<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=us-ascii">
<meta name=Generator content="Microsoft Word 11 (filtered medium)">
<!--[if !mso]>
<style>
v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style>
<![endif]-->
<style>
<!--
/* Font Definitions */
@font-face
{font-family:"MS Mincho";
panose-1:2 2 6 9 4 2 5 8 3 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
{font-family:"\@MS Mincho";
panose-1:0 0 0 0 0 0 0 0 0 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman";}
a:link, span.MsoHyperlink
{color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{color:blue;
text-decoration:underline;}
span.EmailStyle18
{mso-style-type:personal-reply;
font-family:Arial;
color:navy;}
@page Section1
{size:8.5in 11.0in;
margin:1.0in 1.25in 1.0in 1.25in;}
div.Section1
{page:Section1;}
-->
</style>
<!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang=EN-US link=blue vlink=blue>
<div class=Section1>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'>Good info. Indeed, looks like this is what
glibc actually does:<o:p></o:p></span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'><a
href="http://www.cygwin.com/ml/libc-alpha/2007-06/msg00097.html">http://www.cygwin.com/ml/libc-alpha/2007-06/msg00097.html</a><o:p></o:p></span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>
<p class=MsoNormal><font size=2 color=navy face=Arial><span style='font-size:
10.0pt;font-family:Arial;color:navy'><o:p> </o:p></span></font></p>
<div>
<div class=MsoNormal align=center style='text-align:center'><font size=3
face="Times New Roman"><span style='font-size:12.0pt'>
<hr size=2 width="100%" align=center tabindex=-1>
</span></font></div>
<p class=MsoNormal><b><font size=2 face=Tahoma><span style='font-size:10.0pt;
font-family:Tahoma;font-weight:bold'>From:</span></font></b><font size=2
face=Tahoma><span style='font-size:10.0pt;font-family:Tahoma'>
tinghich@gmail.com [mailto:tinghich@gmail.com] <b><span style='font-weight:
bold'>On Behalf Of </span></b>Nguyen Nguyen<br>
<b><span style='font-weight:bold'>Sent:</span></b> Tuesday, September 11, 2007
8:15 PM<br>
<b><span style='font-weight:bold'>To:</span></b> Ilya Lipovsky<br>
<b><span style='font-weight:bold'>Cc:</span></b> Benedict, Michael;
linuxppc-embedded@ozlabs.org<br>
<b><span style='font-weight:bold'>Subject:</span></b> Re: futex priority based
wakeup</span></font><o:p></o:p></p>
</div>
<p class=MsoNormal><font size=3 face="Times New Roman"><span style='font-size:
12.0pt'><o:p> </o:p></span></font></p>
<p class=MsoNormal style='margin-bottom:12.0pt'><font size=3
face="Times New Roman"><span style='font-size:12.0pt'>I have seen something
similar before. Our fix was to use
pthread_attr_setinheritsched(&attr, PTHREAD_EXPLICIT_SCHED) so child
threads wouldn't inherit attribute from parent. Hope it helps.<br>
<br>
<o:p></o:p></span></font></p>
<div>
<p class=MsoNormal><span class=gmailquote><font size=3 face="Times New Roman"><span
style='font-size:12.0pt'>On 9/11/07, <b><span style='font-weight:bold'>Ilya
Lipovsky</span></b> <<a href="mailto:lipovsky@cs.bu.edu">lipovsky@cs.bu.edu</a>>
wrote:</span></font></span><o:p></o:p></p>
<p class=MsoNormal style='margin-bottom:12.0pt'><font size=3
face="Times New Roman"><span style='font-size:12.0pt'>Hmm. Just for kicks -
inside the important thread could you add:<br>
<br>
int curpolicy;<br>
struct sched_param sp;<br>
pthread_getschedparam (pthread_self (), &curpolicy, &sp)<br>
printf("important's policy is %d and priority is %d\n", curpolicy, <br>
sp.__sched_priority);<br>
<br>
before the very first futex syscall and after your "printf("important
got<br>
futex!\n");" line.<br>
<br>
Do similar for the unimportant thread, and see if you get anything weird - <br>
e.g. priorities come out to be the same for threads.<br>
<br>
<br>
-----Original Message-----<br>
From: linuxppc-embedded-bounces+lipovsky=<a href="mailto:cs.bu.edu@ozlabs.org">cs.bu.edu@ozlabs.org</a><br>
[mailto:<a href="mailto:linuxppc-embedded-bounces+lipovsky=cs.bu.edu@ozlabs.org">
linuxppc-embedded-bounces+lipovsky=cs.bu.edu@ozlabs.org</a>] On Behalf<br>
Of Benedict, Michael<br>
Sent: Monday, September 10, 2007 5:41 PM<br>
To: <a href="mailto:linuxppc-embedded@ozlabs.org">linuxppc-embedded@ozlabs.org</a>
<br>
Subject: RE: futex priority based wakeup<br>
<br>
Ilya Lipovsky wrote:<br>
> Your code looks correct to me, so if the kernel developers<br>
> did their job<br>
> correctly, the only potentially weak link is glibc. <br>
><br>
<br>
Well, either the kernel developers didn't do their job, or I am missing<br>
something. The following also fails, and it should be bypassing
glibc:<br>
<br>
#define _XOPEN_SOURCE 600<br>
<br>
#include <linux/futex.h> <br>
#include <sys/time.h><br>
#include <asm/atomic.h><br>
<br>
#include <sys/syscall.h><br>
#include <unistd.h><br>
<br>
#include <sched.h><br>
#include <stdint.h><br>
#include <stdio.h><br>
#include <stdlib.h><br>
<br>
int myfutex = 0;<br>
<br>
void *important(void *ign)<br>
{<br>
sleep(1);<br>
printf("important waiting
for futex\n");<br>
fflush(stdout);<br>
if(syscall(SYS_futex,
&myfutex, FUTEX_WAIT, 0, NULL)) { <br>
perror("futex");<br>
exit(1);<br>
} else {<br>
printf("important
got futex!\n");<br>
fflush(stdout);<br>
syscall(SYS_futex,
&myfutex, FUTEX_WAKE, 1, NULL); <br>
}<br>
<br>
return NULL;<br>
}<br>
<br>
<br>
void *unimportant(void *ign)<br>
{<br>
printf("unimportant waiting
for futex\n");<br>
fflush(stdout);<br>
if(syscall(SYS_futex,
&myfutex, FUTEX_WAIT, 0, NULL)) { <br>
perror("futex");<br>
exit(1);<br>
} else {<br>
printf("unimportant
got futex!\n");<br>
fflush(stdout);<br>
syscall(SYS_futex,
&myfutex, FUTEX_WAKE, 1, NULL); <br>
}<br>
<br>
return NULL;<br>
}<br>
<br>
int main()<br>
{<br>
struct sched_param p;<br>
pthread_attr_t attr;<br>
pthread_t i, u;<br>
<br>
p.__sched_priority =
sched_get_priority_min(SCHED_FIFO); <br>
if(-1 == p.__sched_priority) {<br>
perror("sched_get_priority_min");<br>
return
1;<br>
}<br>
pthread_attr_init(&attr);<br>
pthread_attr_setschedpolicy(&attr,
SCHED_FIFO); <br>
pthread_attr_setschedparam(&attr,
&p);<br>
pthread_create(&u,
&attr, unimportant, NULL);<br>
<br>
p.__sched_priority =
sched_get_priority_max(SCHED_FIFO);<br>
pthread_attr_setschedparam(&attr,
&p); <br>
pthread_create(&i,
&attr, important, NULL);<br>
<br>
sleep(5);<br>
printf("futex
FUTEX_WAKE\n");<br>
fflush(stdout);<br>
syscall(SYS_futex,
&myfutex, FUTEX_WAKE, 1, NULL); <br>
<br>
pthread_join(u, NULL);<br>
pthread_join(i, NULL);<br>
<br>
return 0;<br>
}<br>
<br>
Which produces:<br>
unimportant waiting for futex<br>
important waiting for futex<br>
futex FUTEX_WAKE<br>
unimportant got futex! <br>
important got futex!<br>
<br>
<br>
Could someone with 2.6.22 please verify?<br>
<br>
_______________________________________________<br>
Linuxppc-embedded mailing list<br>
<a href="mailto:Linuxppc-embedded@ozlabs.org">Linuxppc-embedded@ozlabs.org </a><br>
<a href="https://ozlabs.org/mailman/listinfo/linuxppc-embedded">https://ozlabs.org/mailman/listinfo/linuxppc-embedded</a><br>
<br>
_______________________________________________<br>
Linuxppc-embedded mailing list<br>
<a href="mailto:Linuxppc-embedded@ozlabs.org">Linuxppc-embedded@ozlabs.org</a><br>
<a href="https://ozlabs.org/mailman/listinfo/linuxppc-embedded">https://ozlabs.org/mailman/listinfo/linuxppc-embedded</a><o:p></o:p></span></font></p>
</div>
<p class=MsoNormal><font size=3 face="Times New Roman"><span style='font-size:
12.0pt'><o:p> </o:p></span></font></p>
</div>
</body>
</html>