<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><br>
</p>
<div class="moz-cite-prefix">On 02/10/2017 05:21 PM, Dehao Chen
wrote:<br>
</div>
<blockquote
cite="mid:CAO2gOZWqNmyO+ffXRT3+LRsdpymRFHTL8Ni3znesw3SyO1ar0Q@mail.gmail.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<div dir="ltr">Thanks every for the comments.
<div><br>
</div>
<div>Do we have a decision here?</div>
</div>
</blockquote>
<br>
You're good to go as far as I'm concerned.<br>
<br>
-Hal<br>
<br>
<blockquote
cite="mid:CAO2gOZWqNmyO+ffXRT3+LRsdpymRFHTL8Ni3znesw3SyO1ar0Q@mail.gmail.com"
type="cite">
<div dir="ltr">
<div><br>
</div>
<div>Dehao</div>
</div>
<div class="gmail_extra"><br>
<div class="gmail_quote">On Tue, Feb 7, 2017 at 10:24 PM, Hal
Finkel <span dir="ltr"><<a moz-do-not-send="true"
href="mailto:hfinkel@anl.gov" target="_blank">hfinkel@anl.gov</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000"><span class="">
<p><br>
</p>
<div class="m_5543670116359683412moz-cite-prefix">On
02/07/2017 05:29 PM, Sanjay Patel via llvm-dev wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr">Sorry if I missed it, but what
machine/CPU are you using to collect the perf
numbers? <br>
<br>
I am concerned that what may be a win on a CPU that
keeps a couple of hundred instructions in-flight and
has many MB of caches will not hold for a small
core. <br>
</div>
</blockquote>
<br>
</span> In my experience, unrolling tends to help weaker
cores even more than stronger ones because it allows the
instruction scheduler more opportunities to hide latency.
Obviously, instruction-cache pressure is an important
consideration, but the code size changes here seems small.<span
class=""><br>
<br>
<blockquote type="cite">
<div dir="ltr"><br>
Is the proposed change universal? Is there a way to
undo it?<br>
</div>
</blockquote>
<br>
</span> All of the unrolling thresholds should be
target-adjustable using the TTI::getUnrollingPreferences
hook.<br>
<br>
-Hal
<div>
<div class="h5"><br>
<br>
<blockquote type="cite">
<div class="gmail_extra"><br>
<div class="gmail_quote">On Tue, Feb 7, 2017 at
3:26 PM, Dehao Chen via llvm-dev <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org"
target="_blank">llvm-dev@lists.llvm.org</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0
0 0 .8ex;border-left:1px #ccc
solid;padding-left:1ex">
<div dir="ltr">Ping... with the updated code
size impact data, any more comments? Any
more data that would be interesting to
collect?
<div><br>
</div>
<div>Thanks,</div>
<div>Dehao</div>
</div>
<div class="m_5543670116359683412HOEnZb">
<div class="m_5543670116359683412h5">
<div class="gmail_extra"><br>
<div class="gmail_quote">On Thu, Feb 2,
2017 at 2:07 PM, Dehao Chen <span
dir="ltr"><<a
moz-do-not-send="true"
href="mailto:dehao@google.com"
target="_blank">dehao@google.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote"
style="margin:0 0 0
.8ex;border-left:1px #ccc
solid;padding-left:1ex">
<div dir="ltr"> Here is the code
size impact for clang, chrome and
24 google internal benchmarks
(name omited, 14 15 16 are
encoding/decoding benchmarks
similar as h264). There are 2
columns, for threshold 300 and 450
respectively.
<div><br>
</div>
<div>I also tested the llvm test
suite. Changing the threshold to
300/450 does not affect code gen
for any binary in the test
suite.</div>
<div><br>
</div>
<div><br>
<table border="0"
cellspacing="0">
<colgroup span="3" width="85"></colgroup>
<tbody>
<tr>
<td align="left"
height="17"><br>
</td>
<td align="right">300</td>
<td align="right">450</td>
</tr>
<tr>
<td align="left"
height="17">clang</td>
<td align="right">0.30%</td>
<td align="right">0.63%</td>
</tr>
<tr>
<td align="left"
height="17">chrome</td>
<td align="right">0.00%</td>
<td align="right">0.00%</td>
</tr>
<tr>
<td align="right"
height="17">1</td>
<td align="right">0.27%</td>
<td align="right">0.67%</td>
</tr>
<tr>
<td align="right"
height="17">2</td>
<td align="right">0.44%</td>
<td align="right">0.93%</td>
</tr>
<tr>
<td align="right"
height="17">3</td>
<td align="right">0.44%</td>
<td align="right">0.93%</td>
</tr>
<tr>
<td align="right"
height="17">4</td>
<td align="right">0.26%</td>
<td align="right">0.53%</td>
</tr>
<tr>
<td align="right"
height="17">5</td>
<td align="right">0.74%</td>
<td align="right">2.21%</td>
</tr>
<tr>
<td align="right"
height="17">6</td>
<td align="right">0.74%</td>
<td align="right">2.21%</td>
</tr>
<tr>
<td align="right"
height="17">7</td>
<td align="right">0.74%</td>
<td align="right">2.21%</td>
</tr>
<tr>
<td align="right"
height="17">8</td>
<td align="right">0.46%</td>
<td align="right">1.05%</td>
</tr>
<tr>
<td align="right"
height="17">9</td>
<td align="right">0.35%</td>
<td align="right">0.86%</td>
</tr>
<tr>
<td align="right"
height="17">10</td>
<td align="right">0.35%</td>
<td align="right">0.86%</td>
</tr>
<tr>
<td align="right"
height="17">11</td>
<td align="right">0.40%</td>
<td align="right">0.83%</td>
</tr>
<tr>
<td align="right"
height="17">12</td>
<td align="right">0.32%</td>
<td align="right">0.65%</td>
</tr>
<tr>
<td align="right"
height="17">13</td>
<td align="right">0.31%</td>
<td align="right">0.64%</td>
</tr>
<tr>
<td align="right"
height="17">14</td>
<td align="right">4.52%</td>
<td align="right">8.23%</td>
</tr>
<tr>
<td align="right"
height="17">15</td>
<td align="right">9.90%</td>
<td align="right">19.38%</td>
</tr>
<tr>
<td align="right"
height="17">16</td>
<td align="right">9.90%</td>
<td align="right">19.38%</td>
</tr>
<tr>
<td align="right"
height="17">17</td>
<td align="right">0.68%</td>
<td align="right">1.97%</td>
</tr>
<tr>
<td align="right"
height="17">18</td>
<td align="right">0.21%</td>
<td align="right">0.48%</td>
</tr>
<tr>
<td align="right"
height="17">19</td>
<td align="right">0.99%</td>
<td align="right">3.44%</td>
</tr>
<tr>
<td align="right"
height="17">20</td>
<td align="right">0.19%</td>
<td align="right">0.46%</td>
</tr>
<tr>
<td align="right"
height="17">21</td>
<td align="right">0.57%</td>
<td align="right">1.62%</td>
</tr>
<tr>
<td align="right"
height="17">22</td>
<td align="right">0.37%</td>
<td align="right">1.05%</td>
</tr>
<tr>
<td align="right"
height="17">23</td>
<td align="right">0.78%</td>
<td align="right">1.30%</td>
</tr>
<tr>
<td align="right"
height="17">24</td>
<td align="right">0.51%</td>
<td align="right">1.54%</td>
</tr>
</tbody>
</table>
</div>
</div>
<div
class="m_5543670116359683412m_-4814779693185880395HOEnZb">
<div
class="m_5543670116359683412m_-4814779693185880395h5">
<div class="gmail_extra"><br>
<div class="gmail_quote">On
Wed, Feb 1, 2017 at 6:08 PM,
Mikhail Zolotukhin via
llvm-dev <span dir="ltr"><<a
moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org"
target="_blank">llvm-dev@lists.llvm.org</a>></span>
wrote:<br>
<blockquote
class="gmail_quote"
style="margin:0 0 0
.8ex;border-left:1px #ccc
solid;padding-left:1ex">
<div
style="word-wrap:break-word">
<div>
<div><span>
<blockquote
type="cite">
<div>On Feb 1,
2017, at 4:57
PM, Xinliang
David Li via
llvm-dev <<a
moz-do-not-send="true" href="mailto:llvm-dev@lists.llvm.org"
target="_blank">llvm-dev@lists.llvm.org</a>>
wrote:</div>
<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851Apple-interchange-newline">
<div>
<div dir="ltr">clang,
chrome, and
some internal
large apps are
good
candidates for
size metrics.</div>
</div>
</blockquote>
</span>I'd also add
the standard LLVM
testsuite just
because it's the
suite everyone in
the community can
use.
<div><br>
</div>
<div>Michael<br>
</div>
<div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058h5">
<blockquote
type="cite">
<div>
<div dir="ltr">
<div><br>
</div>
<div>David</div>
</div>
<div
class="gmail_extra"><br>
<div
class="gmail_quote">On
Wed, Feb 1,
2017 at 4:47
PM, Chandler
Carruth via
llvm-dev <span
dir="ltr"><<a
moz-do-not-send="true" href="mailto:llvm-dev@lists.llvm.org"
target="_blank">llvm-dev@lists.llvm.org</a>></span>
wrote:<br>
<blockquote
class="gmail_quote"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr">I
had suggested
having size
metrics from
somewhat
larger
applications
such as
Chrome,
Webkit, or
Firefox; clang
itself; and
maybe some of
our internal
binaries with
rough size
brackets?</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851HOEnZb">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851h5"><br>
<div
class="gmail_quote">
<div dir="ltr">On
Wed, Feb 1,
2017 at 4:33
PM Dehao Chen
<<a
moz-do-not-send="true"
href="mailto:dehao@google.com" target="_blank">dehao@google.com</a>>
wrote:<br>
</div>
<blockquote
class="gmail_quote"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">With
the new data
points, any
comments on
whether this
can justify
setting fully
inline
threshold to
300 (or any
other number)
in O2? I can
collect more
data points if
it's helpful.
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Thanks,</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Dehao</div>
</div>
<div
class="m_5543670116359683412gmail_extra
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">On
Tue, Jan 31,
2017 at 3:20
PM, Dehao Chen
<span
dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><<a
moz-do-not-send="true" href="mailto:dehao@google.com"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
target="_blank">dehao@google.com</a>></span> wrote:<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Recollected
the data from
trunk head
with stddev
data and more
threshold data
points
attached:
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Performance:</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<table
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
border="0"
cellspacing="0">
<colgroup
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
span="6"
width="85"></colgroup>
<tbody
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="left"
height="17"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="left">stddev/mean</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">300</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">450</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">600</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">750</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">403</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.37%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.11%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.11%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.09%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.79%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">433</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.14%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.51%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.25%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">-0.63%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">-0.29%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">445</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.08%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.48%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.89%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.12%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.83%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">447</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.16%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">3.50%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.69%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">3.66%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">3.59%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">453</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.11%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">1.49%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.45%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">-0.07%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.78%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">464</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.17%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.75%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">1.80%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">1.86%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">1.54%</td>
</tr>
</tbody>
</table>
<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Code
size:</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<table
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
border="0"
cellspacing="0">
<colgroup
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
span="5"
width="85"></colgroup>
<tbody
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="left"
height="17"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">300</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">450</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">600</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">750</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">403</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.56%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.41%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.74%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">3.75%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">433</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.96%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.84%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">4.19%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">4.87%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">445</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.16%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">3.62%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">4.48%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">5.88%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">447</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.96%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">5.09%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">6.74%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">8.89%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">453</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">0.94%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">1.67%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.73%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">2.96%</td>
</tr>
<tr
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right"
height="17">464</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">8.02%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">13.50%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">20.51%</td>
<td
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
align="right">26.59%</td>
</tr>
</tbody>
</table>
<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Compile
time is
proportional
in the
experiments
and more
noisy, so I
did not
include it.</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">We
have >2%
speedup on
some google
internal
benchmarks
when switching
the threshold
from 150 to
300.</div>
<span
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241HOEnZbm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><font
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
color="#888888">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">Dehao</div>
</font></span></div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241HOEnZbm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241h5">
<div
class="m_5543670116359683412gmail_extra
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">On
Mon, Jan 30,
2017 at 5:06
PM, Chandler
Carruth <span
dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><<a
moz-do-not-send="true" href="mailto:chandlerc@google.com"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
target="_blank">chandlerc@google.com</a>></span> wrote:<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><span
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">On
Mon, Jan 30,
2017 at 4:59
PM Mehdi Amini
<<a
moz-do-not-send="true"
href="mailto:mehdi.amini@apple.com"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
target="_blank">mehdi.amini@apple.com</a>> wrote:<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
</span><span
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div
style="word-wrap:break-word"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<blockquote
type="cite"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
</blockquote>
</div>
</div>
<div
style="word-wrap:break-word"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<blockquote
type="cite"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="font-family:Helvetica;font-size:12px;font-style:normal;font-variant-caps:normal;font-weight:normal;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945h5
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945m_-4486181801685859403gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="word-wrap:break-word">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945m_-4486181801685859403gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945m_-4486181801685859403gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945m_-4486181801685859403gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655m_-6293897042106820945m_-4486181801685859403gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">Another
question is
about PGO
integration:
is it already
hooked there?
Should we have
a more
aggressive
threshold in a
hot function?
(Assuming
we’re willing
to spend some
binary size
there but not
on the cold
path).</div>
</div>
</div>
</blockquote>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
</div>
</div>
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">I
would even
wire the
*unrolling*
the other way:
just suppress
unrolling in
cold paths to
save binary
size. rolled
loops seem
like a
generally good
thing in cold
code unless
they are
having some
larger impact
(IE, the loop
itself is more
expensive than
the unrolled
form).</div>
</div>
</div>
</blockquote>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">Agree
that we could
suppress
unrolling in
cold path to
save code
size. But
that's
orthogonal
with the
propose here.
This proposal
focuses on O2
performance:
shall we have
different
(higher) fully
unroll
threshold than
dynamic/partial unroll.</div>
</div>
</div>
</blockquote>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
</div>
</div>
</div>
<div
style="word-wrap:break-word"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">I
agree that
this is (to
some extent)
orthogonal,
and it makes
sense to me to
differentiate
the threshold
for full
unroll and the
dynamic/partial case.</div>
</div>
</div>
</blockquote>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
</span>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">There
is one issue
that makes
these not
orthogonal.</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">If
even *static*
profile hints
will reduce
some of the
code size
increase
caused by
higher
unrolling
thresholds for
non-cold code,
we should
factor that
into the
tradeoff in
picking where
the threshold
goes.</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"><br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">However,
getting PGO
into the full
unroller is
currently
challenging
outside of the
new pass
manager. We
already have
some
unfortunate
hacks around
this in
LoopUnswitch
that are
making the
port of it to
the new PM
more annoying.</div>
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg"
style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div
style="word-wrap:break-word"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<blockquote
type="cite"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msgm_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="font-family:Helvetica;font-size:12px;font-style:normal;font-variant-caps:normal;font-weight:normal;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px">
<blockquote
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"
style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<div dir="ltr"
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg">
<div
class="m_5543670116359683412gmail_quote
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"><span
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg
m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588m_1694617720154745241m_6119381627642025574m_-8823773316722963655gmail_msg"></span></div>
</div>
</blockquote>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
</div>
<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
</div>
</div>
</blockquote>
</div>
<br
class="m_5543670116359683412m_-4814779693185880395m_8897563994913029058m_6216879383400100851m_8709703343599191588gmail_msg">
</div>
</blockquote>
</div>
</div>
</div>
<br>
______________________________<wbr>_________________<br>
LLVM
Developers
mailing list<br>
<a
moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a
moz-do-not-send="true"
href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
rel="noreferrer"
target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-dev</a><br>
<br>
</blockquote>
</div>
<br>
</div>
______________________________<wbr>_________________<br>
LLVM
Developers
mailing list<br>
<a
moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a
moz-do-not-send="true"
href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-dev</a><br>
</div>
</blockquote>
</div>
</div>
</div>
<br>
</div>
</div>
<br>
______________________________<wbr>_________________<br>
LLVM Developers mailing
list<br>
<a moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a moz-do-not-send="true"
href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
rel="noreferrer"
target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-dev</a><br>
<br>
</blockquote>
</div>
<br>
</div>
</div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</div>
<br>
______________________________<wbr>_________________<br>
LLVM Developers mailing list<br>
<a moz-do-not-send="true"
href="mailto:llvm-dev@lists.llvm.org"
target="_blank">llvm-dev@lists.llvm.org</a><br>
<a moz-do-not-send="true"
href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev"
rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-dev</a><br>
<br>
</blockquote>
</div>
<br>
</div>
<br>
<fieldset
class="m_5543670116359683412mimeAttachmentHeader"></fieldset>
<br>
<pre>______________________________<wbr>_________________
LLVM Developers mailing list
<a moz-do-not-send="true" class="m_5543670116359683412moz-txt-link-abbreviated" href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>
<a moz-do-not-send="true" class="m_5543670116359683412moz-txt-link-freetext" href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-dev</a>
</pre>
</blockquote>
</div></div><span class="HOEnZb"><font color="#888888"><pre class="m_5543670116359683412moz-signature" cols="72">--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory</pre>
</font></span></div>
</blockquote></div>
</div>
</blockquote>
<pre class="moz-signature" cols="72">--
Hal Finkel
Lead, Compiler Technology and Programming Languages
Leadership Computing Facility
Argonne National Laboratory</pre></body></html>