* [Qemu-devel] 答复: Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop
@ 2017-08-10 9:17 wang.yong155
2017-08-10 9:52 ` Paolo Bonzini
0 siblings, 1 reply; 2+ messages in thread
From: wang.yong155 @ 2017-08-10 9:17 UTC (permalink / raw)
To: pbonzini
Cc: stefanha, famz, jasowang, wang.guang55, zhangchen.fnst,
zhang.zhanghailiang, lizhijian, qemu-devel
>Nice. Just a note, I think an iothread should have its own (optional)
>GMainContext accessed with iothread_get_g_main_context(iothread). When
>you call it for the first time, the iothread:
>
>1) creates a GMainContext
>
>2) adds the AioContext as a GSource in the GMainContext
>
>3) asks iothread_run to switch from the AioContext loop to the
>GMainContext loop
>
>
>To simplify thread-safety:
>
>1) the GMainContext can be wrapped with a GOnce
>
>2) the GOnce callback can leave steps 2 and 3 to a bottom half.
>
Thanks, I will submit a patch v2.
WangYong
原始邮件
发件人: <pbonzini@redhat.com>
收件人:王勇10170530
抄送人: <stefanha@redhat.com> <famz@redhat.com> <jasowang@redhat.com>王广10165992 <zhangchen.fnst@cn.fujitsu.com> <zhang.zhanghailiang@huawei.com> <lizhijian@cn.fujitsu.com> <qemu-devel@nongnu.org>
日 期 :2017年08月10日 16:51
主 题 :Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop
----- Original Message -----
> From: "Wang yong" <wang.yong155@zte.com.cn>
> To: pbonzini@redhat.com, stefanha@redhat.com, famz@redhat.com, jasowang@redhat.com
> Cc: "wang yong155" <wang.yong155@zte.com.cn>, "wang guang55" <wang.guang55@zte.com.cn>, "zhangchen fnst"
> <zhangchen.fnst@cn.fujitsu.com>, "zhang zhanghailiang" <zhang.zhanghailiang@huawei.com>, lizhijian@cn.fujitsu.com,
> qemu-devel@nongnu.org
> Sent: Friday, August 11, 2017 2:29:15 AM
> Subject: [PATCH] qemu-iothread: IOThread supports the GMainContext event loop
>
> From: Wang Yong<wang.yong155@zte.com.cn>
>
> IOThread uses AioContext event loop and does not run a GMainContext.
> Therefore,chardev cannot work in IOThread,such as the chardev is
> used for colo-compare packets reception.
>
> This patch makes the IOThread run the GMainContext event loop,
> chardev and IOThread can work together.
>
> Signed-off-by: Wang Yong<wang.yong155@zte.com.cn>
> Signed-off-by: Wang Guang<wang.guang55@zte.com.cn>
Nice. Just a note, I think an iothread should have its own (optional)
GMainContext accessed with iothread_get_g_main_context(iothread). When
you call it for the first time, the iothread:
1) creates a GMainContext
2) adds the AioContext as a GSource in the GMainContext
3) asks iothread_run to switch from the AioContext loop to the
GMainContext loop
To simplify thread-safety:
1) the GMainContext can be wrapped with a GOnce
2) the GOnce callback can leave steps 2 and 3 to a bottom half.
Paolo
> ---
> include/sysemu/iothread.h | 1 +
> iothread.c | 13 +++++++++++++
> 2 files changed, 14 insertions(+)
>
> diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h
> index e6da1a4..ffe4e8a 100644
> --- a/include/sysemu/iothread.h
> +++ b/include/sysemu/iothread.h
> @@ -24,6 +24,7 @@ typedef struct {
>
> QemuThread thread
> AioContext *ctx
> + GMainLoop *loop
> QemuMutex init_done_lock
> QemuCond init_done_cond /* is thread initialization done? */
> bool stopping
> diff --git a/iothread.c b/iothread.c
> index beeb870..b6f3c3c 100644
> --- a/iothread.c
> +++ b/iothread.c
> @@ -46,6 +46,7 @@ AioContext *qemu_get_current_aio_context(void)
> static void *iothread_run(void *opaque)
> {
> IOThread *iothread = opaque
> + GMainContext *context
>
> rcu_register_thread()
>
> @@ -57,6 +58,15 @@ static void *iothread_run(void *opaque)
>
> while (!atomic_read(&iothread->stopping)) {
> aio_poll(iothread->ctx, true)
> +
> + context = iothread->ctx->source.context
> + if (context) {
> + iothread->loop = g_main_loop_new(context, TRUE)
> + g_main_loop_run(iothread->loop)
> +
> + g_main_loop_unref(iothread->loop)
> + g_main_context_unref(context)
> + }
> }
>
> rcu_unregister_thread()
> @@ -72,6 +82,9 @@ static int iothread_stop(Object *object, void *opaque)
> return 0
> }
> iothread->stopping = true
> + if (iothread->loop) {
> + g_main_loop_quit(iothread->loop)
> + }
> aio_notify(iothread->ctx)
> qemu_thread_join(&iothread->thread)
> return 0
> --
> 1.8.3.1
>
>
>
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [Qemu-devel] 答复: Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop
2017-08-10 9:17 [Qemu-devel] 答复: Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop wang.yong155
@ 2017-08-10 9:52 ` Paolo Bonzini
0 siblings, 0 replies; 2+ messages in thread
From: Paolo Bonzini @ 2017-08-10 9:52 UTC (permalink / raw)
To: wang.yong155
Cc: lizhijian, famz, zhang.zhanghailiang, zhangchen.fnst,
wang.guang55, jasowang, qemu-devel, stefanha
On 10/08/2017 11:17, wang.yong155@zte.com.cn wrote:
>> Nice. Just a note, I think an iothread should have its own (optional)
>
>> GMainContext accessed with iothread_get_g_main_context(iothread). When
>
>> you call it for the first time, the iothread:
>
>>
>
>> 1) creates a GMainContext
>
>>
>
>> 2) adds the AioContext as a GSource in the GMainContext
>
>>
>
>> 3) asks iothread_run to switch from the AioContext loop to the
>
>> GMainContext loop
>
>>
>
>>
>
>> To simplify thread-safety:
>
>>
>
>> 1) the GMainContext can be wrapped with a GOnce
>
>>
>
>> 2) the GOnce callback can leave steps 2 and 3 to a bottom half.
>
>>
>
> Thanks, I will submit a patch v2.
>
>
>
>
> WangYong
>
>
>
>
>
>
> 原始邮件
>
>
>
> 发件人: <pbonzini@redhat.com>
> 收件人:王勇10170530
> 抄送人: <stefanha@redhat.com> <famz@redhat.com> <jasowang@redhat.com>王广10165992 <zhangchen.fnst@cn.fujitsu.com> <zhang.zhanghailiang@huawei.com> <lizhijian@cn.fujitsu.com> <qemu-devel@nongnu.org>
> 日 期 :2017年08月10日 16:51
> 主 题 :Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop
>
>
>
>
>
>
>
> ----- Original Message -----
>> From: "Wang yong" <wang.yong155@zte.com.cn>
>> To: pbonzini@redhat.com, stefanha@redhat.com, famz@redhat.com, jasowang@redhat.com
>> Cc: "wang yong155" <wang.yong155@zte.com.cn>, "wang guang55" <wang.guang55@zte.com.cn>, "zhangchen fnst"
>> <zhangchen.fnst@cn.fujitsu.com>, "zhang zhanghailiang" <zhang.zhanghailiang@huawei.com>, lizhijian@cn.fujitsu.com,
>> qemu-devel@nongnu.org
>> Sent: Friday, August 11, 2017 2:29:15 AM
>> Subject: [PATCH] qemu-iothread: IOThread supports the GMainContext event loop
>>
>> From: Wang Yong<wang.yong155@zte.com.cn>
>>
>> IOThread uses AioContext event loop and does not run a GMainContext.
>> Therefore,chardev cannot work in IOThread,such as the chardev is
>> used for colo-compare packets reception.
>>
>> This patch makes the IOThread run the GMainContext event loop,
>> chardev and IOThread can work together.
>>
>> Signed-off-by: Wang Yong<wang.yong155@zte.com.cn>
>> Signed-off-by: Wang Guang<wang.guang55@zte.com.cn>
>
> Nice. Just a note, I think an iothread should have its own (optional)
> GMainContext accessed with iothread_get_g_main_context(iothread). When
> you call it for the first time, the iothread:
>
> 1) creates a GMainContext
>
> 2) adds the AioContext as a GSource in the GMainContext
>
> 3) asks iothread_run to switch from the AioContext loop to the
> GMainContext loop
>
>
> To simplify thread-safety:
>
> 1) the GMainContext can be wrapped with a GOnce
>
> 2) the GOnce callback can leave steps 2 and 3 to a bottom half.
Thanks. Another thing, please add a call to
g_main_context_push_thread_default() and
g_main_context_pop_thread_default().
Thanks,
Paolo
>> ---
>> include/sysemu/iothread.h | 1 +
>> iothread.c | 13 +++++++++++++
>> 2 files changed, 14 insertions(+)
>>
>> diff --git a/include/sysemu/iothread.h b/include/sysemu/iothread.h
>> index e6da1a4..ffe4e8a 100644
>> --- a/include/sysemu/iothread.h
>> +++ b/include/sysemu/iothread.h
>> @@ -24,6 +24,7 @@ typedef struct {
>>
>> QemuThread thread
>> AioContext *ctx
>> + GMainLoop *loop
>> QemuMutex init_done_lock
>> QemuCond init_done_cond /* is thread initialization done? */
>> bool stopping
>> diff --git a/iothread.c b/iothread.c
>> index beeb870..b6f3c3c 100644
>> --- a/iothread.c
>> +++ b/iothread.c
>> @@ -46,6 +46,7 @@ AioContext *qemu_get_current_aio_context(void)
>> static void *iothread_run(void *opaque)
>> {
>> IOThread *iothread = opaque
>> + GMainContext *context
>>
>> rcu_register_thread()
>>
>> @@ -57,6 +58,15 @@ static void *iothread_run(void *opaque)
>>
>> while (!atomic_read(&iothread->stopping)) {
>> aio_poll(iothread->ctx, true)
>> +
>> + context = iothread->ctx->source.context
>> + if (context) {
>> + iothread->loop = g_main_loop_new(context, TRUE)
>> + g_main_loop_run(iothread->loop)
>> +
>> + g_main_loop_unref(iothread->loop)
>> + g_main_context_unref(context)
>> + }
>> }
>>
>> rcu_unregister_thread()
>> @@ -72,6 +82,9 @@ static int iothread_stop(Object *object, void *opaque)
>> return 0
>> }
>> iothread->stopping = true
>> + if (iothread->loop) {
>> + g_main_loop_quit(iothread->loop)
>> + }
>> aio_notify(iothread->ctx)
>> qemu_thread_join(&iothread->thread)
>> return 0
>> --
>> 1.8.3.1
>>
>>
>>
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-08-10 9:53 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-08-10 9:17 [Qemu-devel] 答复: Re: [PATCH] qemu-iothread: IOThread supports the GMainContext eventloop wang.yong155
2017-08-10 9:52 ` Paolo Bonzini
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).