Open winglian opened 1 year ago
anyone else see this error during fine-tuning? Cuda backend works without issue
Traceback (most recent call last): File "<string>", line 21, in matmul_248_kernel KeyError: ('2-.-0-.-0-d82511111ad128294e9d31a6ac684238-d6252949da17ceb5f3a278a70250af13-1af5134066c618146d2cd009138944a0-8c8cc25952b3bfbcd6169fbc5cc570c6-3498c340fd4b6ee7805fd54b882a04f5-e1f133f98d04093da2078dfc51c36b72-b26258bf01f839199e39d64851821f26-d7c0 6e3b46e708006c15224aac7a1378-f585402118c8a136948ce0a49cfe122c', (torch.float16, torch.int32, torch.float16, torch.float16, torch.float16, None, 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32', 'i32'), (256, 64, 32, 8), (Tr ue, True, True, True, True, (False,), (True, False), (True, False), (True, False), (False, False), (False, False), (True, False), (False, True), (True, False), (False, True), (True, False), (False, True), (False, True), (False, True))) During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 937, in build_triton_ir generator.visit(fn.parse()) File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 855, in visit return super().visit(node) ... File "<string>", line 41, in matmul_248_kernel File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 1620, in compile next_module = compile(module) File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 1549, in <lambda> lambda src: ast_to_ttir(src, signature, configs[0], constants)), File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 962, in ast_to_ttir mod, _ = build_triton_ir(fn, signature, specialization, constants) File "/opt/conda/lib/python3.8/site-packages/triton/compiler.py", line 942, in build_triton_ir raise CompilationError(fn.src, node) from e triton.compiler.CompilationError: at 39:21: def matmul_248_kernel(a_ptr, b_ptr, c_ptr, scales_ptr, zeros_ptr, g_ptr, M, N, K, bits, maxq, stride_am, stride_ak, stride_bk, stride_bn, stride_cm, stride_cn, stride_scales, stride_zeros, BLOCK_SIZE_M: tl.constexpr, BLOCK_SIZE_N: tl.constexpr, BLOCK_SIZE_K: tl.constexpr, GROUP_SIZE_M: tl.constexpr): """ Compute the matrix multiplication C = A x B. A is of shape (M, K) float16 B is of shape (K//8, N) int32 C is of shape (M, N) float16 scales is of shape (G, N) float16 zeros is of shape (G, N) float16 g_ptr is of shape (K) int32 """ infearure_per_bits = 32 // bits pid = tl.program_id(axis=0) num_pid_m = tl.cdiv(M, BLOCK_SIZE_M) num_pid_n = tl.cdiv(N, BLOCK_SIZE_N) num_pid_k = tl.cdiv(K, BLOCK_SIZE_K) num_pid_in_group = GROUP_SIZE_M * num_pid_n group_id = pid // num_pid_in_group first_pid_m = group_id * GROUP_SIZE_M group_size_m = min(num_pid_m - first_pid_m, GROUP_SIZE_M) pid_m = first_pid_m + (pid % group_size_m) pid_n = (pid % num_pid_in_group) // group_size_m offs_am = pid_m * BLOCK_SIZE_M + tl.arange(0, BLOCK_SIZE_M) offs_bn = pid_n * BLOCK_SIZE_N + tl.arange(0, BLOCK_SIZE_N) offs_k = tl.arange(0, BLOCK_SIZE_K) a_ptrs = a_ptr + (offs_am[:, None] * stride_am + offs_k[None, :] * stride_ak) # (BLOCK_SIZE_M, BLOCK_SIZE_K) a_mask = (offs_am[:, None] < M) # b_ptrs is set up such that it repeats elements along the K axis 8 times b_ptrs = b_ptr + ((offs_k[:, None] // infearure_per_bits) * stride_bk + offs_bn[None, :] * stride_bn) # (BLOCK_SIZE_K, BLOCK_SIZE_N) g_ptrs = g_ptr + offs_k ^
anyone else see this error during fine-tuning? Cuda backend works without issue