Skip to content
Snippets Groups Projects
  1. Sep 08, 2022
    • Emma Anholt's avatar
      spirv: Lower RelaxedPrecision tex outputs to 16 bits. · 7a50fab9
      Emma Anholt authored
      The drivers have a path to notice when a tex output is lowered to 16 bits
      and then fold the conversions into the tex instr.  However, just doing the
      lowering at spirv time seems to cause a lot of change (including +waves
      and -instrs on some shaders) that suggests that that wasn't quite doing
      what we want all the time.
      
       PERCENTAGE DELTAS                   Shaders  MaxWaves   Instrs   CodeSize    NOPs      MOVs      Full      (ss)      (sy)   (ss)-stall (sy)-stall    Cat0      Cat1      Cat2      Cat3      Cat5
       angle-aliexpress                    45          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-among_us                      45          .       -0.06%      .       -2.62%    -3.12%    +1.07%      .      -10.00%   +100.00%    -31.74%    -2.30%   +14.52%    -0.15%      .         .
       angle-angry_birds_2_1500            72          .       +0.52%    +0.07%    +1.29%      .       -0.24%      .         .          .          .       +1.16%   +15.00%      .         .         .
       angle-angry_birds_2_launch          99          .       +0.44%      .       +2.49%      .         .         .         .          .          .       +2.21%      .       -0.15%      .         .
       angle-arena_of_valor                84          .       +0.25%    -0.33%    +2.17%      .       +0.24%      .         .          .          .       +1.83%    -2.94%      .         .         .
       angle-asphalt_8                     207         .       +0.03%    -0.15%    +0.42%      .         .         .         .       +0.83%        .       +0.37%    -0.73%      .       -0.07%    -2.27%
       angle-asphalt_9                     909         .       -0.00%    -0.01%      .         .         .         .         .       +0.02%        .         .       -0.02%      .         .         .
       angle-avakin_life                   87          .       +0.49%    +0.54%    +0.21%      .       +1.01%    +4.55%      .          .       -0.97%     +0.18%   +21.59%      .         .         .
       angle-aztec_ruins                   167         .       -0.06%    +0.22%    -0.74%    +3.40%      .       +0.60%      .       +0.62%     +1.01%     -0.68%    +1.77%      .         .         .
       angle-badland                       33          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-beach_buggy_racing            162         .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-blockman_go                   60          .       +0.32%    -1.88%    +7.76%      .       -1.24%      .         .          .          .       +6.39%   -25.00%      .         .         .
       angle-brawl_stars                   102         .       -0.05%      .       +0.23%      .         .         .         .          .          .       +0.18%   -12.00%      .         .         .
       angle-bricks_breaker_quest          18          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-bubble_shooter                21          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-bus_simulator_indonesia       156       +0.16%    +1.72%    +0.52%    +2.79%    +6.90%    +0.08%    +2.40%      .       +3.67%     +49.59%    +2.48%   +18.86%    +3.98%    -1.97%      .
       angle-call_break_offline_card_game  9           .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-candy_crush_500               21          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-candy_crush_soda_saga         57          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-car_parking_multiplayer       312         .       +0.37%    +0.09%      .       +3.77%    +0.65%    -0.17%      .       -1.71%     -3.66%       .      +14.69%    +0.01%    -0.08%      .
       angle-clash_of_clans                27          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-clash_royale                  36          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-cod_mobile                    165         .       +0.22%    -0.05%    +1.59%    -6.04%    -0.18%    +0.42%    -3.41%    -0.19%     -3.27%     +1.49%    -0.15%    -0.01%      .         .
       angle-coin_master                   60          .       +0.26%      .         .         .       +0.40%      .         .          .          .         .      +45.45%      .         .         .
       PERCENTAGE DELTAS                   Shaders  MaxWaves   Instrs   CodeSize    NOPs      MOVs      Full      (ss)      (sy)   (ss)-stall (sy)-stall    Cat0      Cat1      Cat2      Cat3      Cat5
       angle-command_and_conquer_rivals    183         .       +0.32%    +0.27%    +0.25%    -1.89%    +0.25%    -0.35%    -2.35%    +1.47%     +17.06%    +0.18%    +5.46%    +0.13%    -0.15%      .
       angle-cookie_run_kingdom            171         .       +1.23%    +0.61%    +0.43%      .       +0.71%   -18.75%      .          .       -0.54%     +0.35%   +105.00%   -0.82%      .         .
       angle-disney_tsum_tsum              42          .       +0.24%      .       +3.33%      .         .         .         .          .          .       +2.58%   -50.00%      .         .         .
       angle-dota_underlords               426         .       -0.15%    -0.01%    -0.41%      .       -0.07%    +0.06%      .       -0.12%     +0.00%     -0.38%    -0.99%    -0.08%      .         .
       angle-dr_driving                    18          .       +1.38%      .       +0.99%      .         .         .      -83.33%       .        +inf%     +0.89%   +15.38%      .         .         .
       angle-dragon_ball_legends           69          .       +0.81%    -0.33%    +1.55%    -1.20%    +1.69%   -16.67%      .          .       +9.76%     +1.27%   +15.38%      .         .         .
       angle-dragon_raja                   225         .       -0.01%    +0.04%    +0.12%    -0.64%      .       -0.68%      .       +0.75%     +0.89%     +0.11%    -0.59%      .         .         .
       angle-efootball_pes_2021            84          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-egypt_1500                    165         .       +0.07%      .       +0.43%      .         .       +4.76%      .       -3.45%        .       +0.36%      .         .         .         .
       angle-eight_ball_pool               33          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-extreme_car_driving_simulator 90        +0.28%    +0.66%    +0.54%    +0.21%   +11.43%      .       +5.88%    +2.78%    -4.59%     -1.07%     +0.18%   +49.32%    -0.17%    -0.34%      .
       angle-fallout_shelter_online        102       -0.49%    -0.16%    +5.08%    -1.57%      .       +0.92%      .         .          .          .       -1.23%      .         .         .         .
       angle-farm_heroes_saga              27          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-fate_grand_order              45          .       +2.23%    +0.76%      .         .       +4.65%      .         .          .          .         .      +175.00%   +1.62%    -0.62%      .
       angle-fifa_mobile                   105       +0.99%    +2.16%    +3.25%    -5.13%    -1.10%    -0.76%    -4.02%    -1.04%    +7.89%     -11.23%    -4.85%   +44.10%      .         .         .
       angle-final_fantasy                 96          .       +0.02%    -0.02%    -3.28%      .       +1.95%   +18.18%      .       +8.33%        .       -2.91%   +32.00%    +0.06%      .         .
       angle-free_fire                     126         .       +0.55%    +0.20%    +1.34%      .       -0.13%      .         .       +0.61%        .       +1.14%    +6.52%      .         .         .
       angle-gardenscapes                  27          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-genshin_impact                1130      +0.28%    +0.79%    +0.76%    +1.77%    -0.25%    -0.40%    -0.39%    -0.36%    +0.33%     -1.32%     +1.66%    +7.16%    -0.17%    -0.02%    -0.10%
       angle-google_maps                   111         .       +6.17%    +6.72%    -1.73%      .         .      +14.29%      .          .       -2.74%     -1.52%   +70.26%      .         .         .
       angle-happy_color                   9           .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-hay_day                       18          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-hearthstone                   147         .       +0.06%    +0.39%    -2.94%      .         .       +2.27%      .          .       +0.14%     -2.46%    +9.09%      .         .         .
       angle-higgs_domino_island           30          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       PERCENTAGE DELTAS                   Shaders  MaxWaves   Instrs   CodeSize    NOPs      MOVs      Full      (ss)      (sy)   (ss)-stall (sy)-stall    Cat0      Cat1      Cat2      Cat3      Cat5
       angle-hill_climb_racing             21          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-homescapes                    24          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-idle_heroes                   27          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-junes_journey                 27          .       +0.78%    +1.16%    -3.23%      .       +4.00%      .         .          .          .       -2.50%   +32.26%      .         .         .
       angle-kartrider_rush                165         .       +1.16%    +0.51%    +7.25%      .       -0.19%      .       +6.35%    -23.15%    -34.02%    +5.91%   +20.75%    +0.03%      .         .
       angle-klondike_adventures           15          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-league_of_legends_wild_rift   255       +0.34%    +1.53%    +0.57%    +2.20%    -0.99%    -0.40%    -4.67%    -1.38%    +16.85%    -5.48%     +1.89%   +15.13%    +0.17%    -0.21%      .
       angle-lego_legacy                   117         .       +0.18%    +0.08%    +0.57%    -0.68%      .         .         .          .       +10.23%    +0.52%    +1.65%      .         .         .
       angle-lineage_m                     84          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-ludo_king                     21          .       +0.79%      .         .         .       +1.18%      .         .          .          .         .      +71.43%      .         .         .
       angle-magic_tiles_3                 18          .       +1.59%      .         .         .       +2.63%      .         .          .          .         .       +inf%       .         .         .
       angle-manhattan_10                  122         .       +0.39%    +0.36%    +0.58%    -0.87%      .         .       -2.67%    -1.57%     -2.63%     +0.50%    +6.19%      .         .         .
       angle-manhattan_31                  187         .       -0.16%    +0.04%    -0.86%    +1.64%      .       +7.98%    -0.77%    +1.00%     -1.06%     -0.77%    +0.45%      .         .         .
       angle-mario_kart_tour               144         .       +0.15%    +0.10%    -0.52%      .       +0.59%      .         .          .          .       -0.48%    +6.74%      .         .         .
       angle-marvel_contest_of_champions   168         .       +0.59%    +0.50%    -0.25%      .       +0.53%      .         .       +0.59%     -6.37%     -0.22%   +26.52%    -0.08%      .       -8.00%
       angle-messenger_lite                36          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-minecraft                     78          .       -0.91%    -1.48%    -0.49%      .       -0.51%      .         .          .          .       -1.39%    -7.67%    -0.12%    +0.90%      .
       angle-mini_world                    87          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-mobile_legends                117         .       +1.26%    +0.82%    +0.28%      .       +0.84%      .       +6.52%       .       -3.76%     +0.24%   +135.71%   -0.05%      .         .
       angle-my_talking_tom2               105         .       +2.57%    +1.35%    +1.10%    +7.44%    +2.45%    +4.11%    -2.63%    -2.03%     -62.16%    +1.00%   +54.51%    -0.12%      .      -12.50%
       angle-my_talking_tom_friends        120         .       +2.51%    +5.04%    +2.17%    -1.75%    -0.13%    -2.00%    -2.22%    -6.50%        .       +1.73%   +44.05%      .         .         .
       angle-nba2k20_800                   651       +0.06%    +1.96%    +2.01%    -2.45%    -1.09%    +0.56%    -0.96%    -0.72%    -1.43%     -11.82%    -2.25%   +50.38%    -0.01%      .       -1.48%
       angle-nier_reincarnation            165         .       -0.11%    -0.12%    -0.56%    -0.58%    +0.18%    -1.23%    +1.82%    -0.98%     +0.11%     -0.53%    +0.78%      .         .         .
       angle-one_punch_man                 51          .       +0.43%    -0.07%    +6.28%      .         .         .         .          .          .       +4.80%    -5.41%    +0.29%      .         .
       PERCENTAGE DELTAS                   Shaders  MaxWaves   Instrs   CodeSize    NOPs      MOVs      Full      (ss)      (sy)   (ss)-stall (sy)-stall    Cat0      Cat1      Cat2      Cat3      Cat5
       angle-plants_vs_zombies_2           15          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-pokemon_go                    198       +0.39%    -0.24%    +0.04%    +0.63%    +2.32%    +0.14%    -1.16%    +2.38%    -10.78%    +0.63%     +0.79%    -5.11%    -0.28%      .         .
       angle-pokemon_unite                 396         .       -0.42%    -0.31%    +1.36%    -3.66%    -0.52%    +0.90%      .       -4.68%     -1.06%     +1.27%   -13.06%    -0.18%    -0.05%      .
       angle-professional_baseball_spirits 129         .         .         .         .         .       -0.12%      .         .          .          .         .         .         .         .         .
       angle-pubg_mobile_battle_royale     267         .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-pubg_mobile_launch            24          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-pubg_mobile_lite              135         .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-pubg_mobile_skydive           99          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-ragnarok_m_eternal_love       207         .       +0.16%    -0.82%    +3.19%      .       +0.12%      .         .       -2.88%     +16.09%    +2.83%    -5.56%    +0.53%    -0.85%      .
       angle-raid_shadow_legends           78          .       +1.24%    +0.30%    +2.89%   -16.36%    +0.43%   -12.50%    -3.23%    -7.94%     -16.83%    +2.59%   +29.46%    -0.05%      .         .
       angle-real_commando_secret_mission  66          .       +1.23%    +1.44%    -1.00%      .       +1.28%      .         .          .          .       -0.81%   +87.50%      .         .         .
       angle-real_cricket_20               73          .       +0.41%    -0.23%    +4.53%      .       +0.60%    -6.49%      .       -22.48%    +13.87%    +4.09%    -5.14%    -0.04%      .         .
       angle-real_gangster_crime           162         .       +0.19%    +0.36%    -1.39%      .         .         .       +5.56%    -0.73%     -25.00%    -1.24%   +20.35%      .         .         .
       angle-respawnables                  78          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-rise_of_kingdoms              87          .       -0.21%    -0.10%    +0.27%      .         .         .         .          .          .       +0.21%   -41.18%      .         .         .
       angle-romancing_saga                30          .       +1.82%    +0.32%    +8.12%    -4.76%    +0.55%    +9.09%    +6.67%    -30.43%    +14.63%    +6.32%   +25.00%    +0.13%      .         .
       angle-rope_hero_vice_town           258         .       +0.50%    +0.89%    +0.29%      .       +0.35%    +0.31%    +1.45%    -0.08%     -2.64%     +0.27%   +30.49%    +0.02%      .         .
       angle-saint_seiya_awakening         123         .       +0.45%    +0.01%    +0.72%      .       +0.15%    +1.00%    +1.06%    -3.41%     +0.05%     +0.65%   +10.99%    -0.16%      .         .
       angle-sakura_school_simulator       120         .       +0.79%    -0.02%    +7.64%    +5.81%    +0.89%    +0.85%      .       +17.34%    +7.23%     +6.79%    -5.88%    +0.36%    -0.18%      .
       angle-scrabble_go                   72          .       +0.19%    +0.15%    -2.81%      .       +1.07%      .         .          .          .       -2.24%   +20.63%      .         .         .
       angle-shadow_fight_2                24          .       +0.70%      .         .         .       +1.12%      .         .          .          .         .      +41.67%      .         .         .
       angle-slingshot_test1               148       +0.26%    +4.02%    +3.12%    +1.79%    +1.43%    -0.43%    +2.43%    -3.16%    -4.11%     -2.17%     +1.67%   +50.96%      .         .         .
       angle-slingshot_test2               163       +0.40%    +3.80%    +4.02%    -0.71%    -2.62%    -0.63%    +1.69%    +0.90%    -7.64%     -2.96%     -0.67%   +41.14%      .         .         .
       angle-sniper_3d                     85          .       +0.22%    +0.35%      .         .         .         .         .       +0.24%     -4.91%       .       +8.98%      .         .         .
       PERCENTAGE DELTAS                   Shaders  MaxWaves   Instrs   CodeSize    NOPs      MOVs      Full      (ss)      (sy)   (ss)-stall (sy)-stall    Cat0      Cat1      Cat2      Cat3      Cat5
       angle-sonic_the_hedgehog            9           .       +1.38%      .       +0.99%      .         .         .      -83.33%       .        +inf%     +0.89%   +15.38%      .         .         .
       angle-standoff_2                    51          .       -0.58%    +0.38%    -6.52%      .         .       -5.88%    +5.26%    +7.92%     +23.93%    -5.61%    +2.99%    +0.72%    -0.27%      .
       angle-subway_princess_runner        51          .       +1.31%      .       +1.45%      .       +1.15%      .         .          .          .       +1.16%   +71.43%      .         .         .
       angle-subway_surfers                48          .       -0.38%    +0.04%    +2.74%      .       +1.29%      .         .          .       +19.64%    +2.25%   -29.41%    +1.49%    -1.22%      .
       angle-summoners_war                 109         .       +1.14%    +1.20%    +0.37%      .         .         .      -81.08%       .      +5100.00%   +0.33%   +14.47%      .         .         .
       angle-talking_tom_hero_dash         120       +0.21%    +0.31%    +0.21%    -1.15%      .       -0.29%      .       -3.70%    -8.97%     -8.91%     -1.05%   +19.02%    +0.20%      .         .
       angle-temple_run_2                  36          .       +2.68%    +3.08%      .         .       +3.61%      .         .          .          .         .      +194.12%     .         .         .
       angle-temple_run_300                21          .       -1.25%    +2.70%    +4.55%      .       -2.25%      .         .          .          .       +3.67%   -24.00%    -2.51%      .         .
       angle-toon_blast                    24          .       +0.85%    -0.23%      .         .       +1.87%      .         .          .          .         .      +19.44%      .         .         .
       angle-township                      30          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-trex_200                      189         .       +0.28%      .       +1.37%      .       +0.24%      .         .          .          .       +1.18%    +2.76%      .         .         .
       angle-war_planet_online             105         .       +0.31%    +0.58%    -2.28%      .         .       -6.67%    -4.65%    -11.11%    -4.96%     -2.10%    +6.60%    -0.04%      .         .
       angle-whatsapp                      60          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-words_with_friends_2          24          .         .       +1.77%    -3.20%      .         .         .         .          .          .       -2.65%    +5.97%      .         .         .
       angle-wordscapes                    21          .       +1.21%      .       +0.85%      .         .         .      -83.33%       .        +inf%     +0.77%   +13.33%      .         .         .
       angle-world_cricket_championship_2  186         .       +0.57%    -0.30%    +3.93%    +1.04%    +0.75%    -2.04%      .       -1.66%     +4.37%     +3.47%    -5.69%    +0.66%    -0.35%      .
       angle-world_of_kings                309       +0.49%    +1.21%    +0.87%    +3.65%    -4.76%    -0.37%    -1.15%    +0.84%    +26.21%    +5.57%     +3.37%    +9.31%    +0.19%    -0.04%      .
       angle-world_of_tanks_blitz          108         .       +3.17%    +1.61%    +1.13%      .       +1.60%      .         .       +0.24%        .       +1.02%   +98.21%      .         .         .
       angle-world_war_doh                 127         .       +0.16%    -0.16%    +2.45%    -7.10%    +1.29%      .         .       -1.34%        .       +2.14%    -5.84%    +0.04%      .         .
       angle-worms_zone_io                 21          .         .         .         .         .         .         .         .          .          .         .         .         .         .         .
       angle-zillow                        22          .       +1.26%      .       +0.84%      .         .         .      -81.40%       .      +1050.00%   +0.77%   +14.14%      .         .         .
       gfxbench-vk-5-high                  243         .       -0.03%    -0.10%      .         .       -0.17%    +0.36%      .       +0.80%        .         .       -0.40%      .         .         .
       gfxbench-vk-5-normal                186         .       -0.04%    -0.12%      .         .       -0.23%    +0.47%      .       +0.88%        .         .       -0.47%      .         .         .
       -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
       All affected                        1764      +0.53%    +3.06%    +2.82%    +2.26%    -1.34%    +0.95%    -0.07%    -8.34%    -1.71%     +3.43%     +2.07%   +32.13%    +0.10%    -0.45%    -0.27%
       -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
       Total                               14375     +0.06%    +0.50%    +0.42%    +0.41%    -0.23%    +0.10%    -0.02%    -3.19%    -0.39%     +1.10%     +0.37%    +8.53%    +0.02%    -0.06%    -0.08%
      7a50fab9
    • David Heidelberg's avatar
      ci/lava: print set-job-env-vars.sh as other setups do · 86d6c60e
      David Heidelberg authored and Marge Bot's avatar Marge Bot committed
      
      Let's be consistent.
      Also needed for parsing device and traces yml file from logs.
      
      Reviewed-by: default avatarGuilherme Gallo <guilherme.gallo@collabora.com>
      Signed-off-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18422>
      86d6c60e
    • David Heidelberg's avatar
      ci: print env as other setups do · 96243ca6
      David Heidelberg authored and Marge Bot's avatar Marge Bot committed
      
      Let's be consistent.
      Also needed for parsing device and traces yml file from logs.
      
      Reviewed-by: default avatarGuilherme Gallo <guilherme.gallo@collabora.com>
      Signed-off-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18422>
      96243ca6
    • Emma Anholt's avatar
      zink: Don't lower indirect derefs of temp arrays. · e1f032ac
      Emma Anholt authored and Marge Bot's avatar Marge Bot committed
      
      nir_to_spirv can handle it.  Cuts instructions in a turnip CS shader on
      Aztec Ruins from 36k to 3k.
      
      Part of #6115
      
      Reviewed-by: default avatarMike Blumenkrantz <michael.blumenkrantz@gmail.com>
      Part-of: <mesa/mesa!18374>
      e1f032ac
    • Emma Anholt's avatar
      zink: Don't upload shader immediate arrays through UBO 0. · 09f6acc4
      Emma Anholt authored and Marge Bot's avatar Marge Bot committed
      
      Trust the host vulkan driver to load it through whatever way it finds to
      be most efficient.  Saves upload on the frontend when other uniforms
      change.
      
      Reviewed-by: default avatarMike Blumenkrantz <michael.blumenkrantz@gmail.com>
      Part-of: <mesa/mesa!18374>
      09f6acc4
    • Mike Blumenkrantz's avatar
      zink: flag all assigned output slots as mapped · a0f6fecc
      Mike Blumenkrantz authored and Marge Bot's avatar Marge Bot committed
      
      this ensures types which consume more than 1 slot are effectively tagged
      so that the next stage inputs are also assigned properly
      
      fixes:
      spec@arb_enhanced_layouts@execution@component-layout@vs-fs-array-dvec3
      
      cc: mesa-stable
      
      Reviewed-by: default avatarTimur Kristóf <timur.kristof@gmail.com>
      Part-of: <mesa/mesa!18444>
      a0f6fecc
    • James Park's avatar
      vulkan: Augment _WIN32 stub comparison · c48c53c2
      James Park authored and Marge Bot's avatar Marge Bot committed
      
      Make current check robust to incremental linking.
      
      Compare JMP targets if the first byte is opcode 0xE9.
      
      Reviewed-by: Jesse Natalie's avatarJesse Natalie <jenatali@microsoft.com>
      Part-of: <mesa/mesa!18400>
      c48c53c2
    • Kenneth Graunke's avatar
      intel/compiler: Use subgroup invocation for ICP handle loads · 19fc870a
      Kenneth Graunke authored and Marge Bot's avatar Marge Bot committed
      
      When loading a TCS or GS input, we generate some code to read the URB
      handle for a particular input control point (ICP handle), which often
      involves indirect addressing due to a non-constant vertex.
      
      For example:
      
         mov(8) vgrf148+0.0:UW, 76543210V
         shl(8) vgrf149:UD, vgrf148+0.0:UW, 2u
         shl(8) vgrf150:UD, vgrf145:UD, 5u
         add(8) vgrf151:UD, vgrf150:UD, vgrf149:UD
         mov_indirect(8) vgrf147:UD, g2:UD, vgrf151:UD, 96u
      
      Unfortunately, the first load with 76543210V is considered a partial
      write because the 8 channels of 16-bit UW data doesn't fill an entire
      register, and we can't allocate VGRFs at sub-register granularity.
      
      This causes none of the above math to be CSE'd, even though the first
      two instructions are common to *all* input loads, and the rest may be
      reused sometimes as well.
      
      To work around this, we stop emitting 76543210V to a temporary, and
      instead use nir_system_values[SYSTEM_VALUE_SUBGROUP_INVOCATION], which
      already contains this value, and is unconditionally set up for us.
      With all input loads using the same register for the sequence, our
      CSE pass is able to eliminate the rest of the common math.
      
      shader-db results on Tigerlake:
      
         total instructions in shared programs: 20748243 -> 20744844 (-0.02%)
         instructions in affected programs: 73410 -> 70011 (-4.63%)
         helped: 242 / HURT: 21
         helped stats (abs) min: 1 max: 37 x̄: 14.17 x̃: 15
         helped stats (rel) min: 0.17% max: 19.58% x̄: 6.13% x̃: 6.32%
         HURT stats (abs)   min: 1 max: 4 x̄: 1.38 x̃: 1
         HURT stats (rel)   min: 0.18% max: 1.31% x̄: 0.58% x̃: 0.58%
         95% mean confidence interval for instructions value: -13.73 -12.12
         95% mean confidence interval for instructions %-change: -6.00% -5.19%
         Instructions are helped.
      
         total cycles in shared programs: 785828951 -> 785788480 (<.01%)
         cycles in affected programs: 597593 -> 557122 (-6.77%)
         helped: 227 / HURT: 13
         helped stats (abs) min: 6 max: 624 x̄: 182.19 x̃: 185
         helped stats (rel) min: 0.24% max: 18.22% x̄: 7.85% x̃: 7.80%
         HURT stats (abs)   min: 2 max: 153 x̄: 68.08 x̃: 36
         HURT stats (rel)   min: 0.03% max: 7.79% x̄: 2.97% x̃: 1.25%
         95% mean confidence interval for cycles value: -182.55 -154.71
         95% mean confidence interval for cycles %-change: -7.84% -6.69%
         Cycles are helped.
      
      Reviewed-by: default avatarCaio Oliveira <caio.oliveira@intel.com>
      Part-of: <mesa/mesa!18455>
      19fc870a
    • Georg Lehmann's avatar
      nir/opt_algebraic: Optimize unpacking of upcasts to 64bit integers. · 4d7fe94f
      Georg Lehmann authored and Marge Bot's avatar Marge Bot committed
      
      Foz-DB Navi21:
      Totals from 7 (0.01% of 134913) affected shaders:
      CodeSize: 213364 -> 213028 (-0.16%)
      Instrs: 38347 -> 38319 (-0.07%)
      Latency: 780148 -> 779776 (-0.05%)
      InvThroughput: 520098 -> 519851 (-0.05%)
      
      Signed-off-by: default avatarGeorg Lehmann <dadschoorse@gmail.com>
      Reviewed-by: default avatarRhys Perry <pendingchaos02@gmail.com>
      Reviewed-by: Emma Anholt's avatarEmma Anholt <emma@anholt.net>
      Part-of: <mesa/mesa!18435>
      4d7fe94f
    • Tomeu Vizoso's avatar
      ci: Stop explicitly passing env vars to FDO_DISTRIBUTION_EXEC command · 3601c286
      Tomeu Vizoso authored and Marge Bot's avatar Marge Bot committed
      
      ci-templates will now pass all env vars to the command.
      
      Reviewed-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18467>
      3601c286
    • Tomeu Vizoso's avatar
      ci: Install sysvinit-core without --no-remove · 7dc8a780
      Tomeu Vizoso authored and Marge Bot's avatar Marge Bot committed
      
      As it conflicts now with some packages already installed but not
      necessary, such as:
      
      libpam-systemd packagekit packagekit-tools policykit-1 systemd-sysv
      
      Reviewed-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18467>
      7dc8a780
    • Tomeu Vizoso's avatar
      ci: Use --no-install-recommends to avoid problems with --no-remove · 86e2078b
      Tomeu Vizoso authored and Marge Bot's avatar Marge Bot committed
      
      Some packages that are being installed via recommends are conflicting
      with already installed packages, causing this error:
      
      E: Packages need to be removed but remove is disabled.
      
      We dont need these packages, so don't install them.
      
      Reviewed-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18467>
      86e2078b
    • Tomeu Vizoso's avatar
      ci: Uprev ci-templates · 55724c2a
      Tomeu Vizoso authored and Marge Bot's avatar Marge Bot committed
      
      Uprev it to:
      
      d5aa3941aa03 ("freebsd: Move from 13.0 to 13.1")
      
      Reviewed-by: default avatarDavid Heidelberg <david.heidelberg@collabora.com>
      Part-of: <mesa/mesa!18467>
      55724c2a
    • Karmjit Mahil's avatar
      pvr: Remove unimplemented push descriptor code. · b36e9b61
      Karmjit Mahil authored and Marge Bot's avatar Marge Bot committed
      
      Push descriptors are part of VK_KHR_push_descriptor.
      Not supporting it for now.
      
      Signed-off-by: default avatarKarmjit Mahil <Karmjit.Mahil@imgtec.com>
      Reviewed-by: default avatarFrank Binns <frank.binns@imgtec.com>
      Part-of: <mesa/mesa!18430>
      b36e9b61
    • Karol Herbst's avatar
      nv50: properly flush the TSC cache on 3D · 54709efd
      Karol Herbst authored and Marge Bot's avatar Marge Bot committed
      
      The change didn't make any sense. `s` will always be
      `NV50_SHADER_STAGE_COMPUTE`, because it's used to loop over all shader
      stages. And the TSC cache on the compute side is already flushed in
      `nv50_compute_validate_samplers`.
      
      Fixes spurious `CACHE_ERROR` dmesg messages.
      
      Fixes: ba6ba8c9 ("nv50: adapt texture and constbuf paths for compute shaders")
      Signed-off-by: Karol Herbst's avatarKarol Herbst <kherbst@redhat.com>
      Reviewed-by: default avatarM Henning <drawoc@darkrefraction.com>
      Part-of: <mesa/mesa!18382>
      54709efd
    • Karol Herbst's avatar
      nv50/ir: fix OP_UNION resolving when used for vector values · b23b94fb
      Karol Herbst authored and Marge Bot's avatar Marge Bot committed
      When an OP_UNION def takes part in a vector source e.g. for a tex
      instruction we failed to clean up the OP_UNION instruction as rep() points
      towards the coalesced value instead.
      
      This fixes a regression on nv50 moving to NIR, but also potentially issues
      with nvc0.
      
      The main reason this is common in nv50 is, that we lower OP_SLCT to a set,
      predicated movs and a union.
      
      Closes: mesa/mesa#6406
      Closes: mesa/mesa#7117
      
      
      Cc: mesa-stable
      Signed-off-by: Karol Herbst's avatarKarol Herbst <kherbst@redhat.com>
      Reviewed-by: default avatarM Henning <drawoc@darkrefraction.com>
      Part-of: <mesa/mesa!18377>
      b23b94fb
    • Frank Binns's avatar
      pvr: finish render job sample count setup · cbc477d0
      Frank Binns authored and Marge Bot's avatar Marge Bot committed
      
      Signed-off-by: default avatarFrank Binns <frank.binns@imgtec.com>
      Reviewed-by: default avatarKarmjit Mahil <Karmjit.Mahil@imgtec.com>
      Part-of: <mesa/mesa!18437>
      cbc477d0
    • Thomas Andersen's avatar
      util: avoid deprecated builtin has_trivial_destructor · 1980827a
      Thomas Andersen authored and Marge Bot's avatar Marge Bot committed
      
      From clang 16 has_trivial_destructor is deprecated.
      Use the replacement __is_trivially_destructible if it
      is available.
      
      Fixes new warnings with clang 16 like:
      
      ../src/compiler/glsl/list.h:58:4: warning: builtin __has_trivial_destructor is deprecated; use __is_trivially_destructible instead [-Wdeprecated-builtins]
      ../src/util/ralloc.h:551:4: note: expanded from macro 'DECLARE_RZALLOC_CXX_OPERATORS'
         DECLARE_ALLOC_CXX_OPERATORS_TEMPLATE(type, rzalloc_size)
         ^
      ../src/util/ralloc.h:542:12: note: expanded from macro 'DECLARE_ALLOC_CXX_OPERATORS_TEMPLATE'
            if (!HAS_TRIVIAL_DESTRUCTOR(TYPE))                                 \
                 ^
      ../src/util/macros.h:233:44: note: expanded from macro 'HAS_TRIVIAL_DESTRUCTOR'
      
      Reviewed-by: default avatarEric Engestrom <eric@igalia.com>
      Part-of: <mesa/mesa!18423>
      1980827a
    • Karmjit Mahil's avatar
      pvr: Finish setting up job resolve info. · 61d265cf
      Karmjit Mahil authored and Marge Bot's avatar Marge Bot committed
      
      Signed-off-by: default avatarKarmjit Mahil <Karmjit.Mahil@imgtec.com>
      Reviewed-by: default avatarFrank Binns <frank.binns@imgtec.com>
      Part-of: <mesa/mesa!18431>
      61d265cf
    • Karmjit Mahil's avatar
      pvr: Set descriptor dirty flag based on other flags. · c6113def
      Karmjit Mahil authored and Marge Bot's avatar Marge Bot committed
      
      Set the flag if the descriptor set is updated.
      Set the fragment descriptor dirty flag if blend consts are updated.
      
      Signed-off-by: default avatarKarmjit Mahil <Karmjit.Mahil@imgtec.com>
      Reviewed-by: default avatarFrank Binns <frank.binns@imgtec.com>
      Part-of: <mesa/mesa!18429>
      c6113def
    • Mark Collins's avatar
      tu: Expose VK_EXT_tooling_info using common implementation · dd19da31
      Mark Collins authored and Marge Bot's avatar Marge Bot committed
      
      Signed-off-by: default avatarMark Collins <mark@igalia.com>
      Part-of: <mesa/mesa!18390>
      dd19da31
    • Mark Collins's avatar
      tu: Clamp priority in DRM submitqueue creation · c82249aa
      Mark Collins authored and Marge Bot's avatar Marge Bot committed
      
      The kernel driver has a range of valid priority values that can
      be supplied to it, submitting any priority value outside these
      bounds will result in `-EINVAL`. To avoid this, the priority
      value is now clamped to the range that the kernel supports.
      
      Fixes: 0c6fbfca
      Signed-off-by: default avatarMark Collins <mark@igalia.com>
      Part-of: <mesa/mesa!18389>
      c82249aa
    • Pavel Ondračka's avatar
      r300: allow presubtract when both ADD sources are negative · c3f51a5d
      Pavel Ondračka authored and Marge Bot's avatar Marge Bot committed
      
      Current code doesn't handle this, however it is easy to make it work
      by moving the negate to the presubtract source. Minor win in shader-db,
      mostly with Unigine shaders.
      
      Shader-db RV530:
      total instructions in shared programs: 136382 -> 136236 (-0.11%)
      instructions in affected programs: 9911 -> 9765 (-1.47%)
      total temps in shared programs: 18939 -> 18942 (0.02%)
      temps in affected programs: 37 -> 40 (8.11%)
      
      Reviewed-by: default avatarFilip Gawin <filip@gawin.net>
      Signed-off-by: Pavel Ondračka's avatarPavel Ondračka <pavel.ondracka@gmail.com>
      Part-of: <mesa/mesa!18289>
      c3f51a5d
    • Gert Wollny's avatar
      virgl: Add some formats that the CTS uses · 1d8627de
      Gert Wollny authored and Marge Bot's avatar Marge Bot committed
      
      Otherwise running the CTS emits lots of warnings about
      these formats missing in the drivers format table.
      
      Signed-off-by: default avatarGert Wollny <gert.wollny@collabora.com>
      Part-of: <mesa/mesa!18462>
      1d8627de
    • Rob Clark's avatar
      egl: Relax locking · bbef3cb9
      Rob Clark authored
      Now that we have the rwlock TerminateLock protecting us against
      eglTerminate() yanking the rug from under us, drop the BDL across
      calls to driver (or at least the main ones that can potentially
      block).
      
      Closes: mesa/mesa#7039
      
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Acked-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      bbef3cb9
    • Rob Clark's avatar
      egl: Introduce rwlock to protect eglTerminate() · 5d99e8cc
      Rob Clark authored
      
      eglTerminate() must be serialized against all other EGL calls.  But in
      most cases, other EGL calls do not need to be serialized against each
      other.  Which fits rather well with a rwlock.
      
      One would be tempted to simply replace the existing BDL with a rwlock,
      but several portability and debuggability limitations of the rwlock
      implementation prevent that, as described in the TerminateLock comment
      block.
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Acked-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      5d99e8cc
    • Rob Clark's avatar
      egl: Make RefCount atomic · 7ba2784b
      Rob Clark authored
      
      Once we relax the locking, we will be doing _eglPutFoo() outside of the
      big display lock.
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Acked-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      7ba2784b
    • Rob Clark's avatar
      egl/dri2: Add display lock · f1efe037
      Rob Clark authored
      
      In preperation of relaxing eglapi to not hold a lock across driver
      calls, but instead only for protecting it's own state, add our own
      lock to protect code paths that need locking or have not been audited
      yet.  The blocking calls (ClientWaitSyncKHR) or critical path and/or
      blocking (MakeCurrent, SwapBuffers*) are lockless, as they have already
      been audited for thread safety.
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Acked-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      f1efe037
    • Rob Clark's avatar
      egl/dri2: Make ref_count atomic · fc528128
      Rob Clark authored
      
      In particular, MakeCurrent can be called on multiple threads in
      parallel.
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Reviewed-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      fc528128
    • Rob Clark's avatar
      egl/wgl: Make ref_count atomic · a2d6dee4
      Rob Clark authored
      
      Looks like wgl doesn't have much display state to protect.  But it's
      ref_count should be atomic before we start removing locking from eglapi
      to protect against MakeCurrent being called in parallel on multiple
      threads.
      
      Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
      Acked-by: default avatarEric Engestrom <eric@igalia.com>
      Reviewed-by: Adam Jackson's avatarAdam Jackson <ajax@redhat.com>
      Part-of: <mesa/mesa!18050>
      a2d6dee4
    • Timothy Arceri's avatar
      glsl: remove GLSL IR inverse comparison optimisations · f182b195
      Timothy Arceri authored and Marge Bot's avatar Marge Bot committed
      
      As per 7d85dc4f GLSL IR is not smart enough to handle this
      correctly for NANs.
      
      Shader-db radeonsi (RX 6800):
      
      Totals from affected shaders:
      SGPRS: 26848 -> 26848 (0.00 %)
      VGPRS: 13552 -> 13552 (0.00 %)
      Spilled SGPRs: 134 -> 134 (0.00 %)
      Spilled VGPRs: 0 -> 0 (0.00 %)
      Private memory VGPRs: 0 -> 0 (0.00 %)
      Scratch size: 0 -> 0 (0.00 %) dwords per thread
      Code Size: 635000 -> 630988 (-0.63 %) bytes
      Max Waves: 5474 -> 5474 (0.00 %)
      
      Shader-db iris (BDW):
      
      total instructions in shared programs: 17538859 -> 17539018 (<.01%)
      instructions in affected programs: 29369 -> 29528 (0.54%)
      helped: 3
      HURT: 126
      helped stats (abs) min: 1 max: 1 x̄: 1.00 x̃: 1
      helped stats (rel) min: 0.49% max: 0.49% x̄: 0.49% x̃: 0.49%
      HURT stats (abs)   min: 1 max: 2 x̄: 1.29 x̃: 1
      HURT stats (rel)   min: 0.27% max: 1.32% x̄: 0.61% x̃: 0.54%
      95% mean confidence interval for instructions value: 1.13 1.33
      95% mean confidence interval for instructions %-change: 0.54% 0.63%
      Instructions are HURT.
      
      total loops in shared programs: 4866 -> 4866 (0.00%)
      loops in affected programs: 0 -> 0
      helped: 0
      HURT: 0
      
      total cycles in shared programs: 858548230 -> 858548915 (<.01%)
      cycles in affected programs: 1331737 -> 1332422 (0.05%)
      helped: 0
      HURT: 92
      HURT stats (abs)   min: 2 max: 49 x̄: 7.45 x̃: 6
      HURT stats (rel)   min: 0.01% max: 1.90% x̄: 0.12% x̃: 0.05%
      95% mean confidence interval for cycles value: 5.72 9.17
      95% mean confidence interval for cycles %-change: 0.05% 0.19%
      Cycles are HURT.
      
      Note: With the addition of "nir/comparison_pre: See through an inot to
      apply the optimization", idr's shader-db results are:
      
      All Broadwell and newer Intel platforms had similar results. (Ice Lake shown)
      total instructions in shared programs: 19940805 -> 19940802 (<.01%)
      instructions in affected programs: 582 -> 579 (-0.52%)
      helped: 3 / HURT: 0
      
      total cycles in shared programs: 858431633 -> 858431747 (<.01%)
      cycles in affected programs: 4938 -> 5052 (2.31%)
      helped: 0 / HURT: 3
      
      All older Intel platforms had similar results. (Haswell shown)
      total instructions in shared programs: 16715626 -> 16715670 (<.01%)
      instructions in affected programs: 9496 -> 9540 (0.46%)
      helped: 0 / HURT: 44
      
      total cycles in shared programs: 881224396 -> 881232314 (<.01%)
      cycles in affected programs: 600610 -> 608528 (1.32%)
      helped: 6 / HURT: 44
      
      Reviewed-by: default avatarIan Romanick <ian.d.romanick@intel.com>
      Part-of: <mesa/mesa!18006>
      f182b195
    • Ian Romanick's avatar
      nir/comparison_pre: See through an inot to apply the optimization · 54735367
      Ian Romanick authored and Marge Bot's avatar Marge Bot committed
      
      This also prevents some small regressions in "glsl: remove GLSL IR
      inverse comparison optimisations".
      
      shader-db results:
      
      All Sandy Bridge and newer Intel platforms had similar results. (Ice Lake shown)
      total instructions in shared programs: 19941025 -> 19940805 (<.01%)
      instructions in affected programs: 52431 -> 52211 (-0.42%)
      helped: 188 / HURT: 6
      
      total cycles in shared programs: 858451784 -> 858431633 (<.01%)
      cycles in affected programs: 2119134 -> 2098983 (-0.95%)
      helped: 183 / HURT: 12
      
      LOST:   2
      GAINED: 0
      
      Iron Lake and GM45 had similar results. (Iron Lake shown)
      total instructions in shared programs: 8364668 -> 8364670 (<.01%)
      instructions in affected programs: 753 -> 755 (0.27%)
      helped: 2 / HURT: 4
      
      total cycles in shared programs: 248752572 -> 248752238 (<.01%)
      cycles in affected programs: 87290 -> 86956 (-0.38%)
      helped: 2 / HURT: 4
      
      fossil-db results:
      
      Skylake, Ice Lake, and Tiger Lake had similar results. (Ice Lake shown)
      Instructions in all programs: 144909184 -> 144909130 (-0.0%)
      Instructions helped: 6
      
      Cycles in all programs: 9138641740 -> 9138640984 (-0.0%)
      Cycles helped: 8
      
      Reviewed-by: default avatarTimothy Arceri <tarceri@itsqueeze.com>
      Part-of: <mesa/mesa!18006>
      54735367
    • Timothy Arceri's avatar
      nir: support loop unrolling with inot conditions · 61c3438b
      Timothy Arceri authored and Marge Bot's avatar Marge Bot committed
      
      Ever since 4246c286 and 7d85dc4f loop unrolling can no
      longer depend on inot being eliminated from the loop
      terminator condition so we need to be able to handle it.
      
      This change avoids 292 loop unrolling regressions with shader-db
      once the following patch is applied.
      
      Reviewed-by: default avatarIan Romanick <ian.d.romanick@intel.com>
      Part-of: <mesa/mesa!18006>
      61c3438b
    • Timothy Arceri's avatar
      nir: update nir_is_supported_terminator_condition() · 96c19d23
      Timothy Arceri authored and Marge Bot's avatar Marge Bot committed
      
      Ever since 4246c286 and 7d85dc4f loop unrolling can no
      longer depend on inot being eliminated from the loop
      terminator condition so we need to be able to handle it.
      
      Here we simply check to see if the inot contains a simple
      terminator condition we previously handled. We also update
      the previous users of this function to use a newly name
      copy of the previous behaviour
      nir_is_terminator_condition_with_two_inputs().
      
      Reviewed-by: default avatarIan Romanick <ian.d.romanick@intel.com>
      Part-of: <mesa/mesa!18006>
      96c19d23
  2. Sep 07, 2022
Loading