blob: 5b5dbf2adc8a458b9845ace4b380f12c7bd2ba09 [file] [log] [blame]
<!DOCTYPE html>
<html lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<meta name="generator" content="AsciiDoc 8.6.9">
<title>cl_intel_subgroups</title>
<style type="text/css">
/* Shared CSS for AsciiDoc xhtml11 and html5 backends */
/* Default font. */
body {
font-family: Georgia,serif;
}
/* Title font. */
h1, h2, h3, h4, h5, h6,
div.title, caption.title,
thead, p.table.header,
#toctitle,
#author, #revnumber, #revdate, #revremark,
#footer {
font-family: Arial,Helvetica,sans-serif;
}
body {
margin: 1em 5% 1em 5%;
}
a {
color: blue;
text-decoration: underline;
}
a:visited {
color: fuchsia;
}
em {
font-style: italic;
color: navy;
}
strong {
font-weight: bold;
color: #083194;
}
h1, h2, h3, h4, h5, h6 {
color: #527bbd;
margin-top: 1.2em;
margin-bottom: 0.5em;
line-height: 1.3;
}
h1, h2, h3 {
border-bottom: 2px solid silver;
}
h2 {
padding-top: 0.5em;
}
h3 {
float: left;
}
h3 + * {
clear: left;
}
h5 {
font-size: 1.0em;
}
div.sectionbody {
margin-left: 0;
}
hr {
border: 1px solid silver;
}
p {
margin-top: 0.5em;
margin-bottom: 0.5em;
}
ul, ol, li > p {
margin-top: 0;
}
ul > li { color: #aaa; }
ul > li > * { color: black; }
.monospaced, code, pre {
font-family: "Courier New", Courier, monospace;
font-size: inherit;
color: navy;
padding: 0;
margin: 0;
}
pre {
white-space: pre-wrap;
}
#author {
color: #527bbd;
font-weight: bold;
font-size: 1.1em;
}
#email {
}
#revnumber, #revdate, #revremark {
}
#footer {
font-size: small;
border-top: 2px solid silver;
padding-top: 0.5em;
margin-top: 4.0em;
}
#footer-text {
float: left;
padding-bottom: 0.5em;
}
#footer-badges {
float: right;
padding-bottom: 0.5em;
}
#preamble {
margin-top: 1.5em;
margin-bottom: 1.5em;
}
div.imageblock, div.exampleblock, div.verseblock,
div.quoteblock, div.literalblock, div.listingblock, div.sidebarblock,
div.admonitionblock {
margin-top: 1.0em;
margin-bottom: 1.5em;
}
div.admonitionblock {
margin-top: 2.0em;
margin-bottom: 2.0em;
margin-right: 10%;
color: #606060;
}
div.content { /* Block element content. */
padding: 0;
}
/* Block element titles. */
div.title, caption.title {
color: #527bbd;
font-weight: bold;
text-align: left;
margin-top: 1.0em;
margin-bottom: 0.5em;
}
div.title + * {
margin-top: 0;
}
td div.title:first-child {
margin-top: 0.0em;
}
div.content div.title:first-child {
margin-top: 0.0em;
}
div.content + div.title {
margin-top: 0.0em;
}
div.sidebarblock > div.content {
background: #ffffee;
border: 1px solid #dddddd;
border-left: 4px solid #f0f0f0;
padding: 0.5em;
}
div.listingblock > div.content {
border: 1px solid #dddddd;
border-left: 5px solid #f0f0f0;
background: #f8f8f8;
padding: 0.5em;
}
div.quoteblock, div.verseblock {
padding-left: 1.0em;
margin-left: 1.0em;
margin-right: 10%;
border-left: 5px solid #f0f0f0;
color: #888;
}
div.quoteblock > div.attribution {
padding-top: 0.5em;
text-align: right;
}
div.verseblock > pre.content {
font-family: inherit;
font-size: inherit;
}
div.verseblock > div.attribution {
padding-top: 0.75em;
text-align: left;
}
/* DEPRECATED: Pre version 8.2.7 verse style literal block. */
div.verseblock + div.attribution {
text-align: left;
}
div.admonitionblock .icon {
vertical-align: top;
font-size: 1.1em;
font-weight: bold;
text-decoration: underline;
color: #527bbd;
padding-right: 0.5em;
}
div.admonitionblock td.content {
padding-left: 0.5em;
border-left: 3px solid #dddddd;
}
div.exampleblock > div.content {
border-left: 3px solid #dddddd;
padding-left: 0.5em;
}
div.imageblock div.content { padding-left: 0; }
span.image img { border-style: none; vertical-align: text-bottom; }
a.image:visited { color: white; }
dl {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
dt {
margin-top: 0.5em;
margin-bottom: 0;
font-style: normal;
color: navy;
}
dd > *:first-child {
margin-top: 0.1em;
}
ul, ol {
list-style-position: outside;
}
ol.arabic {
list-style-type: decimal;
}
ol.loweralpha {
list-style-type: lower-alpha;
}
ol.upperalpha {
list-style-type: upper-alpha;
}
ol.lowerroman {
list-style-type: lower-roman;
}
ol.upperroman {
list-style-type: upper-roman;
}
div.compact ul, div.compact ol,
div.compact p, div.compact p,
div.compact div, div.compact div {
margin-top: 0.1em;
margin-bottom: 0.1em;
}
tfoot {
font-weight: bold;
}
td > div.verse {
white-space: pre;
}
div.hdlist {
margin-top: 0.8em;
margin-bottom: 0.8em;
}
div.hdlist tr {
padding-bottom: 15px;
}
dt.hdlist1.strong, td.hdlist1.strong {
font-weight: bold;
}
td.hdlist1 {
vertical-align: top;
font-style: normal;
padding-right: 0.8em;
color: navy;
}
td.hdlist2 {
vertical-align: top;
}
div.hdlist.compact tr {
margin: 0;
padding-bottom: 0;
}
.comment {
background: yellow;
}
.footnote, .footnoteref {
font-size: 0.8em;
}
span.footnote, span.footnoteref {
vertical-align: super;
}
#footnotes {
margin: 20px 0 20px 0;
padding: 7px 0 0 0;
}
#footnotes div.footnote {
margin: 0 0 5px 0;
}
#footnotes hr {
border: none;
border-top: 1px solid silver;
height: 1px;
text-align: left;
margin-left: 0;
width: 20%;
min-width: 100px;
}
div.colist td {
padding-right: 0.5em;
padding-bottom: 0.3em;
vertical-align: top;
}
div.colist td img {
margin-top: 0.3em;
}
@media print {
#footer-badges { display: none; }
}
#toc {
margin-bottom: 2.5em;
}
#toctitle {
color: #527bbd;
font-size: 1.1em;
font-weight: bold;
margin-top: 1.0em;
margin-bottom: 0.1em;
}
div.toclevel0, div.toclevel1, div.toclevel2, div.toclevel3, div.toclevel4 {
margin-top: 0;
margin-bottom: 0;
}
div.toclevel2 {
margin-left: 2em;
font-size: 0.9em;
}
div.toclevel3 {
margin-left: 4em;
font-size: 0.9em;
}
div.toclevel4 {
margin-left: 6em;
font-size: 0.9em;
}
span.aqua { color: aqua; }
span.black { color: black; }
span.blue { color: blue; }
span.fuchsia { color: fuchsia; }
span.gray { color: gray; }
span.green { color: green; }
span.lime { color: lime; }
span.maroon { color: maroon; }
span.navy { color: navy; }
span.olive { color: olive; }
span.purple { color: purple; }
span.red { color: red; }
span.silver { color: silver; }
span.teal { color: teal; }
span.white { color: white; }
span.yellow { color: yellow; }
span.aqua-background { background: aqua; }
span.black-background { background: black; }
span.blue-background { background: blue; }
span.fuchsia-background { background: fuchsia; }
span.gray-background { background: gray; }
span.green-background { background: green; }
span.lime-background { background: lime; }
span.maroon-background { background: maroon; }
span.navy-background { background: navy; }
span.olive-background { background: olive; }
span.purple-background { background: purple; }
span.red-background { background: red; }
span.silver-background { background: silver; }
span.teal-background { background: teal; }
span.white-background { background: white; }
span.yellow-background { background: yellow; }
span.big { font-size: 2em; }
span.small { font-size: 0.6em; }
span.underline { text-decoration: underline; }
span.overline { text-decoration: overline; }
span.line-through { text-decoration: line-through; }
div.unbreakable { page-break-inside: avoid; }
/*
* xhtml11 specific
*
* */
div.tableblock {
margin-top: 1.0em;
margin-bottom: 1.5em;
}
div.tableblock > table {
border: 3px solid #527bbd;
}
thead, p.table.header {
font-weight: bold;
color: #527bbd;
}
p.table {
margin-top: 0;
}
/* Because the table frame attribute is overriden by CSS in most browsers. */
div.tableblock > table[frame="void"] {
border-style: none;
}
div.tableblock > table[frame="hsides"] {
border-left-style: none;
border-right-style: none;
}
div.tableblock > table[frame="vsides"] {
border-top-style: none;
border-bottom-style: none;
}
/*
* html5 specific
*
* */
table.tableblock {
margin-top: 1.0em;
margin-bottom: 1.5em;
}
thead, p.tableblock.header {
font-weight: bold;
color: #527bbd;
}
p.tableblock {
margin-top: 0;
}
table.tableblock {
border-width: 3px;
border-spacing: 0px;
border-style: solid;
border-color: #527bbd;
border-collapse: collapse;
}
th.tableblock, td.tableblock {
border-width: 1px;
padding: 4px;
border-style: solid;
border-color: #527bbd;
}
table.tableblock.frame-topbot {
border-left-style: hidden;
border-right-style: hidden;
}
table.tableblock.frame-sides {
border-top-style: hidden;
border-bottom-style: hidden;
}
table.tableblock.frame-none {
border-style: hidden;
}
th.tableblock.halign-left, td.tableblock.halign-left {
text-align: left;
}
th.tableblock.halign-center, td.tableblock.halign-center {
text-align: center;
}
th.tableblock.halign-right, td.tableblock.halign-right {
text-align: right;
}
th.tableblock.valign-top, td.tableblock.valign-top {
vertical-align: top;
}
th.tableblock.valign-middle, td.tableblock.valign-middle {
vertical-align: middle;
}
th.tableblock.valign-bottom, td.tableblock.valign-bottom {
vertical-align: bottom;
}
/*
* manpage specific
*
* */
body.manpage h1 {
padding-top: 0.5em;
padding-bottom: 0.5em;
border-top: 2px solid silver;
border-bottom: 2px solid silver;
}
body.manpage h2 {
border-style: none;
}
body.manpage div.sectionbody {
margin-left: 3em;
}
@media print {
body.manpage div#toc { display: none; }
}
@media screen {
body {
max-width: 50em; /* approximately 80 characters wide */
margin-left: 16em;
}
#toc {
position: fixed;
top: 0;
left: 0;
bottom: 0;
width: 13em;
padding: 0.5em;
padding-bottom: 1.5em;
margin: 0;
overflow: auto;
border-right: 3px solid #f8f8f8;
background-color: white;
}
#toc .toclevel1 {
margin-top: 0.5em;
}
#toc .toclevel2 {
margin-top: 0.25em;
display: list-item;
color: #aaaaaa;
}
#toctitle {
margin-top: 0.5em;
}
}
</style>
<script type="text/javascript">
/*<![CDATA[*/
var asciidoc = { // Namespace.
/////////////////////////////////////////////////////////////////////
// Table Of Contents generator
/////////////////////////////////////////////////////////////////////
/* Author: Mihai Bazon, September 2002
* http://students.infoiasi.ro/~mishoo
*
* Table Of Content generator
* Version: 0.4
*
* Feel free to use this script under the terms of the GNU General Public
* License, as long as you do not remove or alter this notice.
*/
/* modified by Troy D. Hanson, September 2006. License: GPL */
/* modified by Stuart Rackham, 2006, 2009. License: GPL */
// toclevels = 1..4.
toc: function (toclevels) {
function getText(el) {
var text = "";
for (var i = el.firstChild; i != null; i = i.nextSibling) {
if (i.nodeType == 3 /* Node.TEXT_NODE */) // IE doesn't speak constants.
text += i.data;
else if (i.firstChild != null)
text += getText(i);
}
return text;
}
function TocEntry(el, text, toclevel) {
this.element = el;
this.text = text;
this.toclevel = toclevel;
}
function tocEntries(el, toclevels) {
var result = new Array;
var re = new RegExp('[hH]([1-'+(toclevels+1)+'])');
// Function that scans the DOM tree for header elements (the DOM2
// nodeIterator API would be a better technique but not supported by all
// browsers).
var iterate = function (el) {
for (var i = el.firstChild; i != null; i = i.nextSibling) {
if (i.nodeType == 1 /* Node.ELEMENT_NODE */) {
var mo = re.exec(i.tagName);
if (mo && (i.getAttribute("class") || i.getAttribute("className")) != "float") {
result[result.length] = new TocEntry(i, getText(i), mo[1]-1);
}
iterate(i);
}
}
}
iterate(el);
return result;
}
var toc = document.getElementById("toc");
if (!toc) {
return;
}
// Delete existing TOC entries in case we're reloading the TOC.
var tocEntriesToRemove = [];
var i;
for (i = 0; i < toc.childNodes.length; i++) {
var entry = toc.childNodes[i];
if (entry.nodeName.toLowerCase() == 'div'
&& entry.getAttribute("class")
&& entry.getAttribute("class").match(/^toclevel/))
tocEntriesToRemove.push(entry);
}
for (i = 0; i < tocEntriesToRemove.length; i++) {
toc.removeChild(tocEntriesToRemove[i]);
}
// Rebuild TOC entries.
var entries = tocEntries(document.getElementById("content"), toclevels);
for (var i = 0; i < entries.length; ++i) {
var entry = entries[i];
if (entry.element.id == "")
entry.element.id = "_toc_" + i;
var a = document.createElement("a");
a.href = "#" + entry.element.id;
a.appendChild(document.createTextNode(entry.text));
var div = document.createElement("div");
div.appendChild(a);
div.className = "toclevel" + entry.toclevel;
toc.appendChild(div);
}
if (entries.length == 0)
toc.parentNode.removeChild(toc);
},
/////////////////////////////////////////////////////////////////////
// Footnotes generator
/////////////////////////////////////////////////////////////////////
/* Based on footnote generation code from:
* http://www.brandspankingnew.net/archive/2005/07/format_footnote.html
*/
footnotes: function () {
// Delete existing footnote entries in case we're reloading the footnodes.
var i;
var noteholder = document.getElementById("footnotes");
if (!noteholder) {
return;
}
var entriesToRemove = [];
for (i = 0; i < noteholder.childNodes.length; i++) {
var entry = noteholder.childNodes[i];
if (entry.nodeName.toLowerCase() == 'div' && entry.getAttribute("class") == "footnote")
entriesToRemove.push(entry);
}
for (i = 0; i < entriesToRemove.length; i++) {
noteholder.removeChild(entriesToRemove[i]);
}
// Rebuild footnote entries.
var cont = document.getElementById("content");
var spans = cont.getElementsByTagName("span");
var refs = {};
var n = 0;
for (i=0; i<spans.length; i++) {
if (spans[i].className == "footnote") {
n++;
var note = spans[i].getAttribute("data-note");
if (!note) {
// Use [\s\S] in place of . so multi-line matches work.
// Because JavaScript has no s (dotall) regex flag.
note = spans[i].innerHTML.match(/\s*\[([\s\S]*)]\s*/)[1];
spans[i].innerHTML =
"[<a id='_footnoteref_" + n + "' href='#_footnote_" + n +
"' title='View footnote' class='footnote'>" + n + "</a>]";
spans[i].setAttribute("data-note", note);
}
noteholder.innerHTML +=
"<div class='footnote' id='_footnote_" + n + "'>" +
"<a href='#_footnoteref_" + n + "' title='Return to text'>" +
n + "</a>. " + note + "</div>";
var id =spans[i].getAttribute("id");
if (id != null) refs["#"+id] = n;
}
}
if (n == 0)
noteholder.parentNode.removeChild(noteholder);
else {
// Process footnoterefs.
for (i=0; i<spans.length; i++) {
if (spans[i].className == "footnoteref") {
var href = spans[i].getElementsByTagName("a")[0].getAttribute("href");
href = href.match(/#.*/)[0]; // Because IE return full URL.
n = refs[href];
spans[i].innerHTML =
"[<a href='#_footnote_" + n +
"' title='View footnote' class='footnote'>" + n + "</a>]";
}
}
}
},
install: function(toclevels) {
var timerId;
function reinstall() {
asciidoc.footnotes();
if (toclevels) {
asciidoc.toc(toclevels);
}
}
function reinstallAndRemoveTimer() {
clearInterval(timerId);
reinstall();
}
timerId = setInterval(reinstall, 500);
if (document.addEventListener)
document.addEventListener("DOMContentLoaded", reinstallAndRemoveTimer, false);
else
window.onload = reinstallAndRemoveTimer;
}
}
asciidoc.install(1);
/*]]>*/
</script>
</head>
<body class="article">
<div id="header">
<h1>cl_intel_subgroups</h1>
<div id="toc">
<div id="toctitle">Table of Contents</div>
<noscript><p><b>JavaScript must be enabled in your browser to display the table of contents.</b></p></noscript>
</div>
</div>
<div id="content">
<div class="sect1">
<h2 id="_name_strings">Name Strings</h2>
<div class="sectionbody">
<div class="paragraph"><p><span class="monospaced">cl_intel_subgroups</span></p></div>
</div>
</div>
<div class="sect1">
<h2 id="_contact">Contact</h2>
<div class="sectionbody">
<div class="paragraph"><p>Ben Ashbaugh, Intel (ben <em>dot</em> ashbaugh <em>at</em> intel <em>dot</em> com)</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_contributors">Contributors</h2>
<div class="sectionbody">
<div class="paragraph"><p>Ben Ashbaugh, Intel<br>
Allen Hux, Intel<br>
Pranayini Gudali, Intel<br>
Dawid Dominiak, Intel<br>
Biju George, Intel</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_notice">Notice</h2>
<div class="sectionbody">
<div class="paragraph"><p>Copyright (c) 2019 Intel Corporation. All rights reserved.</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_status">Status</h2>
<div class="sectionbody">
<div class="paragraph"><p>Final Draft</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_version">Version</h2>
<div class="sectionbody">
<div class="paragraph"><p>Built On: 2019-01-15<br>
Revision: 7</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_dependencies">Dependencies</h2>
<div class="sectionbody">
<div class="paragraph"><p>OpenCL 1.2 is required.
Some features (<span class="monospaced">get_num_enqueued_sub_groups()</span> and the <span class="monospaced">sub_group_barrier()</span> function that accept a memory scope) require OpenCL 2.0.</p></div>
<div class="paragraph"><p>This extension is written against revision 24 of the OpenCL 2.0 API specification, against revision 24 of the OpenCL 2.0 OpenCL C specification, and against revision 24 of the OpenCL 2.0 extension specification.</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_overview">Overview</h2>
<div class="sectionbody">
<div class="paragraph"><p>The goal of this extension is to allow programmers to improve the performance of their applications by taking advantage of the fact that some work items in a work group execute together as a group (a "subgroup"), and that work items in a subgroup can take advantage of hardware features that are not available to work items in a work group.
Specifically, this extension is designed to allow work items in a subgroup to share data without the use of local memory and work group barriers, and to utilize specialized hardware to load and store blocks of data.</p></div>
<div class="paragraph"><p>There is a large amount of overlap between the functionality in this extension and the functionality in the Khronos subgroups extension <span class="monospaced">cl_khr_subgroups</span>, so this extension reuses many of the names, concepts, and functions already described by the <span class="monospaced">cl_khr_subgroups</span> extension.
The key differences between the Intel subgroups extension and the Khronos subgroups extension are:</p></div>
<div class="ulist"><ul>
<li>
<p>
The Khronos subgroups extension requires OpenCL 2.0, but the Intel subgroups extension may be available on OpenCL 1.2 devices.
</p>
</li>
<li>
<p>
The Khronos subgroups extension guarantees that subgroups in a work group will make independent forward progress, but the Intel extension does not guarantee that subgroups in a work group will make independent forward progress.
</p>
</li>
<li>
<p>
The Intel extension adds a rich set of subgroup "shuffle" functions to allow work items within a work group to interchange data without the use of local memory and work group barriers.
</p>
</li>
<li>
<p>
The Intel extension adds a set of subgroup "block read and write" functions to take advantage of specialized hardware to read or write blocks of data from or to buffers or images.
</p>
</li>
<li>
<p>
The Intel subgroups extension does not include the subgroup pipes functions that are included as part of the Khronos subgroups extension.
</p>
</li>
<li>
<p>
The Intel subgroups extension does not include the device-side kernel query functions for subgroups that are included as part of the Khronos subgroups extension.
</p>
</li>
</ul></div>
</div>
</div>
<div class="sect1">
<h2 id="_new_api_functions">New API Functions</h2>
<div class="sectionbody">
<div class="dlist"><dl>
<dt class="hdlist1">
This function is copied unchanged from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">cl_int</span> <span style="font-weight: bold"><span style="color: #000000">clGetKernelSubGroupInfoKHR</span></span><span style="color: #990000">(</span>
<span style="color: #008080">cl_kernel</span> kernel<span style="color: #990000">,</span>
<span style="color: #008080">cl_device_id</span> device<span style="color: #990000">,</span>
<span style="color: #008080">cl_kernel_sub_group_info</span> param_name<span style="color: #990000">,</span>
<span style="color: #008080">size_t</span> input_value_size<span style="color: #990000">,</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #009900">void</span><span style="color: #990000">*</span> input_value<span style="color: #990000">,</span>
<span style="color: #008080">size_t</span> param_value_size<span style="color: #990000">,</span>
<span style="color: #009900">void</span><span style="color: #990000">*</span> param_value<span style="color: #990000">,</span>
size_t<span style="color: #990000">*</span> param_value_size_ret<span style="color: #990000">)</span></tt></pre></div></div>
</div></div>
</dd>
</dl></div>
</div>
</div>
<div class="sect1">
<h2 id="_new_api_enums">New API Enums</h2>
<div class="sectionbody">
<div class="dlist"><dl>
<dt class="hdlist1">
These enums are copied unchanged from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>Accepted as the <em>param_name</em> parameter of <strong>clGetKernelSubGroupInfoKHR</strong>:</p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt>CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE_KHR <span style="color: #993399">0x2033</span>
CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE_KHR <span style="color: #993399">0x2034</span></tt></pre></div></div>
</div></div>
</dd>
</dl></div>
</div>
</div>
<div class="sect1">
<h2 id="_new_opencl_c_functions">New OpenCL C Functions</h2>
<div class="sectionbody">
<div class="dlist"><dl>
<dt class="hdlist1">
These built-in functions are copied unchanged from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_size</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_max_sub_group_size</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_num_sub_groups</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_id</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_local_id</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_barrier</span></span><span style="color: #990000">(</span> <span style="color: #008080">cl_mem_fence_flags</span> flags <span style="color: #990000">);</span>
<span style="color: #009900">int</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_all</span></span><span style="color: #990000">(</span> <span style="color: #009900">int</span> predicate <span style="color: #990000">);</span>
<span style="color: #009900">int</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_any</span></span><span style="color: #990000">(</span> <span style="color: #009900">int</span> predicate <span style="color: #990000">);</span></tt></pre></div></div>
<div class="paragraph"><p>If OpenCL 2.0 is supported:</p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_enqueued_num_sub_groups</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_barrier</span></span><span style="color: #990000">(</span> <span style="color: #008080">cl_mem_fence_flags</span> flags<span style="color: #990000">,</span> <span style="color: #008080">memory_scope</span> scope <span style="color: #990000">);</span></tt></pre></div></div>
<div class="paragraph"><p>For the sub_group_broadcast functions, <span class="monospaced">gentype</span> is <span class="monospaced">int</span>, <span class="monospaced">uint</span>, <span class="monospaced">long</span>, <span class="monospaced">ulong</span>, or <span class="monospaced">float</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp16 is supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">half</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp64 or doubles are supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">double</span>.</p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_broadcast</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">,</span> <span style="color: #008080">uint</span> sub_group_local_id <span style="color: #990000">);</span></tt></pre></div></div>
<div class="paragraph"><p>For the sub_group_reduce, sub_group_scan_exclusive, and sub_group_scan_inclusive functions, <span class="monospaced">gentype</span> is <span class="monospaced">int</span>, <span class="monospaced">uint</span>, <span class="monospaced">long</span>, <span class="monospaced">ulong</span>, or <span class="monospaced">float</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp16 is supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">half</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp64 or doubles are supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">double</span>.</p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span></tt></pre></div></div>
</div></div>
</dd>
<dt class="hdlist1">
These built-in functions are unique to the Intel subgroups extension and are not part of the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>For the sub_group_shuffle, sub_group_shuffle_down, sub_group_shuffle_up, and sub_group_shuffle_xor functions, <span class="monospaced">gentype</span> is <span class="monospaced">float</span>, <span class="monospaced">float2</span>, <span class="monospaced">float4</span>, <span class="monospaced">float8</span>, <span class="monospaced">float16</span>, <span class="monospaced">int</span>, <span class="monospaced">int2</span>, <span class="monospaced">int4</span>, <span class="monospaced">int8</span>, <span class="monospaced">int16</span>, <span class="monospaced">uint</span>, <span class="monospaced">uint2</span>,<span class="monospaced">uint4</span>, <span class="monospaced">uint8</span>, <span class="monospaced">uint16</span>, <span class="monospaced">long</span>, or <span class="monospaced">ulong</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp16 is supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">half</span>.</p></div>
<div class="paragraph"><p>If cl_khr_fp64 or doubles are supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">double</span>.</p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> data<span style="color: #990000">,</span> <span style="color: #008080">uint</span> c <span style="color: #990000">);</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_down</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> current<span style="color: #990000">,</span> <span style="color: #008080">gentype</span> next<span style="color: #990000">,</span> <span style="color: #008080">uint</span> delta <span style="color: #990000">);</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_up</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> previous<span style="color: #990000">,</span> <span style="color: #008080">gentype</span> current<span style="color: #990000">,</span> <span style="color: #008080">uint</span> delta <span style="color: #990000">);</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_xor</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> data<span style="color: #990000">,</span> <span style="color: #008080">uint</span> value <span style="color: #990000">);</span></tt></pre></div></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read</span></span><span style="color: #990000">(</span> <span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">);</span>
<span style="color: #008080">uint2</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read2</span></span><span style="color: #990000">(</span> <span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">);</span>
<span style="color: #008080">uint4</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read4</span></span><span style="color: #990000">(</span> <span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">);</span>
<span style="color: #008080">uint8</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read8</span></span><span style="color: #990000">(</span> <span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">);</span>
<span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord <span style="color: #990000">);</span>
<span style="color: #008080">uint2</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read2</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord <span style="color: #990000">);</span>
<span style="color: #008080">uint4</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read4</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord <span style="color: #990000">);</span>
<span style="color: #008080">uint8</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read8</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write</span></span><span style="color: #990000">(</span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write2</span></span><span style="color: #990000">(</span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint2</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write4</span></span><span style="color: #990000">(</span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint4</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write8</span></span><span style="color: #990000">(</span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint8</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write2</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint2</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write4</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint4</span> data <span style="color: #990000">);</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write8</span></span><span style="color: #990000">(</span> <span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span> <span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint8</span> data <span style="color: #990000">);</span></tt></pre></div></div>
</div></div>
</dd>
</dl></div>
</div>
</div>
<div class="sect1">
<h2 id="_modifications_to_the_opencl_api_specification">Modifications to the OpenCL API Specification</h2>
<div class="sectionbody">
<div class="sect2">
<h3 id="_modifications_to_section_2_glossary">Modifications to Section 2 - "Glossary"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
Add memory_scope_sub_group to the description of Memory Scopes:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="dlist"><dl>
<dt class="hdlist1">
Memory Scopes
</dt>
<dd>
<p>
Memory scopes define a hierarchy of visibilities when analyzing the ordering constraints of memory operations.
They are defined by the values of the <span class="monospaced">memory_scope</span> enumeration constant.
Current values are <span class="monospaced">memory_scope_work_item</span> (memory constraints only apply to a single work item and in practice only apply to image operations), <span class="monospaced">memory_scope_sub_group</span> (memory-ordering constraints only apply to work items executing in a subgroup), <span class="monospaced">memory_scope_work_group</span> &#8230;
</p>
</dd>
</dl></div>
</div></div>
</dd>
<dt class="hdlist1">
Add memory_scope_sub_group to the description of Scope inclusion:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="dlist"><dl>
<dt class="hdlist1">
Scope inclusion
</dt>
<dd>
<p>
Two actions <strong>A</strong> and <strong>B</strong> are defined to have an inclusive scope if they have the same scope <strong>P</strong> such that: (1) if <strong>P</strong> is <span class="monospaced">memory_scope_sub_group</span>, and <strong>A</strong> and <strong>B</strong> are executed by work items within the same subgroup, or (2) if <strong>P</strong> is <span class="monospaced">memory_scope_work_group</span>, and <strong>A</strong> and <strong>B</strong> are executed by work items within the same workgroup &#8230;
</p>
</dd>
</dl></div>
</div></div>
</dd>
<dt class="hdlist1">
Change the description for Subgroups to:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="dlist"><dl>
<dt class="hdlist1">
Subgroup
</dt>
<dd>
<p>
Subgroups are an implementation-dependent grouping of work items within a
work group.
The size and number of subgroups is implementation-defined and not exposed in the core OpenCL 2.0 feature set.
Subgroups execute concurrently within a work group, but are not guaranteed to make independent forward progress.
Subgroups may synchronize internally using subgroup barrier operations without synchronizing with other subgroups.
</p>
</dd>
</dl></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_modifications_to_section_3_2_1_execution_model_mapping_work_items_onto_an_ndrange">Modifications to Section 3.2.1 - "Execution Model: Mapping Work Items Onto an NDRange"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
Change the paragraph describing subgroups to:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>An implementation of OpenCL may divide each work group into one or more subgroups.
The size and number of subgroups is implementation-defined and not exposed in the
core OpenCL 2.0 feature set.</p></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_modifications_to_section_3_2_2_execution_model_execution_of_kernel_instances">Modifications to Section 3.2.2 - "Execution Model: Execution of Kernel Instances"</h3>
<div class="paragraph"><p>Remove the last paragraph describing subgroups and independent forward progress.</p></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_3_2_execution_model">Additions to Section 3.2 - "Execution Model"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
This text is largely the same as the text in the Khronos subgroups extension. Only the sentence about independent forward progress has been modified:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>Within a work group, work items may be divided into subgroups in an implementation-
defined fashion. The mapping of work items to subgroups is implementation-defined
and may be queried at runtime. While subgroups may be used in multi-dimensional
work groups, each subgroup is 1-dimensional and any given work item may query which
subgroup it is a member of.</p></div>
<div class="paragraph"><p>Work items are mapped into subgroups through a combination of compile-time decisions
and the parameters of the dispatch. The mapping to subgroups is invariant for the
duration of a kernel&#8217;s execution, across dispatches of a given kernel with the same
launch parameters, and from one work group to another within the dispatch (excluding
the trailing edge work groups in the presence of non-uniform work group sizes). In
addition, all subgroups within a work group will be the same size, apart from the
subgroup with the maximum index, which may be smaller if the size of the work group
is not evenly divisible by the size of the subgroups.</p></div>
<div class="paragraph"><p>Subgroups execute concurrently within a given work group. Similar to work items
within a work group, subgroups executing within a work group are not guaranteed to make
independent forward progress. Work items in a subgroup can internally synchronize
using subgroup barrier operations without synchronizing with other subgroups.</p></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_3_3_4_memory_model_memory_consistency_model">Additions to Section 3.3.4 - "Memory Model: Memory Consistency Model"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
Add memory_scope_sub_group to the bulleted descriptions of memory scopes:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="ulist"><ul>
<li>
<p>
<span class="monospaced">memory_scope_sub_group</span>: memory-ordering constraints only apply to work items executing within a single subgroup.
</p>
</li>
<li>
<p>
<span class="monospaced">memory_scope_work_group</span>: &#8230;
</p>
</li>
</ul></div>
</div></div>
</dd>
<dt class="hdlist1">
In the paragraph after the bulleted descriptions of memory scopes, include memory_scope_sub_group as a valid memory scope for local memory:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>... For local memory, <span class="monospaced">memory_scope_sub_group</span> and <span class="monospaced">memory_scope_work_group</span> are valid, and may constrain visibility to the subgroup or workgroup.</p></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_3_3_5_memory_model_overview_of_atomic_and_fence_operations">Additions to Section 3.3.5 - "Memory Model: Overview of atomic and fence operations"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
Add memory_scope_sub_group to the definition of inclusive scope:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="ulist"><ul>
<li>
<p>
<strong>P</strong> is <span class="monospaced">memory_scope_sub_group</span> and <strong>A</strong> and <strong>B</strong> are executed by work items within the same subgroup.
</p>
</li>
<li>
<p>
<strong>P</strong> is <span class="monospaced">memory_scope_work_group</span> &#8230;
</p>
</li>
</ul></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_5_9_3_kernel_object_queries">Additions to Section 5.9.3 - "Kernel Object Queries"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
This addition is copied unchanged from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>The function</p></div>
<div class="paragraph"><p></p></div>
<div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">cl_int</span> <span style="font-weight: bold"><span style="color: #000000">clGetKernelSubGroupInfoKHR</span></span><span style="color: #990000">(</span><span style="color: #008080">cl_kernel</span> kernel<span style="color: #990000">,</span>
<span style="color: #008080">cl_device_id</span> device<span style="color: #990000">,</span>
<span style="color: #008080">cl_kernel_sub_group_info</span> param_name<span style="color: #990000">,</span>
<span style="color: #008080">size_t</span> input_value_size<span style="color: #990000">,</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #009900">void</span> <span style="color: #990000">*</span>input_value<span style="color: #990000">,</span>
<span style="color: #008080">size_t</span> param_value_size<span style="color: #990000">,</span>
<span style="color: #009900">void</span> <span style="color: #990000">*</span>param_value<span style="color: #990000">,</span>
<span style="color: #008080">size_t</span> <span style="color: #990000">*</span>param_value_size_ret<span style="color: #990000">)</span></tt></pre></div></div>
<div class="paragraph"><p>returns information about the kernel object.</p></div>
<div class="paragraph"><p><em>kernel</em> specifies the kernel object being queried.</p></div>
<div class="paragraph"><p><em>device</em> identifies a specific device in the list of devices associated with
<em>kernel</em>.
The list of devices is the list of devices in the OpenCL context that is
associated with <em>kernel</em>.
If the list of devices associated with <em>kernel</em> is a single device, <em>device</em>
can be a <span class="monospaced">NULL</span> value.</p></div>
<div class="paragraph"><p><em>param_name</em> specifies the information to query.
The list of supported <em>param_name</em> types and the information returned in
<em>param_value</em> by <strong>clGetKernelSubGroupInfoKHR</strong> is described in the table below.</p></div>
<div class="paragraph"><p><em>input_value_size</em> is used to specify the size in bytes of memory pointed to
by <em>input_value</em>.
This size must be equal to the size of input type as described in the table below.</p></div>
<div class="paragraph"><p><em>input_value</em> is a pointer to memory where the appropriate parameterization
of the query is passed from.
If <em>input_value</em> is <span class="monospaced">NULL</span>, it is ignored.</p></div>
<div class="paragraph"><p><em>param_value</em> is a pointer to memory where the appropriate result being
queried is returned.
If <em>param_value</em> is <span class="monospaced">NULL</span>, it is ignored.</p></div>
<div class="paragraph"><p><em>param_value_size</em> is used to specify the size in bytes of memory pointed to
by <em>param_value</em>.
This size must be greater than or equal to the size of the return type as described in the
table below.</p></div>
<div class="paragraph"><p><em>param_value_size_ret</em> returns the actual size in bytes of data being
queried by <em>param_name</em>.
If <em>param_value_size_ret</em> is <span class="monospaced">NULL</span>, it is ignored.</p></div>
<table class="tableblock frame-all grid-all" id="cl_khr_subgroups-kernel-subgroup-info-table"
style="
width:100%;
">
<caption class="title">Table 1. <strong>clGetKernelSubGroupInfoKHR</strong> parameter queries</caption>
<col style="width:25%;">
<col style="width:25%;">
<col style="width:25%;">
<col style="width:25%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>cl_kernel_sub_group_info</strong> </th>
<th class="tableblock halign-left valign-top" > Input Type </th>
<th class="tableblock halign-left valign-top" > Return Type </th>
<th class="tableblock halign-left valign-top" > Info. returned in <em>param_value</em></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><p class="tableblock"><strong>CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE_KHR</strong></p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">size_t *</p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">size_t</p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the maximum sub-group size for this kernel.
All sub-groups must be the same size, while the last subgroup in
any work-group (i.e. the subgroup with the maximum index) could
be the same or smaller size.</p>
<p class="tableblock"> The <em>input_value</em> must be an array of size_t values
corresponding to the local work size parameter of the intended
dispatch.
The number of dimensions in the ND-range will be inferred from
the value specified for <em>input_value_size</em>.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><p class="tableblock"><strong>CL_KERNEL_SUB_GROUP_COUNT_FOR_NDRANGE_KHR</strong></p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">size_t *</p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">size_t</p></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the number of sub-groups that will be present in each
work-group for a given local work size.
All workgroups, apart from the last work-group in each dimension
in the presence of non-uniform work-group sizes, will have the
same number of sub-groups.</p>
<p class="tableblock"> The <em>input_value</em> must be an array of size_t values
corresponding to the local work size parameter of the intended
dispatch.
The number of dimensions in the ND-range will be inferred from
the value specified for <em>input_value_size</em>.</p></td>
</tr>
</tbody>
</table>
<div class="paragraph"><p><strong>clGetKernelSubGroupInfoKHR</strong> returns CL_SUCCESS if the function is executed
successfully.
Otherwise, it returns one of the following errors:</p></div>
<div class="ulist"><ul>
<li>
<p>
<span class="monospaced">CL_INVALID_DEVICE</span> if <em>device</em> is not in the list of devices associated
with <em>kernel</em> or if <em>device</em> is <span class="monospaced">NULL</span> but there is more than one device
associated with <em>kernel</em>.
</p>
</li>
<li>
<p>
<span class="monospaced">CL_INVALID_VALUE</span> if <em>param_name</em> is not valid, or if size in bytes
specified by <em>param_value_size</em> is less than the size of return type as described in
the table above and <em>param_value</em> is not <span class="monospaced">NULL</span>.
</p>
</li>
<li>
<p>
<span class="monospaced">CL_INVALID_VALUE</span> if <em>param_name</em> is
<span class="monospaced">CL_KERNEL_MAX_SUB_GROUP_SIZE_FOR_NDRANGE</span> and the size in bytes specified by
<em>input_value_size</em> is not valid or if <em>input_value</em> is <span class="monospaced">NULL</span>.
</p>
</li>
<li>
<p>
<span class="monospaced">CL_INVALID_KERNEL</span> if <em>kernel</em> is a not a valid kernel object.
</p>
</li>
<li>
<p>
<span class="monospaced">CL_OUT_OF_RESOURCES</span> if there is a failure to allocate resources required
by the OpenCL implementation on the device.
</p>
</li>
<li>
<p>
<span class="monospaced">CL_OUT_OF_HOST_MEMORY</span> if there is a failure to allocate resources
required by the OpenCL implementation on the host.
</p>
</li>
</ul></div>
</div></div>
</dd>
</dl></div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_modifications_to_the_opencl_c_specification">Modifications to the OpenCL C Specification</h2>
<div class="sectionbody">
<div class="sect2">
<h3 id="_additions_to_section_6_13_1_work_item_functions">Additions to section 6.13.1 - "Work Item Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
These additions are copied unchanged from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:50%;">
<col style="width:50%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_size</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the number of work items in the subgroup.
This value is no more than the maximum subgroup size and is implementation-defined based on a combination of the compiled kernel and the dispatch dimensions.
This will be a constant value for the lifetime of the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_max_sub_group_size</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the maximum size of a subgroup with the dispatch.
This value will be invariant for a given set of dispatch dimensions and a kernel object compiled for a given device.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_num_sub_groups</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the number of subgroups that the current work group is divided into.</p>
<p class="tableblock">This number will be constant for the duration of a work group&#8217;s execution.
If the kernel is executed with a non-uniform work group size in any dimension, calls to this built-in may return a different values for some work groups than for other work groups.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_id</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the subgroup ID, which is a number from zero to <strong>get_num_sub_groups</strong> - 1.</p>
<p class="tableblock">For <strong>clEnqueueTask</strong>, this returns 0.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_sub_group_local_id</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the unique work item ID within the current subgroup.
The mapping from <strong>get_local_id</strong> to <strong>get_sub_group_local_id</strong> will be invariant for the lifetime of the work group.</p></td>
</tr>
</tbody>
</table>
<div class="paragraph"><p>If OpenCL 2.0 is supported:</p></div>
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:50%;">
<col style="width:50%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">get_enqueued_num_sub_groups</span></span><span style="color: #990000">(</span> <span style="color: #009900">void</span> <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the same value as that returned by <strong>get_num_sub_groups</strong> if the kernel is executed with a uniform work group size. This value will be constant for the entire NDRange.</p>
<p class="tableblock">If the kernel is executed with a non-uniform work group size, returns the number of subgroups in a work group that makes up the uniform region of the global NDRange.</p></td>
</tr>
</tbody>
</table>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_6_13_8_synchronization_functions">Additions to Section 6.13.8 - "Synchronization Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
These additions are mostly unchanged from the Khronos subgroups extension, with only minor edits for clarity:
</dt>
<dd>
<div class="openblock">
<div class="content">
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:50%;">
<col style="width:50%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_barrier</span></span><span style="color: #990000">(</span>
<span style="color: #008080">cl_mem_fence_flags</span> flags <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">All work items in a subgroup executing the kernel on a processor must execute this function before any are allowed to continue execution beyond the subgroup barrier.
This function must be encountered by all work items in a subgroup executing the kernel.
These rules apply to NDRanges implemented with uniform and non-uniform work groups.</p>
<p class="tableblock">If <strong>sub_group_barrier</strong> is inside a conditional statement then all work items within the subgroup must enter the conditional if any work item in the subgroup enters the conditional statement and executes the <strong>sub_group_barrier</strong>.</p>
<p class="tableblock">If <strong>sub_group_barrier</strong> is inside a loop, all work items within the subgroup must execute the <strong>sub_group_barrier</strong> for each iteration of the loop before any are allowed to continue execution beyond the <strong>sub_group_barrier</strong>.</p>
<p class="tableblock">The <strong>sub_group_barrier</strong> function also queues a memory fence (reads and writes) to ensure correct ordering of memory operations to local or global memory.</p>
<p class="tableblock">The flags argument specifies the memory address space and can be set to a combination of the following values:</p>
<p class="tableblock"><span class="monospaced">CLK_LOCAL_MEM_FENCE</span> - The <strong>sub_group_barrier</strong> function will either flush any variables stored in local memory or queue a memory fence to ensure correct ordering of memory operations to local memory.</p>
<p class="tableblock"><span class="monospaced">CLK_GLOBAL_MEM_FENCE</span> - The <strong>sub_group_barrier</strong> function will queue a memory fence to ensure correct ordering of memory operations to global memory.
This can be useful when work items, for example, write to buffer objects and then want to read the updated data from these buffer objects.</p></td>
</tr>
</tbody>
</table>
<div class="paragraph"><p>If OpenCL 2.0 is supported, add the following to the table above:</p></div>
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:50%;">
<col style="width:50%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_barrier</span></span><span style="color: #990000">(</span>
<span style="color: #008080">cl_mem_fence_flags</span> flags<span style="color: #990000">,</span>
<span style="color: #008080">memory_scope</span> scope <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">&#8230;</p>
<p class="tableblock">The <strong>sub_group_barrier</strong> function also supports a variant that specifies the memory scope.
For the sub_group_barrier variant that does not take a memory scope, the scope is <span class="monospaced">memory_scope_sub_group</span>.</p>
<p class="tableblock">The scope argument specifies whether the memory accesses of work items in the subgroup to memory address space(s) identified by flags become visible to all work items in the subgroup, the work group, the device, or all SVM devices.</p>
<p class="tableblock">&#8230;</p>
<p class="tableblock"><span class="monospaced">CLK_IMAGE_MEM_FENCE</span> - The <strong>sub_group_barrier</strong> function will queue a memory fence to ensure correct ordering of memory operations to image objects. This can be useful when work items, for example, write to image objects and then want to read the updated data from these image objects.</p></td>
</tr>
</tbody>
</table>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_6_13_11_atomic_functions">Additions to Section 6.13.11 - "Atomic Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
Modify the bullet describing behavior for functions that do not have a memory_scope argument to say:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="ulist"><ul>
<li>
<p>
The subgroup functions that do not have a <em>memory_scope</em> argument have the same semantics as the corresponding functions with the <em>memory_scope</em> argument set to <span class="monospaced">memory_scope_sub_group</span>.
Other functions that do not have a <em>memory_scope</em> argument have the same semantics as the corresponding functions with the <em>memory_scope</em> argument set to <span class="monospaced">memory_scope_device</span>.
</p>
</li>
</ul></div>
</div></div>
</dd>
<dt class="hdlist1">
The following addition is copied unchanged from the Khronos subgroups extension:
</dt>
<dt class="hdlist1">
Add the following new value to the enumerated type memory_scope defined in Section 6.13.11.4:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="listingblock">
<div class="content monospaced">
<pre>memory_scope_sub_group</pre>
</div></div>
<div class="paragraph"><p>The <span class="monospaced">memory_scope_sub_group</span> specifies that the memory ordering constraints
given by <span class="monospaced">memory_order</span> apply to work items in a subgroup.
This memory scope can be used when performing atomic operations to global or
local memory.</p></div>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_additions_to_section_6_13_15_work_group_functions">Additions to Section 6.13.15 - "Work Group Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
These additions are copied from the Khronos subgroups extension:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>The OpenCL C programming language implements the following built-in
functions that operate on a subgroup level.
These built-in functions must be encountered by all work items in a subgroup
executing the kernel.
We use the generic type name <span class="monospaced">gentype</span> to indicate the built-in data types
<span class="monospaced">int</span>, <span class="monospaced">uint</span>, <span class="monospaced">long</span>, <span class="monospaced">ulong</span>, or <span class="monospaced">float</span> as the type for the arguments.</p></div>
<div class="paragraph"><p>If <span class="monospaced">cl_khr_fp16</span> is supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">half</span>.</p></div>
<div class="paragraph"><p>If <span class="monospaced">cl_khr_fp64</span> or doubles are supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">double</span>.</p></div>
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:66%;">
<col style="width:33%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">int</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_all</span></span><span style="color: #990000">(</span> <span style="color: #009900">int</span> predicate <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Evaluates <em>predicate</em> for all work items in the subgroup and returns a
non-zero value if <em>predicate</em> evaluates to non-zero for all work items in
the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">int</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_any</span></span><span style="color: #990000">(</span> <span style="color: #009900">int</span> predicate <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Evaluates <em>predicate</em> for all work items in the subgroup and returns a
non-zero value if <em>predicate</em> evaluates to non-zero for any work items in
the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_broadcast</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> x<span style="color: #990000">,</span>
<span style="color: #008080">uint</span> sub_group_local_id <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Broadcasts the value of <em>x</em> for work item identified by <em>sub_group_local_id</em> (value returned by <strong>get_sub_group_local_id</strong>) to all work items in the subgroup.
<em>sub_group_local_id</em> must be the same value for all work items in the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_reduce_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Returns the result of the specified reduction operation for all values of <em>x</em> specified by work items in a subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_exclusive_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Performs the specified exclusive scan operation of all values <em>x</em> specified by work items in a subgroup.
The scan results are returned for each work item.</p>
<p class="tableblock">The scan order is defined by increasing subgroup local ID within the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_add</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_min</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span>
<span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">sub_group_scan_inclusive_max</span></span><span style="color: #990000">(</span> <span style="color: #008080">gentype</span> x<span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Performs the specified inclusive scan operation of all values <em>x</em> specified by work items in a subgroup.
The scan results are returned for each work item.</p>
<p class="tableblock">The scan order is defined by increasing subgroup local ID within the subgroup.</p></td>
</tr>
</tbody>
</table>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_add_a_new_section_6_13_x_sub_group_shuffle_functions">Add a new Section 6.13.X - "Sub Group Shuffle Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
These are new functions:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>The OpenCL C programming language implements the following built-in functions to allow data to be exchanged among work items in a subgroup.
These built-in functions need not be encountered by all work items in a subgroup executing the kernel, however, data may only be shuffled among work items encountering the subgroup shuffle function.
Shuffling data from a work item that does not encounter the subgroup shuffle function will produce undefined results.
For these functions, <span class="monospaced">gentype</span> is <span class="monospaced">float</span>, <span class="monospaced">float2</span>, <span class="monospaced">float4</span>, <span class="monospaced">float8</span>, <span class="monospaced">float16</span>, <span class="monospaced">int</span>, <span class="monospaced">int2</span>, <span class="monospaced">int4</span>, <span class="monospaced">int8</span>, <span class="monospaced">int16</span>, <span class="monospaced">uint</span>, <span class="monospaced">uint2</span>, <span class="monospaced">uint4</span>, <span class="monospaced">uint8</span>, <span class="monospaced">uint16</span>, <span class="monospaced">long</span>, or <span class="monospaced">ulong</span>.</p></div>
<div class="paragraph"><p>If <span class="monospaced">cl_khr_fp16</span> is supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">half</span>.</p></div>
<div class="paragraph"><p>If <span class="monospaced">cl_khr_fp64</span> or doubles are supported, <span class="monospaced">gentype</span> also includes <span class="monospaced">double</span>.</p></div>
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:50%;">
<col style="width:50%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" > <strong>Function</strong></th>
<th class="tableblock halign-left valign-top" > <strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> data<span style="color: #990000">,</span>
<span style="color: #008080">uint</span> sub_group_local_id <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Allows data to be arbitrarily transferred between work items in a subgroup.
The data that is returned for this work item is the value of <em>data</em> for the work item identified by <em>sub_group_local_id</em>.</p>
<p class="tableblock"><em>sub_group_local_id</em> need not be the same value for all work items in the subgroup.
There is no defined behavior for out-of-range <em>sub_group_local_ids</em>.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_down</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> current<span style="color: #990000">,</span>
<span style="color: #008080">gentype</span> next<span style="color: #990000">,</span>
<span style="color: #008080">uint</span> delta <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Allows data to be transferred from a work item in the subgroup with a higher sub_group_local_id down to a work item in the subgroup with a lower sub_group_local_id.</p>
<p class="tableblock">There are two data sources to this built-in function: <em>current</em> and <em>next</em>.
To determine the result of this built-in function, first let the unsigned shuffle index be equivalent to the sum of this work item&#8217;s sub_group_local_id plus the specified <em>delta</em>:</p>
<p class="tableblock">If the shuffle index is less than the max_sub_group_size, the result of this built-in function is the value of the <em>current</em> data source for the work item with sub_group_local_id equal to the shuffle index.</p>
<p class="tableblock">If the shuffle index is greater than or equal to the max_sub_group_size but less than twice the max_sub_group_size, the result of this built-in function is the value of the <em>next</em> data source for the work item with sub_group_local_id equal to the shuffle index minus the max_sub_group_size.</p>
<p class="tableblock">All other values of the shuffle index are considered to be out-of-range.
There is no defined behavior for out-of-range indices.</p>
<p class="tableblock"><em>delta</em> need not be the same value for all work items in the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_up</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> previous<span style="color: #990000">,</span>
<span style="color: #008080">gentype</span> current<span style="color: #990000">,</span>
<span style="color: #008080">uint</span> delta <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Allows data to be transferred from a work item in the subgroup with a lower sub_group_local_id up to a work item in the subgroup with a higher sub_group_local_id.</p>
<p class="tableblock">There are two data sources to this built-in function: <em>previous</em> and <em>current</em>.
To determine the result of this built-in function, first let the signed shuffle index be equivalent to this work item&#8217;s sub_group_local_id minus the specified <em>delta</em>:</p>
<p class="tableblock">If the shuffle index is greater than or equal to zero and less than the max_sub_group_size, the result of this built-in function is the value of the <em>current</em> data source for the work item with sub_group_local_id equal to the shuffle index.</p>
<p class="tableblock">If the shuffle index is less than zero but greater than or equal to the negative max_sub_group_size, the result of this built-in function is the value of the previous data source for the work item with sub_group_local_id equal to the shuffle index plus the max_sub_group_size.</p>
<p class="tableblock">All other values of the shuffle index are considered to be out-of-range.
There is no defined behavior for out-of-range indices.</p>
<p class="tableblock"><em>delta</em> need not be the same value for all work items in the subgroup.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">gentype</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_shuffle_xor</span></span><span style="color: #990000">(</span>
<span style="color: #008080">gentype</span> data<span style="color: #990000">,</span>
<span style="color: #008080">uint</span> value <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Allows data to be transferred between work items in a subgroup as a function of the work item&#8217;s sub_group_local_id.
The data that is returned for this work item is the value of <em>data</em> for the work item with sub_group_local_id equal to this work item&#8217;s sub_group_local_id XOR&#8217;d with the specified <em>value</em>.
If the result of the XOR is greater than max_sub_group_size then it is considered out-of-range.</p>
<p class="tableblock"><em>value</em> need not be the same for all work items in the subgroup.
There is no defined behavior for out-of-range indices.</p></td>
</tr>
</tbody>
</table>
</div></div>
</dd>
</dl></div>
</div>
<div class="sect2">
<h3 id="_add_a_new_section_6_13_x_sub_group_read_and_write_functions">Add a new Section 6.13.X - "Sub Group Read and Write Functions"</h3>
<div class="dlist"><dl>
<dt class="hdlist1">
These are new functions:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>The OpenCL C programming language implements the following built-in functions to allow data to be read or written as a block by all work items in a subgroup.
These built-in functions must be encountered by all work items in a subgroup executing the kernel.
Furthermore, since these are block operations, the <em>pointer</em>, <em>image</em>, and <em>coordinate</em> arguments to these built-in functions must be the same for all work items in the subgroup (when applicable, only the <em>data</em> argument may be different).</p></div>
<table class="tableblock frame-all grid-all"
style="
width:100%;
">
<col style="width:55%;">
<col style="width:44%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" ><strong>Function</strong></th>
<th class="tableblock halign-left valign-top" ><strong>Description</strong></th>
</tr>
</thead>
<tbody>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read</span></span><span style="color: #990000">(</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">)</span>
<span style="color: #008080">uint2</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read2</span></span><span style="color: #990000">(</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">)</span>
<span style="color: #008080">uint4</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read4</span></span><span style="color: #990000">(</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">)</span>
<span style="color: #008080">uint8</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read8</span></span><span style="color: #990000">(</span>
<span style="font-weight: bold"><span style="color: #0000FF">const</span></span> <span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Reads 1, 2, 4, or 8 uints of data for each work item in the subgroup from the specified pointer as a block operation.
The data is read strided, so the first value read is:</p>
<p class="tableblock"><span class="monospaced">p[ sub_group_local_id ]</span></p>
<p class="tableblock">and the second value read is:</p>
<p class="tableblock"><span class="monospaced">p[ sub_group_local_id + max_sub_group_size ]</span></p>
<p class="tableblock">etc.</p>
<p class="tableblock"><em>p</em> must be aligned to a 32-bit (4-byte) boundary.</p>
<p class="tableblock">There is no defined out-of-range behavior for these functions.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #008080">uint</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord <span style="color: #990000">)</span>
<span style="color: #008080">uint2</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read2</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord <span style="color: #990000">)</span>
<span style="color: #008080">uint4</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read4</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord <span style="color: #990000">)</span>
<span style="color: #008080">uint8</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_read8</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Reads 1, 2, 4, or 8 uints of data for each work item in the subgroup from the specified <em>image</em> at the specified coordinate as a block operation.
Note that the coordinate is a byte coordinate, not an image element coordinate.
Also note that the image data is read without format conversion, so each work item may read multiple image elements
(for images with element size smaller than 16-bits).</p>
<p class="tableblock">The data is read row-by-row, so the first value read is from the row specified in the y-component of the provided <em>byte_coord</em>, the second value is read from the y-component of the provided <em>byte_coord</em> plus one, etc.</p>
<p class="tableblock">Please see the note below describing out-of-bounds behavior for these functions.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write</span></span><span style="color: #990000">(</span>
<span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write2</span></span><span style="color: #990000">(</span>
<span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint2</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write4</span></span><span style="color: #990000">(</span>
<span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint4</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write8</span></span><span style="color: #990000">(</span>
<span style="color: #008080">__global</span> uint<span style="color: #990000">*</span> p<span style="color: #990000">,</span> <span style="color: #008080">uint8</span> data <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Writes 1, 2, 4, or 8 uints of data for each work item in the subgroup to the specified pointer as a block operation.
The data is written strided, so the first value is written to:</p>
<p class="tableblock"><span class="monospaced">p[ sub_group_local_id ]</span></p>
<p class="tableblock">and the second value is written to:</p>
<p class="tableblock"><span class="monospaced">p[ sub_group_local_id + max_sub_group_size ]</span></p>
<p class="tableblock">etc.</p>
<p class="tableblock"><em>p</em> must be aligned to a 128-bit (16-byte) boundary.</p>
<p class="tableblock">There is no defined out-of-range behavior for these functions.</p></td>
</tr>
<tr>
<td class="tableblock halign-left valign-top" ><div><div class="listingblock">
<div class="content"><!-- Generator: GNU source-highlight 3.1.8
by Lorenzo Bettini
http://www.lorenzobettini.it
http://www.gnu.org/software/src-highlite -->
<pre><tt><span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write2</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint2</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write4</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint4</span> data <span style="color: #990000">)</span>
<span style="color: #009900">void</span> <span style="font-weight: bold"><span style="color: #000000">intel_sub_group_block_write8</span></span><span style="color: #990000">(</span>
<span style="color: #008080">image2d_t</span> image<span style="color: #990000">,</span>
<span style="color: #008080">int2</span> byte_coord<span style="color: #990000">,</span> <span style="color: #008080">uint8</span> data <span style="color: #990000">)</span></tt></pre></div></div></div></td>
<td class="tableblock halign-left valign-top" ><p class="tableblock">Writes 1, 2, 4, or 8 uints of data for each work item in the subgroup to the specified <em>image</em> at the specified coordinate as a block operation.
Note that the coordinate is a byte coordinate, not an image element coordinate.
Unlike the image block read function, which may read from any arbitrary byte offset, the x-component of the byte coordinate for the image block write functions must be a multiple of four;
in other words, the write must begin at 32-bit boundary.
There is no restriction on the y-component of the coordinate.
Also, note that the image <em>data</em> is written without format conversion, so each work item may write multiple image elements (for images with element size smaller than 8-bits).</p>
<p class="tableblock">The data is written row-by-row, so the first value written is from the row specified by the y-component of the provided <em>byte_coord</em>, the second value is written from the y-component of the provided <em>byte_coord</em> plus one, etc.</p>
<p class="tableblock">Please see the note below describing out-of-bounds behavior for these functions.</p></td>
</tr>
</tbody>
</table>
<div class="paragraph"><p>Note: The subgroup image block read and write built-ins do support bounds checking, however these built-ins bounds-check to the image width in units of uints, not in units of image elements.
This means:</p></div>
<div class="ulist"><ul>
<li>
<p>
If the image has an element size equal to the size of a uint (four bytes, for example <span class="monospaced">CL_RGBA</span> + <span class="monospaced">CL_UNORM_INT8</span>), the image will be correctly bounds-checked.
In this case, out-of-bounds reads will return the edge image element (the equivalent of <span class="monospaced">CLK_ADDRESS_CLAMP_TO_EDGE</span>), and out-of-bounds writes will be ignored.
</p>
</li>
<li>
<p>
If the image has element size less than the size of a uint (such as <span class="monospaced">CL_R</span> + <span class="monospaced">CL_UNSIGNED_INT8</span>), the entire image is addressable, however bounds checking will occur too late.
For this reason, extra care should be taken to avoid out-of-bounds reads and writes, since out-of-bounds reads may return invalid data and out-of-bounds writes may corrupt other images or buffers unpredictably.
</p>
</li>
</ul></div>
</div></div>
</dd>
<dt class="hdlist1">
Add a new sub-section 6.13.X.1 - Restrictions:
</dt>
<dd>
<div class="openblock">
<div class="content">
<div class="paragraph"><p>The following restrictions apply to the subgroup buffer block read and write functions:</p></div>
<div class="ulist"><ul>
<li>
<p>
The pointer <em>p</em> must be 32-bit (4-byte) aligned for reads, and must be 128-bit (16-byte) aligned for writes.
</p>
</li>
<li>
<p>
If the pointer <em>p</em> is computed from a kernel argument that is a cl_mem that was created with <span class="monospaced">CL_MEM_USE_HOST_PTR</span>, then the <em>host_ptr</em> must be 32-bit (4-byte) aligned for reads, and must be 128-bit (16-byte) aligned for writes.
</p>
</li>
<li>
<p>
If the pointer <em>p</em> is computed from a kernel argument that is a cl_mem that is a sub-buffer, then the <em>origin</em> defining the sub-buffer offset into the <em>buffer</em> must be a multiple of 4 bytes for reads, and must be a multiple of 16 bytes for write, in addition to the <span class="monospaced">CL_DEVICE_MEM_BASE_ADDR_ALIGN</span> requirements.
Additionally, if the <em>buffer</em> that the sub-buffer is created from was created with <span class="monospaced">CL_MEM_USE_HOST_PTR</span>, then the <em>host_ptr</em> for the <em>buffer</em> must be 32-bit (4-byte) aligned for reads, and must be 128-bit(16-byte) aligned for writes.
</p>
</li>
<li>
<p>
If the pointer <em>p</em> is computed from an SVM pointer kernel argument, then the SVM pointer kernel argument must be 32-bit (4-byte) aligned for reads, and must be 128-bit (16-byte) aligned for writes.
</p>
</li>
</ul></div>
<div class="paragraph"><p>The following restrictions apply to the subgroup image block read and write functions:</p></div>
<div class="ulist"><ul>
<li>
<p>
The behavior of the subgroup image block read and write built-ins is undefined for images with an element size greater than four bytes (such as <span class="monospaced">CL_RGBA</span> + <span class="monospaced">CL_FLOAT</span>).
</p>
</li>
<li>
<p>
When reading or writing a 2D image created from a buffer with the subgroup block read and write built-ins, the image row pitch is required to be a multiple of 64-bytes, in addition to the <span class="monospaced">CL_DEVICE_IMAGE_PITCH_ALIGNMENT</span> requirements.
</p>
</li>
<li>
<p>
When reading or writing a 2D image created from a buffer with the subgroup block read and write built-ins, if the buffer is a cl_mem that was created with <span class="monospaced">CL_MEM_USE_HOST_PTR</span>, then the <em>host_ptr</em> must be 256-bit (32-byte) aligned.
</p>
</li>
<li>
<p>
When reading or writing a 2D image created from a buffer with the subgroup block read and write built-ins, if the buffer is a cl_mem that is a sub-buffer, then the <em>origin</em> must be a multiple of 32-bytes.
Additionally, if the <em>buffer</em> that the sub-buffer is created from was created with CL_MEM_USE_HOST_PTR, then the <em>host_ptr</em> for the <em>buffer</em> must be 256-bit (32-byte) aligned.
</p>
</li>
</ul></div>
</div></div>
</dd>
</dl></div>
</div>
</div>
</div>
<div class="sect1">
<h2 id="_issues">Issues</h2>
<div class="sectionbody">
<div class="paragraph"><p>None.</p></div>
</div>
</div>
<div class="sect1">
<h2 id="_revision_history">Revision History</h2>
<div class="sectionbody">
<table class="tableblock frame-all grid-rows"
style="
width:100%;
">
<col style="width:4%;">
<col style="width:14%;">
<col style="width:14%;">
<col style="width:66%;">
<thead>
<tr>
<th class="tableblock halign-left valign-top" >Rev</th>
<th class="tableblock halign-left valign-top" >Date</th>
<th