<br><font size=2 face="sans-serif">Yeah, you're right. I had the
same thought. This option is no good.</font>
<br><font size=2 face="sans-serif"><br>
Dean Burdick<br>
Cell Software Development<br>
IBM Austin, TX<br>
512-838-0264 T/L: 678-0264<br>
</font>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td width=40%><font size=1 face="sans-serif"><b>Arnd Bergmann <arnd@arndb.de></b>
</font>
<p><font size=1 face="sans-serif">10/22/2007 10:19 AM</font>
<td width=59%>
<table width=100%>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">To</font></div>
<td><font size=1 face="sans-serif">cbe-oss-dev@ozlabs.org</font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">cc</font></div>
<td><font size=1 face="sans-serif">Dean Burdick/Austin/IBM@IBMUS, Uzi Shvadron
<SHVADRON@il.ibm.com></font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">Subject</font></div>
<td><font size=1 face="sans-serif">Re: [Cbe-oss-dev] Fwd: adding virtual
SPU timers</font></table>
<br>
<table>
<tr valign=top>
<td>
<td></table>
<br></table>
<br>
<br>
<br><tt><font size=2>On Monday 22 October 2007, Dean Burdick wrote:<br>
> I like the idea of maintaining both the wall time and a delta. This
would <br>
> allow us to provide both real time (for trace timestamps) and virtual
time <br>
> (for measurements/timers) at the same time with the library.<br>
<br>
Any ideas on how the SPU code should get at the offset? If we require<br>
a syscall for every timer operation, that will significantly limit the<br>
accuracy and add latency to the library functions.<br>
<br>
Also, I'm still not sure what times should be included in the offset<br>
between virtual time by the kernel (see the list from my original<br>
mail).<br>
<br>
Arnd <><<br>
</font></tt>
<br>