Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

xehpg: jit: gemm: remove unwanted k-blocking #2422

Merged
merged 1 commit into from
Jan 16, 2025
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions src/gpu/intel/jit/gemm/selector/db/kernel.db
Original file line number Diff line number Diff line change
Expand Up @@ -173,10 +173,10 @@ auto _CATALOG_ = kcatalog::toFlatCatalog({
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {4097, 1, -1}, {-1, 1, 2048}, {4097, 1, -1}, {-1, 1, 2048}, {4, 4, 1}, "H"}, "aB64 aB32x2 aB wg 2x1x8 ikr nse hi ar sb64 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 2}, {65536, 16384, 16777216}, {65536, 16384, 16777216}, {4, 1, 64}, {2, 1, 8}, 1, (WGType) 0, 4357, 0, 64, {4, 4, 4}, {true, true, true}}, {'W', 1, {4}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {1025, 1, -1}, {4096, 1, 4096}, {1025, 1, -1}, {4096, 1, 4096}, {4, 4, 1}, "H"}, "aB16x2 aB128 aB wg 2x1x4 ikr wx4 nse hi ar sb128 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 2}, {32768, 16384, 16777216}, {32768, 16384, 16777216}, {2, 1, 128}, {2, 1, 4}, 4, (WGType) 1, 4357, 0, 64, {4, 4, 4}, {true, true, true}}, {'W', 1, {2}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {1025, 1, 4097}, {4096, 1, -1}, {1025, 1, 4097}, {4096, 1, -1}, {4, 4, 1}, "H"}, "aB16 aB128 aB wg 2x1x16 ikr nse hi ar sb128 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 2}, {65536, 16384, 16777216}, {65536, 16384, 16777216}, {4, 1, 128}, {2, 1, 16}, 1, (WGType) 0, 4357, 0, 64, {4, 4, 4}, {true, true, true}}, {'W', 1, {4}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ioxyz"}, "sS64x2 sB16 sB wg 16x2 cb4 ks32 xaf dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 64}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {978625, 470744, 0, 0, 0, 0, 1.5997, 5.18308, 6.10237, 17.1883, 0.0161902, 0.0142461, 0.0050822, 0.870423, 1.37496, 1.20524, 9.37477e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixyz"}, "sS32x2 sB16 sB wg 16x1x2 ikr cb4 ks64 xaf dw vav bo sr dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 2}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 64}, {16, 1, 2}, 1, (WGType) 1, 4357, 32768, 32768, {4, 4, 4}, {false, false, false}}, {'E', 17, {849827, -83056.2, 214309, 623119, 0, 0, 1.34807, 5.05315, 5.40586, 15.1632, 0.0173594, 0.0309966, 0.00618382, 0.767366, 1.37973, 1.22172, 6.63552e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixyz"}, "sS32x2 sB16 sB wg 16x2 cb4 ks32 xaf dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {982213, 473301, 0, 0, 0, 0, 1.74644, 5.1767, 6.10829, 17.1708, 0.0167439, 0.0136956, 0.00599404, 0.999577, 1.37511, 1.22059, 7.53689e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Isxyz"}, "sS64 sB16x2 sB wg 16x2 cb4 ks64 xaf fx dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 786432, 16777216}, {262144, 786432, 16777216}, {16, 48, 64}, {16, 2, 1}, 1, (WGType) 1, 257, 49152, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {981842, 458926, 0, 0, 0, 0, 1.5015, 5.00498, 6.31005, 16.9024, 0.0169639, 0.0400974, 0, 0.719651, 1.35848, 1.1845, 9.89658e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ioxyz"}, "sS64x2 sB16 sB wg 16x2 cb4 ks32 xaf dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 64}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {978625, 470744, 0, 0, 0, 0, 1.5997, 5.18308, 6.10237, 17.1883, 0.0161902, 0.0142461, 0.0050822, 0.870423, 1.37496, 1.20524, 9.37477e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixyz"}, "sS32x2 sB16 sB wg 16x1x2 ikr cb4 ks64 xaf dw vav bo sr bk0 dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 2}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 64}, {16, 1, 2}, 1, (WGType) 1, 4357, 32768, 32768, {4, 4, 4}, {false, false, false}}, {'E', 17, {849827, -83056.2, 214309, 623119, 0, 0, 1.34807, 5.05315, 5.40586, 15.1632, 0.0173594, 0.0309966, 0.00618382, 0.767366, 1.37973, 1.22172, 6.63552e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixyz"}, "sS32x2 sB16 sB wg 16x2 cb4 ks32 xaf dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {982213, 473301, 0, 0, 0, 0, 1.74644, 5.1767, 6.10829, 17.1708, 0.0167439, 0.0136956, 0.00599404, 0.999577, 1.37511, 1.22059, 7.53689e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Isxyz"}, "sS64 sB16x2 sB wg 16x2 cb4 ks64 xaf fx dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 786432, 16777216}, {262144, 786432, 16777216}, {16, 48, 64}, {16, 2, 1}, 1, (WGType) 1, 257, 49152, 0, {4, 4, 4}, {false, false, false}}, {'E', 17, {981842, 458926, 0, 0, 0, 0, 1.5015, 5.00498, 6.31005, 16.9024, 0.0169639, 0.0400974, 0, 0.719651, 1.35848, 1.1845, 9.89658e-13}}},
{{'E', "gemm", {"F", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, 33, -1}, {-1, -1, -1}, {4, 4, 1}, "xyIs"}, "sB16 sB32 aB wg 4x8 cab3x2 ks32 xaf st dw vav bo sr bk0 dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {524288, 393216, 16777216}, {524288, 393216, 16777216}, {32, 24, 32}, {4, 8, 1}, 1, (WGType) 1, 257, 61440, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {1.01302e+06, 570829, 0, 0, 0, 0, 3.67307, 6.66635, 6.86396, 18.2302, 0.0202076, 0.0155595, 0.00597746, 1, 1.56109, 1.12816, 4.46535e-12}}},
{{'E', "gemm", {"F", "O", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, 8, -1}, {-1, -1, -1}, {-1, 8, -1}, {4, 4, 1}, "Iqxy"}, "sB64 sS16 aS wg 2x1x8 ikr af acb sr bk0 bm0 sys pab grf256 rc0", {8, (LoopType) 0, 256, {(LoopType) 0, (LoopType) 1, (LoopType) 2}, {16777216, 8192, 16777216}, {8192, 8192, 16777216}, {16, 8, 128}, {2, 1, 8}, 1, (WGType) 0, 4357, 0, 1024, {4, 4, 4}, {false, false, true}}, {'E', 17, {3.5449e+06, 60571.4, -243099, 15595.1, 0, 0, 1.78243, 2.8889, 2.76679, 6.10171, 0.051381, 0.0216118, 0.0510683, 1, 1.21576, 1.21633, -9.23968e-14}}},
{{'E', "gemm", {"F", "O", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, 8, -1}, {-1, -1, -1}, {-1, 8, -1}, {4, 4, 1}, "Iqxy"}, "sB64 sS16 aS wg 2x1x8 ikr af acb sr bk0 bm0 sys pab grf256 rc0", {8, (LoopType) 0, 256, {(LoopType) 0, (LoopType) 1, (LoopType) 2}, {16777216, 8192, 16777216}, {8192, 8192, 16777216}, {16, 8, 128}, {2, 1, 8}, 1, (WGType) 0, 4357, 0, 1024, {4, 4, 4}, {false, false, true}}, {'E', 17, {3.5449e+06, 60571.4, -243099, 15595.1, 0, 0, 1.78243, 2.8889, 2.76679, 6.10171, 0.051381, 0.0216118, 0.0510683, 1, 1.21576, 1.21633, -9.23968e-14}}},
Expand Down Expand Up @@ -287,10 +287,10 @@ auto _CATALOG_ = kcatalog::toFlatCatalog({
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {1025, 1, 4097}, {-1, 1, -1}, {1025, 1, 4097}, {-1, 1, -1}, {1, 1, 1}, "H"}, "aB128 aB64x2 aB wg 4x1x8 ikr nse hi ar sb128 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 2}, {32768, 16384, 16777216}, {32768, 16384, 16777216}, {2, 1, 128}, {2, 1, 8}, 1, (WGType) 0, 4357, 0, 64, {1, 2, 4}, {true, true, true}}, {'W', 1, {2}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {1025, 1, 513}, {-1, 1, 4096}, {1025, 1, 513}, {-1, 1, 4096}, {1, 1, 1}, "H"}, "aB128x2 aB128x2 aB wg 16x1 wx2 nse hi ar sb128 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 255}, {16384, 16384, 16777216}, {16384, 16384, 16777216}, {1, 1, 128}, {16, 1, 1}, 2, (WGType) 1, 257, 0, 0, {1, 2, 4}, {true, true, true}}, {'W', 1, {1}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, 2, -1}, {4095, 2, 4095}, {-1, 2, -1}, {4095, 2, 4095}, {4, 4, 1}, "xy"}, "sB128 sB64 aB wg 8x1x4 ikr nse hi ar sb128 bk0 dot", {8, (LoopType) 0, 128, {(LoopType) 144, (LoopType) 255, (LoopType) 2}, {32768, 32768, 16777216}, {32768, 32768, 16777216}, {2, 2, 128}, {8, 1, 4}, 1, (WGType) 0, 4357, 0, 256, {4, 4, 4}, {false, false, true}}, {'W', 1, {128}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "I"}, "aB32 aB16 aB ca3 ks64 wg 2x4x4 kr sys dw af k192 grf256 sm vav di dm sr cc fm", {8, (LoopType) 0, 256, {(LoopType) 0, (LoopType) 1, (LoopType) 2}, {8192, 8192, 2048}, {8192, 8192, 2048}, {16, 16, 192}, {2, 4, 4}, 1, (WGType) 1, 5, 12288, 12288, {4, 4, 2}, {true, true, true}}, {'E', 17, {4.488e+06, 120208, 122348, 327.657, 0, 0, 3.41911, 6.33998, 2.77024, 6.84323, 0.0582208, 0.0145417, 0.0574819, 0.511227, 1.20562, 1.20018, 5.11052e-14}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Isxy"}, "sS32x2 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 786432, 16777216}, {262144, 786432, 16777216}, {16, 48, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 24576, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {995629, 475244, 0, 0, 0, 0, 2.57281, 4.9973, 6.41839, 16.8374, 0.018098, 0.0100962, 0.0118445, 0.996594, 1.41614, 1.19695, 9.60059e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, 2048}, {-1, 1024, -1}, {4, 4, 1}, "Ixy"}, "sS32x2 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 524288, 16777216}, {262144, 524288, 16777216}, {16, 32, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 16384, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {985865, 449373, 0, 0, 0, 0, 2.52021, 5.18432, 6.41111, 16.2959, 0.0196104, 0.0078899, 0.0176455, 0.975623, 1.37619, 1.19224, 7.58759e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixy"}, "sS32 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {996386, 490462, 0, 0, 0, 0, 2.66142, 5.16946, 6.13711, 17.1234, 0.0173418, 0.00858928, 0.0131118, 0.74711, 1.38311, 1.23925, 7.08247e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "I"}, "aB32 aB16 aB ca3 ks64 wg 2x4x4 kr sys dw af k192 grf256 sm vav di dm sr cc fm", {8, (LoopType) 0, 256, {(LoopType) 0, (LoopType) 1, (LoopType) 2}, {8192, 8192, 16777216}, {8192, 8192, 16777216}, {16, 16, 192}, {2, 4, 4}, 1, (WGType) 1, 261, 12288, 12288, {4, 4, 4}, {true, true, true}}, {'E', 17, {4.488e+06, 120208, 122348, 327.657, 0, 0, 3.41911, 6.33998, 2.77024, 6.84323, 0.0582208, 0.0145417, 0.0574819, 0.511227, 1.20562, 1.20018, 5.11052e-14}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Isxy"}, "sS32x2 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 786432, 16777216}, {262144, 786432, 16777216}, {16, 48, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 24576, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {995629, 475244, 0, 0, 0, 0, 2.57281, 4.9973, 6.41839, 16.8374, 0.018098, 0.0100962, 0.0118445, 0.996594, 1.41614, 1.19695, 9.60059e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, 2048}, {-1, 1024, -1}, {4, 4, 1}, "Ixy"}, "sS32x2 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 524288, 16777216}, {262144, 524288, 16777216}, {16, 32, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 16384, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {985865, 449373, 0, 0, 0, 0, 2.52021, 5.18432, 6.41111, 16.2959, 0.0196104, 0.0078899, 0.0176455, 0.975623, 1.37619, 1.19224, 7.58759e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Ixy"}, "sS32 sB16 aB wg 16x2 cb4 ks32 xaf dw vav bo sr bk0 sn dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 1048576, 16777216}, {262144, 1048576, 16777216}, {16, 64, 32}, {16, 2, 1}, 1, (WGType) 1, 257, 32768, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {996386, 490462, 0, 0, 0, 0, 2.66142, 5.16946, 6.13711, 17.1234, 0.0173418, 0.00858928, 0.0131118, 0.74711, 1.38311, 1.23925, 7.08247e-13}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {4, 4, 1}, "Isxy"}, "sB32 sB16 aB wg 8x4 cab3x2 ks32 xaf fx dw vav bo sr bk0 dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 786432, 16777216}, {262144, 786432, 16777216}, {16, 48, 32}, {8, 4, 1}, 1, (WGType) 1, 257, 61440, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {1.05802e+06, 544339, 0, 0, 0, 0, 3.68162, 5.47347, 6.43127, 16.927, 0.0189051, 0.00830613, 0.0176655, 0.737745, 1.33714, 1.26074, -3.55719e-14}}},
{{'E', "gemm", {"O", "H", "S"}, {"T", "N", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, 33, -1}, {-1, -1, -1}, {4, 4, 1}, "xyI"}, "sB32 sB16 aB wg 8x4 cab4x2 ks32 xaf st dw vav bo sr bk0 dm grf256 sys pab", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {262144, 524288, 16777216}, {262144, 524288, 16777216}, {16, 32, 32}, {8, 4, 1}, 1, (WGType) 1, 257, 65536, 0, {4, 4, 4}, {false, false, true}}, {'E', 17, {966409, 495159, 0, 0, 0, 0, 3.17717, 6.26529, 6.93941, 17.3959, 0.0210983, 0.0147392, 0.00938379, 1, 1.72262, 1.08677, 5.49901e-12}}},
{{'E', "gemm", {"O", "O", "I"}, {"A4#8,32", "B32", "N"}}, {-1, -1, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {-1, -1, -1}, {1, 1, 1}, "xyzdsm"}, "sB16 sB16 sb fs wg 4x4 bo acb bk8192 sr", {8, (LoopType) 0, 256, {(LoopType) 128, (LoopType) 255, (LoopType) 255}, {524288, 786432, 8192}, {8192, 8192, 8192}, {32, 48, 32}, {4, 4, 1}, 1, (WGType) 1, 256, 32256, 0, {128, 128, 4}, {false, false, false}}, {'W', 1, {1536}}},
Expand Down
Loading