<html><head><meta http-equiv="Content-Type" content="text/html charset=big5"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div class=""><div><blockquote type="cite" class=""><div class="">Anastasia Stulova <<a href="mailto:Anastasia.Stulova@arm.com" class="">Anastasia.Stulova@arm.com</a>> 於 2016年3月20日 下午5:37 寫道：</div><br class="Apple-interchange-newline"><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Hi Bekket,<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">> 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 class=""></o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">This is not required for OpenCL, but we reuse ObjC implementation as much as possible at the moment. Of course, it doesn’t mean it can’t be change for OpenCL.<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">> 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 class=""></o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Could you elaborate here what your problem is exactly? Would adding address spaces in generated IR help?</span></div></div></div></blockquote><div><br class=""></div><div>The current approach would transform block into block_literal struct, one of the fields, block_description, contains captured variables.</div><div>Another filed in block_literal is the function pointer to the “real” 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. </div><div>Here’s the problem: clang use alloca IR instruction to allocate space for block_literal instance. By default, it would be put on stack. </div><div>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. </div><br class=""><blockquote type="cite" class=""><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">> 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 class=""></o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">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></div></div></div></blockquote><div><br class=""></div><div>One of our ideas is to “flatten” all of the captured variables. That is, by the time a block variable is defined, copy all of the captured variables’s value into the invoke function. Since we can determine those variables’ value at that moment (“<span style="font-family: 'Times New Roman'; font-size: 11.5px;" class="">captured by the Block as const copies</span>” as the spec say).</div><div><br class=""></div><div>OpenCL-C’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 “capture”(binding) actions are performed at the time block variables are defined. This would cause a behavior difference:</div><div>Here is the pseudo code:</div><div><br class=""></div><div>int x = 1;</div><div>Block_t myBlock = ^(void)(void){</div><div><span class="Apple-tab-span" style="white-space:pre"> </span>print x+1;</div><div>};</div><div>x = 2;</div><div>myBlock();</div><div><br class=""></div><div>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’s why we think our “flatten” approach could work: Copy captured variables as constants just one time would make a lot easier. </div><div><br class=""></div><div>We’d just come out this idea few days ago, so we haven’t produce any useful code. We’re also considering using builtins to implement this idea.</div><br class=""><blockquote type="cite" class=""><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""> I am not aware of any work on blocks codegen to IR though. Is there anything in particular you need?<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">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></div></div></div></blockquote><div><br class=""></div><div>Thank you very much. </div><div>We’re also interesting on your approach. Perhaps we can work on this topic together?</div><div><br class=""></div><div>Cheers,</div><div><br class=""></div><div>McClane</div><div><br class=""></div><blockquote type="cite" class=""><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class=""><o:p class=""> </o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Cheers,<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif; color: rgb(31, 73, 125);" class="">Anastasia<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span lang="EN-US" style="font-size: 10pt; font-family: Tahoma, sans-serif;" class="">From:</span></b><span lang="EN-US" style="font-size: 10pt; font-family: Tahoma, sans-serif;" class=""><span class="Apple-converted-space"> </span>cfe-dev [<a href="mailto:cfe-dev-bounces@lists.llvm.org" class="">mailto:cfe-dev-bounces@lists.llvm.org</a>]<span class="Apple-converted-space"> </span><b class="">On Behalf Of<span class="Apple-converted-space"> </span></b>Bekket McClane via cfe-dev<br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>17 March 2016 02:52<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span><a href="mailto:cfe-dev@lists.llvm.org" class="">cfe-dev@lists.llvm.org</a><br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>[cfe-dev] About OpenCL 2.x Dynamic Parallelism<o:p class=""></o:p></span></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Hi,<o:p class=""></o:p></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Our lab is working on compiling opencl 2.x into NVPTX<o:p class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">But we have encountered some problems in dynamic parallelism. Blocks, particularly.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">In short, clang would try to use Objective-C's approach to compile blocks in CL code.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">It would generate block_literal struct to hold block information and block_descriptor to hold captured variables.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">My first question is:<o:p class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">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 class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">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 class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">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 class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">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 class=""></o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0cm 0cm 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Best Regards,</div></div></div></div></blockquote></div><br class=""></div></body></html>