From 49ee0dc6472e47a9ae3bd4584ca7e558cf1599db Mon Sep 17 00:00:00 2001 From: Disty0 Date: Tue, 14 Jul 2026 02:16:33 +0300 Subject: [PATCH] set group_size_m to 8 on intel too --- modules/sdnq/kernels/triton_mm.py | 2 +- modules/sdnq/kernels/triton_scaled_mm.py | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/modules/sdnq/kernels/triton_mm.py b/modules/sdnq/kernels/triton_mm.py index c13167cd3..ac35a410c 100644 --- a/modules/sdnq/kernels/triton_mm.py +++ b/modules/sdnq/kernels/triton_mm.py @@ -12,7 +12,7 @@ matmul_configs = [ for BM in [int(BM) for BM in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_M_LIST", "64,128").replace(" ","").split(",")] for BN in [int(BN) for BN in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_N_LIST", "64,128,256").replace(" ","").split(",")] for BK in [int(BK) for BK in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_K_LIST", "32,64,128").replace(" ","").split(",")] - for GM in [int(GM) for GM in os.environ.get("SDNQ_TRITON_MM_GROUP_SIZE_M_LIST", "16" if torch.xpu.is_available() else "8").replace(" ","").split(",")] + for GM in [int(GM) for GM in os.environ.get("SDNQ_TRITON_MM_GROUP_SIZE_M_LIST", "8").replace(" ","").split(",")] for w in [int(w) for w in os.environ.get("SDNQ_TRITON_MM_NUM_WARPS_LIST", "16" if torch.xpu.is_available() else "4").replace(" ","").split(",")] for s in [int(s) for s in os.environ.get("SDNQ_TRITON_MM_NUM_STAGES_LIST", "2").replace(" ","").split(",")] ] diff --git a/modules/sdnq/kernels/triton_scaled_mm.py b/modules/sdnq/kernels/triton_scaled_mm.py index 60f6e1c7d..18779e74d 100644 --- a/modules/sdnq/kernels/triton_scaled_mm.py +++ b/modules/sdnq/kernels/triton_scaled_mm.py @@ -17,7 +17,7 @@ matmul_configs = [ for BM in [int(BM) for BM in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_M_LIST", "64,128").replace(" ","").split(",")] for BN in [int(BN) for BN in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_N_LIST", "64,128,256").replace(" ","").split(",")] for BK in [int(BK) for BK in os.environ.get("SDNQ_TRITON_MM_BLOCK_SIZE_K_LIST", "32,64,128").replace(" ","").split(",")] - for GM in [int(GM) for GM in os.environ.get("SDNQ_TRITON_MM_GROUP_SIZE_M_LIST", "16" if torch.xpu.is_available() else "8").replace(" ","").split(",")] + for GM in [int(GM) for GM in os.environ.get("SDNQ_TRITON_MM_GROUP_SIZE_M_LIST", "8").replace(" ","").split(",")] for w in [int(w) for w in os.environ.get("SDNQ_TRITON_MM_NUM_WARPS_LIST", "16" if torch.xpu.is_available() else "4").replace(" ","").split(",")] for s in [int(s) for s in os.environ.get("SDNQ_TRITON_MM_NUM_STAGES_LIST", "2").replace(" ","").split(",")] ]