[Math-atlas-commits] CVS: AtlasBase/Clint/misc/ARCHDEF/P4E64SSE3/gcc/gemm atlas_cacheedge.h,NONE,1.1
Brought to you by:
rwhaley,
tonyc040457
Update of /cvsroot/math-atlas/AtlasBase/Clint/misc/ARCHDEF/P4E64SSE3/gcc/gemm In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv20197/Clint/misc/ARCHDEF/P4E64SSE3/gcc/gemm Added Files: atlas_cacheedge.h cCleanK cCleanM cCleanN cMMRES cNB cNCNB cXover.h cbestNN_0x0x0 cbestNN_0x0x40 cbestNN_40x40x40 cbestNT_0x0x0 cbestNT_0x0x40 cbestNT_40x40x40 cbestTN_0x0x0 cbestTN_0x0x40 cbestTN_40x40x40 cbestTT_0x0x0 cbestTT_0x0x40 cbestTT_40x40x40 cgMMRES cuCleanK cuCleanKF cuCleanM cuCleanMF cuCleanN cuCleanNF cuMMRES dCleanK dCleanM dCleanN dMMRES dNB dNCNB dXover.h dbestNN_0x0x0 dbestNN_0x0x36 dbestNN_36x36x36 dbestNT_0x0x0 dbestNT_0x0x36 dbestNT_36x36x36 dbestTN_0x0x0 dbestTN_0x0x36 dbestTN_36x36x36 dbestTT_0x0x0 dbestTT_0x0x36 dbestTT_36x36x36 dgMMRES dnreg duCleanK duCleanKF duCleanM duCleanMF duCleanN duCleanNF duMMRES sCleanK sCleanM sCleanN sMMRES sNB sNCNB sXover.h sbestNN_0x0x0 sbestNN_0x0x48 sbestNN_48x48x48 sbestNT_0x0x0 sbestNT_0x0x48 sbestNT_48x48x48 sbestTN_0x0x0 sbestTN_0x0x48 sbestTN_48x48x48 sbestTT_0x0x0 sbestTT_0x0x48 sbestTT_48x48x48 sgMMRES snreg suCleanK suCleanKF suCleanM suCleanMF suCleanN suCleanNF suMMRES zCleanK zCleanM zCleanN zMMRES zNB zNCNB zXover.h zbestNN_0x0x0 zbestNN_0x0x24 zbestNN_24x24x24 zbestNT_0x0x0 zbestNT_0x0x24 zbestNT_24x24x24 zbestTN_0x0x0 zbestTN_0x0x24 zbestTN_24x24x24 zbestTT_0x0x0 zbestTT_0x0x24 zbestTT_24x24x24 zgMMRES zuCleanK zuCleanKF zuCleanM zuCleanMF zuCleanN zuCleanNF zuMMRES Log Message: 3.7.10 commit --- NEW FILE: atlas_cacheedge.h --- #ifndef ATLAS_CACHEEDGE_H #define ATLAS_CACHEEDGE_H #define CacheEdge 393216 #endif --- NEW FILE: cCleanK --- KB MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH GEN-MFLOP SPC-MFLOP 48 0 4 48 5 1 48 0 5 1 -2575.24 -2509.21 47 0 4 48 5 1 47 0 5 1 -2589.13 -2491.56 46 0 4 48 5 1 46 0 5 1 -2594.77 -2494.38 45 0 4 48 5 1 45 0 5 1 -2590.51 -2496.15 44 0 4 48 5 1 44 0 5 1 -2591.50 -2495.64 43 0 4 48 5 1 43 0 5 1 -2592.42 -2498.55 42 0 4 48 5 1 42 0 5 1 -2591.49 -2498.74 41 0 4 48 5 1 41 0 5 1 -2590.51 -2500.65 40 0 4 48 5 1 40 0 5 1 -2584.09 -2493.11 39 0 4 48 5 1 39 0 5 1 -2586.00 -2485.69 38 0 4 48 5 1 38 0 5 1 -2582.95 -2469.23 37 0 4 48 5 1 37 0 5 1 -2580.91 -2476.65 36 0 4 48 5 1 36 0 5 1 -2576.73 -2466.47 35 0 4 48 5 1 35 0 5 1 -2578.67 -2465.97 34 0 4 48 5 1 34 0 5 1 -2571.52 -2482.54 33 0 4 48 5 1 33 0 5 1 -2569.22 -2449.02 32 0 4 48 5 1 32 0 5 1 -2546.37 -2460.22 31 0 4 48 5 1 31 0 5 1 -2560.40 -2438.48 30 0 4 48 5 1 30 0 5 1 -2557.68 -2437.47 29 0 4 48 5 1 29 0 5 1 -2553.01 -2428.38 28 0 4 48 5 1 28 0 5 1 -2544.70 -2426.11 27 0 4 48 5 1 27 0 5 1 -2547.75 -2416.86 26 0 4 48 5 1 26 0 5 1 -2537.66 -2411.03 25 0 4 48 5 1 25 0 5 1 -2531.46 -2405.04 24 0 4 48 5 1 24 0 5 1 -2520.48 -2391.38 23 0 4 48 5 1 23 0 5 1 -2521.33 -2381.96 22 0 4 48 5 1 22 0 5 1 -2504.66 -2348.19 21 0 4 48 5 1 21 0 5 1 -2475.28 -2350.58 20 0 4 48 5 1 20 0 5 1 -2458.37 -2324.89 19 0 4 48 5 1 19 0 5 1 -2445.19 -2289.72 18 0 4 48 5 1 18 0 5 1 -2456.98 -2284.78 17 0 4 48 5 1 17 0 5 1 -2450.90 -2254.34 16 0 4 48 5 1 16 0 5 1 -2445.81 -2260.89 15 0 4 48 5 1 15 0 5 1 -2450.25 -2205.52 14 0 4 48 5 1 14 0 5 1 -2439.35 -2196.95 13 0 4 48 5 1 13 0 5 1 -2407.59 -2205.51 12 0 4 48 5 1 12 0 5 1 -2383.94 -2135.64 11 0 4 48 5 1 11 0 5 1 -2356.94 -2094.75 10 0 4 48 5 1 10 0 5 1 -2304.49 -2077.27 9 0 4 48 5 1 9 0 5 1 -2299.52 -2015.44 8 0 4 48 5 1 8 0 5 1 -2239.05 -1962.63 7 0 4 48 5 1 7 0 5 1 -2144.63 -1877.59 6 0 4 48 5 1 6 0 5 1 -2119.33 -1877.59 5 0 4 48 5 1 5 0 5 1 -1970.40 -1877.59 0 0 5 48 5 1 1 0 5 1 -1800.02 -1877.59 --- NEW FILE: cCleanM --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2579.51 --- NEW FILE: cCleanN --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2719.68 --- NEW FILE: cMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2556.61 ICASE NB MFLOP ROUT AUTHOR 220 48 -7933.08 "ATL_smm6x1x80_sse.c" "R. Clint Whaley" --- NEW FILE: cNB --- 1 48 --- NEW FILE: cNCNB --- 40 --- NEW FILE: cXover.h --- #ifndef CXOVER_H #define CXOVER_H #define ATL_3NB 144 #define NN_MNK_M 4800 #define NN_MNK_N 4800 #define NN_MNK_MN 23040 #define NN_MNK_K 43200 #define NN_MNK_GE 13824 #define NT_MNK_M 4800 #define NT_MNK_N 10800 #define NT_MNK_MN 23040 #define NT_MNK_K 10800 #define NT_MNK_GE 3375 #define TN_MNK_M 4800 #define TN_MNK_N 4800 #define TN_MNK_MN 23040 #define TN_MNK_K 10800 #define TN_MNK_GE 27000 #define TT_MNK_M 4800 #define TT_MNK_N 10800 #define TT_MNK_MN 23040 #define TT_MNK_K 10800 #define TT_MNK_GE 13824 #endif --- NEW FILE: cbestNN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 40 5 1 16 0 5 1 -2453.59 --- NEW FILE: cbestNN_0x0x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2620.24 --- NEW FILE: cbestNN_40x40x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2624.54 --- NEW FILE: cbestNT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 40 5 1 16 0 5 1 -2220.93 --- NEW FILE: cbestNT_0x0x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2406.57 --- NEW FILE: cbestNT_40x40x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2412.75 --- NEW FILE: cbestTN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 40 5 1 16 0 5 1 -2510.36 --- NEW FILE: cbestTN_0x0x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2634.29 --- NEW FILE: cbestTN_40x40x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2640.55 --- NEW FILE: cbestTT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 40 5 1 16 0 5 1 -2207.56 --- NEW FILE: cbestTT_0x0x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2319.35 --- NEW FILE: cbestTT_40x40x40 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 40 5 1 40 0 5 1 -2328.21 --- NEW FILE: cgMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 40 5 1 40 0 5 1 -2760.42 --- NEW FILE: cuCleanK --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 2 1 1 0 48 42 24 16 -1760.50 ATL_mm4x4x2US.c 4 220 1 48 40 24 16 -6438.09 ATL_smm6x1x80_sse.c --- NEW FILE: cuCleanKF --- MULT ICASE FIXED NB 1 4 220 1 48 --- NEW FILE: cuCleanM --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 4 1 311 0 48 42 24 16 -1262.14 ATL_mm6x8x8_1p.c 4 211 1 48 40 24 16 -5233.37 ATL_smm_sse1_K.c 6 220 0 48 42 24 18 -7832.48 ATL_smm6x1x80_sse.c 14 218 0 48 42 14 28 -7977.07 ATL_smm14x1x84_sseCU.c --- NEW FILE: cuCleanMF --- MULT ICASE FIXED NB 3 4 211 1 48 6 220 0 48 14 218 0 48 --- NEW FILE: cuCleanN --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 1 220 0 48 42 24 16 -7545.36 ATL_smm6x1x80_sse.c --- NEW FILE: cuCleanNF --- MULT ICASE FIXED NB 1 1 220 0 48 --- NEW FILE: cuMMRES --- CASE NB MFLOP ROUTINE 220 48 -7933.08 "ATL_smm6x1x80_sse.c" "R. Clint Whaley" --- NEW FILE: dCleanK --- KB MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH GEN-MFLOP SPC-MFLOP 36 0 1 36 4 1 36 0 2 2 -2042.69 -1818.30 35 0 1 36 4 1 35 0 2 2 -2060.25 -1849.68 34 0 1 36 4 1 34 0 2 2 -2046.48 -1845.07 33 0 1 36 4 1 33 0 2 2 -2060.70 -1829.14 32 0 1 36 4 1 32 0 2 2 -2044.48 -1808.38 31 0 1 36 4 1 31 0 2 2 -2050.60 -1794.93 30 0 1 36 4 1 30 0 2 2 -2037.51 -1800.50 29 0 1 36 4 1 29 0 2 2 -2050.91 -1815.58 28 0 1 36 4 1 28 0 2 2 -2042.39 -1793.07 27 0 1 36 4 1 27 0 2 2 -2042.78 -1783.94 26 0 1 36 4 1 26 0 2 2 -2034.91 -1766.42 25 0 1 36 4 1 25 0 2 2 -2048.24 -1738.15 24 0 1 36 4 1 24 0 2 2 -2014.70 -1716.06 23 0 1 36 4 1 23 0 2 2 -2032.44 -1702.34 22 0 1 36 4 1 22 0 2 2 -2022.45 -1712.03 21 0 1 36 4 1 21 0 2 2 -2026.78 -1733.26 20 0 1 36 4 1 20 0 2 2 -2008.67 -1675.67 19 0 1 36 4 1 19 0 2 2 -2023.57 -1729.39 18 0 1 36 4 1 18 0 2 2 -2000.64 -1704.24 17 0 1 36 4 1 17 0 2 2 -2008.30 -1713.66 16 0 1 36 4 1 16 0 2 2 -1975.09 -1683.16 15 0 1 36 4 1 15 0 2 2 -1978.26 -1653.74 14 0 1 36 4 1 14 0 2 2 -1967.66 -1610.69 13 0 1 36 4 1 13 0 2 2 -1963.32 -1583.34 12 0 1 36 4 1 12 0 2 2 -1933.96 -1540.59 11 0 1 36 4 1 11 0 2 2 -1948.92 -1561.18 10 0 1 36 4 1 10 0 2 2 -1915.58 -1511.39 9 0 1 36 4 1 9 0 2 2 -1900.57 -1444.83 8 0 1 36 4 1 8 0 2 2 -1850.09 -1374.15 7 0 1 36 4 1 7 0 2 2 -1871.31 -1290.12 6 0 1 36 4 1 6 0 2 2 -1834.18 -1192.58 5 0 1 36 4 1 5 0 2 2 -1847.30 -1171.96 4 0 1 36 4 1 4 0 2 2 -1751.78 -1046.34 3 0 1 36 4 1 3 0 2 2 -1686.27 -896.76 2 0 1 36 4 1 2 0 2 2 -1569.21 -782.27 1 0 1 36 4 1 1 0 2 2 -919.48 -782.27 0 0 1 36 4 1 1 0 2 2 -919.48 -782.27 --- NEW FILE: dCleanM --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2043.02 --- NEW FILE: dCleanN --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2043.37 --- NEW FILE: dMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2046.04 ICASE NB MFLOP ROUT AUTHOR 228 36 -4342.37 "ATL_dmm6x1x72_sse2.c" "R. Clint Whaley" --- NEW FILE: dNB --- 1 36 --- NEW FILE: dNCNB --- 36 --- NEW FILE: dXover.h --- #ifndef DXOVER_H #define DXOVER_H #define ATL_3NB 108 #define NN_MNK_M 20736 #define NN_MNK_N 51984 #define NN_MNK_MN 12960 #define NN_MNK_K 51984 #define NN_MNK_GE 54872 #define NT_MNK_M 3600 #define NT_MNK_N 3600 #define NT_MNK_MN 12960 #define NT_MNK_K 32400 #define NT_MNK_GE 3375 #define TN_MNK_M 51984 #define TN_MNK_N 51984 #define TN_MNK_MN 12960 #define TN_MNK_K 51984 #define TN_MNK_GE 54872 #define TT_MNK_M 3600 #define TT_MNK_N 3600 #define TT_MNK_MN 12960 #define TT_MNK_K 32400 #define TT_MNK_GE 13824 #endif --- NEW FILE: dbestNN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 18 0 2 2 -1715.18 --- NEW FILE: dbestNN_0x0x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1842.43 --- NEW FILE: dbestNN_36x36x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1840.84 --- NEW FILE: dbestNT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 18 0 2 2 -1484.56 --- NEW FILE: dbestNT_0x0x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1568.90 --- NEW FILE: dbestNT_36x36x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1571.03 --- NEW FILE: dbestTN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 18 0 2 2 -1817.69 --- NEW FILE: dbestTN_0x0x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2021.62 --- NEW FILE: dbestTN_36x36x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2025.44 --- NEW FILE: dbestTT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 18 0 2 2 -1617.66 --- NEW FILE: dbestTT_0x0x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1697.41 --- NEW FILE: dbestTT_36x36x36 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1696.08 --- NEW FILE: dgMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2046.04 --- NEW FILE: dnreg --- 13 --- NEW FILE: duCleanK --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 2 1 3 1 36 32 18 16 -1055.55 ATL_mm4x4x2_1_prefCU.c 2 228 1 36 32 18 16 -4048.30 ATL_dmm6x1x72_sse2.c --- NEW FILE: duCleanKF --- MULT ICASE FIXED NB 1 2 228 1 36 --- NEW FILE: duCleanM --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 6 228 0 36 30 18 0 -4371.78 ATL_dmm6x1x72_sse2.c --- NEW FILE: duCleanMF --- MULT ICASE FIXED NB 1 6 228 0 36 --- NEW FILE: duCleanN --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 1 228 0 36 32 18 16 -3902.63 ATL_dmm6x1x72_sse2.c --- NEW FILE: duCleanNF --- MULT ICASE FIXED NB 1 1 228 0 36 --- NEW FILE: duMMRES --- CASE NB MFLOP ROUTINE 228 36 -4342.37 "ATL_dmm6x1x72_sse2.c" "R. Clint Whaley" --- NEW FILE: sCleanK --- KB MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH GEN-MFLOP SPC-MFLOP 48 0 4 48 5 1 48 0 5 1 -2586.89 -2515.66 47 0 4 48 5 1 47 0 5 1 -2609.40 -2527.27 46 0 4 48 5 1 46 0 5 1 -2615.44 -2532.08 45 0 4 48 5 1 45 0 5 1 -2606.65 -2518.76 44 0 4 48 5 1 44 0 5 1 -2609.85 -2529.90 43 0 4 48 5 1 43 0 5 1 -2622.36 -2536.66 42 0 4 48 5 1 42 0 5 1 -2603.81 -2512.40 41 0 4 48 5 1 41 0 5 1 -2612.62 -2512.74 40 0 4 48 5 1 40 0 5 1 -2604.15 -2526.73 39 0 4 48 5 1 39 0 5 1 -2608.61 -2512.33 38 0 4 48 5 1 38 0 5 1 -2606.29 -2508.27 37 0 4 48 5 1 37 0 5 1 -2611.57 -2511.20 36 0 4 48 5 1 36 0 5 1 -2603.41 -2504.32 35 0 4 48 5 1 35 0 5 1 -2608.08 -2496.08 34 0 4 48 5 1 34 0 5 1 -2606.37 -2500.17 33 0 4 48 5 1 33 0 5 1 -2611.00 -2497.85 32 0 4 48 5 1 32 0 5 1 -2578.91 -2474.86 31 0 4 48 5 1 31 0 5 1 -2604.25 -2483.91 30 0 4 48 5 1 30 0 5 1 -2593.65 -2462.02 29 0 4 48 5 1 29 0 5 1 -2588.90 -2457.87 28 0 4 48 5 1 28 0 5 1 -2589.19 -2466.46 27 0 4 48 5 1 27 0 5 1 -2589.75 -2447.73 26 0 4 48 5 1 26 0 5 1 -2586.95 -2470.90 25 0 4 48 5 1 25 0 5 1 -2582.25 -2436.84 24 0 4 48 5 1 24 0 5 1 -2573.19 -2420.35 23 0 4 48 5 1 23 0 5 1 -2579.51 -2425.70 22 0 4 48 5 1 22 0 5 1 -2570.25 -2412.64 21 0 4 48 5 1 21 0 5 1 -2542.19 -2401.75 20 0 4 48 5 1 20 0 5 1 -2529.49 -2393.10 19 0 4 48 5 1 19 0 5 1 -2534.51 -2349.39 18 0 4 48 5 1 18 0 5 1 -2506.75 -2336.37 17 0 4 48 5 1 17 0 5 1 -2559.09 -2318.92 16 0 4 48 5 1 16 0 5 1 -2560.50 -2309.44 15 0 4 48 5 1 15 0 5 1 -2540.31 -2259.54 14 0 4 48 5 1 14 0 5 1 -2558.97 -2261.28 13 0 4 48 5 1 13 0 5 1 -2549.69 -2274.78 12 0 4 48 5 1 12 0 5 1 -2515.50 -2202.60 11 0 4 48 5 1 11 0 5 1 -2517.10 -2190.21 10 0 4 48 5 1 10 0 5 1 -2484.19 -2160.54 9 0 4 48 5 1 9 0 5 1 -2488.94 -2102.66 8 0 4 48 5 1 8 0 5 1 -2463.49 -2054.75 7 0 4 48 5 1 7 0 5 1 -2374.44 -1971.80 6 0 4 48 5 1 6 0 5 1 -2295.49 -1889.23 5 0 4 48 5 1 5 0 5 1 -2240.37 -1889.23 4 0 4 48 5 1 4 0 5 1 -2060.01 -1889.23 0 0 5 48 5 1 1 0 5 1 -1877.35 -1889.23 --- NEW FILE: sCleanM --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2591.98 --- NEW FILE: sCleanN --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2753.08 --- NEW FILE: sMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2771.94 ICASE NB MFLOP ROUT AUTHOR 220 48 -8552.81 "ATL_smm6x1x80_sse.c" "R. Clint Whaley" --- NEW FILE: sNB --- 1 48 --- NEW FILE: sNCNB --- 48 --- NEW FILE: sXover.h --- #ifndef SXOVER_H #define SXOVER_H #define ATL_3NB 144 #define NN_MNK_M 4800 #define NN_MNK_N 69312 #define NN_MNK_MN 23040 #define NN_MNK_K 106032 #define NN_MNK_GE 103823 #define NT_MNK_M 4800 #define NT_MNK_N 10800 #define NT_MNK_MN 23040 #define NT_MNK_K 43200 #define NT_MNK_GE 3375 #define TN_MNK_M 106032 #define TN_MNK_N 106032 #define TN_MNK_MN 23040 #define TN_MNK_K 106032 #define TN_MNK_GE 103823 #define TT_MNK_M 27648 #define TT_MNK_N 27648 #define TT_MNK_MN 23040 #define TT_MNK_K 69312 #define TT_MNK_GE 54872 #endif --- NEW FILE: sbestNN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 48 5 1 24 0 5 1 -2439.22 --- NEW FILE: sbestNN_0x0x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2539.71 --- NEW FILE: sbestNN_48x48x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2704.04 --- NEW FILE: sbestNT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 48 5 1 24 0 5 1 -2314.20 --- NEW FILE: sbestNT_0x0x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2396.06 --- NEW FILE: sbestNT_48x48x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2585.07 --- NEW FILE: sbestTN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 48 5 1 16 0 5 1 -2529.94 --- NEW FILE: sbestTN_0x0x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2594.09 --- NEW FILE: sbestTN_48x48x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 48 5 1 48 0 5 1 -2744.89 --- NEW FILE: sbestTT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 5 48 5 1 24 0 5 1 -2332.34 --- NEW FILE: sbestTT_0x0x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2425.89 --- NEW FILE: sbestTT_48x48x48 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 8 48 5 1 48 0 5 1 -2579.85 --- NEW FILE: sgMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 4 40 5 1 40 0 5 1 -2771.94 --- NEW FILE: snreg --- 11 --- NEW FILE: suCleanK --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 2 1 1 0 48 42 24 16 -1772.70 ATL_mm4x4x2US.c 4 220 1 48 40 24 16 -7340.67 ATL_smm6x1x80_sse.c --- NEW FILE: suCleanKF --- MULT ICASE FIXED NB 1 4 220 1 48 --- NEW FILE: suCleanM --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 4 1 311 0 48 42 24 16 -1277.11 ATL_mm6x8x8_1p.c 4 211 1 48 40 24 16 -5363.44 ATL_smm_sse1_K.c 6 220 0 48 42 24 18 -8536.80 ATL_smm6x1x80_sse.c 14 218 0 48 42 14 28 -8166.02 ATL_smm14x1x84_sseCU.c --- NEW FILE: suCleanMF --- MULT ICASE FIXED NB 3 4 211 1 48 6 220 0 48 14 218 0 48 --- NEW FILE: suCleanN --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 1 220 0 48 42 24 16 -7895.44 ATL_smm6x1x80_sse.c --- NEW FILE: suCleanNF --- MULT ICASE FIXED NB 1 1 220 0 48 --- NEW FILE: suMMRES --- CASE NB MFLOP ROUTINE 220 48 -8552.81 "ATL_smm6x1x80_sse.c" "R. Clint Whaley" --- NEW FILE: zCleanK --- KB MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH GEN-MFLOP SPC-MFLOP 36 0 1 36 4 1 36 0 2 2 -2048.22 -1833.41 35 0 1 36 4 1 35 0 2 2 -2058.49 -1833.91 34 0 1 36 4 1 34 0 2 2 -2053.62 -1801.41 33 0 1 36 4 1 33 0 2 2 -2046.89 -1846.39 32 0 1 36 4 1 32 0 2 2 -2043.29 -1836.73 31 0 1 36 4 1 31 0 2 2 -2044.35 -1803.89 30 0 1 36 4 1 30 0 2 2 -2037.49 -1805.35 29 0 1 36 4 1 29 0 2 2 -2039.92 -1815.19 28 0 1 36 4 1 28 0 2 2 -2039.35 -1797.52 27 0 1 36 4 1 27 0 2 2 -2021.20 -1808.64 26 0 1 36 4 1 26 0 2 2 -2011.47 -1780.99 25 0 1 36 4 1 25 0 2 2 -2023.02 -1774.12 24 0 1 36 4 1 24 0 2 2 -2002.13 -1748.04 23 0 1 36 4 1 23 0 2 2 -2003.71 -1736.34 22 0 1 36 4 1 22 0 2 2 -1992.56 -1702.90 21 0 1 36 4 1 21 0 2 2 -1999.55 -1686.19 20 0 1 36 4 1 20 0 2 2 -1970.37 -1710.41 19 0 1 36 4 1 19 0 2 2 -1967.24 -1708.41 18 0 1 36 4 1 18 0 2 2 -1959.59 -1644.25 17 0 1 36 4 1 17 0 2 2 -1951.79 -1720.31 16 0 1 36 4 1 16 0 2 2 -1926.09 -1659.00 15 0 1 36 4 1 15 0 2 2 -1921.31 -1649.79 14 0 1 36 4 1 14 0 2 2 -1908.27 -1601.58 13 0 1 36 4 1 13 0 2 2 -1870.23 -1569.71 12 0 1 36 4 1 12 0 2 2 -1852.31 -1524.31 11 0 1 36 4 1 11 0 2 2 -1853.96 -1531.54 10 0 1 36 4 1 10 0 2 2 -1805.62 -1471.75 9 0 1 36 4 1 9 0 2 2 -1769.03 -1439.53 8 0 1 36 4 1 8 0 2 2 -1727.67 -1347.79 7 0 1 36 4 1 7 0 2 2 -1709.34 -1277.03 6 0 1 36 4 1 6 0 2 2 -1595.56 -1160.09 5 0 1 36 4 1 5 0 2 2 -1621.88 -1085.55 4 0 1 36 4 1 4 0 2 2 -1481.19 -1000.38 3 0 1 36 4 1 3 0 2 2 -1289.60 -805.24 2 0 1 36 4 1 2 0 2 2 -1120.59 -779.71 0 0 1 36 4 1 1 0 2 2 -773.73 -779.71 --- NEW FILE: zCleanM --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2045.77 --- NEW FILE: zCleanN --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2052.41 --- NEW FILE: zMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -1887.77 ICASE NB MFLOP ROUT AUTHOR 228 36 -4200.46 "ATL_dmm6x1x72_sse2.c" "R. Clint Whaley" --- NEW FILE: zNB --- 1 36 --- NEW FILE: zNCNB --- 24 --- NEW FILE: zXover.h --- #ifndef ZXOVER_H #define ZXOVER_H #define ATL_3NB 108 #define NN_MNK_M 3600 #define NN_MNK_N 3600 #define NN_MNK_MN 12960 #define NN_MNK_K 8100 #define NN_MNK_GE 3375 #define NT_MNK_M 3600 #define NT_MNK_N 3600 #define NT_MNK_MN 12960 #define NT_MNK_K 8100 #define NT_MNK_GE 3375 #define TN_MNK_M 8100 #define TN_MNK_N 3600 #define TN_MNK_MN 12960 #define TN_MNK_K 8100 #define TN_MNK_GE 3375 #define TT_MNK_M 3600 #define TT_MNK_N 3600 #define TT_MNK_MN 12960 #define TT_MNK_K 8100 #define TT_MNK_GE 3375 #endif --- NEW FILE: zbestNN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 12 0 2 2 -1631.54 --- NEW FILE: zbestNN_0x0x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1819.72 --- NEW FILE: zbestNN_24x24x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1814.67 --- NEW FILE: zbestNT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 12 0 2 2 -1524.58 --- NEW FILE: zbestNT_0x0x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1691.86 --- NEW FILE: zbestNT_24x24x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1694.20 --- NEW FILE: zbestTN_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 12 0 2 2 -1702.20 --- NEW FILE: zbestTN_0x0x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1938.32 --- NEW FILE: zbestTN_24x24x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1934.37 --- NEW FILE: zbestTT_0x0x0 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 12 0 2 2 -1588.92 --- NEW FILE: zbestTT_0x0x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1701.32 --- NEW FILE: zbestTT_24x24x24 --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 24 4 1 24 0 2 2 -1695.00 --- NEW FILE: zgMMRES --- MULADD LAT NB MU NU KU FFTCH IFTCH NFTCH MFLOP 0 1 36 4 1 36 0 2 2 -2060.14 --- NEW FILE: zuCleanK --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 2 1 3 1 36 32 18 16 -1057.12 ATL_mm4x4x2_1_prefCU.c 2 228 1 36 32 18 16 -3888.58 ATL_dmm6x1x72_sse2.c --- NEW FILE: zuCleanKF --- MULT ICASE FIXED NB 1 2 228 1 36 --- NEW FILE: zuCleanM --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 6 228 0 36 30 18 0 -4194.45 ATL_dmm6x1x72_sse2.c --- NEW FILE: zuCleanMF --- MULT ICASE FIXED NB 1 6 228 0 36 --- NEW FILE: zuCleanN --- MULT ICASE FIXED NB NB0 NB1 NB2 MFLOP ROUT 1 1 228 0 36 32 18 16 -4005.26 ATL_dmm6x1x72_sse2.c --- NEW FILE: zuCleanNF --- MULT ICASE FIXED NB 1 1 228 0 36 --- NEW FILE: zuMMRES --- CASE NB MFLOP ROUTINE 228 36 -4200.46 "ATL_dmm6x1x72_sse2.c" "R. Clint Whaley" |