error MSB3073 exited with code -1

[lurchinms]

Fresh install on Ubuntu 20.04. I also reinstall Ubuntu 20.04 and have the same error.

rm -f librandomarq.so g++ -shared -pthread -L. -Wl,-whole-archive librandomx.a -Wl,-no-whole-archive -o librandomarq.so -lstdc++ -lgcc -lc /home/XXXXXXXX/miningcore/src/Miningcore/Miningcore.csproj(123,9): error MSB3073: The command "(cd /home/XXXXXXXX/miningcore/src/Miningcore/ && sh build-libs-linux.sh bin/Release/net6.0/)" exited with code -1.

[oliverw]

All dependencies installed? What happens if you run make inside src/Native/librandomarq?

[lurchinms]

Used the build-ubuntu-20.04.sh

[lurchinms]

Same error on manual build as well.

[oliverw]

Same error on manual build as well.

I can't be the same error as MSBUILD is not involved when running make individually.

[lurchinms]

I am sorry, when I say manual build, I meant: dotnet publish -c Release --framework net6.0 -o ../../build

[lurchinms]

Won't build on Ubuntu 21 either. Same error MSB3073.

[lurchinms]

/usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h(913,1): error G2ABF96FA: inlining failed in call to 'always_inline' '__m256i _mm256_xor_si256(__m256i, __m256i)': target specific option mismatch [/home/XXXXXXXX /miningcore/src/Miningcore/Miningcore.csproj]

[lurchinms]

Must be a Microsoft dotnet issue, maybe some update. I tried build it on a different server using Ubuntu 18.04 and get the same error.

[oliverw]

/usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h(913,1): error G2ABF96FA: inlining failed in call to 'always_inline' '__m256i _mm256_xor_si256(__m256i, __m256i)': target specific option mismatch [/home/XXXXXXXX /miningcore/src/Miningcore/Miningcore.csproj]

That is a GCC error. Not a Microsoft error. Could you please try what I outlined above?

[oliverw]

Does it still happen with the latest dev branch?

[lurchinms]

I perform fresh Ubuntu 21.04 install and installed using the build-ubuntu-21.04.sh and it performed the same error.

[lurchinms]

I think the problem is inside the XMR build. from xmrig/crypto/cn/gpu/cn_gpu_avx.cpp:30: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h: In function ‘void double_compute_wrap(const __m256&, const __m256&, const __m256&, const __m256&, float, float, const __m256&, __m256&, __m256i&) [with long unsigned int rot = 0]’: /usr/lib/gcc/x86_64-linux-gnu/11/include/avx2intrin.h:913:1: error: inlining failed in call to ‘always_inline’ ‘__m256i _mm256_xor_si256(__m256i, __m256i)’: target specific option mismatch 913 | _mm256_xor_si256 (__m256i __A, __m256i __B) | ^~~~~~~~~~~~~~~~ xmrig/crypto/cn/gpu/cn_gpu_avx.cpp:131:27: note: called from here 131 | out = _mm256_xor_si256(out, r); | ~~~~~~~~~~~~~~~~^~~~~~~~ make: *** [: xmrig/crypto/cn/gpu/cn_gpu_avx.o] Error 1

[oliverw]

I cannot reproduce this problem on any system.

[lurchinms]

I cannot reproduce this problem on any system.

On fresh server installs?

[lurchinms]

I think the error is in the build-libs-linux.sh for RandomX. I will test further when I get home from work this evening.

[makertronic]

hello, I'm on debian 11 and i got this error too : 3randomx_cache]+0x28): undefined reference to randomx::JitCompilerX86::~JitCompilerX86()' /usr/bin/ld: librandomx.a(dataset.cpp.o): in function void randomx::deallocCacherandomx::LargePageAllocator(randomx_cache*)': dataset.cpp:(.text._ZN7randomx12deallocCacheINS_18LargePageAllocatorEEEvP13randomx_cache[_ZN7randomx12deallocCacheINS_18LargePageAllocatorEEEvP13randomx_cache]+0x28): undefined reference to `randomx::JitCompilerX86::~JitCompilerX86()' collect2 : error : ld returned 1 exit status [/home/nicola/miningcore/src/Miningcore/Miningcore.csproj] make[2]: *** [CMakeFiles/randomx-benchmark.dir/build.make:119: randomx-benchmark] Error 1 make[1]: *** [CMakeFiles/Makefile2:101: CMakeFiles/randomx-benchmark.dir/all] Error 2 make: *** [Makefile:149: all] Error 2 /home/nicola/miningcore/src/Miningcore/Miningcore.csproj(123,9): error MSB3073: La commande "(cd /home/nicola/miningcore/src/Miningcore/ && sh build-libs-linux.sh bin/Release/net6.0/)" s'est arrêtée avec le code 2.

[oliverw]

I suspect both of you are running on CPUs which are missing instruction extensions required to compile RandomX.

[makertronic]

maybe, here my CPU : processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 60 model name : Intel(R) Core(TM) i5-4590 CPU @ 3.30GHz stepping : 3 microcode : 0x28 cpu MHz : 2537.699 cache size : 6144 KB physical id : 0 siblings : 4 core id : 3 cpu cores : 4 apicid : 6 initial apicid : 6 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti ssbd ibrs ibpb stibp fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts md_clear flush_l1d bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds

[makertronic]

anyway, can we remove support for RandomX ?

[oliverw]

Sure, just comment out the last two lines inside build-libs-linux.sh and try again. Let me know how it went and obviously don't try to use any Cryptonight Family coin (Monero etc) with your pool.

[makertronic]

it worked ! So maybe it was my poor cpu :)

[makertronic]

[lurchinms]

Hmmm. I will do more research on this issue. I pretty sure this can run it.

[oliverw]

Hmmm. I will do more research on this issue. I pretty sure this can run it.

I agree.

[lurchinms]

I ended up going to an older build. Before Commits on March 11, 2022

[minershaven]

I ended up going to an older build. Before Commits on March 11, 2022

i had to do the same thing. got that same error using ubuntu 18 and 20.

[minershaven]

I have tried the new version on Ubuntu 18 and 20 as well as trying to build it on windows and get the same error each time. Something with the Randomx algorithm is stopping it.

[3djc]

I faced the same issue, but reading the previous comment, it became clear the issue was with the CPU not handling all the required instruction set for Randomx? Since it was a self hosted VM, I changed CPU type from "kvm64" to "host", and that game me the required CPU functions and it did compile (and run) successfully. Of course, one could hope a much clearer error message, or conditional inclusion of randomx based on CPU generation

[minershaven]

my cpu should be fine. ive tried it on my servers which have Xeon(R) E3-1231 v3 @ 3.40GHz cpu's and i even tried the windows build on my home pc which has a i9-10980xe so that shouldn't be whats preventing me from building it. I even tried taking out the RandomX part of the install as Oliver suggested earlier but still nothing. The error i'm getting, g++ -shared -pthread -L. -Wl,-whole-archive librandomx.a -Wl,-no-whole-archive -o librandomarq.so -lstdc++ -lgcc -lc /home/XXXXXXXX/miningcore/src/Miningcore/Miningcore.csproj(123,9): error MSB3073: The command "(cd /home/XXXXXXXX/miningcore/src/Miningcore/ && sh build-libs-linux.sh bin/Release/net6.0/)" exited with code -1.

looks to me like its an issue with the bin/Release/net6.0/ folder. dont know if it's a permission issue or if it doesnt exsist but it looks like the issue is there.

Or i could be completely wrong, who knows

[minershaven]

havent been able to fix it but i did find a workaround for it. everything gets built to /Miningcore/bin/Release/net6.0 folder so i just copied everything in there to the build folder and everything is working fine now.