From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 14/43] tcg: synchronize exit_request and tcg_current_cpu accesses
Date: Wed, 9 Sep 2015 15:49:44 +0200 [thread overview]
Message-ID: <1441806613-13775-15-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1441806613-13775-1-git-send-email-pbonzini@redhat.com>
Synchronize the remaining pair of accesses in cpu_signal. These should
be necessary on Windows as well, at least in theory. Probably
SuspendProcess and ResumeProcess introduce some implicit memory
barrier.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
cpu-exec.c | 2 +-
cpus.c | 14 ++++++++++----
2 files changed, 11 insertions(+), 5 deletions(-)
diff --git a/cpu-exec.c b/cpu-exec.c
index e24c640..ef9d745 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -372,7 +372,7 @@ int cpu_exec(CPUState *cpu)
atomic_mb_set(&tcg_current_cpu, cpu);
rcu_read_lock();
- if (unlikely(exit_request)) {
+ if (unlikely(atomic_mb_read(&exit_request))) {
cpu->exit_request = 1;
}
diff --git a/cpus.c b/cpus.c
index 6cebb7a..dd2fc29 100644
--- a/cpus.c
+++ b/cpus.c
@@ -663,11 +663,15 @@ static void cpu_handle_guest_debug(CPUState *cpu)
static void cpu_signal(int sig)
{
- CPUState *cpu = atomic_mb_read(&tcg_current_cpu);
+ CPUState *cpu;
+ /* Ensure whatever caused the exit has reached the CPU threads before
+ * writing exit_request.
+ */
+ atomic_mb_set(&exit_request, 1);
+ cpu = atomic_mb_read(&tcg_current_cpu);
if (cpu) {
cpu_exit(cpu);
}
- exit_request = 1;
}
#ifdef CONFIG_LINUX
@@ -1063,7 +1067,7 @@ static void *qemu_tcg_cpu_thread_fn(void *arg)
}
/* process any pending work */
- exit_request = 1;
+ atomic_mb_set(&exit_request, 1);
while (1) {
tcg_exec_all();
@@ -1441,7 +1445,9 @@ static void tcg_exec_all(void)
break;
}
}
- exit_request = 0;
+
+ /* Pairs with smp_wmb in qemu_cpu_kick. */
+ atomic_mb_set(&exit_request, 0);
}
void list_cpus(FILE *f, fprintf_function cpu_fprintf, const char *optarg)
--
2.4.3
next prev parent reply other threads:[~2015-09-09 13:50 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-09-09 13:49 [Qemu-devel] [PULL 00/43] First batch of misc changes for 2.5 (2015-09-09) Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 01/43] qemu-thread: handle spurious futex_wait wakeups Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 02/43] seqlock: add missing 'inline' to seqlock_read_retry Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 03/43] seqlock: read sequence number atomically Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 04/43] cpus.c: qemu_mutex_lock_iothread fix race condition at cpu thread init Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 05/43] iohandler: Use aio API Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 06/43] block/iscsi: validate block size returned from target Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 07/43] Added generic panic handler qemu_system_guest_panicked() Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 08/43] i8257: rewrite DMA_schedule to avoid hooking into the CPU loop Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 09/43] i8257: remove cpu_request_exit irq Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 10/43] tcg: introduce tcg_current_cpu Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 11/43] remove qemu/tls.h Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 12/43] tcg: assign cpu->current_tb in a simpler place Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 13/43] tcg: synchronize cpu->exit_request and cpu->tcg_exit_req accesses Paolo Bonzini
2015-09-09 13:49 ` Paolo Bonzini [this message]
2015-09-09 13:49 ` [Qemu-devel] [PULL 15/43] use qemu_cpu_kick instead of cpu_exit or qemu_cpu_kick_thread Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 16/43] tcg: signal-free qemu_cpu_kick Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 17/43] Move RAMBlock and ram_list to ram_addr.h Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 18/43] Makefile.target: include top level build dir in vpath Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 19/43] rcu: init rcu_registry_lock after fork Paolo Bonzini
2015-09-16 12:37 ` Gerd Hoffmann
2015-09-16 12:38 ` Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 20/43] rcu: fix comment with s/rcu_gp_lock/rcu_registry_lock/ Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 21/43] linux-user: call rcu_(un)register_thread on pthread_(exit|create) Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 22/43] translate-all: remove obsolete comment about l1_map Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 23/43] cutils: Add qemu_strtol() wrapper Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 24/43] cutils: Add qemu_strtoul() wrapper Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 25/43] cutils: Add qemu_strtoll() wrapper Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 26/43] cutils: Add qemu_strtoull() wrapper Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 27/43] qmp: Add example usage of strto*l() qemu wrapper Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 28/43] CODING_STYLE: update mixed declaration rules Paolo Bonzini
2015-09-09 13:49 ` [Qemu-devel] [PULL 29/43] checkpatch: adapt some tests to QEMU Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 30/43] checkpatch: remove tests that are not relevant outside the kernel Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 31/43] vhost-scsi: fix wrong vhost-scsi firmware path Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 32/43] configure: factor out adding disas configure Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 33/43] add macro file for coccinelle Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 34/43] configure: Add support for jemalloc Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 35/43] scripts/dump-guest-memory.py: fix after RAMBlock change Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 36/43] cpus: protect work list with work_mutex Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 37/43] cpus: remove tcg_halt_cond and tcg_cpu_thread globals Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 38/43] replace spinlock by QemuMutex Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 39/43] remove unused spinlock Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 40/43] tcg: add memory barriers in page_find_alloc accesses Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 41/43] tcg: comment on which functions have to be called with mmap_lock held Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 42/43] exec: make mmap_lock/mmap_unlock globally available Paolo Bonzini
2015-09-09 13:50 ` [Qemu-devel] [PULL 43/43] cpu-exec: fix lock hierarchy for user-mode emulation Paolo Bonzini
2015-09-09 18:41 ` [Qemu-devel] [PULL 00/43] First batch of misc changes for 2.5 (2015-09-09) Peter Maydell
2015-09-09 19:25 ` Paolo Bonzini
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=1441806613-13775-15-git-send-email-pbonzini@redhat.com \
--to=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).