<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=gb2312">
<style>
<!--
body
{line-height:1.5}
blockquote
{margin-top:0px;
margin-bottom:0px;
margin-left:0.5em}
body
{font-size:16pt;
font-family:���ź�;
color:rgb(0,0,0);
line-height:1.5}
-->
</style>
</head>
<body>
<blockquote style="margin-top:0px; margin-bottom:0px; margin-left:0.5em">
<div class="FoxDiv20150630100730998087">
<div><font face="Courier New">Hi</font></div>
<div><font face="Courier New"><br>
</font></div>
<div><font face="Courier New">1. In the callback o2net_sendpage -> sendpage<span style="line-height:1.5">()</span></font></div>
<div><font face="Courier New"><span style="line-height:1.5">2. I</span><span style="line-height:1.5">f </span>sendpage<span style="line-height:1.5"> (tcp</span><span style="line-height:1.5"> socket send) continuously returns </span><span style="line-height:1.5">-EAGAIN</span><span style="line-height:1.5">,
t</span><span style="line-height:1.5">hen </span><span style="line-height:1.5">get into an endless
</span><span style="line-height:1.5">loop,</span></font></div>
<div><font face="Courier New"><span style="line-height:1.5"> </span><span style="line-height:1.5">even though </span><span style="line-height:1.5">the function of
</span><span style="line-height:1.5">cond_resched() </span><span style="line-height:1.5">have </span><span style="line-height:1.5">already </span><span style="line-height:1.5">used</span><span style="line-height:1.5">.</span></font></div>
<div><font face="Courier New"><span style="line-height:1.5">3. I think it is not reasonable, </span><span style="line-height:1.5">try to </span><span style="line-height:1.5">continuously </span><span style="line-height:1.5">send </span><span style="line-height:1.5">20
times </span><span style="line-height:1.5">returns </span><span style="line-height:1.5">-EAGAIN</span><span style="line-height:1.5"> ,</span></font></div>
<div><font face="Courier New"><span style="line-height:1.5"> </span><span style="line-height:1.5">shutdown the socket to avoid
</span><span style="line-height:1.5">affecting </span><span style="line-height:1.5">the </span><span style="line-height:1.5">entire </span><span style="line-height:1.5">cluster.</span></font></div>
<div><font face="Courier New"><span style="line-height:1.5"><br>
</span></font></div>
<div><span style="font-family:''; font-size:16pt; line-height:1.5; background-color:window"> </span><span style="font-size:16pt; line-height:1.5; background-color:window"> </span><span style="font-family:'Courier New'; font-size:16pt; line-height:1.5; background-color:window">Finally,
any feedback about this process </span><span style="font-family:'Courier New'; font-size:16pt; line-height:1.5; background-color:window">(positive or negative) would be greatly appreciated.</span></div>
<div>
<pre><span style=""><font face="Courier New">--- tcp.c 2015-06-30 11:46:54.727447919 +0800
<br>+++ tcp.c.diff 2015-06-30 11:52:12.823447881 +0800
<br>@@ -949,6 +949,7 @@
<br> {
<br> struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num);
<br> ssize_t ret;
<br>+ int send_fails = 20;
<br>
<br> while (1) {
<br> mutex_lock(&sc->sc_send_lock);
<br>@@ -959,10 +960,11 @@
<br> mutex_unlock(&sc->sc_send_lock);
<br> if (ret == size)
<br> break;
<br>- if (ret == (ssize_t)-EAGAIN) {
<br>+ if (ret == (ssize_t)-EAGAIN && send_fails > 0) {
<br> mlog(0, "sendpage of size %zu to " SC_NODEF_FMT
<br> " returned EAGAIN\n", size, SC_NODEF_ARGS(sc));
<br> cond_resched();
<br>+ --send_fails;
<br> continue;
<br> }
<br> mlog(ML_ERROR, "sendpage of size %zu to " SC_NODEF_FMT</font></span></pre>
<pre><span style=""><font face="Courier New"><br></font></span></pre>
<pre><span style=""><font face="Courier New">syslog:</font></span></pre>
<pre><span style="">/var/log/syslog:Jun 29 09:32:58 cvk47 kernel: [156022.769539] (kworker/u130:1,12041,9):o2net_sendpage:1026 sendpage of size 24 to node cvk61 (num 5) at 172.16.202.61:7100 returned EAGAIN
<br>/var/log/syslog:Jun 29 09:32:58 cvk47 kernel: [156022.769542] (kworker/u130:1,12041,9):o2net_sendpage:1026 sendpage of size 24 to node cvk61 (num 5) at 172.16.202.61:7100 returned EAGAIN
<br>/var/log/syslog:Jun 29 09:32:58 cvk47 kernel: [156022.769544] (kworker/u130:1,12041,9):o2net_sendpage:1026 sendpage of size 24 to node cvk61 (num 5) at 172.16.202.61:7100 returned EAGAIN
<br>/var/log/syslog:Jun 29 09:32:58 cvk47 kernel: [156022.769546] (kworker/u130:1,12041,9):o2net_sendpage:1026 sendpage of</span></pre>
</div>
<hr color="#b5c4df" size="1" align="left" style="width:210px; height:1px">
<div><span>
<div style="margin:10px; font-family:verdana; font-size:10pt">
<div>zhangguanghui 10102</div>
</div>
</span></div>
</div>
</blockquote>
<span style="font-size:7.5pt; font-family:����ϸ��; color:gray"><span lang="EN-US">-------------------------------------------------------------------------------------------------------------------------------------<br>
</span>���ʼ����丽�����к��ݻ���ͨ�ż�������˾�ı�����Ϣ�������ڷ��������ַ���г�<span lang="EN-US"><br>
</span>�ĸ��˻�Ⱥ�顣��ֹ�κ����������κ���ʽʹ�ã�������������ȫ���ֵ�й¶�����ơ�<span lang="EN-US"><br>
</span>��ɢ�������ʼ��е���Ϣ������������˱��ʼ������������绰���ʼ�֪ͨ�����˲�ɾ����<span lang="EN-US"><br>
</span>�ʼ���<span lang="EN-US"><br>
</span></span><span lang="EN-US" style="font-size:7.5pt; font-family:"Arial","sans-serif"; color:gray">This e-mail and its attachments contain confidential information from H3C, which is
<br>
intended only for the person or entity whose address is listed above. Any use of the
<br>
information contained herein in any way (including, but not limited to, total or partial
<br>
disclosure, reproduction, or dissemination) by persons other than the intended <br>
recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender
<br>
by phone or email immediately and delete it!</span>
</body>
</html>