<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Aug 17, 2016, at 2:08 PM, Zhuravlyov, Konstantin <<a href="mailto:Konstantin.Zhuravlyov@amd.com" class="">Konstantin.Zhuravlyov@amd.com</a>> wrote:</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-caps: 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: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>Why not going with a metadata attachment directly and kill the "singlethread" keyword? Something like:<o:p class=""></o:p></span></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>Something like:<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">> cmpxchg i32* %addr, i32 42, i32 0 monotonic monotonic, 3, !memory.scope{!42}<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">> cmpxchg i32* %addr, i32 42, i32 0 monotonic monotonic, 3, !memory.scope{!43}<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>...<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>!42 = !{"singlethread"}<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>!43 = !{"L2"}<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">>It is not clear to me if there is any correctness issues to dropping metadata?<o:p class=""></o:p></span></div></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class="">Yes, we cannot use the metadata approach since this metadata can be dropped during the processing of one module but not dropped in the processing of a second module, potentially resulting in inconsistent scopes for synchronizing operations leading to data races and subsequently leading to correctness issues.<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div></div></div></blockquote><div><br class=""></div><div>Right, I saw Sameer's explanation for that earlier, and we shouldn’t move forward (without Philip’s opinion on the topic as he expressed concerns).</div><div><br class=""></div><div>But you stripped out the second part of my email where I wrote "It seems you’re going back to integer, which I don’t really like for reasons mentioned earlier in this thread, and that I don’t feel you addressed here”. Why can’t `synchscope` take a string literal?</div><div><br class=""></div><div><br class=""></div><div>— </div><div>Mehdi</div><div><br class=""></div><div><br class=""></div><div><br class=""></div><div><br class=""></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-caps: 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: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""><br class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; background-color: white;" class=""><span style="font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></span></div><div class="MsoNormal" align="center" style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif; text-align: center;"><hr size="2" width="98%" align="center" class=""></div><div id="divRplyFwdMsg" class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class="">From:</span></b><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span class="Apple-converted-space"> </span>Zhuravlyov, Konstantin<br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Wednesday, August 17, 2016 4:29:30 PM<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>Sameer Sahasrabuddhe; Philip Reames<br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span>Mehdi Amini; Liu, Yaxun (Sam); Ke Bai; Mekhanoshin, Stanislav; Sumner, Brian; <a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>; Tye, Tony<br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>RE: [llvm-dev] Memory scope proposal</span><o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""> <o:p class=""></o:p></div></div></div><div class=""><div style="margin: 0in 0in 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,<o:p class=""></o:p></span></div><div style="margin: 0in 0in 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: 0in 0in 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 have updated the review here:<o:p class=""></o:p></span></div><div style="margin: 0in 0in 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=""><a href="https://reviews.llvm.org/D21723" style="color: purple; text-decoration: underline;" class="">https://reviews.llvm.org/D21723</a><o:p class=""></o:p></span></div><div style="margin: 0in 0in 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: 0in 0in 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="">As Sameer pointed out, the motivation is:<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">In OpenCL 2.x, two atomic operations on the same atomic object need to have the same scope to prevent a data race. This derives from the definition of "inclusive scope" in OpenCL 2.x. Encoding OpenCL 2.x scope as metadata in LLVM IR would be a problem because there cannot be a "safe default value" to be used when the metadata is dropped. If the "largest" scope is used as the default, then the optimizer must guarantee that the metadata is dropped from every atomic operation in the whole program, or not dropped at all.<o:p class=""></o:p></div><div style="margin: 0in 0in 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: 0in 0in 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="">Thanks,<o:p class=""></o:p></span></div><div style="margin: 0in 0in 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="">Konstantin<o:p class=""></o:p></span></div><div style="margin: 0in 0in 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: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><b class=""><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class="">From:</span></b><span style="font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span class="Apple-converted-space"> </span>Sameer Sahasrabuddhe [<a href="mailto:sameer@sbuddhe.net" class="">mailto:sameer@sbuddhe.net</a>]<span class="Apple-converted-space"> </span><br class=""><b class="">Sent:</b><span class="Apple-converted-space"> </span>Sunday, July 10, 2016 4:06 AM<br class=""><b class="">To:</b><span class="Apple-converted-space"> </span>Philip Reames <<a href="mailto:listmail@philipreames.com" class="">listmail@philipreames.com</a>><br class=""><b class="">Cc:</b><span class="Apple-converted-space"> </span>Mehdi Amini <<a href="mailto:mehdi.amini@apple.com" class="">mehdi.amini@apple.com</a>>; Liu, Yaxun (Sam) <<a href="mailto:Yaxun.Liu@amd.com" class="">Yaxun.Liu@amd.com</a>>; Ke Bai <<a href="mailto:kebai613@gmail.com" class="">kebai613@gmail.com</a>>; Mekhanoshin, Stanislav <<a href="mailto:Stanislav.Mekhanoshin@amd.com" class="">Stanislav.Mekhanoshin@amd.com</a>>; Sumner, Brian <<a href="mailto:Brian.Sumner@amd.com" class="">Brian.Sumner@amd.com</a>>; <a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>; Zhuravlyov, Konstantin <<a href="mailto:Konstantin.Zhuravlyov@amd.com" class="">Konstantin.Zhuravlyov@amd.com</a>>; Tye, Tony <<a href="mailto:Tony.Tye@amd.com" class="">Tony.Tye@amd.com</a>><br class=""><b class="">Subject:</b><span class="Apple-converted-space"> </span>Re: [llvm-dev] Memory scope proposal<o:p class=""></o:p></span></div><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class=""><o:p class=""> </o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">On Mon, Jul 4, 2016 at 5:09 AM, Philip Reames via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank" style="color: purple; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a>> wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-color: rgb(204, 204, 204); border-left-width: 1pt; padding: 0in 0in 0in 6pt; margin: 5pt 0in 5pt 4.8pt;" class=""><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">I will comment - as one of the few people actually working on llvm's atomic implementation with any regularity - that I am opposed to extending the instructions without a strong motivating case. I don't care anywhere near as much about metadata based schemes, but extending the instruction semantics imposes a much larger burden on the rest of the community. That burden has to be well justified and supported. <o:p class=""></o:p></div></div></div></blockquote><div class=""><p class="MsoNormal" style="margin: 0in 0in 12pt; font-size: 12pt; font-family: 'Times New Roman', serif;"><br class="">In OpenCL 2.x, two atomic operations on the same atomic object need to have the same scope to prevent a data race. This derives from the definition of "inclusive scope" in OpenCL 2.x. Encoding OpenCL 2.x scope as metadata in LLVM IR would be a problem because there cannot be a "safe default value" to be used when the metadata is dropped. If the "largest" scope is used as the default, then the optimizer must guarantee that the metadata is dropped from every atomic operation in the whole program, or not dropped at all.<o:p class=""></o:p></p></div><div class=""><p class="MsoNormal" style="margin: 0in 0in 12pt; font-size: 12pt; font-family: 'Times New Roman', serif;">Hence the original attempt to extend LLVM atomic instructions with a broader scope field.<o:p class=""></o:p></p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 12pt; font-family: 'Times New Roman', serif;" class="">Sameer.</div></div></div></div></div></div></div></div></blockquote></div><br class=""></body></html>