Before:
 29.63%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs_b
 14.46%  ffmpeg_g  ffmpeg_g           [.] decode_b
  8.19%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_16_ssse3.loop
  5.32%  ffmpeg_g  ffmpeg_g           [.] inter_recon.isra.5
  4.89%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_8_ssse3.loop
  3.82%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs.isra.7
  3.29%  ffmpeg_g  ffmpeg_g           [.] find_ref_mvs
  3.21%  ffmpeg_g  ffmpeg_g           [.] fill_mv
  2.70%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_4_ssse3.loop
  2.52%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_16_ssse3.loop
  2.42%  ffmpeg_g  ffmpeg_g           [.] decode_sb
  1.79%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs_b32.constprop.14
  1.68%  ffmpeg_g  ffmpeg_g           [.] intra_recon.isra.4
  1.39%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_8_ssse3.loop
  1.05%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_v_16_ssse3.loop
  0.82%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_idct_idct_8x8_add_avx.idctfull
  0.80%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_4_ssse3.loop
  0.58%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_h_16_ssse3.loop
  0.51%  ffmpeg_g  ffmpeg_g           [.] vp9_decode_frame
  0.47%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_v_8_ssse3.loop
  0.45%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_idct_idct_4x4_add_ssse3.idctfull
  0.38%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put16_sse.loop


After:
 29.52%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs_b
 14.81%  ffmpeg_g  ffmpeg_g           [.] decode_b
  5.65%  ffmpeg_g  ffmpeg_g           [.] inter_recon.isra.5
  5.45%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_16_ssse3.loop
  5.05%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_8_ssse3.loop
  3.85%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs.isra.7
  3.21%  ffmpeg_g  ffmpeg_g           [.] find_ref_mvs
  3.08%  ffmpeg_g  ffmpeg_g           [.] fill_mv
  2.82%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_4_ssse3.loop
  2.48%  ffmpeg_g  ffmpeg_g           [.] decode_sb
  2.01%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_h_32_avx2.loop
  1.93%  ffmpeg_g  ffmpeg_g           [.] decode_coeffs_b32.constprop.14
  1.83%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_16_ssse3.loop
  1.60%  ffmpeg_g  ffmpeg_g           [.] intra_recon.isra.4
  1.59%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_8_ssse3.loop
  0.80%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_idct_idct_8x8_add_avx.idctfull
  0.79%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_4_ssse3.loop
  0.60%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_v_16_ssse3.loop
  0.57%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_v_8_ssse3.loop
  0.51%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put_8tap_1d_v_32_avx2.loop
  0.47%  ffmpeg_g  ffmpeg_g           [.] vp9_decode_frame
  0.46%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_idct_idct_4x4_add_ssse3.idctfull
  0.41%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_put16_sse.loop
  0.33%  ffmpeg_g  ffmpeg_g           [.] ff_vp9_avg_8tap_1d_h_16_ssse3.loop