<div dir="ltr">I'm sure.<div><br></div><div>That said, you've got some warnings in the code with a ToT clang now:<br><br><div>[1961/3415] Building CXX object lib/Target/Hexagon/MCTargetDesc/CMakeFiles/LLVMHexagonDesc.dir/HexagonMCDuplexInfo.cpp.o</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp: In function ‘llvm::MCInst llvm::HexagonMCInstrInfo::deriveSubInst(const llvm::MCInst&)’:</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:786:5: warning: this statement may fall through [-Wimplicit-fallthrough=]</div><div>     if (Value == 2) {</div><div>     ^~</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:792:3: note: here</div><div>   case Hexagon::A4_combineir:</div><div>   ^~~~</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:898:12: warning: this statement may fall through [-Wimplicit-fallthrough=]</div><div>     } else if (Value == 1) {</div><div>            ^~</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:904:3: note: here</div><div>   case Hexagon::S2_storerb_io:</div><div>   ^~~~</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:934:12: warning: this statement may fall through [-Wimplicit-fallthrough=]</div><div>     } else if (Inst.getOperand(0).getReg() == Hexagon::R29) {</div><div>            ^~</div><div>/usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCDuplexInfo.cpp:940:3: note: here</div><div>   case Hexagon::S2_storeri_io:</div><div>   ^~~~</div><div>[2042/3415] Building CXX object lib/Target/Hexagon/AsmParser/CMakeFiles/LLVMHexagonAsmParser.dir/HexagonAsmParser.cpp.o</div><div>In file included from /usr/local/google/home/echristo/sources/llvm/include/llvm/ADT/StringRef.h:13:0,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/include/llvm/ADT/StringMap.h:17,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/include/llvm/Support/Host.h:17,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/include/llvm/ADT/Hashing.h:49,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/include/llvm/ADT/ArrayRef.h:13,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/include/llvm/Support/CommandLine.h:23,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.h:17,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/Hexagon.h:45,</div><div>                 from /usr/local/google/home/echristo/sources/llvm/lib/Target/Hexagon/AsmParser/HexagonAsmParser.cpp:12:</div><div>/usr/local/google/home/echristo/sources/llvm/include/llvm/ADT/STLExtras.h:886:6: warning: mangled name for ‘decltype (llvm::adl_begin(Range)) llvm::remove_if(R&&, UnaryPredicate) [with R = std::__cxx11::basic_string<char>&; UnaryPredicate = int (*)(int) throw ()]’ will change in C++17 because the exception specification is part of a function type [-Wnoexcept-type]</div><div> auto remove_if(R &&Range, UnaryPredicate P) -> decltype(adl_begin(Range)) {</div><div>      ^~~~~~~~~</div></div><div><br></div><div>I looked at trying to fix, but I don't know how your processor works. :)</div><div><br></div><div>-eric</div></div><br><div class="gmail_quote"><div dir="ltr">On Tue, Mar 20, 2018 at 11:53 AM Krzysztof Parzyszek <<a href="mailto:kparzysz@codeaurora.org">kparzysz@codeaurora.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I've tried. Going from V62 to V65 was quite a large change, and locally<br>
we had lots of interdependent patches implementing it, intertwined with<br>
unrelated changes to the same components. In the end it saved me a lot<br>
of time to just lump it all together.<br>
<br>
-Krzysztof<br>
<br>
On 3/20/2018 1:44 PM, Eric Christopher wrote:<br>
> For the record you really should have broken this up into separate commits.<br>
><br>
>   69 files changed, 13364 insertions(+), 5905 deletions(-)<br>
><br>
> -eric<br>
><br>
<br>
--<br>
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,<br>
hosted by The Linux Foundation<br>
</blockquote></div>