Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Codegen tests fail on modern x86 CPUs due to v prefix #931

Open
DevJPM opened this issue Oct 17, 2020 · 3 comments
Open

Codegen tests fail on modern x86 CPUs due to v prefix #931

DevJPM opened this issue Oct 17, 2020 · 3 comments

Comments

@DevJPM
Copy link
Contributor

DevJPM commented Oct 17, 2020

I have access to an Intel Ice Lake CPU which has AVX-512 (and VAES / VPCLMULQDQ) support, it was this CPU that produced the below results.

When I follow the updated CONTRIBUTING.MD (#390) with -C target-cpu=native I get quite a few codegen tests for core_arch to fail. The reason seems to be that LLVM (or the disassembler) for some of these instructions doesn't report back their normal name, e.g. aesenc but rather their vectorized counter-part, e.g. vaesenc even though the right instruction was generated. This confuses the tool that checks the disassembly for the instruction and lets the test fail. I think this happens on all CPUs that support AVX-512 (i.e. starting from Skylake-X for Intel).

For completeness: I'm running on Arch Linux with rust nightly (cargo 1.49.0-nightly (79b397d72 2020-10-15)) and this invocation (on an Ice Lake CPU): RUSTFLAGS="-C target-cpu=native" TARGET=x86_64-unknown-linux-gnu cargo +nightly test --release -p core_arch

Here's a structural example:

---- core_arch::x86::sse::assert__mm_stream_ps_movntps stdout ----
disassembly for stdarch_test_shim__mm_stream_ps_movntps:
         0: lea 0x11d316(%rip),%rax # 20cd5d <anon.264e4d374f3812f9c2d5070410b589dc.255.llvm.10815413451089272223+0x7>
         1: lea 0x1e75c2(%rip),%rcx # 2d7010 <_ZN12stdarch_test11_DONT_DEDUP17h5804a9238811d034E>
         2: mov %rax,(%rcx)
         3: vmovntps %xmm0,(%rdi)
         4: retq
         5: nopw %cs:0x0(%rax,%rax,1)
         6:
thread 'core_arch::x86::sse::assert__mm_stream_ps_movntps' panicked at 'failed to find instruction `movntps` in the disassembly', crates/stdarch-test/src/lib.rs:144:9

And here's the list of tests that fail:

failures:
    core_arch::x86::aes::assert__mm_aesdec_si128_aesdec
    core_arch::x86::aes::assert__mm_aesdeclast_si128_aesdeclast
    core_arch::x86::aes::assert__mm_aesenc_si128_aesenc
    core_arch::x86::aes::assert__mm_aesenclast_si128_aesenclast
    core_arch::x86::aes::assert__mm_aesimc_si128_aesimc
    core_arch::x86::aes::assert__mm_aeskeygenassist_si128_aeskeygenassist
    core_arch::x86::avx512f::assert__kandn_mask16_not
    core_arch::x86::avx512f::assert__mm512_abs_pd_vpandq
    core_arch::x86::avx512f::assert__mm512_abs_ps_vpandq
    core_arch::x86::avx512f::assert__mm512_and_epi32_vpandq
    core_arch::x86::avx512f::assert__mm512_and_epi64_vpandq
    core_arch::x86::avx512f::assert__mm512_and_si512_vpandq
    core_arch::x86::avx512f::assert__mm512_andnot_epi32_vpandnq
    core_arch::x86::avx512f::assert__mm512_andnot_epi64_vpandnq
    core_arch::x86::avx512f::assert__mm512_andnot_si512_vpandnq
    core_arch::x86::avx512f::assert__mm512_kandn_not
    core_arch::x86::avx512f::assert__mm512_mask_abs_pd_vpandq
    core_arch::x86::avx512f::assert__mm512_mask_abs_ps_vpandd
    core_arch::x86::avx512f::assert__mm512_mask_and_epi32_vpandd
    core_arch::x86::avx512f::assert__mm512_mask_and_epi64_vpandq
    core_arch::x86::avx512f::assert__mm512_mask_andnot_epi32_vpandnd
    core_arch::x86::avx512f::assert__mm512_mask_andnot_epi64_vpandnq
    core_arch::x86::avx512f::assert__mm512_mask_or_epi32_vpord
    core_arch::x86::avx512f::assert__mm512_mask_or_epi64_vporq
    core_arch::x86::avx512f::assert__mm512_mask_xor_epi32_vpxord
    core_arch::x86::avx512f::assert__mm512_mask_xor_epi64_vpxorq
    core_arch::x86::avx512f::assert__mm512_maskz_and_epi32_vpandd
    core_arch::x86::avx512f::assert__mm512_maskz_and_epi64_vpandq
    core_arch::x86::avx512f::assert__mm512_maskz_andnot_epi32_vpandnd
    core_arch::x86::avx512f::assert__mm512_maskz_andnot_epi64_vpandnq
    core_arch::x86::avx512f::assert__mm512_maskz_or_epi32_vpord
    core_arch::x86::avx512f::assert__mm512_maskz_or_epi64_vporq
    core_arch::x86::avx512f::assert__mm512_maskz_xor_epi32_vpxord
    core_arch::x86::avx512f::assert__mm512_maskz_xor_epi64_vpxorq
    core_arch::x86::avx512f::assert__mm512_or_epi32_vporq
    core_arch::x86::avx512f::assert__mm512_or_epi64_vporq
    core_arch::x86::avx512f::assert__mm512_or_si512_vporq
    core_arch::x86::avx512f::assert__mm512_xor_epi32_vpxorq
    core_arch::x86::avx512f::assert__mm512_xor_epi64_vpxorq
    core_arch::x86::avx512f::assert__mm512_xor_si512_vpxorq
    core_arch::x86::avx::assert__mm256_permute2f128_ps_vperm2f128
    core_arch::x86::avx::assert__mm256_permute2f128_si256_vperm2f128
    core_arch::x86::avx::assert__mm256_set1_epi16_vinsertf128
    core_arch::x86::avx::assert__mm256_set1_epi64x_vinsertf128
    core_arch::x86::avx::assert__mm256_set1_epi8_vinsertf128
    core_arch::x86::avx::assert__mm256_set1_epi8_vpshufb
    core_arch::x86::pclmulqdq::assert__mm_clmulepi64_si128_pclmulhqhqdq
    core_arch::x86::pclmulqdq::assert__mm_clmulepi64_si128_pclmulhqlqdq
    core_arch::x86::pclmulqdq::assert__mm_clmulepi64_si128_pclmullqhqdq
    core_arch::x86::pclmulqdq::assert__mm_clmulepi64_si128_pclmullqlqdq
    core_arch::x86::sse2::assert__mm_add_epi16_paddw
    core_arch::x86::sse2::assert__mm_add_epi32_paddd
    core_arch::x86::sse2::assert__mm_add_epi64_paddq
    core_arch::x86::sse2::assert__mm_add_epi8_paddb
    core_arch::x86::sse2::assert__mm_add_pd_addpd
    core_arch::x86::sse2::assert__mm_add_sd_addsd
    core_arch::x86::sse2::assert__mm_adds_epi16_paddsw
    core_arch::x86::sse2::assert__mm_adds_epi8_paddsb
    core_arch::x86::sse2::assert__mm_adds_epu16_paddusw
    core_arch::x86::sse2::assert__mm_adds_epu8_paddusb
    core_arch::x86::sse2::assert__mm_and_pd_andps
    core_arch::x86::sse2::assert__mm_and_si128_andps
    core_arch::x86::sse2::assert__mm_andnot_pd_andnps
    core_arch::x86::sse2::assert__mm_andnot_si128_andnps
    core_arch::x86::sse2::assert__mm_avg_epu16_pavgw
    core_arch::x86::sse2::assert__mm_avg_epu8_pavgb
    core_arch::x86::sse2::assert__mm_bslli_si128_pslldq
    core_arch::x86::sse2::assert__mm_bsrli_si128_psrldq
    core_arch::x86::sse2::assert__mm_cmpeq_epi16_pcmpeqw
    core_arch::x86::sse2::assert__mm_cmpeq_epi32_pcmpeqd
    core_arch::x86::sse2::assert__mm_cmpeq_epi8_pcmpeqb
    core_arch::x86::sse2::assert__mm_cmpeq_pd_cmpeqpd
    core_arch::x86::sse2::assert__mm_cmpeq_sd_cmpeqsd
    core_arch::x86::sse2::assert__mm_cmpge_pd_cmplepd
    core_arch::x86::sse2::assert__mm_cmpge_sd_cmplesd
    core_arch::x86::sse2::assert__mm_cmpgt_epi16_pcmpgtw
    core_arch::x86::sse2::assert__mm_cmpgt_epi32_pcmpgtd
    core_arch::x86::sse2::assert__mm_cmpgt_epi8_pcmpgtb
    core_arch::x86::sse2::assert__mm_cmpgt_pd_cmpltpd
    core_arch::x86::sse2::assert__mm_cmpgt_sd_cmpltsd
    core_arch::x86::sse2::assert__mm_cmple_pd_cmplepd
    core_arch::x86::sse2::assert__mm_cmple_sd_cmplesd
    core_arch::x86::sse2::assert__mm_cmplt_epi16_pcmpgtw
    core_arch::x86::sse2::assert__mm_cmplt_epi32_pcmpgtd
    core_arch::x86::sse2::assert__mm_cmplt_epi8_pcmpgtb
    core_arch::x86::sse2::assert__mm_cmplt_pd_cmpltpd
    core_arch::x86::sse2::assert__mm_cmplt_sd_cmpltsd
    core_arch::x86::sse2::assert__mm_cmpneq_pd_cmpneqpd
    core_arch::x86::sse2::assert__mm_cmpneq_sd_cmpneqsd
    core_arch::x86::sse2::assert__mm_cmpnge_pd_cmpnlepd
    core_arch::x86::sse2::assert__mm_cmpnge_sd_cmpnlesd
    core_arch::x86::sse2::assert__mm_cmpngt_pd_cmpnltpd
    core_arch::x86::sse2::assert__mm_cmpngt_sd_cmpnltsd
    core_arch::x86::sse2::assert__mm_cmpnle_pd_cmpnlepd
    core_arch::x86::sse2::assert__mm_cmpnle_sd_cmpnlesd
    core_arch::x86::sse2::assert__mm_cmpnlt_pd_cmpnltpd
    core_arch::x86::sse2::assert__mm_cmpnlt_sd_cmpnltsd
    core_arch::x86::sse2::assert__mm_cmpord_pd_cmpordpd
    core_arch::x86::sse2::assert__mm_cmpord_sd_cmpordsd
    core_arch::x86::sse2::assert__mm_cmpunord_pd_cmpunordpd
    core_arch::x86::sse2::assert__mm_cmpunord_sd_cmpunordsd
    core_arch::x86::sse2::assert__mm_comieq_sd_comisd
    core_arch::x86::sse2::assert__mm_comige_sd_comisd
    core_arch::x86::sse2::assert__mm_comigt_sd_comisd
    core_arch::x86::sse2::assert__mm_comile_sd_comisd
    core_arch::x86::sse2::assert__mm_comilt_sd_comisd
    core_arch::x86::sse2::assert__mm_comineq_sd_comisd
    core_arch::x86::sse2::assert__mm_cvtepi32_pd_cvtdq2pd
    core_arch::x86::sse2::assert__mm_cvtepi32_ps_cvtdq2ps
    core_arch::x86::sse2::assert__mm_cvtpd_epi32_cvtpd2dq
    core_arch::x86::sse2::assert__mm_cvtpd_ps_cvtpd2ps
    core_arch::x86::sse2::assert__mm_cvtps_epi32_cvtps2dq
    core_arch::x86::sse2::assert__mm_cvtps_pd_cvtps2pd
    core_arch::x86::sse2::assert__mm_cvtsd_si32_cvtsd2si
    core_arch::x86::sse2::assert__mm_cvtsd_ss_cvtsd2ss
    core_arch::x86::sse2::assert__mm_cvtsi128_si32_movd
    core_arch::x86::sse2::assert__mm_cvtsi32_sd_cvtsi2sd
    core_arch::x86::sse2::assert__mm_cvtsi32_si128_movd
    core_arch::x86::sse2::assert__mm_cvtss_sd_cvtss2sd
    core_arch::x86::sse2::assert__mm_cvttpd_epi32_cvttpd2dq
    core_arch::x86::sse2::assert__mm_cvttps_epi32_cvttps2dq
    core_arch::x86::sse2::assert__mm_cvttsd_si32_cvttsd2si
    core_arch::x86::sse2::assert__mm_div_pd_divpd
    core_arch::x86::sse2::assert__mm_div_sd_divsd
    core_arch::x86::sse2::assert__mm_extract_epi16_pextrw
    core_arch::x86::sse2::assert__mm_insert_epi16_pinsrw
    core_arch::x86::sse2::assert__mm_load_pd_movaps
    core_arch::x86::sse2::assert__mm_load_sd_movsd
    core_arch::x86::sse2::assert__mm_load_si128_movaps
    core_arch::x86::sse2::assert__mm_loadh_pd_movhps
    core_arch::x86::sse2::assert__mm_loadl_pd_movlps
    core_arch::x86::sse2::assert__mm_loadr_pd_movaps
    core_arch::x86::sse2::assert__mm_loadu_pd_movups
    core_arch::x86::sse2::assert__mm_loadu_si128_movups
    core_arch::x86::sse2::assert__mm_madd_epi16_pmaddwd
    core_arch::x86::sse2::assert__mm_maskmoveu_si128_maskmovdqu
    core_arch::x86::sse2::assert__mm_max_epi16_pmaxsw
    core_arch::x86::sse2::assert__mm_max_epu8_pmaxub
    core_arch::x86::sse2::assert__mm_max_pd_maxpd
    core_arch::x86::sse2::assert__mm_max_sd_maxsd
    core_arch::x86::sse2::assert__mm_min_epi16_pminsw
    core_arch::x86::sse2::assert__mm_min_epu8_pminub
    core_arch::x86::sse2::assert__mm_min_pd_minpd
    core_arch::x86::sse2::assert__mm_min_sd_minsd
    core_arch::x86::sse2::assert__mm_move_epi64_movq
    core_arch::x86::sse2::assert__mm_move_sd_movsd
    core_arch::x86::sse2::assert__mm_movemask_epi8_pmovmskb
    core_arch::x86::sse2::assert__mm_movemask_pd_movmskpd
    core_arch::x86::sse2::assert__mm_mul_epu32_pmuludq
    core_arch::x86::sse2::assert__mm_mul_pd_mulpd
    core_arch::x86::sse2::assert__mm_mul_sd_mulsd
    core_arch::x86::sse2::assert__mm_mulhi_epi16_pmulhw
    core_arch::x86::sse2::assert__mm_mulhi_epu16_pmulhuw
    core_arch::x86::sse2::assert__mm_mullo_epi16_pmullw
    core_arch::x86::sse2::assert__mm_or_pd_orps
    core_arch::x86::sse2::assert__mm_or_si128_orps
    core_arch::x86::sse2::assert__mm_packs_epi16_packsswb
    core_arch::x86::sse2::assert__mm_packs_epi32_packssdw
    core_arch::x86::sse2::assert__mm_packus_epi16_packuswb
    core_arch::x86::sse2::assert__mm_sad_epu8_psadbw
    core_arch::x86::sse2::assert__mm_setzero_pd_xorps
    core_arch::x86::sse2::assert__mm_setzero_si128_xorps
    core_arch::x86::sse2::assert__mm_shuffle_epi32_pshufd
    core_arch::x86::sse2::assert__mm_shuffle_pd_shufps
    core_arch::x86::sse2::assert__mm_shufflehi_epi16_pshufhw
    core_arch::x86::sse2::assert__mm_shufflelo_epi16_pshuflw
    core_arch::x86::sse2::assert__mm_sll_epi16_psllw
    core_arch::x86::sse2::assert__mm_sll_epi32_pslld
    core_arch::x86::sse2::assert__mm_sll_epi64_psllq
    core_arch::x86::sse2::assert__mm_slli_epi16_psllw
    core_arch::x86::sse2::assert__mm_slli_epi32_pslld
    core_arch::x86::sse2::assert__mm_slli_epi64_psllq
    core_arch::x86::sse2::assert__mm_slli_si128_pslldq
    core_arch::x86::sse2::assert__mm_sqrt_pd_sqrtpd
    core_arch::x86::sse2::assert__mm_sqrt_sd_sqrtsd
    core_arch::x86::sse2::assert__mm_sra_epi16_psraw
    core_arch::x86::sse2::assert__mm_sra_epi32_psrad
    core_arch::x86::sse2::assert__mm_srai_epi16_psraw
    core_arch::x86::sse2::assert__mm_srai_epi32_psrad
    core_arch::x86::sse2::assert__mm_srl_epi16_psrlw
    core_arch::x86::sse2::assert__mm_srl_epi32_psrld
    core_arch::x86::sse2::assert__mm_srl_epi64_psrlq
    core_arch::x86::sse2::assert__mm_srli_epi16_psrlw
    core_arch::x86::sse2::assert__mm_srli_epi32_psrld
    core_arch::x86::sse2::assert__mm_srli_epi64_psrlq
    core_arch::x86::sse2::assert__mm_srli_si128_psrldq
    core_arch::x86::sse2::assert__mm_store_pd_movaps
    core_arch::x86::sse2::assert__mm_store_sd_movlps
    core_arch::x86::sse2::assert__mm_store_si128_movaps
    core_arch::x86::sse2::assert__mm_storeh_pd_movhps
    core_arch::x86::sse2::assert__mm_storel_pd_movlps
    core_arch::x86::sse2::assert__mm_storeu_pd_movups
    core_arch::x86::sse2::assert__mm_storeu_si128_movups
    core_arch::x86::sse2::assert__mm_stream_pd_movntps
    core_arch::x86::sse2::assert__mm_stream_si128_movntps
    core_arch::x86::sse2::assert__mm_sub_epi16_psubw
    core_arch::x86::sse2::assert__mm_sub_epi32_psubd
    core_arch::x86::sse2::assert__mm_sub_epi64_psubq
    core_arch::x86::sse2::assert__mm_sub_epi8_psubb
    core_arch::x86::sse2::assert__mm_sub_pd_subpd
    core_arch::x86::sse2::assert__mm_sub_sd_subsd
    core_arch::x86::sse2::assert__mm_subs_epi16_psubsw
    core_arch::x86::sse2::assert__mm_subs_epi8_psubsb
    core_arch::x86::sse2::assert__mm_subs_epu16_psubusw
    core_arch::x86::sse2::assert__mm_subs_epu8_psubusb
    core_arch::x86::sse2::assert__mm_ucomieq_sd_ucomisd
    core_arch::x86::sse2::assert__mm_ucomige_sd_ucomisd
    core_arch::x86::sse2::assert__mm_ucomigt_sd_ucomisd
    core_arch::x86::sse2::assert__mm_ucomile_sd_ucomisd
    core_arch::x86::sse2::assert__mm_ucomilt_sd_ucomisd
    core_arch::x86::sse2::assert__mm_ucomineq_sd_ucomisd
    core_arch::x86::sse2::assert__mm_unpackhi_epi16_punpckhwd
    core_arch::x86::sse2::assert__mm_unpackhi_epi32_unpckhps
    core_arch::x86::sse2::assert__mm_unpackhi_epi64_unpckhpd
    core_arch::x86::sse2::assert__mm_unpackhi_epi8_punpckhbw
    core_arch::x86::sse2::assert__mm_unpackhi_pd_unpckhpd
    core_arch::x86::sse2::assert__mm_unpacklo_epi16_punpcklwd
    core_arch::x86::sse2::assert__mm_unpacklo_epi32_unpcklps
    core_arch::x86::sse2::assert__mm_unpacklo_epi64_movlhps
    core_arch::x86::sse2::assert__mm_unpacklo_epi8_punpcklbw
    core_arch::x86::sse2::assert__mm_unpacklo_pd_movlhps
    core_arch::x86::sse2::assert__mm_xor_pd_xorps
    core_arch::x86::sse2::assert__mm_xor_si128_xorps
    core_arch::x86::sse3::assert__mm_addsub_pd_addsubpd
    core_arch::x86::sse3::assert__mm_addsub_ps_addsubps
    core_arch::x86::sse3::assert__mm_hadd_pd_haddpd
    core_arch::x86::sse3::assert__mm_hadd_ps_haddps
    core_arch::x86::sse3::assert__mm_hsub_pd_hsubpd
    core_arch::x86::sse3::assert__mm_hsub_ps_hsubps
    core_arch::x86::sse3::assert__mm_lddqu_si128_lddqu
    core_arch::x86::sse3::assert__mm_loaddup_pd_movddup
    core_arch::x86::sse3::assert__mm_movedup_pd_movddup
    core_arch::x86::sse3::assert__mm_movehdup_ps_movshdup
    core_arch::x86::sse3::assert__mm_moveldup_ps_movsldup
    core_arch::x86::sse41::assert__mm_blend_epi16_blendps
    core_arch::x86::sse41::assert__mm_blend_pd_blendps
    core_arch::x86::sse41::assert__mm_blend_ps_blendps
    core_arch::x86::sse41::assert__mm_blendv_epi8_pblendvb
    core_arch::x86::sse41::assert__mm_blendv_pd_blendvpd
    core_arch::x86::sse41::assert__mm_blendv_ps_blendvps
    core_arch::x86::sse41::assert__mm_ceil_pd_roundpd
    core_arch::x86::sse41::assert__mm_ceil_ps_roundps
    core_arch::x86::sse41::assert__mm_ceil_sd_roundsd
    core_arch::x86::sse41::assert__mm_ceil_ss_roundss
    core_arch::x86::sse41::assert__mm_cmpeq_epi64_pcmpeqq
    core_arch::x86::sse41::assert__mm_cvtepi16_epi32_pmovsxwd
    core_arch::x86::sse41::assert__mm_cvtepi16_epi64_pmovsxwq
    core_arch::x86::sse41::assert__mm_cvtepi32_epi64_pmovsxdq
    core_arch::x86::sse41::assert__mm_cvtepi8_epi16_pmovsxbw
    core_arch::x86::sse41::assert__mm_cvtepi8_epi32_pmovsxbd
    core_arch::x86::sse41::assert__mm_cvtepi8_epi64_pmovsxbq
    core_arch::x86::sse41::assert__mm_cvtepu16_epi32_pmovzxwd
    core_arch::x86::sse41::assert__mm_cvtepu16_epi64_pmovzxwq
    core_arch::x86::sse41::assert__mm_cvtepu32_epi64_pmovzxdq
    core_arch::x86::sse41::assert__mm_cvtepu8_epi16_pmovzxbw
    core_arch::x86::sse41::assert__mm_cvtepu8_epi32_pmovzxbd
    core_arch::x86::sse41::assert__mm_cvtepu8_epi64_pmovzxbq
    core_arch::x86::sse41::assert__mm_dp_pd_dppd
    core_arch::x86::sse41::assert__mm_dp_ps_dpps
    core_arch::x86::sse41::assert__mm_extract_epi32_extractps
    core_arch::x86::sse41::assert__mm_extract_epi8_pextrb
    core_arch::x86::sse41::assert__mm_extract_ps_extractps
    core_arch::x86::sse41::assert__mm_floor_pd_roundpd
    core_arch::x86::sse41::assert__mm_floor_ps_roundps
    core_arch::x86::sse41::assert__mm_floor_sd_roundsd
    core_arch::x86::sse41::assert__mm_floor_ss_roundss
    core_arch::x86::sse41::assert__mm_insert_epi32_pinsrd
    core_arch::x86::sse41::assert__mm_insert_epi8_pinsrb
    core_arch::x86::sse41::assert__mm_insert_ps_insertps
    core_arch::x86::sse41::assert__mm_max_epi32_pmaxsd
    core_arch::x86::sse41::assert__mm_max_epi8_pmaxsb
    core_arch::x86::sse41::assert__mm_max_epu16_pmaxuw
    core_arch::x86::sse41::assert__mm_max_epu32_pmaxud
    core_arch::x86::sse41::assert__mm_min_epi32_pminsd
    core_arch::x86::sse41::assert__mm_min_epi8_pminsb
    core_arch::x86::sse41::assert__mm_min_epu16_pminuw
    core_arch::x86::sse41::assert__mm_min_epu32_pminud
    core_arch::x86::sse41::assert__mm_minpos_epu16_phminposuw
    core_arch::x86::sse41::assert__mm_mpsadbw_epu8_mpsadbw
    core_arch::x86::sse41::assert__mm_mul_epi32_pmuldq
    core_arch::x86::sse41::assert__mm_mullo_epi32_pmulld
    core_arch::x86::sse41::assert__mm_packus_epi32_packusdw
    core_arch::x86::sse41::assert__mm_round_pd_roundpd
    core_arch::x86::sse41::assert__mm_round_ps_roundps
    core_arch::x86::sse41::assert__mm_round_sd_roundsd
    core_arch::x86::sse41::assert__mm_round_ss_roundss
    core_arch::x86::sse41::assert__mm_test_all_ones_pcmpeqd
    core_arch::x86::sse41::assert__mm_test_all_ones_ptest
    core_arch::x86::sse41::assert__mm_test_all_zeros_ptest
    core_arch::x86::sse41::assert__mm_test_mix_ones_zeros_ptest
    core_arch::x86::sse41::assert__mm_testc_si128_ptest
    core_arch::x86::sse41::assert__mm_testnzc_si128_ptest
    core_arch::x86::sse41::assert__mm_testz_si128_ptest
    core_arch::x86::sse42::assert__mm_cmpestra_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpestrc_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpestri_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpestrm_pcmpestrm
    core_arch::x86::sse42::assert__mm_cmpestro_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpestrs_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpestrz_pcmpestri
    core_arch::x86::sse42::assert__mm_cmpgt_epi64_pcmpgtq
    core_arch::x86::sse42::assert__mm_cmpistra_pcmpistri
    core_arch::x86::sse42::assert__mm_cmpistrc_pcmpistri
    core_arch::x86::sse42::assert__mm_cmpistri_pcmpistri
    core_arch::x86::sse42::assert__mm_cmpistrm_pcmpistrm
    core_arch::x86::sse42::assert__mm_cmpistro_pcmpistri
    core_arch::x86::sse42::assert__mm_cmpistrs_pcmpistri
    core_arch::x86::sse42::assert__mm_cmpistrz_pcmpistri
    core_arch::x86::sse::assert__mm_add_ps_addps
    core_arch::x86::sse::assert__mm_add_ss_addss
    core_arch::x86::sse::assert__mm_and_ps_andps
    core_arch::x86::sse::assert__mm_andnot_ps_andnps
    core_arch::x86::sse::assert__mm_cmpeq_ps_cmpeqps
    core_arch::x86::sse::assert__mm_cmpeq_ss_cmpeqss
    core_arch::x86::sse::assert__mm_cmpge_ps_cmpleps
    core_arch::x86::sse::assert__mm_cmpge_ss_cmpless
    core_arch::x86::sse::assert__mm_cmpgt_ps_cmpltps
    core_arch::x86::sse::assert__mm_cmpgt_ss_cmpltss
    core_arch::x86::sse::assert__mm_cmple_ps_cmpleps
    core_arch::x86::sse::assert__mm_cmple_ss_cmpless
    core_arch::x86::sse::assert__mm_cmplt_ps_cmpltps
    core_arch::x86::sse::assert__mm_cmplt_ss_cmpltss
    core_arch::x86::sse::assert__mm_cmpneq_ps_cmpneqps
    core_arch::x86::sse::assert__mm_cmpneq_ss_cmpneqss
    core_arch::x86::sse::assert__mm_cmpnge_ps_cmpnleps
    core_arch::x86::sse::assert__mm_cmpnge_ss_cmpnless
    core_arch::x86::sse::assert__mm_cmpngt_ps_cmpnltps
    core_arch::x86::sse::assert__mm_cmpngt_ss_cmpnltss
    core_arch::x86::sse::assert__mm_cmpnle_ps_cmpnleps
    core_arch::x86::sse::assert__mm_cmpnle_ss_cmpnless
    core_arch::x86::sse::assert__mm_cmpnlt_ps_cmpnltps
    core_arch::x86::sse::assert__mm_cmpnlt_ss_cmpnltss
    core_arch::x86::sse::assert__mm_cmpord_ps_cmpordps
    core_arch::x86::sse::assert__mm_cmpord_ss_cmpordss
    core_arch::x86::sse::assert__mm_cmpunord_ps_cmpunordps
    core_arch::x86::sse::assert__mm_cmpunord_ss_cmpunordss
    core_arch::x86::sse::assert__mm_comieq_ss_comiss
    core_arch::x86::sse::assert__mm_comige_ss_comiss
    core_arch::x86::sse::assert__mm_comigt_ss_comiss
    core_arch::x86::sse::assert__mm_comile_ss_comiss
    core_arch::x86::sse::assert__mm_comilt_ss_comiss
    core_arch::x86::sse::assert__mm_comineq_ss_comiss
    core_arch::x86::sse::assert__mm_cvt_si2ss_cvtsi2ss
    core_arch::x86::sse::assert__mm_cvt_ss2si_cvtss2si
    core_arch::x86::sse::assert__mm_cvtsi32_ss_cvtsi2ss
    core_arch::x86::sse::assert__mm_cvtss_si32_cvtss2si
    core_arch::x86::sse::assert__mm_cvtt_ss2si_cvttss2si
    core_arch::x86::sse::assert__mm_cvttss_si32_cvttss2si
    core_arch::x86::sse::assert__mm_div_ps_divps
    core_arch::x86::sse::assert__mm_div_ss_divss
    core_arch::x86::sse::assert__mm_getcsr_stmxcsr
    core_arch::x86::sse::assert__mm_load1_ps_movss
    core_arch::x86::sse::assert__mm_load_ps1_movss
    core_arch::x86::sse::assert__mm_load_ps_movaps
    core_arch::x86::sse::assert__mm_load_ss_movss
    core_arch::x86::sse::assert__mm_loadr_ps_movaps
    core_arch::x86::sse::assert__mm_loadu_ps_movups
    core_arch::x86::sse::assert__mm_loadu_si64_movq
    core_arch::x86::sse::assert__mm_max_ps_maxps
    core_arch::x86::sse::assert__mm_max_ss_maxss
    core_arch::x86::sse::assert__mm_min_ps_minps
    core_arch::x86::sse::assert__mm_min_ss_minss
    core_arch::x86::sse::assert__mm_move_ss_movss
    core_arch::x86::sse::assert__mm_movehl_ps_movhlps
    core_arch::x86::sse::assert__mm_movelh_ps_movlhps
    core_arch::x86::sse::assert__mm_movemask_ps_movmskps
    core_arch::x86::sse::assert__mm_mul_ps_mulps
    core_arch::x86::sse::assert__mm_mul_ss_mulss
    core_arch::x86::sse::assert__mm_or_ps_orps
    core_arch::x86::sse::assert__mm_rcp_ps_rcpps
    core_arch::x86::sse::assert__mm_rcp_ss_rcpss
    core_arch::x86::sse::assert__mm_rsqrt_ps_rsqrtps
    core_arch::x86::sse::assert__mm_rsqrt_ss_rsqrtss
    core_arch::x86::sse::assert__mm_set1_ps_shufps
    core_arch::x86::sse::assert__mm_set_ps1_shufps
    core_arch::x86::sse::assert__mm_set_ps_unpcklps
    core_arch::x86::sse::assert__mm_set_ss_movss
    core_arch::x86::sse::assert__mm_setcsr_ldmxcsr
    core_arch::x86::sse::assert__mm_setr_ps_unpcklps
    core_arch::x86::sse::assert__mm_setzero_ps_xorps
    core_arch::x86::sse::assert__mm_shuffle_ps_shufps
    core_arch::x86::sse::assert__mm_sqrt_ps_sqrtps
    core_arch::x86::sse::assert__mm_sqrt_ss_sqrtss
    core_arch::x86::sse::assert__mm_store1_ps_movaps
    core_arch::x86::sse::assert__mm_store_ps1_movaps
    core_arch::x86::sse::assert__mm_store_ps_movaps
    core_arch::x86::sse::assert__mm_store_ss_movss
    core_arch::x86::sse::assert__mm_storer_ps_movaps
    core_arch::x86::sse::assert__mm_storeu_ps_movups
    core_arch::x86::sse::assert__mm_stream_ps_movntps
    core_arch::x86::sse::assert__mm_sub_ps_subps
    core_arch::x86::sse::assert__mm_sub_ss_subss
    core_arch::x86::sse::assert__mm_ucomieq_ss_ucomiss
    core_arch::x86::sse::assert__mm_ucomige_ss_ucomiss
    core_arch::x86::sse::assert__mm_ucomigt_ss_ucomiss
    core_arch::x86::sse::assert__mm_ucomile_ss_ucomiss
    core_arch::x86::sse::assert__mm_ucomilt_ss_ucomiss
    core_arch::x86::sse::assert__mm_ucomineq_ss_ucomiss
    core_arch::x86::sse::assert__mm_unpackhi_ps_unpckhps
    core_arch::x86::sse::assert__mm_unpacklo_ps_unpcklps
    core_arch::x86::sse::assert__mm_xor_ps_xorps
    core_arch::x86::ssse3::assert__mm_abs_epi16_pabsw
    core_arch::x86::ssse3::assert__mm_abs_epi32_pabsd
    core_arch::x86::ssse3::assert__mm_abs_epi8_pabsb
    core_arch::x86::ssse3::assert__mm_alignr_epi8_palignr
    core_arch::x86::ssse3::assert__mm_hadd_epi16_phaddw
    core_arch::x86::ssse3::assert__mm_hadd_epi32_phaddd
    core_arch::x86::ssse3::assert__mm_hadds_epi16_phaddsw
    core_arch::x86::ssse3::assert__mm_hsub_epi16_phsubw
    core_arch::x86::ssse3::assert__mm_hsub_epi32_phsubd
    core_arch::x86::ssse3::assert__mm_hsubs_epi16_phsubsw
    core_arch::x86::ssse3::assert__mm_maddubs_epi16_pmaddubsw
    core_arch::x86::ssse3::assert__mm_mulhrs_epi16_pmulhrsw
    core_arch::x86::ssse3::assert__mm_shuffle_epi8_pshufb
    core_arch::x86::ssse3::assert__mm_sign_epi16_psignw
    core_arch::x86::ssse3::assert__mm_sign_epi32_psignd
    core_arch::x86::ssse3::assert__mm_sign_epi8_psignb
    core_arch::x86_64::sse2::assert__mm_cvtsd_si64_cvtsd2si
    core_arch::x86_64::sse2::assert__mm_cvtsd_si64x_cvtsd2si
    core_arch::x86_64::sse2::assert__mm_cvtsi128_si64_movq
    core_arch::x86_64::sse2::assert__mm_cvtsi128_si64x_movq
    core_arch::x86_64::sse2::assert__mm_cvtsi64_sd_cvtsi2sd
    core_arch::x86_64::sse2::assert__mm_cvtsi64_si128_movq
    core_arch::x86_64::sse2::assert__mm_cvtsi64x_sd_cvtsi2sd
    core_arch::x86_64::sse2::assert__mm_cvtsi64x_si128_movq
    core_arch::x86_64::sse2::assert__mm_cvttsd_si64_cvttsd2si
    core_arch::x86_64::sse2::assert__mm_cvttsd_si64x_cvttsd2si
    core_arch::x86_64::sse41::assert__mm_extract_epi64_pextrq
    core_arch::x86_64::sse41::assert__mm_insert_epi64_pinsrq
    core_arch::x86_64::sse::assert__mm_cvtsi64_ss_cvtsi2ss
    core_arch::x86_64::sse::assert__mm_cvtss_si64_cvtss2si
    core_arch::x86_64::sse::assert__mm_cvttss_si64_cvttss2si

test result: FAILED. 3685 passed; 432 failed; 0 ignored; 0 measured; 0 filtered out
@Amanieu
Copy link
Member

Amanieu commented Oct 18, 2020

For the intrinsics we don't actually make any guarantees on what instructions the compiler will generate, so the tests are actually more picky than they should be. In general you shouldn't be using any -C target-cpu flags when running the tests except for ARM which needs -C target-feature=+neon.

DevJPM added a commit to DevJPM/stdarch that referenced this issue Oct 25, 2020
@DevJPM
Added a reference to issue rust-lang#931
50bfa30 
Added a reference to issue 931 to explain why the manual codegen tests may fail
@DevJPM
Copy link
Contributor Author

DevJPM commented Oct 25, 2020

How would one go about fixing this, i.e. making the assertion accept a v prefix in specific cases?

@bjorn3
Copy link
Member

bjorn3 commented Oct 25, 2020

You should not pass -C target-cpu=native when running cargo test.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@Amanieu @DevJPM @bjorn3