Closed randall-coding closed 3 years ago
Maybe it’s running with python2 by accident? From the source code, repo
wants at least python 3.6. Running with 3.5 prints a warning and everything below aborts.
What version of a repo tool are you using.
I don't have issues with repo 2.12-1 from Debian testing.
Sorry for my late response. Playing around with all these GPU settings really did a number on my machine (I think it was installing some of the proprietary Radeon software that tanked my machine).
I updated my kernel to 5.4 and updated Ubuntu from 18.04 to 20.04. I will look into the points mentioned here and get back to you
Ok, now I'm following the Ubuntu 20.04 instructions. I'm able to get to the repo sync
step now but it started to hang after fetching 7/8.
Fetching: 75% [3 jobs] (6/8) finished llpc Fetching: 87% [2 jobs] (7/8) finished pal ->hangs
On subsequent tries it doesn't print out anything and just hangs there.
Update: When I cancel with alt+ctrl+c, it produces this message
error: Cannot fetch llvm-project from https://github.com/GPUOpen-Drivers/llvm-project Fetching: 100% [1 job] (8/8) finished llvm-projectaborted by user
@baryluk
repo version is 2.13, I can try 2.12 instead
I tried deleting everything in .repo and starting over. Looks like maybe the issue is just a slow download on the last repo. Strangely the log messages don't show up on repo sync -v
(even in verbose mode) until after I abort the operation. On this last attempt I saw messages indicating it was downloading something (just very slowly). Looks like a slow download speed could be the culprit.
Left it running over night and it completed successfully. Turns out it was just a long download.
Fetching: 100% (8/8), done in 5h12m8.467s Updating files: 100% (101166/101166), done. Checking out: 100% (8/8), done in 15.842s repo sync has finished successfully.
Fetching: 100% (8/8), done in 5h12m8.467s
Yes, the llvm one is very big.
Glad you have it working.
You were probably using too old repo or some weird manually installed repo version on the Ubuntu 18.04 before, that caused some issues.
Unfortunately I'm running into errors on one of the build commands that I'm trying to figure out.
ninja -C drivers/xgl/builds/Release64
Do you know if there is a way to re-build everything without having to do that long repo sync
step again?
I think I may need a rebuild because my PC shutoff during the make process by accident.
Here is the error output for reference.
Linking CXX shared library compiler/llpc/llvm/lib/libLTO.so.13git FAILED: compiler/llpc/llvm/lib/libLTO.so.13git : && /usr/bin/c++ -fPIC -fPIC -fvisibility-inlines-hidden -Werror=date-time -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wimplicit-fallthrough -Wno-maybe-uninitialized -Wno-class-memaccess -Wno-redundant-move -Wno-noexcept-type -Wdelete-non-virtual-dtor -Wsuggest-override -Wno-comment -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -Wl,-z,defs -Wl,-z,nodelete -s -Wl,-rpath-link,/home/super/vulkandriver/drivers/xgl/builds/Release64/compiler/llpc/llvm/./lib -Wl,-O3 -Wl,--gc-sections -Wl,--version-script,"/home/super/vulkandriver/drivers/xgl/builds/Release64/compiler/llpc/llvm/tools/lto/LTO.exports" -shared -Wl,-soname,libLTO.so.13git -o compiler/llpc/llvm/lib/libLTO.so.13git compiler/llpc/llvm/tools/lto/CMakeFiles/LTO.dir/LTODisassembler.cpp.o compiler/llpc/llvm/tools/lto/CMakeFiles/LTO.dir/lto.cpp.o -Wl,-rpath,"\$ORIGIN/../lib" compiler/llpc/llvm/lib/libLLVMAMDGPUAsmParser.a compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a compiler/llpc/llvm/lib/libLLVMAMDGPUDesc.a compiler/llpc/llvm/lib/libLLVMAMDGPUDisassembler.a compiler/llpc/llvm/lib/libLLVMAMDGPUInfo.a compiler/llpc/llvm/lib/libLLVMBitReader.a compiler/llpc/llvm/lib/libLLVMCore.a compiler/llpc/llvm/lib/libLLVMCodeGen.a compiler/llpc/llvm/lib/libLLVMLTO.a compiler/llpc/llvm/lib/libLLVMMC.a compiler/llpc/llvm/lib/libLLVMMCDisassembler.a compiler/llpc/llvm/lib/libLLVMSupport.a compiler/llpc/llvm/lib/libLLVMTarget.a compiler/llpc/llvm/lib/libLLVMAsmPrinter.a compiler/llpc/llvm/lib/libLLVMDebugInfoDWARF.a compiler/llpc/llvm/lib/libLLVMGlobalISel.a compiler/llpc/llvm/lib/libLLVMSelectionDAG.a compiler/llpc/llvm/lib/libLLVMMIRParser.a compiler/llpc/llvm/lib/libLLVMAMDGPUDesc.a compiler/llpc/llvm/lib/libLLVMAMDGPUInfo.a compiler/llpc/llvm/lib/libLLVMAMDGPUUtils.a compiler/llpc/llvm/lib/libLLVMCodeGen.a compiler/llpc/llvm/lib/libLLVMPasses.a compiler/llpc/llvm/lib/libLLVMTarget.a compiler/llpc/llvm/lib/libLLVMCoroutines.a compiler/llpc/llvm/lib/libLLVMipo.a compiler/llpc/llvm/lib/libLLVMVectorize.a compiler/llpc/llvm/lib/libLLVMFrontendOpenMP.a compiler/llpc/llvm/lib/libLLVMIRReader.a compiler/llpc/llvm/lib/libLLVMAsmParser.a compiler/llpc/llvm/lib/libLLVMInstrumentation.a compiler/llpc/llvm/lib/libLLVMScalarOpts.a compiler/llpc/llvm/lib/libLLVMBitWriter.a compiler/llpc/llvm/lib/libLLVMAggressiveInstCombine.a compiler/llpc/llvm/lib/libLLVMExtensions.a compiler/llpc/llvm/lib/libLLVMInstCombine.a compiler/llpc/llvm/lib/libLLVMLinker.a compiler/llpc/llvm/lib/libLLVMObjCARCOpts.a compiler/llpc/llvm/lib/libLLVMTransformUtils.a compiler/llpc/llvm/lib/libLLVMAnalysis.a compiler/llpc/llvm/lib/libLLVMProfileData.a compiler/llpc/llvm/lib/libLLVMObject.a compiler/llpc/llvm/lib/libLLVMBitReader.a compiler/llpc/llvm/lib/libLLVMCore.a compiler/llpc/llvm/lib/libLLVMRemarks.a compiler/llpc/llvm/lib/libLLVMBitstreamReader.a compiler/llpc/llvm/lib/libLLVMMCParser.a compiler/llpc/llvm/lib/libLLVMMC.a compiler/llpc/llvm/lib/libLLVMDebugInfoCodeView.a compiler/llpc/llvm/lib/libLLVMDebugInfoMSF.a compiler/llpc/llvm/lib/libLLVMTextAPI.a compiler/llpc/llvm/lib/libLLVMBinaryFormat.a compiler/llpc/llvm/lib/libLLVMSupport.a -lrt -ldl -lpthread -lm compiler/llpc/llvm/lib/libLLVMDemangle.a && : /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in function
llvm::R600Subtarget::R600Subtarget(llvm::Triple const&, llvm::StringRef, llvm::StringRef, llvm::TargetMachine const&)': AMDGPUSubtarget.cpp:(.text._ZN4llvm13R600SubtargetC2ERKNS_6TripleENS_9StringRefES4_RKNS_13TargetMachineE+0x15e): undefined reference to
vtable for llvm::SelectionDAGTargetInfo' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionllvm::GCNSubtarget::GCNSubtarget(llvm::Triple const&, llvm::StringRef, llvm::StringRef, llvm::GCNTargetMachine const&)': AMDGPUSubtarget.cpp:(.text._ZN4llvm12GCNSubtargetC2ERKNS_6TripleENS_9StringRefES4_RKNS_16GCNTargetMachineE+0x19e): undefined reference to
vtable for llvm::SelectionDAGTargetInfo' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionllvm::R600Subtarget::~R600Subtarget()': AMDGPUSubtarget.cpp:(.text._ZN4llvm13R600SubtargetD2Ev[_ZN4llvm13R600SubtargetD5Ev]+0x2e): undefined reference to
llvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionnon-virtual thunk to llvm::R600Subtarget::~R600Subtarget()': AMDGPUSubtarget.cpp:(.text._ZN4llvm13R600SubtargetD2Ev[_ZN4llvm13R600SubtargetD5Ev]+0x15e): undefined reference to
llvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionllvm::GCNSubtarget::~GCNSubtarget()': AMDGPUSubtarget.cpp:(.text._ZN4llvm12GCNSubtargetD2Ev+0x132): undefined reference to
llvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionnon-virtual thunk to llvm::R600Subtarget::~R600Subtarget()': AMDGPUSubtarget.cpp:(.text._ZN4llvm13R600SubtargetD0Ev[_ZN4llvm13R600SubtargetD5Ev]+0x3b): undefined reference to
llvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUSubtarget.cpp.o): in functionllvm::R600Subtarget::~R600Subtarget()': AMDGPUSubtarget.cpp:(.text._ZN4llvm13R600SubtargetD0Ev[_ZN4llvm13R600SubtargetD5Ev]+0x16e): undefined reference to
llvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMAMDGPUCodeGen.a(AMDGPUTargetMachine.cpp.o):AMDGPUTargetMachine.cpp:(.text._ZN4llvm17R600TargetMachineD2Ev[_ZN4llvm17R600TargetMachineD5Ev]+0xa4): more undefined references tollvm::SelectionDAGTargetInfo::~SelectionDAGTargetInfo()' follow /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGRRList.cpp.o):(.data.rel.ro._ZTVN12_GLOBAL__N_117ScheduleDAGRRListE+0x40): undefined reference to
llvm::ScheduleDAGSDNodes::getGraphNodeLabel[abi:cxx11](llvm::SUnit const) const' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGRRList.cpp.o):(.data.rel.ro._ZTVN12_GLOBAL__N_117ScheduleDAGRRListE+0x78): undefined reference to `llvm::ScheduleDAGSDNodes::getCustomGraphFeatures(llvm::GraphWriter<llvm::ScheduleDAG>&) const' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGSDNodes.cpp.o):(.data.rel.ro._ZTVN4llvm18ScheduleDAGSDNodesE[_ZTVN4llvm18ScheduleDAGSDNodesE]+0x40): undefined reference tollvm::ScheduleDAGSDNodes::getGraphNodeLabel[abi:cxx11](llvm::SUnit const*) const' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGSDNodes.cpp.o):(.data.rel.ro._ZTVN4llvm18ScheduleDAGSDNodesE[_ZTVN4llvm18ScheduleDAGSDNodesE]+0x78): undefined reference to
llvm::ScheduleDAGSDNodes::getCustomGraphFeatures(llvm::GraphWriter<llvm::ScheduleDAG>&) const' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGVLIW.cpp.o):(.data.rel.ro._ZTVN12_GLOBAL__N_115ScheduleDAGVLIWE+0x40): undefined reference to `llvm::ScheduleDAGSDNodes::getGraphNodeLabel[abi:cxx11](llvm::SUnit const) const' /usr/bin/ld: compiler/llpc/llvm/lib/libLLVMSelectionDAG.a(ScheduleDAGVLIW.cpp.o):(.data.rel.ro._ZTVN12_GLOBAL__N_115ScheduleDAGVLIWE+0x78): undefined reference to `llvm::ScheduleDAGSDNodes::getCustomGraphFeatures(llvm::GraphWriter<llvm::ScheduleDAG*>&) const' collect2: error: ld returned 1 exit status [45/635] Building CXX object compiler/llpc/CMakeFiles/llpc.dir/translator/lib/SPIRV/SPIRVReader.cpp.o
Unfortunately I'm running into errors on one of the build commands that I'm trying to figure out.
ninja -C drivers/xgl/builds/Release64
Just do rm -rf drivers/xgl/builds/Release64
and redo the cmake + ninja steps. Next time open separate issue ticket for separate issues.
Ok thanks again. Will do.
When running this command
repo init -u https://github.com/GPUOpen-Drivers/AMDVLK.git -b master
as part of the installation process, I receive a syntax error in the python script main.pyI have python3 and 2.7 installed at the same time. I'm going to try removing the offending line and try again. Any advice as to why this is happening?
My system is running Ubuntu 18.04 64 bit, AMD Radeon 7730 HD