submission 66965 · KernelIndex

Use it

Vendorable · source mirrored · license unknownView source →

No package. Vendor the mirrored source: 47 lines, June 9 Researcher Reciprocity License v1.0.

submission.py

curl "https://kernelindex.com/api/v1/implementations/kernelbot-vectorsum-v2-66965?include=source"

interfacepython

Compatibility

measured onNVIDIA B200

declared hardwareNVIDIA B200

architecturessm_100

dtypesfp32

Benchmark evidence

1 measurement across 1 GPU, fastest first.

Operation / workload

Hardware

Latency

Rank

Observed

Vector sum reductionsuite of 6 cases

NVIDIA B200

48.2µs

#19 of 88

2025-11-05

Run →

Reported · How evidence levels are derived →

Source and license

sourceavailable

revision digestsha256:7116642cb27fdb77c4d962c60271d3aa2dd8d585c606cdf9075fce6a0294fcc1

license declaredunknown

license concludedunknown

authorsSaint of the Famished

imported2026-08-15

Kernel source

submission.py47 lines

#!POPCORN leaderboard vectorsum_v2

import torch
import triton
import triton.language as tl
from task import input_t, output_t


@triton.jit
def sum_kernel(
    x_ptr,
    partial_sums_ptr,
    n_elements,
    BLOCK_SIZE: tl.constexpr,
):
    pid = tl.program_id(0)
    block_start = pid * BLOCK_SIZE
    offsets = block_start + tl.arange(0, BLOCK_SIZE)
    mask = offsets < n_elements
    x = tl.load(x_ptr + offsets, mask=mask, other=0.0, eviction_policy="evict_first")
    block_sum = tl.sum(x, axis=0)
    tl.store(partial_sums_ptr + pid, block_sum)


def custom_kernel(data: input_t) -> output_t:
    input, output = data
    n_elements = input.numel()

    if n_elements >= 10_000_000:
        BLOCK_SIZE = 8192
    elif n_elements >= 1_000_000:
        BLOCK_SIZE = 4096
    elif n_elements >= 100_000:
        BLOCK_SIZE = 2048
    elif n_elements >= 10_000:
        BLOCK_SIZE = 1024
    else:
        BLOCK_SIZE = 512

    n_blocks = triton.cdiv(n_elements, BLOCK_SIZE)
    partial_sums = torch.empty(n_blocks, device=input.device, dtype=input.dtype)

    grid = (n_blocks,)
    sum_kernel[grid](input, partial_sums, n_elements, BLOCK_SIZE=BLOCK_SIZE)

    return partial_sums.sum()

scrolls · 47 lines total

Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0

Changes from previous submission

Against this author's previous submission submission 66749.

  ⋯ 6 unchanged lines
   
   
  @triton.jit
- def sum_kernel_optimized(
+ def sum_kernel(
      x_ptr,
      partial_sums_ptr,
      n_elements,
  ⋯ 8 unchanged lines
      tl.store(partial_sums_ptr + pid, block_sum)
   
   
- @triton.jit
- def sum_kernel_atomic(
-     x_ptr,
-     output_ptr,
-     n_elements,
-     BLOCK_SIZE: tl.constexpr,
- ):
-     pid = tl.program_id(0)
-     block_start = pid * BLOCK_SIZE
-     offsets = block_start + tl.arange(0, BLOCK_SIZE)
-     mask = offsets < n_elements
-  
-     x = tl.load(x_ptr + offsets, mask=mask, other=0.0, eviction_policy="evict_first")
-     block_sum = tl.sum(x, axis=0)
-  
-     tl.atomic_add(output_ptr, block_sum)
-  
-  
  def custom_kernel(data: input_t) -> output_t:
      input, output = data
      n_elements = input.numel()
   
-     BLOCK_SIZE = 2**10
+     if n_elements >= 10_000_000:
+         BLOCK_SIZE = 8192
+     elif n_elements >= 1_000_000:
+         BLOCK_SIZE = 4096
+     elif n_elements >= 100_000:
+         BLOCK_SIZE = 2048
+     elif n_elements >= 10_000:
+         BLOCK_SIZE = 1024
+     else:
+         BLOCK_SIZE = 512
   
      n_blocks = triton.cdiv(n_elements, BLOCK_SIZE)
      partial_sums = torch.empty(n_blocks, device=input.device, dtype=input.dtype)
   
      grid = (n_blocks,)
-     sum_kernel_optimized[grid](input, partial_sums, n_elements, BLOCK_SIZE=BLOCK_SIZE)
+     sum_kernel[grid](input, partial_sums, n_elements, BLOCK_SIZE=BLOCK_SIZE)
   
-     while partial_sums.numel() > BLOCK_SIZE:
-         n_elements = partial_sums.numel()
-         n_blocks = triton.cdiv(n_elements, BLOCK_SIZE)
-         next_level = torch.empty(n_blocks, device=input.device, dtype=input.dtype)
-  
-         grid = (n_blocks,)
-         sum_kernel_optimized[grid](partial_sums, next_level, n_elements, BLOCK_SIZE=BLOCK_SIZE)
-         partial_sums = next_level
-  
-     if partial_sums.numel() <= 128:
-         final_output = torch.zeros(1, device=input.device, dtype=input.dtype)
-         n_elements = partial_sums.numel()
-         sum_kernel_atomic[(1,)](partial_sums, final_output, n_elements, BLOCK_SIZE=BLOCK_SIZE)
-         return final_output[0]
-     else:
-         n_elements = partial_sums.numel()
-         n_blocks = triton.cdiv(n_elements, BLOCK_SIZE)
-         next_level = torch.empty(n_blocks, device=input.device, dtype=input.dtype)
-  
-         grid = (n_blocks,)
-         sum_kernel_optimized[grid](partial_sums, next_level, n_elements, BLOCK_SIZE=BLOCK_SIZE)
-         return next_level.sum()
+     return partial_sums.sum()

scrolls · 77 diff lines total

Best evidence level for this revision: reported

JSON