Commit f2cc681
deps: enable AVX-512 OpenSSL asm with clang
Node.js ships two pre-generated sets of OpenSSL assembly: `asm`, which
contains the AVX-512 routines, and `asm_avx2`, which does not. The
set is picked in deps/openssl/openssl.gyp based on `gas_version` or
`nasm_version`, but configure.py only reports `gas_version` when the
compiler is not clang, because clang uses its own integrated
assembler and has no GNU assembler version to report. Consequently
every clang build silently falls back to the AVX-512-less `asm_avx2`
set, with no warning.
The result is that `ossl_vaes_vpclmulqdq_capable()` is assembled as a
stub that always returns 0, so OpenSSL never selects
`ossl_aes_gcm_encrypt_avx512()` and uses the older AES-NI path
instead. On an Intel Xeon Gold 6548N this costs roughly 1.6x on
AES-256-GCM and 1.8x on both ChaCha20-Poly1305 and RSA-2048 signing.
This is not limited to custom builds: BUILDING.md documents that the
official linux-x64 binaries are produced with clang, and the shipped
v25.x and v26.x binaries contain the stub.
Accept `llvm_version` in the condition, the way
deps/openssl/openssl.gypi already does for the AVX2 set. clang's
integrated assembler has handled AVX512IFMA since 3.9 and VAES /
VPCLMULQDQ since 6.0; 8.0 is used as a conservative floor, well below
the clang 19.1 that Node.js already requires.
PR-URL: #65136
Reviewed-By: Aviv Keller <me@aviv.sh>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>1 parent 3cfe194 commit f2cc681
1 file changed
Lines changed: 4 additions & 2 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
@@ -36,7 +36,8 @@ | |||
| 36 | 36 | # VC-WIN64-ARM inherits from VC-noCE-common that has no asms. | |
| 37 | 37 | 'includes': ['./openssl_no_asm.gypi'], | |
| 38 | 38 | }, 'gas_version and v(gas_version) >= v("2.26") or ' | |
| 39 | - 'nasm_version and v(nasm_version) >= v("2.11.8")', { | ||
| 39 | + 'nasm_version and v(nasm_version) >= v("2.11.8") or ' | ||
| 40 | + 'llvm_version and v(llvm_version) >= v("8.0")', { | ||
| 40 | 41 | # Require AVX512IFMA supported. See | |
| 41 | 42 | # https://www.openssl.org/docs/man1.1.1/man3/OPENSSL_ia32cap.html | |
| 42 | 43 | # Currently crypto/poly1305/asm/poly1305-x86_64.pl requires AVX512IFMA. | |
@@ -114,7 +115,8 @@ | |||
| 114 | 115 | # VC-WIN64-ARM inherits from VC-noCE-common that has no asms. | |
| 115 | 116 | 'includes': ['./openssl-fips_no_asm.gypi'], | |
| 116 | 117 | }, 'gas_version and v(gas_version) >= v("2.26") or ' | |
| 117 | - 'nasm_version and v(nasm_version) >= v("2.11.8")', { | ||
| 118 | + 'nasm_version and v(nasm_version) >= v("2.11.8") or ' | ||
| 119 | + 'llvm_version and v(llvm_version) >= v("8.0")', { | ||
| 118 | 120 | # Require AVX512IFMA supported. See | |
| 119 | 121 | # https://www.openssl.org/docs/man1.1.1/man3/OPENSSL_ia32cap.html | |
| 120 | 122 | # Currently crypto/poly1305/asm/poly1305-x86_64.pl requires AVX512IFMA. | |
0 commit comments