee@linaro.org \
    --to=alex.bennee@linaro.org \
    --cc=crosthwaite.peter@gmail.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.