Feature Request: make adaptive or changeable kvarn decode split-128 crossover threshold for more performance
Nadie ha tomado este issue todavía.
Evaluación
- Dificultad
- 4/5
- Tiempo estimado
- 3-5 días
- Aptitud para principiantes
- 25/100
- Tipo de issue
- Nueva funcionalidad
- Claridad
- Bastante claro
- Estado de actividad
- Tranquilo
- Stack tecnológico
- cpp
- Área
- performance
Línea de trabajo
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.
Escrito por el modelo de indexación a partir del texto del issue.
Descripción
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 == 1path, 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.
- Lenguaje dominante
- C++
- Estrellas
- 1.2k
- Forks
- 80
- Merge medio
- 4 d 20 h
- PR fusionados (30 d)
- 2
Preparar el entorno
Este proyecto no incluye contenedor de desarrollo, Dockerfile ni guía de contribución, así que la configuración corre por tu cuenta: empieza por su README y consulta nuestra guía para la primera contribución para los pasos generales.
Primeros pasos
- Lee el issue completo y luego la guía de contribución del proyecto.
- Comenta en el issue que vas a ocuparte — evita que dos personas hagan lo mismo.
- Haz un fork del repositorio y trabaja en una rama.
- Abre un pull request que haga referencia al número del issue.
Más de Anbeeld/beellama.cpp
-
[BUG] KVarN assertion failure in build_inp_mem_hybrid_iswa() due to uninitialized self_kvarn_rotAbierto
Dificultad 2/5 1-3 horas Aptitud para principiantes 68/100
Anbeeld/beellama.cpp#205 ·
-
Dificultad 3/5 Medio día Aptitud para principiantes 58/100
Anbeeld/beellama.cpp#206 ·
-
Dificultad 3/5 1-2 días Aptitud para principiantes 57/100
Anbeeld/beellama.cpp#204 ·
-
Dificultad 4/5 3-5 días Aptitud para principiantes 45/100
Anbeeld/beellama.cpp#203 · 2 comentarios ·
-
Dificultad 4/5 3-5 días Aptitud para principiantes 45/100
Anbeeld/beellama.cpp#202 · 1 comentario ·
Todos los issues de Anbeeld/beellama.cpp
Issues similares
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 72/100
CachyOS/packageinstaller#144 ·
Los mantenedores suelen responder en 1 día
-
documentation
Dificultad 1/5 Menos de una hora Aptitud para principiantes 82/100
TomerAberbach/profiler-md#951 ·
Los mantenedores suelen responder en 1 día
-
[Bug] Default key routing (JavaStringHash) picks a different partition than the Java client for about half of keysPosiblemente ocupada @0z5a la tomó hoy. Abierto
Dificultad 2/5 1-3 horas Aptitud para principiantes 82/100
apache/pulsar-client-go#1575 ·
Los mantenedores suelen responder en 2 días
-
Dificultad 2/5 1-3 horas Aptitud para principiantes 72/100
pocoproject/poco#5764 ·
Los mantenedores suelen responder en 2 días
-
bug cpp-core-guidelines low-priority
Dificultad 2/5 1-3 horas Aptitud para principiantes 62/100
NatronGitHub/openfx-supportext#164 ·
Los mantenedores suelen responder en 1 día