<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-2022-jp">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:PMingLiU;
panose-1:2 2 5 0 0 0 0 0 0 0;}
@font-face
{font-family:PMingLiU;
panose-1:2 2 5 0 0 0 0 0 0 0;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:Tahoma;
panose-1:2 11 6 4 3 5 4 4 2 4;}
@font-face
{font-family:"\@PMingLiU";
panose-1:2 2 5 0 0 0 0 0 0 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"PMingLiU","serif";
mso-fareast-language:ZH-TW;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.MsoAcetate, li.MsoAcetate, div.MsoAcetate
{mso-style-priority:99;
mso-style-link:"Balloon Text Char";
margin:0cm;
margin-bottom:.0001pt;
font-size:8.0pt;
font-family:"Tahoma","sans-serif";
mso-fareast-language:ZH-TW;}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
{mso-style-priority:34;
margin-top:0cm;
margin-right:0cm;
margin-bottom:0cm;
margin-left:36.0pt;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"PMingLiU","serif";
mso-fareast-language:ZH-TW;}
span.BalloonTextChar
{mso-style-name:"Balloon Text Char";
mso-style-priority:99;
mso-style-link:"Balloon Text";
font-family:"Tahoma","sans-serif";
mso-fareast-language:ZH-TW;}
span.apple-tab-span
{mso-style-name:apple-tab-span;}
span.apple-converted-space
{mso-style-name:apple-converted-space;}
span.EmailStyle22
{mso-style-type:personal-reply;
font-family:"Calibri","sans-serif";
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-GB" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Hi Bekket,<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal">> In the normal circumstance, it would print 3. But in OpenCL-C, since we bind x as constant upon definition, it would print 2. That<span lang="ZH-TW">$B!G(B</span>s why we think our
<span lang="ZH-TW">$B!H(B</span>flatten<span lang="ZH-TW">$B!I(B</span> approach could work: Copy captured variables as constants just one time would make a lot easier. <o:p></o:p></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">I am just not sure what you mean by $B!H(Bnormal$B!I(B here, because looking at ObjC code generation it does the following for the example similar to yours:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">typedef int (^const Block_t)(void);<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">void foo(){<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">int x = 1;<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Block_t myBlock = ^(void){<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> return x+1;<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">};<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">x = 2;<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">myBlock();<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">}<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">1. Allocate and initialise x:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> %x = alloca i32<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> store i32 1, i32* %x<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">2. Allocate block on the stack:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> %block = alloca <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }><o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> $B!D(B<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">3. Store captures into the block fields:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> %block.captured = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32 }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i32
}>* %block, i32 0, i32 5<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> %0 = load i32, i32* %x<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> store i32 %0, i32* %block.captured<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">4. Assign the new value to x and the rest of the code:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> store i32 2, i32* %x<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> $B!D(B<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">5. When we are calling the block with:<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> %call = call i32 %6(i8* %4)<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">The value of x will be used from the passed in capture field of the %block variable on the stack and its value is 1 (and not 2). I don$B!G(Bt see any difference
between ObjC and OpenCL in this case.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Regarding enqueue_kernel, for the moment we plan to just pass the local stack block descriptor variable into the generated IR call. The implementation of enqueue_kernel
builtin can decide how to proceed with the values of the stack variable containing all the useful flags and captures. This implementation would however imply it has to be copied in some global buffer to be accessible outside, which isn$B!G(Bt ideal. It would be
nicer to avoid memory copies overall. However, it doesn$B!G(Bt seem easy without significant modification to IR codegen or adding some sort of simplified global heap allocation supported on device.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">However, you are right considering the restrictions to blocks in OpenCL Spec we could modify generated IR to make it more efficient. If you could provide more
details on your $B!H(Bflatten$B!I(B approach or perhaps create the review/share some patches we could look into extending the upstream implementation provided that everyone finds it suitable.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Thanks,<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Anastasia<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"><o:p> </o:p></span></p>
<div>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> Bekket McClane [<a href="mailto:bekket.mcclane@gmail.com">mailto:bekket.mcclane@gmail.com</a>]
<br>
<b>Sent:</b> 20 March 2016 16:19<br>
<b>To:</b> Anastasia Stulova<br>
<b>Cc:</b> <a href="mailto:cfe-dev@lists.llvm.org">cfe-dev@lists.llvm.org</a><br>
<b>Subject:</b> Re: [cfe-dev] About OpenCL 2.x Dynamic Parallelism<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">Anastasia Stulova <<a href="mailto:Anastasia.Stulova@arm.com">Anastasia.Stulova@arm.com</a>>
<span lang="ZH-TW">$B1w(B</span> 2016<span lang="ZH-TW">$BG/(B</span>3<span lang="ZH-TW">$B7n(B</span>20<span lang="ZH-TW">$BF|(B $B2<8a(B</span>5:37
<span lang="ZH-TW">$BUmF;!'(B</span><o:p></o:p></p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Hi Bekket,</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">> There is a field in block_literal called "isa" representing the type of this block. But this field seems to always set to external symbols starts with "_NS", which is Cocoa's symbols.
Is this necessary? <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">This is not required for OpenCL, but we reuse ObjC implementation as much as possible at the moment. Of course, it doesn$B!G(Bt mean it can$B!G(Bt be change for OpenCL.</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">> But it seems that this approach doesn't work well with CL language since it's hard to handling the address space of block_literal instances.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Could you elaborate here what your problem is exactly? Would adding address spaces in generated IR help?</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">The current approach would transform block into block_literal struct, one of the fields, block_description, contains captured variables.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Another filed in block_literal is the function pointer to the
<span lang="ZH-TW">$B!H(B</span>real<span lang="ZH-TW">$B!I(B</span> invoke function. The first parameter would always be the pointer to the block_literal instance, where the invoke function can use it to retrieve captured variables. <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Here<span lang="ZH-TW">$B!G(B</span>s the problem: clang use alloca IR instruction to allocate space for block_literal instance. By default, it would be put on stack. <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">So if we move this situation to OpenCL kernel, alloca instruction would put block_literal instance into private address space by default. Then the block instance we pass to enqueue_kernel would need the pointer of block_literal, which is
passed as the first argument of the invoke function, to access captured variables but result in failures since the pointer is in private address space. <o:p></o:p></p>
</div>
<p class="MsoNormal"><br>
<br>
<o:p></o:p></p>
<div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">> Last but not the least, Is there anyone working on implementing dynamic parallelism? There seems to be no discussions about that on this mailing list.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">We are working on this now. There were two commits to blocks diagnostics past weeks and I am planning to setup the review on enqueue_kernel builtin upcoming
weeks.</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">One of our ideas is to <span lang="ZH-TW">$B!H(B</span>flatten<span lang="ZH-TW">$B!I(B</span> all of the captured variables. That is, by the time a block variable is defined, copy all of the captured variables<span lang="ZH-TW">$B!G(B</span>s value into
the invoke function. Since we can determine those variables<span lang="ZH-TW">$B!G(B</span> value at that moment (<span lang="ZH-TW">$B!H(B</span><span style="font-size:8.5pt;font-family:"Times New Roman","serif"">captured by the Block as const copies</span><span lang="ZH-TW">$B!I(B</span>
as the spec say).<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">OpenCL-C<span lang="ZH-TW">$B!G(B</span>s block is slightly different from the normal one. First, every block variable is const, so each of them need to be defined upon declaration. Second, the
<span lang="ZH-TW">$B!H(B</span>capture<span lang="ZH-TW">$B!I(B</span>(binding) actions are performed at the time block variables are defined. This would cause a behavior difference:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Here is the pseudo code:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">int x = 1;<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">Block_t myBlock = ^(void)(void){<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><span class="apple-tab-span"> </span>print x+1;<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">};<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">x = 2;<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">myBlock();<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">In the normal circumstance, it would print 3. But in OpenCL-C, since we bind x as constant upon definition, it would print 2. That<span lang="ZH-TW">$B!G(B</span>s why we think our
<span lang="ZH-TW">$B!H(B</span>flatten<span lang="ZH-TW">$B!I(B</span> approach could work: Copy captured variables as constants just one time would make a lot easier. <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">We<span lang="ZH-TW">$B!G(B</span>d just come out this idea few days ago, so we haven<span lang="ZH-TW">$B!G(B</span>t produce any useful code. We<span lang="ZH-TW">$B!G(B</span>re also considering using builtins to implement this idea.<o:p></o:p></p>
</div>
<p class="MsoNormal"><br>
<br>
<o:p></o:p></p>
<div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">I am not aware of any work on blocks codegen to IR though. Is there anything in particular you need?</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Also if you have any code you think might be useful to open source Clang that adds new functionality or improves existing work on this topic and you are happy
to share, do let me know.</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Thank you very much. <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">We<span lang="ZH-TW">$B!G(B</span>re also interesting on your approach. Perhaps we can work on this topic together?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Cheers,<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">McClane<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D"> </span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Cheers,</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri","sans-serif";color:#1F497D">Anastasia</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><b><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">From:</span></b><span class="apple-converted-space"><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""> </span></span><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">cfe-dev
[</span><a href="mailto:cfe-dev-bounces@lists.llvm.org"><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">mailto:cfe-dev-bounces@lists.llvm.org</span></a><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">]<span class="apple-converted-space"> </span><b>On
Behalf Of<span class="apple-converted-space"> </span></b>Bekket McClane via cfe-dev<br>
<b>Sent:</b><span class="apple-converted-space"> </span>17 March 2016 02:52<br>
<b>To:</b><span class="apple-converted-space"> </span></span><a href="mailto:cfe-dev@lists.llvm.org"><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif"">cfe-dev@lists.llvm.org</span></a><span lang="EN-US" style="font-size:10.0pt;font-family:"Tahoma","sans-serif""><br>
<b>Subject:</b><span class="apple-converted-space"> </span>[cfe-dev] About OpenCL 2.x Dynamic Parallelism</span><span style="font-family:"Times New Roman","serif""><o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">Hi,<o:p></o:p></span></p>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">Our lab is working on compiling opencl 2.x into NVPTX<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">But we have encountered some problems in dynamic parallelism. Blocks, particularly.<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">In short, clang would try to use Objective-C's approach to compile blocks in CL code.<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">It would generate block_literal struct to hold block information and block_descriptor to hold captured variables.<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">My first question is:<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">There is a field in block_literal called "isa" representing the type of this block. But this field seems to always set to external symbols starts with "_NS", which is Cocoa's symbols.
Is this necessary? <o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">Second, clang put captured variables in block_descriptor and passed the entire block_literal, which also contains a field holding pointer to a block_descriptor instance, as a implicit
"0 th" arguments for the real invoked function. But it seems that this approach doesn't wok well with CL language since it's hard to handling the address space of block_literal instances.<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">We have an idea passing all of the captured variables as function arguments, by value, for the invoked function. What do you folks think?<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">Last but not the least, Is there anyone working on implementing dynamic parallelism? There seems to be no discussions about that on this mailing list.<o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif""> <o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal"><span style="font-family:"Times New Roman","serif"">Best Regards,<o:p></o:p></span></p>
</div>
</div>
</div>
</blockquote>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</div>
</body>
</html>