<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">Is there a reason to have both bcmp and memcmp forms of this? The only different appears to be the nature of the result value, would that be better modeled as an immarg parameter?<div class=""><br class=""></div><div class="">-Chris<br class=""><div><br class=""><blockquote type="cite" class=""><div class="">On Nov 24, 2021, at 2:07 AM, Dmitry Makogon <<a href="mailto:dmakogon@azul.com" class="">dmakogon@azul.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><meta charset="UTF-8" class=""><div class="WordSection1" style="page: WordSection1; caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I propose they have signatures similar to non-atomic/atomic llvm.memcpy:<br class="">llvm_i32_ty llvm.memcmp (llvm_anyptr_ty %lhs, llvm_anyptr_ty %rhs, llvm_anyint_ty %length, llvm_i1_ty %is_volatile),<br class="">llvm_i32_ty llvm.bcmp (llvm_anyptr_ty %lhs, llvm_anyptr_ty %rhs, llvm_anyint_ty %length, llvm_i1_ty %is_volatile),<br class="">llvm_i32_ty llvm.memcmp.element.unordered.atomic (llvm_anyptr_ty %lhs, llvm_anyptr_ty %rhs, llvm_anyint_ty %length, llvm_i32_ty %element_size),<br class="">llvm_i32_ty llvm.bcmp.element.unordered.atomic (llvm_anyptr_ty %lhs, llvm_anyptr_ty %rhs, llvm_anyint_ty %length, llvm_i32_ty %element_size).<br class=""><br class=""><o:p class=""></o:p></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Having such signatures, it’s easy to transform the libc calls to the new intrinsics.<br class=""><br class="">Our main motivation for having these intrinsics is that we want to have an atomic memory comparison semantics, which we can’t express through a libc call.<span class="Apple-converted-space"> </span><br class="">And if we are to add an atomic memcmp intrinsic, it’d natural to transform libc calls to intrinsics and we’d have some common logic work for all these functions.<br class=""><br class="">Dmitry<br class=""><br class=""><o:p class=""></o:p></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="border-style: solid none none; border-top-width: 1pt; border-top-color: rgb(225, 225, 225); padding: 3pt 0in 0in;" class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><b class=""><span lang="RU" class="">От:</span></b><span lang="RU" class=""><span class="Apple-converted-space"> </span>Chris Lattner <<a href="mailto:clattner@nondot.org" style="color: blue; text-decoration: underline;" class="">clattner@nondot.org</a>><span class="Apple-converted-space"> </span><br class=""><b class="">Отправлено:</b><span class="Apple-converted-space"> </span>Tuesday, November 23, 2021 2:29 PM<br class=""><b class="">Кому:</b><span class="Apple-converted-space"> </span>Dmitry Makogon <<a href="mailto:dmakogon@azul.com" style="color: blue; text-decoration: underline;" class="">dmakogon@azul.com</a>><br class=""><b class="">Копия:</b><span class="Apple-converted-space"> </span><a href="mailto:llvm-dev@lists.llvm.org" style="color: blue; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a><br class=""><b class="">Тема:</b><span class="Apple-converted-space"> </span>Re: [llvm-dev] Proposal: Introduce memory comparison intrinsics<o:p class=""></o:p></span></div></div></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><br class=""><br class=""><o:p class=""></o:p></div><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">On Nov 22, 2021, at 6:22 AM, Dmitry Makogon via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" style="color: blue; text-decoration: underline;" class="">llvm-dev@lists.llvm.org</a>> wrote:<o:p class=""></o:p></div></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div class=""><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Hello everyone.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I would like to introduce new intrinsics for memory comparison:<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">memcmp - an equivalent of libc' memcmp,<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">bcmp - an equivalent of libc' bcmp,<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">memcmp.element.unordered.atomic - an element atomic version of memcmp,<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">bcmp.element.unordered.atomic - an element atomic version of bcmp.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Currently there exist some optimizations for memcmp/bcmp libc calls.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">We would like to have these optimizations for element atomic comparisons (atomicity permitted).<o:p class=""></o:p></div></div></div></blockquote><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Could you elaborate on the specific signatures these intrinsics would have?<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">llvm.memcpy and friends exist because we want to capture additional semantics beyond what the memcpy signature does - notably alignment information. What is all the additional information are you planning to capture for these?<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">-Chris<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><br class=""><br class=""><o:p class=""></o:p></div><blockquote style="margin-top: 5pt; margin-bottom: 5pt;" class=""><div class=""><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I suggest we rewrite the existing optimizations to work with on new intrinsics and transform<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">memcmp/bcmp libc calls to the corresponding intrinsics. This is similar to what we do with<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">memcpy library calls.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Having these optimizations work on intrinsics and not on recognized libc calls<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">will allow us to share some existing transforms between atomic and non-atomic variants.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I propose the following plan for introducing the new intrinsics:<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">1. Introduce non-atomic memcmp and bcmp intrinsics.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">2. Reimplement existing transforms for non-atomic memcmp intrinsic,<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">the same way as it's done for memcpy.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">3. Introduce atomic intrinsics and reuse the optimizations.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Please express your concerns about this.<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Dmitry<o:p class=""></o:p></div></div><div style="margin: 0in; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9pt; font-family: Helvetica, sans-serif;" class="">_______________________________________________<br class="">LLVM Developers mailing list<br class=""></span><a href="mailto:llvm-dev@lists.llvm.org" style="color: blue; text-decoration: underline;" class=""><span style="font-size: 9pt; font-family: Helvetica, sans-serif;" class="">llvm-dev@lists.llvm.org</span></a><span style="font-size: 9pt; font-family: Helvetica, sans-serif;" class=""><br class=""></span><a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" style="color: blue; text-decoration: underline;" class=""><span style="font-size: 9pt; font-family: Helvetica, sans-serif;" class="">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</span></a></div></div></blockquote></div></div></div></blockquote></div><br class=""></div></body></html>