[ET-VK] Add nchw_to_int8x4_buffer shader for prepacking int8 staging data#17448
Merged
Conversation
…data Pull Request resolved: #17392 This adds a GLSL compute shader and supporting C++ dispatch logic to transfer int8 tensor data from a staging buffer (in NCHW contiguous order) to a GPU buffer in any PackedInt8 layout (4W, 4C, 4W4C, 4H4W, 4C1W). Previously there was no prepack path for kInt8x4 tensors, so constant int8 tensors (TensorRef inputs) could not be transferred to GPU buffers. This is needed to support constant quantized weights in q8ta operators. The shader uses texel-level dispatch where each thread writes one texel (one int32 = 4 packed int8 values). It decomposes the texel index into block-space coordinates using BufferMetadata strides and hashed_layout dim_order, then reads the corresponding int8 bytes from the staging buffer (interpreted as int32 for device compatibility, avoiding the need for 8-bit buffer support). New files: - nchw_to_int8x4_buffer.glsl: Compute shader handling all PackedInt8 layouts - nchw_to_int8x4_buffer.yaml: Shader variant config - Q8taStaging.h/cpp: C++ dispatch function creating the PrepackNode Modified files: - Staging.cpp: Routes kInt8x4 tensors in prepack_op() to the new function - TestQ8taBinary.cpp: Prepacks TensorRef inputs before quantization - test_q8ta_binary.cpp: Adds const_b test cases for constant tensor B inputs This diff was authored with Claude. ghstack-source-id: 341022576 @exported-using-ghexport Differential Revision: [D93000169](https://our.internmc.facebook.com/intern/diff/D93000169/)
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/17448
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
Bot
added
the
CLA Signed
added 2 commits
February 13, 2026 12:38
Pull Request resolved: #17393 This removes the dynamic dispatch logic in q8ta_conv2d() that selected between the im2col and general convolution paths. The function now unconditionally uses q8ta_conv2d_general(). This simplifies the dispatch since the im2col path selection will be handled upstream by the pattern matcher routing to specialized ops (q8ta_conv2d_pw, q8ta_conv2d_dw, etc.) instead of being decided at runtime. ghstack-source-id: 341022577 @exported-using-ghexport Differential Revision: [D93000164](https://our.internmc.facebook.com/intern/diff/D93000164/)
Pull Request resolved: #17394 This adds fine-grained ET_EVENT_TRACER profiling blocks to the Vulkan backend's execute() method in VulkanBackend.cpp. Previously, only GPU shader timestamps were logged. Now the following phases are individually traced: ETVK_COPY_INPUTS (host-to-GPU input transfer), ETVK_RESIZE (graph resize propagation), ETVK_COMPUTE_GRAPH_EXECUTE (GPU compute dispatch), ETVK_COPY_OUTPUTS (GPU-to-host output transfer), and ETVK_EXECUTE (overall delegate execution). The GPU shader timestamp extraction is also moved to occur right after execute() completes rather than at the end of the function, so it falls within the ETVK_EXECUTE span. ghstack-source-id: 341022578 @exported-using-ghexport Differential Revision: [D93000163](https://our.internmc.facebook.com/intern/diff/D93000163/)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR was created by the merge bot to help merge the original PR into the main branch.
ghstack PR number: #17392 by @SS-JIA
^ Please use this as the source of truth for the PR details, comments, and reviews
ghstack PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/423/base
ghstack PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/423/head
Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/SS-JIA/422/orig
Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/SS-JIA/423/orig
Differential Revision: D93000169
@diff-train-skip-merge