<div dir="ltr">Hi Jingyue,<div><br></div><div>Thanks for the instructions! I instrumented the device code and got a binary of <a href="http://axpy.cu">axpy.cu</a>; however, the resulting executable always fails on the first cudaMalloc call in host code (the kernel had not even been launched yet), with the error code being 30 (cudaErrorUnknown). In my instrumentation pass, I only inserted a hook function upon each access to device memory, with their signatures being: "<span class="">__device__</span><span class=""> </span><span class="">void</span><span class=""> _Cool_MemRead_Hook(</span><span class="">uint64_t</span><span class=""> addr)</span>". I've compiled these hooks functions into a shared object, and linked the axpy binary with it. </div><div><br></div><div>I'm really sorry to bother you again, but I wonder whether any step I did was apparently wrong, or there's any gpucc-specific step I need to do when instrumenting a kernel?</div><div><br></div><div>Thanks!</div><div>yuanfeng </div><div><br></div><div> </div>
</div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Mar 4, 2016 at 7:56 PM, Jingyue Wu <span dir="ltr"><<a href="mailto:jingyue@google.com" target="_blank">jingyue@google.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote"><span class="">On Fri, Mar 4, 2016 at 5:50 PM, Yuanfeng Peng <span dir="ltr"><<a href="mailto:yuanfeng.jack.peng@gmail.com" target="_blank">yuanfeng.jack.peng@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr">Hi Jingyue,<div><br></div><div>My name is Yuanfeng Peng, I'm a PhD student at UPenn. I'm sorry to bother you, but I'm having trouble with gpucc in my project, and I would be really grateful for your help!</div><div> </div><div>Currently we're trying to instrument CUDA code using LLVM 3.9, and I've written a pass to insert hook functions for certain function calls and memory accesses. For example, given a CUDA program, say, <a href="http://axpy.cu" target="_blank">axpy.cu</a>, I first compile it with </div><div> </div><div> clang++ -emit-llvm -c <a href="http://axpy.cu" target="_blank">axpy.cu</a>, </div><div><br></div><div>which gives me two bitcode files, axpy.bc and axpy-sm_20.bc. Then I use opt to load my pass and insert the hook functions to axpy.bc, which works fine. After inspecting the instrumented axpy.bc, I noticed that the kernel code was not there; rather, it lived inside axpy-sm_20.bc, so I also load my pass to instrument axpy-sm_20.bc. </div></div></blockquote><div><br></div></span><div>Expected. axpy.bc contains host code, and axpy-sm_??.bc contains device code. If you only want to instrument the device side, you don't need to modify axpy.bc. </div><span class=""><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div><br></div><div> However, after instrumenting axpy-sm_20.bc, I don't know how could I combine the host bitcode & device bitcode into a single binary... When I used llc to compile axpy-sm_20.bc into native code, I always got a bunch of errors; if I only do llc axpy.bc -o axpy.s and then link axpy.s with the necessary libraries, I got a working binary, but only the host code was instrumented. </div><div><br></div><div>So what should I do to get a binary where the device code is also instrumented? </div></div></blockquote><div><br></div><div><br></div></span><div>To link the modified axpy-sm_20.bc to the final binary, you need several extra steps:</div><div>1. Compile axpy-sm_20.bc to PTX assembly using llc: llc axpy-sm_20.bc -o axpy-sm_20.ptx -march=<nvptx or nvptx64></div><div>2. Compile the PTX assembly to SASS using ptxas</div><div>3. Make the SASS a fat binary using NVIDIA's fatbinary tool</div><div>4. Link the fat binary to the host code using ld. </div><div><br></div><div>Clang does step 2-4 by invoking subcommands. Therefore, you can use "clang -###" to dump all the subcommands, and then find the ones for step 2-4. For example,</div><div><br></div><div><div>$ clang++ -### -O3 <a href="http://axpy.cu" target="_blank">axpy.cu</a> -I/usr/local/cuda/samples/common/inc -L/usr/local/cuda/lib64 -lcudart_static -lcuda -ldl -lrt -pthread --cuda-gpu-arch=sm_35</div><div><br></div><div>clang version 3.9.0 (<a href="http://llvm.org/git/clang.git" target="_blank">http://llvm.org/git/clang.git</a> 4ce165e39e7b185e394aa713d9adffd920288988) (<a href="http://llvm.org/git/llvm.git" target="_blank">http://llvm.org/git/llvm.git</a> 2550ef485b6f9668bb7a4daa7ab276b6501492df)</div><div>Target: x86_64-unknown-linux-gnu</div><div>Thread model: posix</div><div>InstalledDir: /usr/local/google/home/jingyue/Work/llvm/install/bin</div><div> "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/clang-3.7" "-cc1" "-triple" "nvptx64-nvidia-cuda" "-aux-triple" "x86_64-unknown-linux-gnu" "-fcuda-target-overloads" "-fcuda-disable-target-call-checks" "-S" "-disable-free" "-main-file-name" "<a href="http://axpy.cu" target="_blank">axpy.cu</a>" "-mrelocation-model" "static" "-mthread-model" "posix" "-mdisable-fp-elim" "-fmath-errno" "-no-integrated-as" "-fcuda-is-device" "-mlink-cuda-bitcode" "/usr/local/cuda/nvvm/libdevice/libdevice.compute_35.10.bc" "-target-feature" "+ptx42" "-target-cpu" "sm_35" "-dwarf-column-info" "-debugger-tuning=gdb" "-resource-dir" "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/../lib/clang/3.9.0" "-I" "/usr/local/cuda/samples/common/inc" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8/backward" "-internal-isystem" "/usr/local/include" "-internal-isystem" "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/../lib/clang/3.9.0/include" "-internal-externc-isystem" "/include" "-internal-externc-isystem" "/usr/include" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8/backward" "-internal-isystem" "/usr/local/cuda/include" "-include" "__clang_cuda_runtime_wrapper.h" "-O3" "-fdeprecated-macro" "-fno-dwarf-directory-asm" "-fdebug-compilation-dir" "/usr/local/google/home/jingyue/Work/cuda" "-ferror-limit" "19" "-fmessage-length" "205" "-pthread" "-fobjc-runtime=gcc" "-fcxx-exceptions" "-fexceptions" "-fdiagnostics-show-option" "-fcolor-diagnostics" "-vectorize-loops" "-vectorize-slp" "-o" "/tmp/axpy-a88a72.s" "-x" "cuda" "<a href="http://axpy.cu" target="_blank">axpy.cu</a>"</div><div> "/usr/local/cuda/bin/ptxas" "-m64" "-O3" "--gpu-name" "sm_35" "--output-file" "/tmp/axpy-1dbca7.o" "/tmp/axpy-a88a72.s"</div><div> "/usr/local/cuda/bin/fatbinary" "--cuda" "-64" "--create" "/tmp/axpy-e6057c.fatbin" "--image=profile=sm_35,file=/tmp/axpy-1dbca7.o" "--image=profile=compute_35,file=/tmp/axpy-a88a72.s"</div><div> "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/clang-3.7" "-cc1" "-triple" "x86_64-unknown-linux-gnu" "-aux-triple" "nvptx64-nvidia-cuda" "-fcuda-target-overloads" "-fcuda-disable-target-call-checks" "-emit-obj" "-disable-free" "-main-file-name" "<a href="http://axpy.cu" target="_blank">axpy.cu</a>" "-mrelocation-model" "static" "-mthread-model" "posix" "-fmath-errno" "-masm-verbose" "-mconstructor-aliases" "-munwind-tables" "-fuse-init-array" "-target-cpu" "x86-64" "-momit-leaf-frame-pointer" "-dwarf-column-info" "-debugger-tuning=gdb" "-resource-dir" "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/../lib/clang/3.9.0" "-I" "/usr/local/cuda/samples/common/inc" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/x86_64-linux-gnu/c++/4.8" "-internal-isystem" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../../include/c++/4.8/backward" "-internal-isystem" "/usr/local/include" "-internal-isystem" "/usr/local/google/home/jingyue/Work/llvm/install-git/bin/../lib/clang/3.9.0/include" "-internal-externc-isystem" "/usr/include/x86_64-linux-gnu" "-internal-externc-isystem" "/include" "-internal-externc-isystem" "/usr/include" "-internal-isystem" "/usr/local/cuda/include" "-include" "__clang_cuda_runtime_wrapper.h" "-O3" "-fdeprecated-macro" "-fdebug-compilation-dir" "/usr/local/google/home/jingyue/Work/cuda" "-ferror-limit" "19" "-fmessage-length" "205" "-pthread" "-fobjc-runtime=gcc" "-fcxx-exceptions" "-fexceptions" "-fdiagnostics-show-option" "-fcolor-diagnostics" "-vectorize-loops" "-vectorize-slp" "-o" "/tmp/axpy-48f6b5.o" "-x" "cuda" "<a href="http://axpy.cu" target="_blank">axpy.cu</a>" "-fcuda-include-gpubinary" "/tmp/axpy-e6057c.fatbin"</div><div> "/usr/bin/ld" "-z" "relro" "--hash-style=gnu" "--build-id" "--eh-frame-hdr" "-m" "elf_x86_64" "-dynamic-linker" "/lib64/ld-linux-x86-64.so.2" "-o" "a.out" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../x86_64-linux-gnu/crt1.o" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../x86_64-linux-gnu/crti.o" "/usr/lib/gcc/x86_64-linux-gnu/4.8/crtbegin.o" "-L/usr/local/cuda/lib64" "-L/usr/lib/gcc/x86_64-linux-gnu/4.8" "-L/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../x86_64-linux-gnu" "-L/lib/x86_64-linux-gnu" "-L/lib/../lib64" "-L/usr/lib/x86_64-linux-gnu" "-L/usr/lib/gcc/x86_64-linux-gnu/4.8/../../.." "-L/usr/local/google/home/jingyue/Work/llvm/install-git/bin/../lib" "-L/lib" "-L/usr/lib" "/tmp/axpy-48f6b5.o" "-lcudart_static" "-lcuda" "-ldl" "-lrt" "-lstdc++" "-lm" "-lgcc_s" "-lgcc" "-lpthread" "-lc" "-lgcc_s" "-lgcc" "/usr/lib/gcc/x86_64-linux-gnu/4.8/crtend.o" "/usr/lib/gcc/x86_64-linux-gnu/4.8/../../../x86_64-linux-gnu/crtn.o"</div></div><span class=""><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div><br></div><div>I apologize for the long email and I look forward to hearing from you. </div><div><br></div><div>Thanks!</div><span><font color="#888888"><div>Yuanfeng Peng</div></font></span></div>
</blockquote></span></div><br></div></div>
</blockquote></div><br></div>