Search results

[GPU] linalg.pack op invalid tile factor or output size provided

bug 🐞Something isn't workingSomething isn't working

Status: Open.

#22026 In iree-org/iree;

· dezhiAmd opened on Sep 17, 2025

Attention performance regresses with FullMaskedLoadToConditionalLoad pattern

Status: Open.

#22024 In iree-org/iree;

· Max191 opened on Sep 17, 2025

YOLOv12 Compilation from PyTorch

Status: Open.

#22022 In iree-org/iree;

· jishminor opened on Sep 17, 2025

Scaled MFMA failing at GPUCheckResourceUsagePass

Status: Open.

#22018 In iree-org/iree;

· Muzammiluddin-Syed-ECE opened on Sep 17, 2025

Scaled MFMA failing at GPUVerifyDistributionPass

Status: Open.

#22017 In iree-org/iree;

· Muzammiluddin-Syed-ECE opened on Sep 17, 2025

[CompilationError] : error: 'llvm.extractelement' op operand #0 must be LLVM dialect-compatible vector type, but got '!llvm.array<1 x vector<2xf32>>'

bug 🐞Something isn't workingSomething isn't working

Status: Open.

#22015 In iree-org/iree;

· amd-vivekag opened on Sep 17, 2025

[CPU][RISCV] Failed to compile gather->attention because of large vectors

Status: Open.

#22013 In iree-org/iree;

· hanhanW opened on Sep 17, 2025

Performance improvement for input backward convolutions

Status: Open.

#22009 In iree-org/iree;

· yzhang93 opened on Sep 16, 2025

[Dispatch Creation] Prefer fusing broadcast with attention

Status: Open.

#22005 In iree-org/iree;

· IanWood1 opened on Sep 16, 2025

Export Llama 405 IR with MLIR mxfp4 scaled matmul kernels

Status: Open.

#22002 In iree-org/iree;

· jtuyls opened on Sep 16, 2025

Enable more C++ warnings?

Status: Open.

#21996 In iree-org/iree;

· schuermans-roofline opened on Sep 16, 2025

[GPU][DT] Failed to compile toy_llama.mlir when const-eval is off

Status: Open.

#21990 In iree-org/iree;

· hanhanW opened on Sep 15, 2025