<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="">I’ve submitted a patch that aims to fix this <a href="https://reviews.llvm.org/D70815" class="">https://reviews.llvm.org/D70815</a>, can I ask you to review it?<br class=""><div><br class=""><blockquote type="cite" class=""><div class="">On 20 Nov 2019, at 20:56, Peter Smith <<a href="mailto:peter.smith@linaro.org" class="">peter.smith@linaro.org</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><span style="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; float: none; display: inline !important;" class="">On Wed, 20 Nov 2019 at 17:04, Sergej Jaskiewicz <</span><a href="mailto:jaskiewiczs@icloud.com" 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="">jaskiewiczs@icloud.com</a><span style="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; float: none; display: inline !important;" class="">> wrote:</span><br style="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;" 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; text-decoration: none;" class=""><br class=""><br class=""><br class="">On 18 Nov 2019, at 22:11, Peter Smith <<a href="mailto:peter.smith@linaro.org" class="">peter.smith@linaro.org</a>> wrote:<br class=""><br class="">On Mon, 18 Nov 2019 at 17:06, Sergej Jaskiewicz <<a href="mailto:jaskiewiczs@icloud.com" class="">jaskiewiczs@icloud.com</a>> wrote:<br class=""><br class=""><br class=""><br class=""><br class="">On 18 Nov 2019, at 19:55, Peter Smith <<a href="mailto:peter.smith@linaro.org" class="">peter.smith@linaro.org</a>> wrote:<br class=""><br class="">On Mon, 18 Nov 2019 at 15:23, Sergej Jaskiewicz <<a href="mailto:jaskiewiczs@icloud.com" class="">jaskiewiczs@icloud.com</a>> wrote:<br class=""><br class=""><br class="">Hi Peter,<br class=""><br class="">Thanks for your response.<br class=""><br class="">On 18 Nov 2019, at 17:44, Peter Smith <<a href="mailto:peter.smith@linaro.org" class="">peter.smith@linaro.org</a>> wrote:<br class=""><br class="">On Mon, 18 Nov 2019 at 12:32, Sergej Jaskiewicz via llvm-dev<br class=""><<a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a>> wrote:<br class=""><br class=""><br class="">There’s this bug: <a href="https://bugs.llvm.org/show_bug.cgi?id=38468" class="">https://bugs.llvm.org/show_bug.cgi?id=38468</a>.<br class=""><br class="">I’ve managed to track it down to a configuration issue. The thing is that in order for libunwind to be usable on ARM Linux, it should be built with the -funwind-tables flag. This flag is conditionally set here: <a href="https://github.com/llvm/llvm-project/blob/master/libunwind/CMakeLists.txt#L294" class="">https://github.com/llvm/llvm-project/blob/master/libunwind/CMakeLists.txt#L294</a>, if the compiler “supports” it.<br class=""><br class=""><br class="">Are you sure that libunwind being built without -funwind-tables is the<br class="">cause of the bug?<br class=""><br class=""><br class="">Yep, pretty sure, I’ve found that the cause of the problem is the _Unwind_Backtrace function not executing the provided callback. It isn’t doing so because, since libunwind is compiled without the flag, the information about the stack frame is lost, so, when _Unwind_Backtrace looks for it, it can’t find it (since we’ve entered the _Unwind_Backtrace stack frame, which lives in libunwind, where no unwind info is present).<br class=""><br class="">I’ve looked at the generated assembly of libunwind and found the .cantunwind directives all over the place.<br class=""><br class="">I would only expect that to be a problem if an<br class="">exception were being propagated through a libunwind function, and that<br class="">shouldn't happen unless something has gone badly wrong.<br class=""><br class=""><br class="">Can you explain what you mean?<br class=""><br class="">I'm going from the example provided in the PR:<br class=""><br class="">int main(int argc, char* argv[])<br class="">{<br class=""> try<br class=""> {<br class=""> throw 1;<br class=""> }<br class=""> catch(...)<br class=""> {<br class=""> std::cout << "Exception is caught\n";<br class=""> }<br class="">}<br class=""><br class="">When the exception is thrown via a call to __cxa_throw, what I would<br class="">expect is for the unwinder to look up the PC value at the callsite in<br class="">the .ARM.exidx table. This should contain information for unwinding<br class="">the stack should the exception propagate outside of main. This<br class="">propagation of the exception out of main shouldn't go through a<br class="">function in libunwind as libunwind should not be in the call stack for<br class="">main. This is the case if the search in the table starts from the<br class="">callsite at main, and in this case it shouldn't matter whether<br class="">libunwind has unwind tables or not.<br class=""><br class="">One thing I've not done is go back to the source and check if the<br class="">search does not start in main, but instead starts in a function in<br class="">libunwind, which via a further unwind gets to main where the exception<br class="">is caught. If that were the case then yes we would need unwinding<br class="">information. We'd also need unwinding information in libunwind if it<br class="">called a function that could throw an exception.<br class=""><br class="">I tried the<br class="">example with the armv7l release of clang 8.0 which I happened to have<br class="">installed on an Armv8l machine and the program worked. I was able to<br class="">reproduce the problem with the PR with the default Ubuntu16.04 clang<br class="">(3.8) and libc++-dev package.<br class=""><br class="">I also note that when looking at the link line for the example in the<br class="">PR, clang was linking libgcc_s and not libunwind so I think the<br class="">problem may be somewhere else. There have been quite a lot of fixes<br class="">since clang 3.8 and its libc++ so it may be worth trying a more recent<br class="">clang.<br class=""><br class=""><br class="">I’m using mainline just-built clang to build libunwind. Actually, I’m cross-compiling on Windows for Linux.<br class="">Also, I’m using compiler-rt instead of libgcc.<br class="">And yes, the problem is not with libgcc_s, because, as I’ve said, force-setting the -funwind-tables flag in libunwind configuration makes the problem go away.<br class=""><br class=""><br class="">I've just checked a native build on Arm v8l, apologies I've not got a<br class="">Windows machine to cross-compile on. In my case I am getting unwind<br class="">information generated for libunwind. I'm guessing you are cross<br class="">compiling with a standalone build?<br class=""><br class=""><br class="">No, I’m building libunwind as part of LLVM. Can you please provide the options that you’re using to build libunwind?<br class=""><br class=""><br class="">The one I just checked was a native build where CC and CXX where clang 8.0<br class="">cmake -G Ninja ../llvm \<br class=""> -DCMAKE_BUILD_TYPE=Release\<br class=""> -DLLVM_ENABLE_ASSERTIONS=True\<br class=""> -DLLVM_TARGETS_TO_BUILD="ARM"\<br class=""> -DLLVM_ENABLE_PROJECTS="clang;compiler-rt;lld;libc++;libc++abi;libunwind"\<br class=""> -DLLVM_ENABLE_LLD=True<br class=""><br class="">In that case I'd expect my clang -fuse-unwind-tables to work as it<br class="">would be running natively and any test would pick up the native<br class="">libgcc_s at link time.<br class=""><br class="">readelf -u gave me amongst other things:<br class="">Unwind table index '.ARM.exidx.text._Unwind_Backtrace' at offset 0x49c<br class="">contains 1 entries:<br class=""><br class="">0x0: 0x809b46af<br class="">Compact model index: 0<br class="">0x9b vsp = r11<br class="">0x46 vsp = vsp - 28<br class="">0xaf pop {r4, r5, r6, r7, r8, r9, r10, r11, r14}<br class=""><br class=""><br class="">In the past I have cross-compiled<br class="">libunwind, in summary:<br class="">build compiler-rt builtins<br class="">build libuwind (statically, depending on compiler-rt)<br class="">build libcxxabi depending on libunwind and compiler-rt<br class="">build libcxx depending on libcxxabi and compiler-rt<br class=""><br class=""><br class="">That’s what I’m doing as well.<br class=""><br class=""><br class="">Apologies for the delay in responding, I thought I'd better dig out<br class="">build my script to check to see what unwind tables were present in<br class="">libunwind. I found that when building libunwind I needed<br class="">-DCMAKE_TRY_COMPILE_TARGET_TYPE=STATIC_LIBRARY for the CMake test for<br class="">unwind tables to succeed when cross-compiling. If I took that out the<br class="">test failed and I ended up with no exceptions.<br class=""><br class=""><br class="">Thank you for your time!<br class=""><br class="">I’ve tried CMAKE_TRY_COMPILE_TARGET_TYPE and it works. But setting this option manually seems suboptimal to me, maybe it makes sense to enable it whenever we’re cross-compiling like you suggested?<br class=""><br class="">The CMake documentation, however, says that setting this option to STATIC_LIBRARY avoids running the linker in try_compile() checks. I’m not sure we _always_ want that. Do we?<br class=""><br class=""></blockquote><span style="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; float: none; display: inline !important;" class="">I think it is fine when the output of the build is a static library as</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">there is no link step involved in it. So for linking libunwind.a I</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">think it is reasonable to add the flag. I agree that having to</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">remember to set it when cross-compiling is not ideal.</span><br style="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;" class=""><br style="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;" 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; text-decoration: none;" class=""><br class="">Using qemu-arm I was able to reproduce the failure when there were no<br class="">unwind tables in libunwind. So empirically it seems like unwind tables<br class="">are required. If I get some time tomorrow I'll check the libunwind<br class="">source to see if it starts the unwinding from libunwind itself. If it<br class="">is required for a significant part of libunwind on Arm, then ideally<br class="">we would want to fail the build if the flag were not supported.<br class=""><br class="">Peter<br class=""><br class=""><br class="">They are probably required only when we’re compiling for ARM, and not just any ARM, but when we’re using ARM EHABI, e. g. on Linux (AFAIU Darwin doesn’t use EHABI). But I don’t know a way to detect this at configuration time, only at build time.<br class=""><br class="">I have an idea of the way that we could detect the -funwind-tables flag at build time: the check_cxx_compiler_flag function in CMake sets the LIBUNWIND_SUPPORTS_${flagname}_FLAG variable. Now, if the value of that variable is true, we pass a preprocessor definition like _LIBUNWIND_HAS_UNWIND_TABLES to the compiler when building libunwind. In the libunwind source code, we check if that macro is defined, and if not, raise an #error.<br class=""><br class="">I’m going to prepare a patch so that this could be discussed further. What do you think?<br class=""></blockquote><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">That sounds reasonable, although I'm no CMake expert and there may be</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">a better way of detecting cross-compilation for EHABI and either</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">unconditionally adding -funwind-tables or throwing an error.</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">Compiler-rt has got quite a lot of target detection logic in it,</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">although it is difficult to follow. It may be worth starting a new</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">thread as I suspect that many of the experts won't be looking at this</span><br style="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;" class=""><span style="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; float: none; display: inline !important;" class="">thread anymore.</span><br style="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;" class=""><br style="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;" 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; text-decoration: none;" class=""><br class=""><br class=""><br class=""><br class=""><br class="">Are you trying to build libunwind into a shared library? I don't think<br class="">that is a supported configuration due to the cyclical dependency.<br class=""><br class=""><br class="">I’m building a static library.<br class=""><br class="">I’m building llvm & clang & lld first using the native compiler (MSVC), and then the just-built clang acts as a cross-compiler for the runtime libraries that you’ve mentioned.<br class=""><br class=""><br class="">Peter<br class=""><br class="">So, the main question remains: when we’re configuring libunwind build, CMake checks the -funwind-tables flag and that check fails because the __aeabi_unwind_cpp_pr0 symbol is absent. This symbol should be defined in libunwind, which is not build yet. I’m having a hard time understanding how can this be even possible. Can it be that you are not experiencing the problem because your clang uses libgcc and not compiler-rt?<br class=""><br class=""><br class="">Can I ask that you try and narrow down what the problem is, what your<br class="">environment is, and comment on the PR if it helps reproduce? It is<br class="">definitely worth looking at the output of clang -v to see which<br class="">unwinder it is using, by default it will be libgcc_s on Linux.<br class=""><br class=""><br class="">Peter<br class=""><br class=""><br class="">However, the CMake check fails with the following error:<br class=""><br class="">```<br class="">ld.lld: error: undefined symbol: __aeabi_unwind_cpp_pr0<br class=""><br class="">referenced by src.cxx<br class=""><br class=""><br class=""> CMakeFiles/cmTC_e9739.dir/src.cxx.o:(.ARM.exidx.text.main+0x0)<br class=""><br class=""><br class="">clang++: error: linker command failed with exit code 1 (use -v to see invocation)<br class=""><br class="">ninja: build stopped: subcommand failed.<br class=""><br class="">Source file was:<br class="">int main() { return 0; }<br class="">```<br class=""><br class="">No wonder! __aeabi_unwind_cpp_pr0 is defined in libunwind itself, which we haven’t built yet.<br class=""><br class="">If I instead set the -funwind-tables flag unconditionally using `add_compile_flags(-funwind-tables)` instead of `add_cxx_compile_flags_if_supported(-funwind-tables)`, everything is fine and the aforementioned bug is gone.<br class=""><br class="">I’ve found a PR which seemed to address this: <a href="https://reviews.llvm.org/D31858" class="">https://reviews.llvm.org/D31858</a> (cc’ing @phosek, @compnerd and @beanz as participants of this PR). It mentions that the __aeabi_unwind_cpp_pr0 symbol is provided by the compiler runtime. However, as I’ve already said, it lives in libunwind, so the problem doesn’t seem to be solved.<br class=""><br class="">I’m very tempted to just set the -funwind-tables flag unconditionally, but I’m afraid it’ll break something. What would be the right solution for building libunwind with this flag for ARM Linux?<br class=""><br class="">_______________________________________________<br class="">LLVM Developers mailing list<br class=""><a href="mailto:llvm-dev@lists.llvm.org" class="">llvm-dev@lists.llvm.org</a><br class="">https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</blockquote></div></blockquote></div><br class=""></body></html>