Feature Request: make adaptive or changeable kvarn decode split-128 crossover threshold for more performance

未关闭
#183 3 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
25/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
冷清
技术栈
cpp
领域
performance

调研方向

Start in fattn-mma-kvarn-decode.cuh, in ggml_cuda_fattn_kvarn_decode_consider, where the SPLIT_TOKENS == 128 branch checks n_q == 1 against the hard-coded 32768 crossover for D != 128. The open question is whether that value should come from an environment override or from the device SM count, and the issue's benchmark tables give the depths to reproduce. Done means the threshold is adjustable or derived per device, and llama-bench on CUDA hardware shows the gain without regressing the 2000-depth case.

由索引模型根据 Issue 内容生成。

描述

enhancement
Prerequisites
  • I am running the latest code. Mention the version if possible as well.
  • I carefully followed the README.md.
  • I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
  • I reviewed the Discussions, and have a new and useful enhancement to share.
Feature Description

In fattn-mma-kvarn-decode.cuh the split-128 kernel for n_q == 1, D != 128 is currently not allowed below n_kv = 32768

if (n_q == 1 && ((D == 128 && n_kv < 4096) ||
        (D != 128 && n_kv < 32768))) {
    return;
}

On smaller cards, such as the 5060 TI, some TG performance is therefore left on the table, particularly when MTP is enabled. Making this threshold adaptive, or at least adjustable, could help with testing and tuning, especially on these smaller cards. Maybe it's even worth adjusting it down on 3090 with drafting enabled.

Motivation

I did some performance testing and noticed a visible step in TG performance at 32768 context. This was tracked down by Claude, Opus 5.5, to the split-128 kernel threshold mentioned above. An overridable threshold via an environment variable was added like this:

--- a/ggml/src/ggml-cuda/fattn-mma-kvarn-decode.cuh
+++ b/ggml/src/ggml-cuda/fattn-mma-kvarn-decode.cuh
@@ -5,6 +5,8 @@
 #include "fattn-mma-kvarn-load.cuh"
 #include "mma.cuh"
 
+#include <cstdlib>
+
 using namespace ggml_cuda_mma;
 
 static constexpr int GGML_CUDA_FATTN_KVARN_DECODE_THREADS = 256;
@@ -969,8 +971,14 @@ static void ggml_cuda_fattn_kvarn_decode_consider(
     if constexpr (SPLIT_TOKENS == 128) {
         // A D128 CTA can assign its eight warps to both halves of the record;
         // wider heads keep the measured deep-context crossover.
+        // Experiment: GGML_KVARN_SPLIT128_MIN_KV overrides the crossover for
+        // wider heads; unset keeps the default of 32768.
+        static const int split128_min_kv = [] {
+            const char * value = getenv("GGML_KVARN_SPLIT128_MIN_KV");
+            return value != nullptr ? atoi(value) : 32768;
+        }();
         if (n_q == 1 && ((D == 128 && n_kv < 4096) ||
-                (D != 128 && n_kv < 32768))) {
+                (D != 128 && n_kv < split128_min_kv))) {
             return;
         }
     }

(GGML_KVARN_SPLIT128_MIN_KV=0 allows split-128 at any depth)

and benched afterwards. This gave the following results with llama-bench (no MTP) on RTX 5060 Ti 16 GB, driver 616.92, Windows 11, CUDA build for sm_120, beellama.cpp v0.4.8 @ 49df8fa4, Qwen3.8-27B-UD-Q3_K_XL, -ctk/-ctv kvarn4, tg128 at the given depth

Run the same command twice, once without the variable and once with GGML_KVARN_SPLIT128_MIN_KV=0:

llama-bench -m Qwen3.8-27B-UD-Q3_K_XL.gguf -ngl 99 -fa on -ctk kvarn4 -ctv kvarn4 \
  -p 0 -n 128 -d 2000,10000,22000,30000,34000 -r 3 -o csv
depth default t/s threshold 0 t/s change kvarn_geometry_split_64 (default)
2000 27.48 ± 0.04 27.03 ± 0.03 −1.6 % 208
10000 26.58 ± 0.13 26.51 ± 0.06 −0.2 % 48
22000 24.56 ± 0.02 25.06 ± 0.06 +2.1 % 208
30000 23.41 ± 0.04 23.94 ± 0.06 +2.3 % 48
34000 23.29 ± 0.06 23.31 ± 0.01 ±0 (same kernel) 16

With a threshold of 0, the value of kvarn_geometry_split_64 is 0 at every depth, confirming that the override changes the route. At 34,000, both runs use the same kernel. This row acts as a control and demonstrates that the difference is not due to noise. The reason for those jumping kvarn_geometry_split_64 values is unknown to me.

Finer sweep with llama-server (own benchmark tool)

I also measured with llama-server over /completion with agentic-coding-style prompts, cold cache per point, 512 generated tokens, 3 repeats, temperature 0.6. Below is the change in decode time per step (predicted_ms / (predicted_n − draft_n_accepted)) with threshold 0 against the default. A negative value means split-128 is faster. The spread within each point was 0.1–0.3 %.

ctx no MTP MTP (--spec-type draft-mtp --spec-draft-n-max 2 --spec-draft-p-min 0.0)
1000 +0.9 % +0.9 %
2000 +1.6 % +0.8 %
4000 +0.7 % ±0.0 %
6000 +1.0 % −0.6 %
8000 +1.0 % −1.1 %
10000 +0.3 % / ±0.0 % −1.6 %
14000 ±0.0 % −3.5 %
18000 −1.0 % −4.1 %
22000 −2.3 % −4.8 %
26000 −2.4 % −5.9 %
30000 −2.3 % −6.8 %
34000–42000 ±0.0 % −0.2 … 0 %

At 10000 the no-MTP value was measured in short sweep and long sweep, hence the two values.

  • Without MTP the crossover is at ~14k–16k. Below that, split-64 is up to ~1.6 % faster.
  • With MTP the crossover moves down to ~4k, and the gain grows to ~7 % just below 32768. The reason for this is unknown. My guess is that the MTP draft head's single-token decodes also hit the n_q == 1 path, but I haven't verified that.

Tried to reproduce this on a RTX PRO 6000 Blackwell System via modal to rule out architectural effects, but the system had ~5% measurement noise so effects of 1-2% would be hidden. The ~7 % MTP effect seen on the 5060 Ti did not show up. I'm happy to run further tests as needed.

Possible Implementation

The hypothesis is that this is related to the SM count, as these might have been filled earlier by the split-64 grid. So I suggest a threshold by the device's SM count e.g. for non-MTP maybe 32768 * nsm / 82, or at least an environment override for further testing and feedback.

主要语言
C++
星标
1.2k
派生
80
平均合并
4 天 20 小时
30 天内合并 PR
2

环境准备

这个项目没有提供开发容器、Dockerfile 或贡献指南,环境需要你自己搭建:先看它的 README,通用步骤见我们的新手贡献指南。

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

Anbeeld/beellama.cpp 的其他 Issue

查看 Anbeeld/beellama.cpp 的全部 Issue

相似的 Issue

更多 C++ Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。