Skip to content
Snippets Groups Projects

x86: Add 8-bit ipred z1 AVX-512 (Ice Lake) asm

Merged Henrik Gramner requested to merge gramner/dav1d:ipred_z1_8bpc_avx512icl into master
All threads resolved!

With AVX-512 we can now keep all the data in registers and use vpermb/vpermi2b to load the input pixels instead of having to do memory loads from a stack buffer.

intra_pred_z1_w4_8bpc_c:             204.5 ( 1.00x)
intra_pred_z1_w4_8bpc_ssse3:          37.6 ( 5.43x)
intra_pred_z1_w4_8bpc_avx2:           37.8 ( 5.41x)
intra_pred_z1_w4_8bpc_avx512icl:      23.5 ( 8.70x)

intra_pred_z1_w8_8bpc_c:             342.5 ( 1.00x)
intra_pred_z1_w8_8bpc_ssse3:          58.1 ( 5.90x)
intra_pred_z1_w8_8bpc_avx2:           40.9 ( 8.38x)
intra_pred_z1_w8_8bpc_avx512icl:      27.5 (12.44x)

intra_pred_z1_w16_8bpc_c:           1162.9 ( 1.00x)
intra_pred_z1_w16_8bpc_ssse3:        124.8 ( 9.31x)
intra_pred_z1_w16_8bpc_avx2:          79.5 (14.63x)
intra_pred_z1_w16_8bpc_avx512icl:     45.1 (25.76x)

intra_pred_z1_w32_8bpc_c:           1927.0 ( 1.00x)
intra_pred_z1_w32_8bpc_ssse3:        254.9 ( 7.56x)
intra_pred_z1_w32_8bpc_avx2:         160.5 (12.01x)
intra_pred_z1_w32_8bpc_avx512icl:    110.3 (17.47x)

intra_pred_z1_w64_8bpc_c:           2962.4 ( 1.00x)
intra_pred_z1_w64_8bpc_ssse3:        504.7 ( 5.87x)
intra_pred_z1_w64_8bpc_avx2:         284.4 (10.42x)
intra_pred_z1_w64_8bpc_avx512icl:    245.8 (12.05x)

Merge request reports

Loading
Loading

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
  • Henrik Gramner added 3 commits

    added 3 commits

    • 48035599 - 1 commit from branch videolan:master
    • 8936bab7 - x86: Consolidate some pb_0to31 and pb_0to63 constants
    • 4c012978 - x86: Add 8-bit ipred z1 AVX-512 (Ice Lake) asm

    Compare with previous version

  • Henrik Gramner resolved all threads

    resolved all threads

  • Victorien Le Couviour--Tuffet approved this merge request

    approved this merge request

  • Henrik Gramner changed milestone to %1.4.0

    changed milestone to %1.4.0

  • merged

  • Ronald S. Bultje mentioned in issue #316

    mentioned in issue #316

  • Please register or sign in to reply
    Loading