<div dir="ltr">Hi Amara,<div><br></div><div>Just to follow up, I think we merged the command-line flags etc., but are still waiting for release notes/docs?</div><div><br></div><div>Thanks,</div><div>Hans</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jan 18, 2018 at 4:10 AM, Amara Emerson <span dir="ltr"><<a href="mailto:aemerson@apple.com" target="_blank">aemerson@apple.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">> On 17 Jan 2018, at 21:45, Quentin Colombet <<a href="mailto:qcolombet@apple.com">qcolombet@apple.com</a>> wrote:<br>
><br>
> Hi Hans,<br>
><br>
> Thanks for the reminder.<br>
><br>
> I think we should!<br>
> @Amara, did we get around a flag in clang to disable/enable global-isel, so that people can file more informed bugs or try it with optimizations?<br>
><br>
> Cheers,<br>
> -Quentin<br>
><br>
>> On Jan 17, 2018, at 7:41 AM, Hans Wennborg via llvm-commits <<a href="mailto:llvm-commits@lists.llvm.org">llvm-commits@lists.llvm.org</a>> wrote:<br>
>><br>
>> Do you want to write something about this for the 6.0.0 release notes?<br>
>><br>
>> On Tue, Jan 2, 2018 at 5:30 PM, Amara Emerson via llvm-commits<br>
>> <<a href="mailto:llvm-commits@lists.llvm.org">llvm-commits@lists.llvm.org</a>> wrote:<br>
>>> Author: aemerson<br>
>>> Date: Tue Jan 2 08:30:47 2018<br>
>>> New Revision: 321655<br>
>>><br>
>>> URL: <a href="http://llvm.org/viewvc/llvm-project?rev=321655&view=rev" rel="noreferrer" target="_blank">http://llvm.org/viewvc/llvm-<wbr>project?rev=321655&view=rev</a><br>
>>> Log:<br>
>>> [AArch64][GlobalISel] Enable GlobalISel at -O0 by default<br>
>>><br>
>>> Tests updated to explicitly use fast-isel at -O0 instead of implicitly.<br>
>>><br>
>>> This change also allows an explicit -fast-isel option to override an<br>
>>> implicitly enabled global-isel. Otherwise -fast-isel would have no effect at -O0.<br>
>>><br>
>>> Differential Revision: <a href="https://reviews.llvm.org/D41362" rel="noreferrer" target="_blank">https://reviews.llvm.org/<wbr>D41362</a><br>
>>><br>
>>> Modified:<br>
>>> llvm/trunk/include/llvm/<wbr>CodeGen/TargetPassConfig.h<br>
>>> llvm/trunk/lib/CodeGen/<wbr>TargetPassConfig.cpp<br>
>>> llvm/trunk/lib/Target/AArch64/<wbr>AArch64TargetMachine.cpp<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/GlobalISel/arm64-<wbr>fallback.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/GlobalISel/gisel-<wbr>commandline-option.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/aarch64_f16_be.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/and-mask-removal.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-EXT-undef-mask.<wbr>ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-abi.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-abi_align.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-elf-constpool.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-elf-globals.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>alloca.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-br.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-call.<wbr>ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>conversion-fallback.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>conversion.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-fcmp.<wbr>ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-gv.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-icmp.<wbr>ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>intrinsic.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>materialize.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-<wbr>noconvert.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-rem.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel-ret.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-fast-isel.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-simd-scalar-to-<wbr>vector.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-tls-dynamic-<wbr>together.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/arm64-vcvt_f.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/br-cond-not-merge.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/cmpxchg-O0.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/cxx-tlscc.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/fast-isel-atomic.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/fast-isel-sp-adjust.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/i128-fast-isel-<wbr>fallback.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/preferred-alignment.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/swift-return.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/swifterror.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/swiftself.ll<br>
>>> llvm/trunk/test/CodeGen/<wbr>AArch64/tailcall-fastisel.ll<br>
>>> llvm/trunk/test/DebugInfo/<wbr>AArch64/asan-stack-vars.ll<br>
>>> llvm/trunk/test/DebugInfo/<wbr>AArch64/frameindices.ll<br>
>>> llvm/trunk/test/DebugInfo/<wbr>AArch64/line-header.ll<br>
>>> llvm/trunk/test/DebugInfo/<wbr>AArch64/prologue_end.ll<br>
>> ______________________________<wbr>_________________<br>
>> llvm-commits mailing list<br>
>> <a href="mailto:llvm-commits@lists.llvm.org">llvm-commits@lists.llvm.org</a><br>
>> <a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits" rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/<wbr>mailman/listinfo/llvm-commits</a><br>
><br>
<br>
</div></div>@Hans We can definitely do that. As Quentin said we have a flag which we want to introduce to clang to give a more user friendly way to control the behaviour. I’ll post a patch for that soon.<br>
<br>
Cheers,<br>
Amara</blockquote></div><br></div>