<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 07/25/2013 02:47 PM, Daniel Dunbar
wrote:<br>
</div>
<blockquote
cite="mid:CAEU8z6_baBEPPbfneKZgY3T8D+rJou5DAeB15DuQLnWyha_G3A@mail.gmail.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html;
charset=ISO-8859-1">
<div dir="ltr">Ok, this seems like a pretty custom situation, I'd
rather that have minimal impact on the LNT code paths. It
doesn't generally make any sense to compile "timeit-target" for
the host.<br>
<div>
<br>
</div>
<div>Could you instead implement this as a patch to instead
cause "timeit" to be used instead of "timeit-target". This
should be as simple as adding a new Makefile variable that
will be used in Makefile.programs to cause RUNSAFELY to use
timeit instead of timeit-target. Then you can use LNT's
existing --make-param option to pass that in.</div>
<div><br>
</div>
</div>
</blockquote>
Sounds like a good plan.<br>
<br>
We will make that change.<br>
<br>
<blockquote
cite="mid:CAEU8z6_baBEPPbfneKZgY3T8D+rJou5DAeB15DuQLnWyha_G3A@mail.gmail.com"
type="cite">
<div dir="ltr">
<div> - Daniel</div>
</div>
<div class="gmail_extra"><br>
<br>
<div class="gmail_quote">On Wed, Jul 24, 2013 at 11:37 AM, Doug
Gilmore <span dir="ltr"><<a moz-do-not-send="true"
href="mailto:Doug.Gilmore@imgtec.com" target="_blank">Doug.Gilmore@imgtec.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Tue, 2013-07-23 at 17:17 -0700, an Reed
Kotler wrote:<br>
><br>
> In this case we are not using lnt under Qemu user
mode for benchmarking;<br>
> just as a way to run test-suite to test whether the
code is correct.<br>
><br>
> Qemu user mode emulates target instructions, but when
it gets a Unix<br>
> Kernel trap, it uses the host to emulate those.<br>
><br>
> For example, file I/O.<br>
><br>
> It is possible to run target timeit under qemu and
let it launch the app<br>
> or a wrapper.<br>
> (But it is more limited as to what can be done here
under qemu vs under<br>
> the host OS directly).<br>
><br>
> For time functions, it is also going to use the host
to emulate those.<br>
><br>
> So whether timeit is running under qemu or directly
on the host, the<br>
> answers regarding time will be the same.<br>
><br>
> But running timeit under qemu will be much slower as
far as elapsed time<br>
> than running it on the host directly.<br>
><br>
> We would also need to add some new mechanism to Lnt
or the makefiles to<br>
> also wrap timeit.<br>
</div>
Good point Reed -- thanks!<br>
<div class="im">><br>
> Reed<br>
><br>
><br>
> On 07/23/2013 02:19 PM, Daniel Dunbar wrote:<br>
> > Wouldn't it be a more accurate simulation to run
timeit-target under<br>
> > the emulator as well? Or is that too much to
ask?<br>
> ><br>
> > - Daniel<br>
</div>
Hi Daniel,<br>
<br>
I agree with Reed's discussion of the issues. We are mainly
concerned with<br>
the correctness when running under QEMU, though the the
timing data<br>
might be useful at a very course grain level.<br>
<br>
Doug<br>
> ><br>
> ><br>
> > On Mon, Jul 22, 2013 at 6:47 PM, Reed Kotler
<rkotler at <a moz-do-not-send="true"
href="http://mips.com" target="_blank">mips.com</a><br>
> > <mailto:<a moz-do-not-send="true"
href="mailto:rkotler">rkotler</a> at <a
moz-do-not-send="true" href="http://mips.com"
target="_blank">mips.com</a>>> wrote:<br>
> > ...<br>
<br>
<br>
</blockquote>
</div>
<br>
</div>
</blockquote>
<br>
</body>
</html>