<div dir="ltr"><div dir="ltr">On Thu, Apr 15, 2021, 6:03 PM via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>> wrote:<br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div lang="EN-US">
<div>
<p class="MsoNormal">Reid, I’m not clear why anyone would want to “power down” the alignment-aware optimizations? How does that benefit anyone? For example…<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Let’s postulate a target that has only non-trapping load/store instructions; maybe they go faster on aligned addresses, but don’t trap on unaligned addresses. It has been a few decades but I think VAX worked this way.<u></u><u></u></p>
<p class="MsoNormal">Would you insist we should power-down the alignment-aware optimizations for this target? Just because the hardware couldn’t require aligned data? I hope not.<u></u><u></u></p>
<br>
<p class="MsoNormal">The conclusion must be, then, that there is no relationship between the existence of trapping/non-trapping instruction behavior for a given target, and how the frontend and middle-end should behave.<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Therefore, we can’t insist on the front-end slapping “align 1” on everything just because the target doesn’t trap a non-aligned load.<u></u><u></u></p>
<p class="MsoNormal"><u></u></p></div></div></blockquote><div><br></div><div>Certainly, it's entirely valid for a target to not trap on an unaligned load. We have many such targets. A target trapping on misaligned loads isn't a required feature. (If users want to reliably diagnose misalignment bugs, -fsanitize=alignment is the way to do so.)</div><div></div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div lang="EN-US"><div><p class="MsoNormal"><u></u></p>
<p class="MsoNormal">Therefore, the choice of trapping/non-trapping instruction behavior in the X86 target specifically, has no necessary relationship to how alignment is thought of in the front-end/middle-end.<u></u><u></u></p>
<p class="MsoNormal"><u></u></p></div></div></blockquote><div><br></div><div>If the proposal here had been: "We should switch X86 from using movaps (alignment-checking) to movups (non-alignment-checking), because movups has a smaller encoding size (or is faster to execute on new microarchitectures, or ...), there'd be no problem. </div><div><br></div><div>But, that is <i>not</i> what's being proposed here. This proposal is to switch to movups as a workaround for software that has undefined behavior due to misaligned objects. That is misguided, because the proposed change does not fix such code! That the movaps instruction traps in such programs is like a proverbial "canary in a coal mine". It's a result of your program containing alignment-related UB. Removing the canary prevents you from having a dead canary, but it doesn't prevent the mine from exploding.</div><div><br></div><div><div>I have the feeling folks aren't understanding what exactly I'm talking about w.r.t. alignment-related breakage. There's at least three things LLVM can do with alignment information today.</div><div>1. Most obviously, it allows generation of hardware load instructions that require a certain alignment (MOVAPS on X86, LDM on ARM, etc.).</div><div>2. It enables known-bits analysis on pointers: "ptr & 0x3" is optimized to 0 if ptr is known to have alignment >= 4. Example: `int foo(int& x) { return ((uintptr_t)&x) & 0x3; }` </div><div>3. It can assist with alias analysis: if both addr1 and addr2 have align 8, then a 4-byte load from (addr1 + 0) cannot possibly alias a 4-byte load from (addr2 + 4). This is true even without TBAA, and even if know nothing else about the relationship between addr1 and addr2. (I don't have an example of this -- it looks like llvm may not be doing as good a job here as it could, but I definitely recall reading code which purported to implement this.)</div><div><br></div></div><div>The initial proposal only addresses the first issue, leaving users who depend on this are in an extremely precarious position -- liable to be broken by any future optimization improvement.</div><div><br></div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div lang="EN-US"><div>
<div style="border-top:none;border-right:none;border-bottom:none;border-left:1.5pt solid blue;padding:0in 0in 0in 4pt">
<div>
<div style="border-right:none;border-bottom:none;border-left:none;border-top:1pt solid rgb(225,225,225);padding:3pt 0in 0in">
<p class="MsoNormal"><b>From:</b> Craig Topper <<a href="mailto:craig.topper@gmail.com" rel="noreferrer" target="_blank">craig.topper@gmail.com</a>> <br>
<b>Sent:</b> Thursday, April 15, 2021 4:51 PM<br>
<b>To:</b> Reid Kleckner <<a href="mailto:rnk@google.com" rel="noreferrer" target="_blank">rnk@google.com</a>><br>
<b>Cc:</b> Robinson, Paul <<a href="mailto:paul.robinson@sony.com" rel="noreferrer" target="_blank">paul.robinson@sony.com</a>>; Luo, Yuanke <<a href="mailto:yuanke.luo@intel.com" rel="noreferrer" target="_blank">yuanke.luo@intel.com</a>>; Liu, Chen3 <<a href="mailto:chen3.liu@intel.com" rel="noreferrer" target="_blank">chen3.liu@intel.com</a>>; Maslov, Sergey V <<a href="mailto:sergey.v.maslov@intel.com" rel="noreferrer" target="_blank">sergey.v.maslov@intel.com</a>>; llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a>><br>
<b>Subject:</b> Re: [llvm-dev] [RFC] [X86] Emit unaligned vector moves on avx machine with option control.<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<p class="MsoNormal">What if we didn't use aligned instructions by default like what PS4 did. And then had a command line option that would "enable alignment exceptions" if someone wants them. Maybe that option should also disable memory folding since memory
folding never checks alignment with AVX? Do other targets that have vectors have alignment exceptions like this? We're not obligated to emit code that detects alignment errors. And we already don't if the load gets folded. It seems the problem with the current
proposal is that once you have the exception, setting a flag to make it go away is the wrong response.<u></u><u></u></p>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<div>
<div>
<p class="MsoNormal">~Craig<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
</div>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<p class="MsoNormal">On Thu, Apr 15, 2021 at 1:10 PM Reid Kleckner via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a>> wrote:<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<p class="MsoNormal">Right, I get that this doesn't match what you are doing for PS4, and it doesn't match what Chen3 Liu proposed. To James's point, the -fmax-type-align flag is more principled because it powers down all the other LLVM optimizations that assume
aligned pointers have zeros in the low bits.<u></u><u></u></p>
<div>
<p class="MsoNormal"><u></u> <u></u></p>
</div>
<div>
<p class="MsoNormal">As for how to handle explicit alignment attributes that don't come from type information, maybe we could revisit that behavior, or conditionalize it with a flag. I just mean to say that there is prior art for this direction. We should continue
in the direction of a complete solution from the frontend, rather than adding a workaround in the backend.<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal"><u></u> <u></u></p>
<div>
<div>
<p class="MsoNormal">On Thu, Apr 15, 2021 at 11:54 AM <<a href="mailto:paul.robinson@sony.com" rel="noreferrer" target="_blank">paul.robinson@sony.com</a>> wrote:<u></u><u></u></p>
</div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin-left:4.8pt;margin-right:0in">
<div>
<div>
<p class="MsoNormal">| This sounds like the -fmax-type-align flag:<u></u><u></u></p>
<p class="MsoNormal"> <u></u><u></u></p>
<p class="MsoNormal">Well, no, at least not for the PS4 case. In our case, the type had an alignment attribute but the caller didn’t make sure the allocated memory was aligned properly. The -fmax-type-align
flag explicitly doesn’t do anything in that case, if I’m reading it correctly. (Yes, it’s a bug. Yes, sanitizers or other testing could have found it. No, there is no opportunity to do any of the things that would have fixed it correctly.)<u></u><u></u></p>
<p class="MsoNormal"> <u></u><u></u></p>
<p class="MsoNormal">Really what we did was effectively this: Pretend X86 doesn’t have a VMOVAPS opcode. That’s all. Nothing about memory/operand alignment attributes was modified, IR is unchanged.
Pretend that one machine opcode is missing. Can’t possibly affect anything about IR optimizations, *<b>maybe</b>* something post-ISel would be different but even that is hard to imagine. (As best I can remember, the only test updates we had to make were
to change things like “vmovaps” to “vmov{{u|a}}ps” and done.) It’s like we did <span style="font-family:"Courier New"">
s/movaps/movups/g</span> on the assembly output.<u></u><u></u></p>
<p class="MsoNormal"> <u></u><u></u></p>
<p class="MsoNormal">I still can’t say I think it should be appropriate to do upstream—no real info yet on Intel’s problem case--but I hope this explains why the bigger hammer (i.e., get Clang involved)
doesn’t seem necessary or appropriate.<u></u><u></u></p>
<p class="MsoNormal">--paulr<u></u><u></u></p>
<p class="MsoNormal"> <u></u><u></u></p>
<div style="border-top:none;border-right:none;border-bottom:none;border-left:1.5pt solid blue;padding:0in 0in 0in 4pt">
<div>
<div style="border-right:none;border-bottom:none;border-left:none;border-top:1pt solid rgb(225,225,225);padding:3pt 0in 0in">
<p class="MsoNormal"><b>From:</b> llvm-dev <<a href="mailto:llvm-dev-bounces@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev-bounces@lists.llvm.org</a>>
<b>On Behalf Of </b>Reid Kleckner via llvm-dev<br>
<b>Sent:</b> Thursday, April 15, 2021 12:59 PM<br>
<b>To:</b> James Y Knight <<a href="mailto:jyknight@google.com" rel="noreferrer" target="_blank">jyknight@google.com</a>><br>
<b>Cc:</b> <a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a>; Liu, Chen3 <<a href="mailto:chen3.liu@intel.com" rel="noreferrer" target="_blank">chen3.liu@intel.com</a>>; Luo, Yuanke <<a href="mailto:yuanke.luo@intel.com" rel="noreferrer" target="_blank">yuanke.luo@intel.com</a>>;
Maslov, Sergey V <<a href="mailto:sergey.v.maslov@intel.com" rel="noreferrer" target="_blank">sergey.v.maslov@intel.com</a>><br>
<b>Subject:</b> Re: [llvm-dev] [RFC] [X86] Emit unaligned vector moves on avx machine with option control.<u></u><u></u></p>
</div>
</div>
<p class="MsoNormal"> <u></u><u></u></p>
<div>
<div>
<p class="MsoNormal">On Wed, Apr 14, 2021 at 11:58 AM James Y Knight via llvm-dev <<a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a>> wrote:<u></u><u></u></p>
</div>
<div>
<blockquote style="border-top:none;border-right:none;border-bottom:none;border-left:1pt solid rgb(204,204,204);padding:0in 0in 0in 6pt;margin:5pt 0in 5pt 4.8pt">
<div>
<div>
<p class="MsoNormal">What I suspect you
<i>actually</i> want here is an option to tell Clang not to infer load/store alignments based on object types or alignment attributes -- instead treating everything as being potentially aligned to 1 unless the allocation is seen (e.g. global/local variables).
Clang would still need to use the usual alignment computation for variable definitions and structure layout, but not memory operations. If clang emits "load ... align 1" instructions in LLVM IR, the right thing would then happen in the X86 backend automatically.<u></u><u></u></p>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal"> <u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">This sounds like the -fmax-type-align flag:<u></u><u></u></p>
</div>
<div>
<p class="MsoNormal"><a href="https://urldefense.com/v3/__https:/clang.llvm.org/docs/UsersManual.html*controlling-code-generation__;Iw!!JmoZiZGBv3RvKRSx!uoBVF33nyuM5lbseJ-XKanIeYhdhHW9yOoxyF7zJ56FjUs8jsfdUcuw4AQ96FRBrmA$" rel="noreferrer" target="_blank">https://clang.llvm.org/docs/UsersManual.html#controlling-code-generation</a><u></u><u></u></p>
</div>
<div>
<p class="MsoNormal">Explicit alignment attributes are still honored, so some aligned vector instructions may be generated. However, the documentation describes essentially this exact use case.<u></u><u></u></p>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
<p class="MsoNormal">_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a href="https://urldefense.com/v3/__https:/lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev__;!!JmoZiZGBv3RvKRSx!tTmi5Ot5ypUoBSp2e6p8a3o7U86YV49CFHt2_pW2GwCyapgR-cMMoUAeUQxP8A7xBQ$" rel="noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><u></u><u></u></p>
</blockquote>
</div>
</div>
</div>
</div>
_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org" rel="noreferrer" target="_blank">llvm-dev@lists.llvm.org</a><br>
<a href="https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer noreferrer" target="_blank">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
</blockquote></div></div>
</div>