fix(cuda): handle 64-lane warps in broadcast backward reduction - #216
Open
chen2021673 wants to merge 3 commits into
Open
chen2021673 wants to merge 3 commits into
chen2021673 wants to merge 3 commits into
Conversation
Chamberlain0w0
approved these changes
Sep 14, 2026
kilinchange
requested changes
Sep 17, 2026
| const int physical_lane = tid % warpSize; | ||
| const int logical_base = (physical_lane / kWarpSize) * kWarpSize; | ||
| const WarpMask logical_lane_mask = static_cast<WarpMask>(uint64_t{0xffffffff} << logical_base); | ||
| const WarpMask active_mask = physical_active_mask & logical_lane_mask; |
Collaborator
There was a problem hiding this comment.
这个 pr 的主要解决的问题是,当 physical warp size != 32 时,能够正确得到当前 logical warp 内真正 in_bounds 的线程。
但这里有两个容易混淆的概念,建议区分一下 warpSize 和 kWarpSize 的语义:
- warpSize 是 CUDA built-in variable,表示 hardware physical warp size;
- kWarpSize 实际表示这个 kernel 使用的 logical warp size。
当前实现中的
uint64_t{0xffffffff} << logical_base
和
unsigned logical_active_mask
都隐含了 logical warp 固定为 32 lane 的假设,因此建议显式增加:
static_assert(
kWarpSize == 32,
"BinaryBackwardKernel assumes a 32-lane logical warp");
同时建议将 kWarpSize 重命名为 kLogicalWarpSize,避免与 built-in warpSize 混淆。
Contributor
Author
There was a problem hiding this comment.
已修改,显式增加 static_assert 并将 kWarpSize 重命名为 kLogicalWarpSize。
chen2021673
force-pushed
the
fix/cuda-64-lane-warp-reduction
branch
from
September 18, 2026 05:04
93f8603 to
dea656b
Compare
Rename kWarpSize to kLogicalWarpSize to distinguish it from the CUDA built-in warpSize, which represents the physical hardware warp size. Also add a static assertion documenting that BinaryBackwardKernel assumes a 32-lane logical warp, since its mask construction and active-lane handling currently rely on that layout.
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
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.
背景
BroadCast 反向传播 kernel 之前默认物理 WarpSize 为 32,在 32-lane 设备上运行正常,但在部分 WarpSize 为 64 的设备上会出现规约错误。
修改内容
测试
新增
{2, 64} * {2, 1}Broadcast Mul Backward 用例,验证:grad_a均为2grad_b为{64, 64}maca C550 WarpSize=64 环境中测试:

CUDA 测试:

