Skip to content

feat/hygon: chunk/recurrent deltanet #1377

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
wooway777 merged 1 commit into from
Jul 8, 2026
+28 −4
Merged

feat/hygon: chunk/recurrent deltanet #1377

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 13 additions & 1 deletion src/infiniop/ops/chunk_gated_delta_rule/operator.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
#include "../../handle.h"
#include "infiniop/ops/chunk_gated_delta_rule.h"

#if defined(ENABLE_NVIDIA_API)
#if defined(ENABLE_NVIDIA_API) || defined(ENABLE_HYGON_API)
#include "nvidia/chunk_gated_delta_rule_nvidia.cuh"
#endif

Expand Down Expand Up @@ -41,6 +41,9 @@ __INFINI_C infiniStatus_t infiniopCreateChunkGatedDeltaRuleDescriptor(
#ifdef ENABLE_NVIDIA_API
CREATE(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
CREATE(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand All @@ -61,6 +64,9 @@ __INFINI_C infiniStatus_t infiniopGetChunkGatedDeltaRuleWorkspaceSize(
#ifdef ENABLE_NVIDIA_API
GET(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
GET(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand Down Expand Up @@ -88,6 +94,9 @@ __INFINI_C infiniStatus_t infiniopChunkGatedDeltaRule(
#ifdef ENABLE_NVIDIA_API
CALCULATE(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
CALCULATE(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand All @@ -107,6 +116,9 @@ __INFINI_C infiniStatus_t infiniopDestroyChunkGatedDeltaRuleDescriptor(
#ifdef ENABLE_NVIDIA_API
DESTROY(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
DESTROY(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand Down
4 changes: 2 additions & 2 deletions src/infiniop/ops/recurrent_gated_delta_rule/cuda/kernel.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,9 +35,9 @@ __device__ inline float loadAsFloat<cuda_bfloat16>(const cuda_bfloat16 *ptr, ptr
__device__ inline float warpReduceSum(float value) {
#pragma unroll
for (int offset = 16; offset > 0; offset >>= 1) {
value += __shfl_down_sync(0xffffffff, value, offset);
value += __shfl_down_sync(0xffffffff, value, offset, 32);
}
return __shfl_sync(0xffffffff, value, 0);
return __shfl_sync(0xffffffff, value, 0, 32);
}
template <typename Tdata, typename Tgate, typename Tcompute, size_t Dk, size_t Dv, size_t WARPS_PER_BLOCK>
__device__ void recurrentGatedDeltaRuleIndexedPoolWarpKernel(
Expand Down
14 changes: 13 additions & 1 deletion src/infiniop/ops/recurrent_gated_delta_rule/operator.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
#include "../../handle.h"
#include "infiniop/ops/recurrent_gated_delta_rule.h"

#if defined(ENABLE_NVIDIA_API)
#if defined(ENABLE_NVIDIA_API) || defined(ENABLE_HYGON_API)
#include "nvidia/recurrent_gated_delta_rule_nvidia.cuh"
#endif

Expand Down Expand Up @@ -37,6 +37,9 @@ __INFINI_C infiniStatus_t infiniopCreateRecurrentGatedDeltaRuleDescriptor(
#ifdef ENABLE_NVIDIA_API
CREATE(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
CREATE(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand All @@ -57,6 +60,9 @@ __INFINI_C infiniStatus_t infiniopGetRecurrentGatedDeltaRuleWorkspaceSize(
#ifdef ENABLE_NVIDIA_API
GET(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
GET(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand Down Expand Up @@ -85,6 +91,9 @@ __INFINI_C infiniStatus_t infiniopRecurrentGatedDeltaRule(
#ifdef ENABLE_NVIDIA_API
CALCULATE(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
CALCULATE(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand All @@ -104,6 +113,9 @@ __INFINI_C infiniStatus_t infiniopDestroyRecurrentGatedDeltaRuleDescriptor(
#ifdef ENABLE_NVIDIA_API
DESTROY(INFINI_DEVICE_NVIDIA, nvidia)
#endif
#ifdef ENABLE_HYGON_API
DESTROY(INFINI_DEVICE_HYGON, nvidia)
#endif

default:
return INFINI_STATUS_DEVICE_TYPE_NOT_SUPPORTED;
Expand Down
Loading