All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Alex Bennée" <alex.bennee@linaro.org>
To: pbonzini@redhat.com
Cc: qemu-devel@nongnu.org, mttcg@listserver.greensocs.com,
	fred.konrad@greensocs.com, a.rigo@virtualopensystems.com,
	cota@braap.org, bobby.prani@gmail.com, nikunj@linux.vnet.ibm.com,
	mark.burton@greensocs.com, jan.kiszka@siemens.com,
	serge.fdrv@gmail.com, rth@twiddle.net, peter.maydell@linaro.org,
	claudio.fontana@huawei.com,
	"Alex Bennée" <alex.bennee@linaro.org>,
	"Peter Crosthwaite" <crosthwaite.peter@gmail.com>
Subject: [Qemu-devel] [PATCH v6 03/19] tcg: add kick timer for single-threaded vCPU emulation
Date: Wed,  9 Nov 2016 14:57:32 +0000	[thread overview]
Message-ID: <20161109145748.27282-4-alex.bennee@linaro.org> (raw)
In-Reply-To: <20161109145748.27282-1-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>

---
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
---
 cpus.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 60 insertions(+)

diff --git a/cpus.c b/cpus.c
index 73ff851..b3bf7b8 100644
--- a/cpus.c
+++ b/cpus.c
@@ -736,6 +736,52 @@ 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;
@@ -989,9 +1035,12 @@ 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);
     }
@@ -1191,6 +1240,15 @@ 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;
@@ -1217,6 +1275,8 @@ static void *qemu_tcg_cpu_thread_fn(void *arg)
         }
     }
 
+    start_tcg_kick_timer();
+
     /* process any pending work */
     atomic_mb_set(&exit_request, 1);
 
-- 
2.10.1

  parent reply	other threads:[~2016-11-09 14:58 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-09 14:57 [Qemu-devel] [PATCH v6 00/19] Remaining MTTCG Base patches and ARM enablement Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 01/19] docs: new design document multi-thread-tcg.txt Alex Bennée
2016-11-10 15:00   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 02/19] tcg: add options for enabling MTTCG Alex Bennée
2016-11-09 14:57 ` Alex Bennée [this message]
2016-11-10 15:10   ` [Qemu-devel] [PATCH v6 03/19] tcg: add kick timer for single-threaded vCPU emulation Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 04/19] tcg: rename tcg_current_cpu to tcg_current_rr_cpu Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 05/19] tcg: drop global lock during TCG code execution Alex Bennée
2016-11-10 15:18   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 06/19] tcg: remove global exit_request Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 07/19] tcg: enable tb_lock() for SoftMMU Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 08/19] tcg: enable thread-per-vCPU Alex Bennée
2016-11-10 16:35   ` Richard Henderson
2016-11-10 16:46     ` Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 09/19] tcg: handle EXCP_ATOMIC exception for system emulation Alex Bennée
2016-11-10 16:36   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 10/19] cputlb: add assert_cpu_is_self checks Alex Bennée
2016-11-10 16:39   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 11/19] cputlb: introduce tlb_flush_* async work Alex Bennée
2016-11-10 16:48   ` Richard Henderson
2016-11-10 17:34     ` Alex Bennée
2016-11-10 17:40       ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 12/19] cputlb: tweak qemu_ram_addr_from_host_nofail reporting Alex Bennée
2016-11-10 16:51   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 13/19] cputlb: atomically update tlb fields used by tlb_reset_dirty Alex Bennée
2016-11-09 19:36   ` Pranith Kumar
2016-11-10 16:14     ` Alex Bennée
2016-11-10 17:27       ` Richard Henderson
2016-11-10 18:00         ` Alex Bennée
2016-11-10 18:32           ` Richard Henderson
2016-11-10 17:23   ` Richard Henderson
2016-11-10 18:07     ` Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 14/19] target-arm/powerctl: defer cpu reset work to CPU context Alex Bennée
2016-11-10 17:35   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 15/19] target-arm/cpu: don't reset TLB structures, use cputlb to do it Alex Bennée
2016-11-10 17:48   ` Richard Henderson
2016-11-10 18:08     ` Alex Bennée
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 16/19] target-arm: ensure BQL taken for ARM_CP_IO register access Alex Bennée
2016-11-10 17:54   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 17/19] target-arm: helpers which may affect global state need the BQL Alex Bennée
2016-11-10 17:56   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 18/19] target-arm: don't generate WFE/YIELD calls for MTTCG Alex Bennée
2016-11-10 17:59   ` Richard Henderson
2016-11-09 14:57 ` [Qemu-devel] [PATCH v6 19/19] tcg: enable MTTCG by default for ARM on x86 hosts Alex Bennée
2016-11-10 18:00   ` Richard Henderson
2016-11-10 18:13     ` Alex Bennée
2016-11-10 18:41       ` Richard Henderson
2016-11-09 15:11 ` [Qemu-devel] [PATCH v6 00/19] Remaining MTTCG Base patches and ARM enablement Paolo Bonzini
2016-11-09 18:38   ` Alex Bennée
2016-11-13  5:50 ` no-reply

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=20161109145748.27282-4-alex.bennee@linaro.org \
    --to=alex.bennee@linaro.org \
    --cc=a.rigo@virtualopensystems.com \
    --cc=bobby.prani@gmail.com \
    --cc=claudio.fontana@huawei.com \
    --cc=cota@braap.org \
    --cc=crosthwaite.peter@gmail.com \
    --cc=fred.konrad@greensocs.com \
    --cc=jan.kiszka@siemens.com \
    --cc=mark.burton@greensocs.com \
    --cc=mttcg@listserver.greensocs.com \
    --cc=nikunj@linux.vnet.ibm.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=serge.fdrv@gmail.com \
    /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.