AddingZero
/
op_optimization
forked from metax-maca/op_optimization
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

<docs>(flashinfer): gitlink格式适配

This commit is contained in:
MaseChen 2026-06-25 14:03:51 +08:00
parent a5cbd78004
commit 10d459e808
1 changed files with 9 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
基于AI Agent开发范式的国产GPU大模型推理算子库优化/FlashInfer Benchmark实战：从性能基线到XPU-OJ冒烟提交.md
View File

@ -566,14 +566,14 @@ FlashInfer 方向包含 **4 个可选算子题目**，每个对应独立的 benc
- **固定参数**：`num_qo_heads = 32`、`num_kv_heads = 4`、`head_dim_qk = 128`、`head_dim_vo = 128`、`causal = 1`、数据类型 `bfloat16`
- **可变参数**（共 **12 个测试用例**）：
| batch_size | seq_len | 估算 Q 张量大小 | 估算 KV 张量大小 |
|:---:|:---:|:---:|:---:|
| 1 | 1024 | 1×1024×32×128×2B = 8 MB | 2×1×1024×4×128×2B ≈ 2 MB |
| 1 | 4096 | 32 MB | 8 MB |
| 1 | 8192 | 64 MB | 16 MB |
| 1 | 16384 | 128 MB | 32 MB |
| 4 | 1024~16384 | 32 MB ~ 512 MB | 8 MB ~ 128 MB |
| 16 | 1024~16384 | 128 MB ~ 2 GB | 32 MB ~ 512 MB |
| batch_size | seq_len | 估算 Q 张量大小 | 估算 KV 张量大小 |
|:---:|:---:|:---:|:---:|
| 1 | 1024 | 1×1024×32×128×2B = 8 MB | 2×1×1024×4×128×2B ≈ 2 MB |
| 1 | 4096 | 32 MB | 8 MB |
| 1 | 8192 | 64 MB | 16 MB |
| 1 | 16384 | 128 MB | 32 MB |
| 4 | 1024~16384 | 32 MB ~ 512 MB | 8 MB ~ 128 MB |
| 16 | 1024~16384 | 128 MB ~ 2 GB | 32 MB ~ 512 MB |
- **精度要求**：`torch.allclose(output.float(), output_ref.float(), rtol = 1e-2, atol = 1e-2)`
- **显存上限**：OJ 评测环境设计 `VRAM_SIZE = 48 GB`（见 `testcase_config.py`）
@ -661,7 +661,7 @@ FlashInfer 方向包含 **4 个可选算子题目**，每个对应独立的 benc
![smoke code](https://origin.picgo.net/2026/06/25/-2026-06-25-095143e59315c41ab0319a.png)
**参考 prompt：**
**参考 prompt** 及 **20001 FlashInfer Ragged Prefill 参考冒烟代码**：
```plaintext
# FlashInfer Ragged Prefill CUDA Kernel — Problem 20001
@ -847,12 +847,9 @@ FlashInfer 方向包含 **4 个可选算子题目**，每个对应独立的 benc
15. All kernel pointers `__restrict__`. warp_sum in namespace. Only code output — no markdown.
**Common errors**: (1) `-INFINITY` for m → NaN; use `-1.0e20f`. (2) Missing alpha guard → exp(1e20) overflow. (3) K/V ptr uses 32 instead of 4 → wrong stride (4096 vs 512). (4) Output ptr uses kv_head instead of qo_head → 8 heads write to same location. (5) `sum/t` divides by zero at t=0; use `sum/(t+1)`. (6) Signed `0xffffffff` mask → UB; use `0xffffffffu`.
```
**20001 FlashInfer Ragged Prefill 参考冒烟代码：**
```cpp
#include <stdint.h>