Re: AVX-512 for LAME: code to try, if you have the hardware

Il giorno Sun, 26 Jul 2026 15:24:42 +0200
Alexander Leidinger via Lame-dev <[email protected]> ha
scritto:

> In case someone has an Intel CPU with AVX-512, it would be
> interesting to see the numbers there, it seems to have a 512bit
> datapath.

Hello all,

here an almost-complete run on and Intel Core i5:

$ gcc --version
gcc (Debian 14.2.0-19) 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.

$ uname -a
Linux faye 6.1.0-41-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.158-1
(2025-11-09) x86_64 GNU/Linux

$ cat /proc/cpuinfo 
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 126
model name	: Intel(R) Core(TM) i5-1035G1 CPU @ 1.00GHz
stepping	: 5
microcode	: 0xcc
cpu MHz		: 400.000
cache size	: 6144 KB
physical id	: 0
siblings	: 8
core id		: 0
cpu cores	: 4
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
invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi
flexpriority ept vpid ept_ad fsgsbase tsc_adjust sgx bmi1 avx2 smep
bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma
clflushopt intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec
xgetbv1 xsaves split_lock_detect dtherm ida arat pln pts hwp hwp_notify
hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke
avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg
avx512_vpopcntdq rdpid sgx_lc fsrm md_clear 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 itlb_multihit srbds mmio_stale_data retbleed
eibrs_pbrsb gds bhi its its_native_only bogomips	: 2380.80
clflush size	: 64 cache_alignment	: 64 address
sizes	: 39 bits physical, 48 bits virtual power management:
[cut]

Stage 1

heading messages during encode:

$ ./frontend/lame -V2 audio_10m.wav a.mp3
LAME 4.1 (alpha 0, Jul 26 2026 14:42:26) 64bits
(https://lame.sourceforge.io) warning: alpha versions should be used
for testing only CPU features: MMX, SSE, SSE2, AVX2, AVX-512
vector routines: AVX-512                <----- AVX-512 !!!
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding audio_10m.wav to a.mp3
[cut]

then

[...]
============================================================================
Testsuite summary for lame 4.1
============================================================================
# TOTAL: 13
# PASS:  13
# SKIP:  0
# XFAIL: 0
# FAIL:  0
# XPASS: 0
# ERROR: 0
============================================================================

-V2 & -V2 --noasm avx512
Result: IDENTICAL

-b 320 & -b 320 --noasm avx512
Result: IDENTICAL

-V2

real    0m7,516s
user    0m7,490s
sys     0m0,029s

real    0m7,511s
user    0m7,470s
sys     0m0,045s

real    0m7,618s
user    0m7,588s
sys     0m0,031s

-V2 --noasm avx512

real    0m7,551s
user    0m7,512s
sys     0m0,042s

real    0m7,696s
user    0m7,665s
sys     0m0,036s

real    0m7,547s
user    0m7,520s
sys     0m0,032s

-b 320

real    0m11,793s
user    0m11,773s
sys     0m0,023s

real    0m11,679s
user    0m11,645s
sys     0m0,037s

real    0m11,726s
user    0m11,693s
sys     0m0,036s

-b 320 --noasm avx512

real    0m13,003s
user    0m12,972s
sys     0m0,032s

real    0m12,416s
user    0m12,381s
sys     0m0,039s

real    0m12,672s
user    0m12,643s
sys     0m0,033s

STAGE 2

[...]
PASS: test_quantize_lines_vector
FAIL: test_calc_sfb_noise_vector
============================================================================
Testsuite summary for lame 4.1
============================================================================
# TOTAL: 13
# PASS:  12
# SKIP:  0
# XFAIL: 0
# FAIL:  1
# XPASS: 0
# ERROR: 0
============================================================================
See test/unit/test-suite.log
Please report to [email protected]
============================================================================
[contents of test/unit/test-suite.log later on...]

-V2 & -V2 --noasm avx512
Result: IDENTICAL

-b 320 & -b 320 --noasm avx512
Result: IDENTICAL

-V2

real    0m7,990s
user    0m7,969s
sys     0m0,025s

real    0m7,917s
user    0m7,887s
sys     0m0,034s

real    0m7,984s
user    0m7,943s
sys     0m0,044s

-V2 --noasm avx512

real    0m7,422s
user    0m7,387s
sys     0m0,039s

real    0m7,627s
user    0m7,588s
sys     0m0,042s

real    0m7,614s
user    0m7,587s
sys     0m0,030s

-b 320

real    0m11,917s
user    0m11,889s
sys     0m0,032s

real    0m11,746s
user    0m11,704s
sys     0m0,045s

real    0m11,845s
user    0m11,794s
sys     0m0,055s

-b 320 --noasm avx512

real    0m12,533s
user    0m12,463s
sys     0m0,070s

real    0m12,471s
user    0m12,429s
sys     0m0,044s

real    0m12,542s
user    0m12,504s
sys     0m0,040s

$ cat test/unit/test-suite.log 
========================================
   lame 4.1: test/unit/test-suite.log
========================================

# TOTAL: 13
# PASS:  12
# SKIP:  0
# XFAIL: 0
# FAIL:  1
# XPASS: 0
# ERROR: 0

.. contents:: :depth: 2

FAIL: test_calc_sfb_noise_vector
================================

[==========] tests: Running 5 test(s).
[ RUN      ] test_lengths
[       OK ] test_lengths
[ RUN      ] test_index_boundaries
[       OK ] test_index_boundaries
[ RUN      ] test_all_zero
[       OK ] test_all_zero
[ RUN      ] test_reference_can_disagree
[       OK ] test_reference_can_disagree
[ RUN      ] test_matches_sse2_exactly
[  ERROR   ] --- a == b
[   LINE   ] --- test_calc_sfb_noise_vector.c:253: error: Failure!
[  FAILED  ] test_matches_sse2_exactly
[==========] tests: 5 test(s) run.
[  PASSED  ] 4 test(s).
[  FAILED  ] tests: 1 test(s), listed below:
[  FAILED  ] test_matches_sse2_exactly

 1 FAILED TEST(S)
FAIL test_calc_sfb_noise_vector (exit status: 1)

Hope this can be useful!

Elio