[ExecuTorch][Vulkan] Tests for et_vk.linear_q4gsw_backward#20944
Merged
Conversation
JCNTH
requested review from
SS-JIA,
kirklandsign and
larryliu0820
as code owners
July 14, 2026 21:50
pytorch-bot
Bot
added
the
module: vulkan
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20944
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 Cancelled Job, 1 Unrelated FailureAs of commit 919ce23 with merge base 21554e5 ( CANCELLED JOB - The following job was cancelled. Please retry:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Jul 14, 2026
Merged
[ExecuTorch][WebGPU] Add et_vk.adamw_step (on-GPU AdamW optimizer step) to the WebGPU backend
#20935
Merged
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
JCNTH
added a commit
that referenced
this pull request
Jul 21, 2026
Pull Request resolved: #20944 **Correctness tests for the Vulkan `et_vk.linear_q4gsw_backward` kernel** (stacked above the op diff). **Coverage:** the golden `d_x = d_out @ dequant(W)` is computed with ATen (`matmul` over the library-dequantized weight), mirroring the CPU-eager `linear_q4gsw_backward_impl` in `custom_ops_lib.py` — never a hand-rolled matmul. The packed weight is fed as a constant `tensorref` so the op exercises the real `prepack_q4_w_4x8_nc_buffer` W_4X8 path. Cases: - `test_tile_aligned` — single group, tile-aligned M/N/K. - `test_grouped` — multiple quantization groups along K. - `test_odd_n4_partial_m` — `N % 8 != 0` (odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises the `min()` clamps). Also wires `quantized_linear_backward_test` into `targets.bzl` + `CMakeLists.txt` (mirrors the sibling `linear_q4gsw_dw_test`). ghstack-source-id: 405059117 @exported-using-ghexport Differential Revision: [D111797530](https://our.internmc.facebook.com/intern/diff/D111797530/)
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.
Stack from ghstack (oldest at bottom):
Correctness tests for the Vulkan
et_vk.linear_q4gsw_backwardkernel (stacked above the op diff).Coverage: the golden
d_x = d_out @ dequant(W)is computed with ATen (matmulover the library-dequantized weight), mirroring the CPU-eagerlinear_q4gsw_backward_implincustom_ops_lib.py— never a hand-rolled matmul. The packed weight is fed as a constanttensorrefso the op exercises the realprepack_q4_w_4x8_nc_bufferW_4X8 path.Cases:
test_tile_aligned— single group, tile-aligned M/N/K.test_grouped— multiple quantization groups along K.test_odd_n4_partial_m—N % 8 != 0(odd N4 -> padded W_4X8 stride) plus a partial-M tile (exercises themin()clamps).Also wires
quantized_linear_backward_testintotargets.bzl+CMakeLists.txt(mirrors the siblinglinear_q4gsw_dw_test).@exported-using-ghexport
Differential Revision: D111797530
Differential Revision: D111797530