<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 Jan 2, 2017, at 2:12 PM, James Price <<a href="mailto:J.Price@bristol.ac.uk" class="">J.Price@bristol.ac.uk</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><meta http-equiv="Content-Type" content="text/html charset=utf-8" class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Hi Mehdi,<div class=""><br class=""></div><div class="">Thanks for the reply - I’ve finally got round to trying to fix this based on your suggestion.</div></div></div></blockquote><div><br class=""></div><div>Glad to read this!</div><br class=""><blockquote type="cite" class=""><div class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><div class=""> I’ve got something that mostly works, but I just wanted to double-check something about the regression tests before I post a patch.</div><div class=""><br class=""></div><div class=""><blockquote type="cite" class=""><span class="" style="display: inline !important;">The memcpy is supposed to be equivalent to a sequence of load and store. Here we are just failing to keep the property that the load is performed through addrspace(2).</span></blockquote></div><div class=""><br class=""></div><div class="">Based on this comment, I am suspicious of the validity of a couple of existing instcombine regression tests in `memcpy-from-global.ll`. Specifically, there are two tests that look like this:</div><div class=""><br class=""></div><div class=""> define void @test3_addrspacecast() {<br class=""> %A = alloca %T<br class=""> %a = bitcast %T* %A to i8*<br class=""> call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%T* @G to i8 addrspace(1)*), i64 124, i32 4, i1 false)<br class=""> call void @bar(i8* %a) readonly<br class=""> ; CHECK-LABEL: @test3_addrspacecast(<br class=""> ; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T, %T* @G, i64 0, i32 0))<br class=""> ret void<br class=""> }</div><div class=""><br class=""></div><div class="">Here, there is a global variable in `addrspace(0)`, which is passed as a source operand to `llvm.memcpy` via an `addrspacecast` to `addrspace(1)`. The memcpy destination operand (an `alloca` in `addrspace(0)`) is then passed to a function. The test asserts that this memcpy should be removed and the global should just be passed directly to the function, but doesn’t this lose the property that the load should be performed through `addrspace(1)`, as per your comment above?</div></div></div></blockquote><div><br class=""></div><div>Let’s desugarize memcpy to a sequence of load-store as my previous comment implied (simplified for a single i8, otherwise we need a loop or a load of T, but the reasoning seems the same to me):</div><div><br class=""></div><div><br class=""></div><div> define void @test3_addrspacecast() {<br class=""> %A = alloca %T<br class=""> %a = bitcast %T* %A to i8*</div><div> %tmp = load i8, i8 addrspace(1)* addrspacecast (%T* @G to i8 addrspace(1)*) ; Load from Addrspace 1 </div><div> store i8 %tmp, i8* %a ; Store to the alloca in Addrspace 0, now %A and %G have the same content.<br class=""> call void @bar(i8* %a) readonly<br class=""> ret void<br class=""> }</div><div><br class=""></div><div><br class=""></div><div>If I understand correctly, there is no load involved here after optimization. Before the optimization, the function bar() takes a pointer to the memory location @A, in address space 0. After the optimization, the function still takes a pointer to address space 0. So this part is valid.</div><div><br class=""></div><div>Since the LangRef for addrspacecast says " if the address space conversion is legal then both result and operand refer to the same memory location”, the optimizer is free to assume that a load to @G through either addrspace 0 or 1 yields the same value, and thus can figure that the alloca value is the same as @G. </div><div>Since bar() is read-only, exchanging a pointer for another one in the same address space, when the pointers point to the same value, seems valid to me.</div><div><br class=""></div><div>— </div><div>Mehdi</div><div><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><div class=""><br class=""></div><div class="">I can (and have) fixed this as part of the patch I’m working on, but since this implies that a couple of existing regression tests would be incorrect I just wanted to double-check that I’m not misinterpreting something.</div><div class=""><br class=""></div><div class="">CCing Matt Arsenault who added those specific tests originally in r207054.</div><div class=""><br class=""></div><div class="">James</div><div class=""><br class=""></div><div class=""><blockquote type="cite" class=""><div class="">On 9 Nov 2016, at 21:11, Mehdi Amini <<a href="mailto:mehdi.amini@apple.com" class="">mehdi.amini@apple.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><span style="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; float: none; display: inline !important;" class="">Hi,</span><br style="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;" class=""><br style="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;" class=""><blockquote type="cite" style="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-size-adjust: auto; -webkit-text-stroke-width: 0px;" class="">On Nov 9, 2016, at 10:13 AM, James Price via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>> wrote:<br class=""><br class="">Hi,<br class=""><br class="">I’ve recently encountered an issue where the `instcombine` pass replaces an `llvm.memcpy` between two distinct address spaces with an `addrspacecast` instruction.<br class=""><br class="">As an example, see the trivial OpenCL kernel attached. I’m compiling like this:<br class=""><br class=""> clang -cc1 -triple spir64-unknown-unknown -x cl -O0 -emit-llvm array_init.cl -o before.ll<br class=""><br class="">This yields an `llvm.memcpy` to copy the array initialiser data from the global variable (in `addrspace(2)`) to the `alloca` result (in `addrspace(0)`).<br class=""><br class="">I then apply the `instcombine` pass via:<br class=""><br class=""> opt -S -instcombine before.ll -o after.ll<br class=""><br class="">This results in the memcpy being nuked, and the `addrspace(2)` data is now accessed directly via an `addrspacecast` to `addrspace(0)`.<br class=""><br class=""><br class="">It seems to me that this sort of optimisation is only valid if it is guaranteed that the two address spaces alias for the given target triple (which for SPIR, they do not).<br class=""></blockquote><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">I’m not sure “alias” is the right consideration here.</span><br style="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;" class=""><br style="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;" class=""><blockquote type="cite" style="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-size-adjust: auto; -webkit-text-stroke-width: 0px;" class=""><br class=""><br class="">This particular optimisation is coming from lines ~290-300 of InstCombineLoadStoreAlloca.cpp, although I suspect this isn’t the only case where this might happen.<br class=""><br class="">Adding a check to only perform this replacement if the two address spaces are equal fixes the issue for me, but this is probably too conservative since many targets with flat address spaces will probably benefit from this optimisation.<br class="">It feels like passes should query the target about whether two address spaces alias before introducing an `addrspacecast`, but I’m not familiar enough with LLVM internals to know if this is information that is easy to make available (if it isn’t already).<br class=""><br class="">Is there something we can do here to avoid this sort of optimisation causing problems for targets with segmented address spaces?<br class=""></blockquote><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">This is a bug, we can’t assume any memory layout I believe.</span><br style="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;" class=""><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">The memcpy is supposed to be equivalent to a sequence of load and store. Here we are just failing to keep the property that the load is performed through addrspace(2). The fix is not “trivial” though, the transformation can only be performed if the address of the alloca does not escape, and some rewriting of the uses is needed to propagate the address space through GEPs for instance.</span><br style="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;" class=""><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">—<span class="Apple-converted-space"> </span></span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">Mehdi</span></div></blockquote></div><br class=""></div></div></blockquote></div><br class=""></body></html>