Error when compiling AVX512 related functions

[dnomd343]

Zig Version

0.14.0-dev.2271+f845fa04a

Steps to Reproduce and Observed Behavior

I have the zig c++ command working just fine when compiling most of my C++ code, but there is one module that relies on xxHash, a fast hashing implementation.

On my computer, the following simple C++ code can reproduce this problem.

#include <string>
#include <iostream>

#include "xxHash/xxh3.h"

int main() {
    const std::string data {"hello"};
    const auto digest = XXH_INLINE_XXH3_64bits(data.c_str(), data.size());
    std::cout << std::hex << digest << std::endl;
}

Under g++ and clang it works fine:

> g++ --version
g++ (Ubuntu 14.2.0-4ubuntu2~24.04) 14.2.0
Copyright (C) 2024 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
> g++ demo.cc -o demo
> ./demo
9555e8555c62dcfd

> clang++ --version
Ubuntu clang version 18.1.3 (1ubuntu1)
Target: x86_64-pc-linux-gnu
Thread model: posix
InstalledDir: /usr/bin
> clang++ demo.cc -o demo
> ./demo
9555e8555c62dcfd

But when I use zig c++ instead, it gives the following error:

> zig version
0.14.0-dev.2271+f845fa04a
> zig c++ demo.cc -o demo
In file included from demo.cc:4:
In file included from ./xxHash/xxh3.h:55:
./xxHash/xxhash.h:4587:37: error: always_inline function '_mm512_loadu_si512' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4587 |         __m512i const data_vec    = _mm512_loadu_si512   (input);
      |                                     ^
./xxHash/xxhash.h:4587:37: error: AVX vector return of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4589:37: error: always_inline function '_mm512_loadu_si512' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4589 |         __m512i const key_vec     = _mm512_loadu_si512   (secret);
      |                                     ^
./xxHash/xxhash.h:4589:37: error: AVX vector return of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4591:37: error: always_inline function '_mm512_xor_si512' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4591 |         __m512i const data_key    = _mm512_xor_si512     (data_vec, key_vec);
      |                                     ^
./xxHash/xxhash.h:4591:37: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4593:37: error: always_inline function '_mm512_srli_epi64' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4593 |         __m512i const data_key_lo = _mm512_srli_epi64 (data_key, 32);
      |                                     ^
./xxHash/xxhash.h:4593:37: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4595:37: error: always_inline function '_mm512_mul_epu32' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4595 |         __m512i const product     = _mm512_mul_epu32     (data_key, data_key_lo);
      |                                     ^
./xxHash/xxhash.h:4595:37: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4597:35: error: '__builtin_ia32_pshufd512' needs target feature avx512f,evex512
 4597 |         __m512i const data_swap = _mm512_shuffle_epi32(data_vec, (_MM_PERM_ENUM)_MM_SHUFFLE(1, 0, 3, 2));
      |                                   ^
/usr/local/zig/lib/include/avx512fintrin.h:8784:13: note: expanded from macro '_mm512_shuffle_epi32'
 8784 |   ((__m512i)__builtin_ia32_pshufd512((__v16si)(__m512i)(A), (int)(I)))
      |             ^
In file included from demo.cc:4:
In file included from ./xxHash/xxh3.h:55:
./xxHash/xxhash.h:4598:35: error: always_inline function '_mm512_add_epi64' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4598 |         __m512i const sum       = _mm512_add_epi64(*xacc, data_swap);
      |                                   ^
./xxHash/xxhash.h:4598:35: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4600:17: error: always_inline function '_mm512_add_epi64' requires target feature 'evex512', but would be inlined into function 'XXH3_accumulate_512_avx512' that is compiled without support for 'evex512'
 4600 |         *xacc = _mm512_add_epi64(product, sum);
      |                 ^
./xxHash/xxhash.h:4600:17: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4632:33: error: always_inline function '_mm512_set1_epi32' requires target feature 'evex512', but would be inlined into function 'XXH3_scrambleAcc_avx512' that is compiled without support for 'evex512'
 4632 |         const __m512i prime32 = _mm512_set1_epi32((int)XXH_PRIME32_1);
      |                                 ^
./xxHash/xxhash.h:4632:33: error: AVX vector return of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
./xxHash/xxhash.h:4636:37: error: always_inline function '_mm512_srli_epi64' requires target feature 'evex512', but would be inlined into function 'XXH3_scrambleAcc_avx512' that is compiled without support for 'evex512'
 4636 |         __m512i const shifted     = _mm512_srli_epi64    (acc_vec, 47);
      |                                     ^
./xxHash/xxhash.h:4636:37: error: AVX vector argument of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI
fatal error: too many errors emitted, stopping now [-ferror-limit=]
20 errors generated.

The xxhash.h can find here.

More detail:

> zig c++ demo.cc -o demo -v
clang version 19.1.0 (https://github.com/ziglang/zig-bootstrap 46b9e66db90230fe62404b27b85a378ccf2c82c2)
Target: x86_64-unknown-linux6.8.0-gnu2.39.0
Thread model: posix
InstalledDir: /usr/local/zig
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/13
Found candidate GCC installation: /usr/lib/gcc/x86_64-linux-gnu/14
Selected GCC installation: /usr/lib/gcc/x86_64-linux-gnu/14
Candidate multilib: .;@m64
Selected multilib: .;@m64
 (in-process)
 "/usr/local/zig/zig" -cc1 -triple x86_64-unknown-linux6.8.0-gnu2.39.0 -emit-obj -disable-free -clear-ast-before-backend -disable-llvm-verifier -discard-value-names -main-file-name demo.cc -mrelocation-model pic -pic-level 2 -fhalf-no-semantic-interposition -mframe-pointer=all -fmath-errno -ffp-contract=on -fno-rounding-math -mconstructor-aliases -funwind-tables=2 -target-cpu x86-64 -tune-cpu generic -debug-info-kind=constructor -dwarf-version=4 -debugger-tuning=gdb -gdwarf32 -fdebug-compilation-dir=/root/klotski/third_party -v -fcoverage-compilation-dir=/root/klotski/third_party -nostdsysteminc -nobuiltininc -resource-dir /usr/local/lib/clang/19 -dependency-file /root/.cache/zig/tmp/69dcf0347f794377-demo.o.d -MT /root/.cache/zig/tmp/69dcf0347f794377-demo.o -sys-header-deps -MV -isystem /usr/local/zig/lib/libcxx/include -isystem /usr/local/zig/lib/libcxxabi/include -isystem /usr/local/zig/lib/libunwind/include -isystem /usr/local/zig/lib/include -isystem /usr/include -isystem /usr/include/x86_64-linux-gnu -isystem /usr/local/include -isystem /usr/include/x86_64-linux-gnu -isystem /usr/include -D _LIBCPP_DISABLE_VISIBILITY_ANNOTATIONS -D _LIBCPP_HAS_NO_VENDOR_AVAILABILITY_ANNOTATIONS -D _LIBCXXABI_DISABLE_VISIBILITY_ANNOTATIONS -D _LIBCPP_PSTL_BACKEND_SERIAL -D _LIBCPP_ABI_VERSION=1 -D _LIBCPP_ABI_NAMESPACE=__1 -D _LIBCPP_HARDENING_MODE=_LIBCPP_HARDENING_MODE_DEBUG -D __GLIBC_MINOR__=39 -D _DEBUG -O0 -fdeprecated-macro -ferror-limit 19 -fsanitize=alignment,array-bounds,bool,builtin,enum,float-cast-overflow,integer-divide-by-zero,nonnull-attribute,null,pointer-overflow,return,returns-nonnull-attribute,shift-base,shift-exponent,signed-integer-overflow,unreachable,vla-bound -fsanitize-trap=alignment,array-bounds,bool,builtin,enum,float-cast-overflow,integer-divide-by-zero,nonnull-attribute,null,pointer-overflow,return,returns-nonnull-attribute,shift-base,shift-exponent,signed-integer-overflow,unreachable,vla-bound -fno-sanitize-memory-param-retval -fno-sanitize-address-use-odr-indicator -stack-protector 2 -stack-protector-buffer-size 4 -fgnuc-version=4.2.1 -fskip-odr-check-in-gmf -fcxx-exceptions -fexceptions -fcolor-diagnostics -fno-spell-checking -target-cpu x86-64 -target-feature -16bit-mode -target-feature -32bit-mode -target-feature +64bit -target-feature +adx -target-feature +aes -target-feature -allow-light-256-bit -target-feature -amx-bf16 -target-feature -amx-complex -target-feature -amx-fp16 -target-feature -amx-int8 -target-feature -amx-tile -target-feature +avx -target-feature -avx10.1-256 -target-feature -avx10.1-512 -target-feature +avx2 -target-feature -avx512bf16 -target-feature +avx512bitalg -target-feature +avx512bw -target-feature +avx512cd -target-feature +avx512dq -target-feature +avx512f -target-feature -avx512fp16 -target-feature +avx512ifma -target-feature +avx512vbmi -target-feature +avx512vbmi2 -target-feature +avx512vl -target-feature +avx512vnni -target-feature +avx512vp2intersect -target-feature +avx512vpopcntdq -target-feature -avxifma -target-feature -avxneconvert -target-feature -avxvnni -target-feature -avxvnniint16 -target-feature -avxvnniint8 -target-feature +bmi -target-feature +bmi2 -target-feature -branch-hint -target-feature -branchfusion -target-feature -ccmp -target-feature -cf -target-feature -cldemote -target-feature +clflushopt -target-feature +clwb -target-feature -clzero -target-feature +cmov -target-feature -cmpccxadd -target-feature -crc32 -target-feature +cx16 -target-feature +cx8 -target-feature -egpr -target-feature -enqcmd -target-feature -ermsb -target-feature -evex512 -target-feature +f16c -target-feature -false-deps-getmant -target-feature -false-deps-lzcnt-tzcnt -target-feature -false-deps-mulc -target-feature -false-deps-mullq -target-feature -false-deps-perm -target-feature -false-deps-popcnt -target-feature -false-deps-range -target-feature -fast-11bytenop -target-feature -fast-15bytenop -target-feature -fast-7bytenop -target-feature -fast-bextr -target-feature -fast-dpwssd -target-feature -fast-gather -target-feature -fast-hops -target-feature -fast-imm16 -target-feature -fast-lzcnt -target-feature -fast-movbe -target-feature -fast-scalar-fsqrt -target-feature -fast-scalar-shift-masks -target-feature -fast-shld-rotate -target-feature -fast-variable-crosslane-shuffle -target-feature -fast-variable-perlane-shuffle -target-feature -fast-vector-fsqrt -target-feature -fast-vector-shift-masks -target-feature -faster-shift-than-shuffle -target-feature +fma -target-feature -fma4 -target-feature +fsgsbase -target-feature -fsrm -target-feature +fxsr -target-feature +gfni -target-feature -harden-sls-ijmp -target-feature -harden-sls-ret -target-feature -hreset -target-feature -idivl-to-divb -target-feature +idivq-to-divl -target-feature -inline-asm-use-gpr32 -target-feature +invpcid -target-feature -kl -target-feature -lea-sp -target-feature -lea-uses-ag -target-feature -lvi-cfi -target-feature -lvi-load-hardening -target-feature -lwp -target-feature +lzcnt -target-feature +macrofusion -target-feature +mmx -target-feature +movbe -target-feature +movdir64b -target-feature +movdiri -target-feature -mwaitx -target-feature -ndd -target-feature -nf -target-feature -no-bypass-delay -target-feature -no-bypass-delay-blend -target-feature -no-bypass-delay-mov -target-feature -no-bypass-delay-shuffle -target-feature +nopl -target-feature -pad-short-functions -target-feature +pclmul -target-feature -pconfig -target-feature +pku -target-feature +popcnt -target-feature -ppx -target-feature -prefer-128-bit -target-feature -prefer-256-bit -target-feature -prefer-mask-registers -target-feature -prefer-movmsk-over-vtest -target-feature -prefer-no-gather -target-feature -prefer-no-scatter -target-feature -prefetchi -target-feature +prfchw -target-feature -ptwrite -target-feature -push2pop2 -target-feature -raoint -target-feature +rdpid -target-feature -rdpru -target-feature +rdrnd -target-feature +rdseed -target-feature -retpoline -target-feature -retpoline-external-thunk -target-feature -retpoline-indirect-branches -target-feature -retpoline-indirect-calls -target-feature -rtm -target-feature +sahf -target-feature -sbb-dep-breaking -target-feature -serialize -target-feature -seses -target-feature -sgx -target-feature +sha -target-feature -sha512 -target-feature +shstk -target-feature +slow-3ops-lea -target-feature +slow-incdec -target-feature -slow-lea -target-feature -slow-pmaddwd -target-feature -slow-pmulld -target-feature -slow-shld -target-feature -slow-two-mem-ops -target-feature -slow-unaligned-mem-16 -target-feature -slow-unaligned-mem-32 -target-feature -sm3 -target-feature -sm4 -target-feature +sse -target-feature +sse2 -target-feature +sse3 -target-feature +sse4.1 -target-feature +sse4.2 -target-feature -sse4a -target-feature -sse-unaligned-mem -target-feature +ssse3 -target-feature -tagged-globals -target-feature -tbm -target-feature -tsxldtrk -target-feature -tuning-fast-imm-vector-shift -target-feature -uintr -target-feature -use-glm-div-sqrt-costs -target-feature -use-slm-arith-costs -target-feature -usermsr -target-feature +vaes -target-feature +vpclmulqdq -target-feature +vzeroupper -target-feature -waitpkg -target-feature -wbnoinvd -target-feature -widekl -target-feature +x87 -target-feature -xop -target-feature +xsave -target-feature +xsavec -target-feature +xsaveopt -target-feature +xsaves -target-feature -zu -faddrsig -D__GCC_HAVE_DWARF2_CFI_ASM=1 -o /root/.cache/zig/tmp/69dcf0347f794377-demo.o -x c++ demo.cc
clang -cc1 version 19.1.0 based upon LLVM 19.1.0 default target x86_64-linux-musl
ignoring duplicate directory "/usr/include/x86_64-linux-gnu"
ignoring duplicate directory "/usr/include"
#include "..." search starts here:
#include <...> search starts here:
 /usr/local/zig/lib/libcxx/include
 /usr/local/zig/lib/libcxxabi/include
 /usr/local/zig/lib/libunwind/include
 /usr/local/zig/lib/include
 /usr/include
 /usr/include/x86_64-linux-gnu
 /usr/local/include
End of search list.

My CPU info (with avx512f support):

> cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 140
model name      : 11th Gen Intel(R) Core(TM) i3-1115G4 @ 3.00GHz
stepping        : 1
microcode       : 0xb8
cpu MHz         : 1860.770
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 0
cpu cores       : 2
apicid          : 0
initial apicid  : 0
fpu             : yes
fpu_exception   : yes
cpuid level     : 27
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 art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx 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 3dnowprefetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2intersect md_clear ibt flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple pml ept_mode_based_exec tsc_scaling
bugs            : spectre_v1 spectre_v2 spec_store_bypass swapgs eibrs_pbrsb gds bhi
bogomips        : 5990.40
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 1
vendor_id       : GenuineIntel
cpu family      : 6
model           : 140
model name      : 11th Gen Intel(R) Core(TM) i3-1115G4 @ 3.00GHz
stepping        : 1
microcode       : 0xb8
cpu MHz         : 400.000
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 1
cpu cores       : 2
apicid          : 2
initial apicid  : 2
fpu             : yes
fpu_exception   : yes
cpuid level     : 27
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 art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx 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 3dnowprefetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2intersect md_clear ibt flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple pml ept_mode_based_exec tsc_scaling
bugs            : spectre_v1 spectre_v2 spec_store_bypass swapgs eibrs_pbrsb gds bhi
bogomips        : 5990.40
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 2
vendor_id       : GenuineIntel
cpu family      : 6
model           : 140
model name      : 11th Gen Intel(R) Core(TM) i3-1115G4 @ 3.00GHz
stepping        : 1
microcode       : 0xb8
cpu MHz         : 1872.220
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 0
cpu cores       : 2
apicid          : 1
initial apicid  : 1
fpu             : yes
fpu_exception   : yes
cpuid level     : 27
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 art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx 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 3dnowprefetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2intersect md_clear ibt flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple pml ept_mode_based_exec tsc_scaling
bugs            : spectre_v1 spectre_v2 spec_store_bypass swapgs eibrs_pbrsb gds bhi
bogomips        : 5990.40
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 6
model           : 140
model name      : 11th Gen Intel(R) Core(TM) i3-1115G4 @ 3.00GHz
stepping        : 1
microcode       : 0xb8
cpu MHz         : 1888.402
cache size      : 6144 KB
physical id     : 0
siblings        : 4
core id         : 1
cpu cores       : 2
apicid          : 3
initial apicid  : 3
fpu             : yes
fpu_exception   : yes
cpuid level     : 27
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 art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx 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 3dnowprefetch cpuid_fault epb cat_l2 cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2intersect md_clear ibt flush_l1d arch_capabilities
vmx flags       : vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple pml ept_mode_based_exec tsc_scaling
bugs            : spectre_v1 spectre_v2 spec_store_bypass swapgs eibrs_pbrsb gds bhi
bogomips        : 5990.40
clflush size    : 64
cache_alignment : 64
address sizes   : 39 bits physical, 48 bits virtual
power management:

Expected Behavior

Use Zig as a drop-in C++ compiler.

[alexrp]

Zig uses Clang 19, so you should compare to that.

[dnomd343]

Hi alexrp, I tried clang-19 and it also seems to be normal.

I also tried clang-20 and it seems to work fine too.

[smups]

I've encountered this issue too when using Eigen. Manually setting the target cpu to one that supports AVX512 got rid of the compiler error for me (e.g. -Dcpu=x86_64_v4).

[smups]

I've encountered this issue too when using Eigen. Manually setting the target cpu to one that supports AVX512 got rid of the compiler error for me (e.g. -Dcpu=x86_64_v4).

Clarification: I am using a zig build script, passing it a -Dcpu=x86_64_v4 option worked. I have not tested passing -target-cpu=x86_64_v4 to zig c++ directly.

[dnomd343]

Thanks for @smups, I tried adding the schema specification and it seems to work now. (no need for x86_64_v4, just use x86_64)

# work fine
> zig c++ -march=x86_64 demo.cc -o demo -v

# the above issues
> zig c++ -march=native demo.cc -o demo -v

Comparing the compilation options, we can find that x86_64 turns off the avx512-related target-feature:

-target-feature -avx512bf16 -target-feature -avx512bitalg -target-feature -avx512bw -target-feature -avx512cd -target-feature -avx512dq -target-feature -avx512f -target-feature -avx512fp16 -target-feature -avx512ifma -target-feature -avx512vbmi -target-feature -avx512vbmi2 -target-feature -avx512vl -target-feature -avx512vnni -target-feature -avx512vp2intersect -target-feature -avx512vpopcntdq

While these are enabled when the native architecture is specified:

-target-feature -avx512bf16 -target-feature +avx512bitalg -target-feature +avx512bw -target-feature +avx512cd -target-feature +avx512dq -target-feature +avx512f -target-feature -avx512fp16 -target-feature +avx512ifma -target-feature +avx512vbmi -target-feature +avx512vbmi2 -target-feature +avx512vl -target-feature +avx512vnni -target-feature +avx512vp2intersect -target-feature +avx512vpopcntdq

[dnomd343]

I also tried similar options for clang:

# work fine
> clang++-19 -march=x86-64 demo.cc -o demo -v

# work fine
> clang++-19 -march=native demo.cc -o demo -v

When clang specifies x86_64, no specific target-feature field is generated. When native is specified, avx512-related features are similar to zig's native and are also enabled, but no compilation errors are generated.

[tmm1]

I'm having a similar problem trying to compile nodejs with zig c++:

  zig c++ -target x86_64-linux-gnu.2.17 -fno-sanitize=undefined -o /home/runner/_work/out/Release/obj.target/simdutf/deps/simdutf/simdutf.o ../deps/simdutf/simdutf.cpp '-D_GLIBCXX_USE_CXX11_ABI=1' '-D_FILE_OFFSET_BITS=64' '-DNODE_OPENSSL_CONF_NAME=nodejs_conf' '-DICU_NO_USER_DATA_OVERRIDE' '-D__STDC_FORMAT_MACROS' '-DOPENSSL_NO_PINSHARED' '-DOPENSSL_THREADS' -I../deps/simdutf  -pthread -Wall -Wextra -Wno-unused-parameter -m64 -O3 -fno-omit-frame-pointer -fno-rtti -fno-exceptions -fno-strict-aliasing -std=gnu++20 -MMD -MF /home/runner/_work/out/Release/.deps//home/runner/_work/out/Release/obj.target/simdutf/deps/simdutf/simdutf.o.d.raw   -c

../deps/simdutf/simdutf.cpp:30672:26: error: always_inline function '_mm512_set1_epi32' requires target feature 'evex512', but would be inlined into function 'detect_encodings' that is compiled without support for 'evex512'

../deps/simdutf/simdutf.cpp:30674:30: error: always_inline function '_mm512_setzero_si512' requires target feature 'evex512', but would be inlined into function 'detect_encodings' that is compiled without support for 'evex512'
 30674 |   __m512i currentoffsetmax = _mm512_setzero_si512();
       |                              ^

../deps/simdutf/simdutf.cpp:30674:30: error: AVX vector return of type '__m512i' (vector of 8 'long long' values) without 'evex512' enabled changes the ABI

../deps/simdutf/simdutf.cpp:30679:26: error: always_inline function '_mm512_loadu_si512' requires target feature 'evex512', but would be inlined into function 'detect_encodings' that is compiled without support for 'evex512'
 30679 |     const __m512i data = _mm512_loadu_si512((const __m512i *)ptr);
       |                          ^

There was some discussion here including a recommendation to use:

-Xclang -target-feature -Xclang +evex512

However that didn't immediately work for me.

[angt]

@tmm1, try with -mevex512 maybe.