Cpp2 terminated by signal 11: error with newer NVHPC releases like 23.1, 22.11

[pramod.s.kumbhar]

Dear All,

For the NEURON simulator project we have been using NVHPC 22.3 until now without issue. We are trying to use latest releases like 22.11 and 23.1 but seeing segfault in the following configurations:

• -O2 or -O1 -g compiles fine
• -O2 -g segfaults with following error:

$ /path.../Linux_x86_64/22.11/compilers/bin/nvc++ ivocvect.i -c -O2 -g

/gpfs-path//Linux_x86_64/22.11/compilers/bin/tools/cpp2 TERMINATED by signal 11
Arguments to /gpfs-path//Linux_x86_64/22.11/compilers/bin/tools/cpp2
/gpfs-path//Linux_x86_64/22.11/compilers/bin/tools/cpp2 /gpfs/bbp.cscs.ch/home/kumbhar/ivocvect.i -debug -x 120 0x8000 -opt 2 -terse 1 -inform warn -x 119 0xa10000 -x 122 0x40 -x 123 0x1000 -x 127 4 -x 127 17 -x 19 0x400000 -x 28 0x40000 -x 120 0x10000000 -x 70 0x8000 -x 122 1 -x 125 0x20000 -quad -vect 56 -y 34 16 -x 37 0x480000 -x 34 0x8 -x 32 259522

56 -y 19 8 -y 35 0 -x 42 0x30 -x 39 0x40 -x 42 0x20000 -x 39 0x80 -x 59 4 -x 129 2 -y 129 0x8000 -tp skylake-avx512 -astype 0 -x 121 1 -fn /gpfs/bbp.cscs.ch/home/kumbhar/ivocvect.i -il /tmp/nvc++Zwtiz-GrG7NQ.il -x 117 0x200 -x 123 0x80000000 -x 123 4 -x 119 0x20 -def __pgnu_vsn=110200 -x 70 0x40000000 -x 183 4 -x 121 0x800 -x 6 0x20000 -autoinl 10 -x 168 400 -x 174 128000 -x 14 0x200000 -x 14 0x400000 -autoinl 10 -x 168 400 -x 174 128000 -x 14 0x200000 -x 14 0x400000 -x 249 140 -x 120 0x200000 -x 70 0x40000000 -x 8 0x40000000 -x 164 0x800000 -x 71 0x2000 -x 71 0x4000 -x 34 0x40000000 -x 83 0x1 -x 85 0x1 -x 15 0x1000000 -x 206 0x02 -x 120 0x1000000 -x 68 0x1 -x 39 4 -x 56 0x10 -x 26 0x10 -x 26 1 -x 56 0x4000 -x 197 0 -x 175 0 -x 203 0 -x 204 0 -y 163 0xc0000000 -x 192 0x40000000 -x 189 0x10 -y 189 0x4000000 -x 60 512 -x 9 1 -x 72 0x1 -x 136 0x11 -x 9 1 -x 72 0x1 -x 136 0x11 -gnuvsn 110200 -x 69 0x200 -x 123 0x400 -x 119 0x08 -cmdline '+nvc++ /tmp/nvc++Zwtiz-GrG7NQ.il -D__GCC_ATOMIC_TEST_AND_SET_TRUEVAL=1 -c -O2 -Mvect=simd -Mno-signed-zeros -g' -asm /tmp/nvc++JwtiPoWT2K1Z.ll

I have attached the preprocessed source file ivocvect.i from our project that you can use to reproduce the issue.

Please let me know if you need additional information.

Thank you!

P.S. I will upload here smaller reproducer if I get one via cvise/c-reduce

[MatColgrove]

Hi Pramod,

Thanks for the report. I was able to reproduce the error with 23.1 but fortunately don’t see the error in our pre-release 23.3 version.

Though, in looking at the changes that went in when this started passing, they seem to me to be unrelated. Hence I went ahead and added a problem report, TPR #33056, asking engineering to ensure that the changes did indeed fix the issue as opposed to masking it.

-Mat

[pramod.s.kumbhar]

Thanks for the quick update, Mat!

@MatColgrove

but fortunately don’t see the error in our pre-release 23.3 version.

Do you know when 22.3 will be out?

[MatColgrove]

Do you know when 22.3 will be out?

Not specifically, but most likely some time in March. Though occasionally things slip.