llama.cpp/src at 6d476d6ccc6fcffb145956e110b0cb238829f109 - llama.cpp - Gitea: Git with a cup of tea

Lumpiasty/llama.cpp

Files

T

History

Lumpiasty 6d476d6ccc llama: free backend compute scratch on cold; guard memory_breakdown

On device release with KV eviction, also call ggml_backend_free_scratch()
on each backend so a cold model drops its Vulkan compute preallocations
(reallocated lazily on the next compute via restore).

Also guard llama_context::memory_breakdown() against a freed scheduler:
evict_kv release resets sched to null, so a cold model that is then torn
down (e.g. terminated by the process manager) hit GGML_ASSERT(sched) in
ggml_backend_sched_get_buffer_type. Skip the compute-buffer accounting
when sched is null.

Assisted-by: Claude

2026-07-26 21:58:43 +02:00

..

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

CMakeLists.txt

DeepSeek V4 (#24162 )

2026-06-29 16:58:51 +08:00

llama-adapter.cpp

hparams : refactor hparams.n_layer (#24060 )

2026-06-05 11:09:36 +03:00

llama-adapter.h

llama : re-enable manual LoRA adapter free (#19983 )

2026-03-18 12:03:26 +02:00

llama-arch.cpp

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-arch.h

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-batch.cpp

llama-batch: fix allowed decreasing pos in a seq (#25449 )

2026-07-08 19:24:34 +03:00

llama-batch.h

llama-batch: add n_keep_tail in split_equal for recurrent models (#25278 )

2026-07-08 15:55:19 +08:00

llama-chat.cpp

chat : add Granite 4.1 chat template (#23518 )

2026-05-28 13:13:33 +02:00

llama-chat.h

chat : add Granite 4.1 chat template (#23518 )

2026-05-28 13:13:33 +02:00

llama-context.cpp

llama: free backend compute scratch on cold; guard memory_breakdown

2026-07-26 21:58:43 +02:00

llama-context.h

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-cparams.cpp

cparams : rename LLAMA_MAX_PARALLEL_SEQUENCES to LLAMA_MAX_SEQ (#14188 )

2025-06-15 10:08:58 +03:00

llama-cparams.h

DeepseekV4: Add fused hyper-connection ops (#25585 )

2026-07-17 00:33:33 +08:00

llama-ext.h

spec : Support Step3.5/3.7 flash mtp3 (#24340 )

2026-06-21 11:33:18 +03:00

llama-grammar.cpp

llama: various bug fixes (#26051 )

2026-07-24 18:56:42 +02:00

llama-grammar.h

common/grammar : replace problematic backtracking regex [\s\S]* (#18342 )

2026-01-03 16:02:43 -06:00

llama-graph.cpp

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-graph.h

DeepseekV4: Add fused hyper-connection ops (#25585 )

2026-07-17 00:33:33 +08:00

llama-hparams.cpp

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-hparams.h

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-impl.cpp

llama : correct platform-independent loading of BOOL metadata (#21428 )

2026-04-06 01:40:38 +02:00

llama-impl.h

llama: refactor fused ops (#24646 )

2026-07-08 18:18:09 +08:00

llama-io.cpp

server : avoid checkpoint data host copies (#22558 )

2026-05-02 18:03:25 +03:00

llama-io.h

llama : add option to save memory in device buffers (#22679 )

2026-05-05 06:35:07 +03:00

llama-kv-cache-dsa.cpp

llama-batch: add n_keep_tail in split_equal for recurrent models (#25278 )

2026-07-08 15:55:19 +08:00

llama-kv-cache-dsa.h

model : support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation (#23346 )

2026-05-29 10:15:17 +02:00

llama-kv-cache-dsv4.cpp

llama_dsv4: write only used rows in state (#25325 )

2026-07-20 22:43:39 +08:00

llama-kv-cache-dsv4.h

DeepseekV4: fix seq_rm (#25588 )

2026-07-14 21:45:36 +08:00

llama-kv-cache-iswa.cpp

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-kv-cache-iswa.h

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-kv-cache.cpp

llama: guard buffer iteration against released device buffers

2026-07-26 21:58:43 +02:00

llama-kv-cache.h

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-kv-cells.h

kv-cache : avoid kv cells copies (#24277 )

2026-06-07 21:42:54 +03:00

llama-memory-hybrid-iswa.cpp

llama-batch: add n_keep_tail in split_equal for recurrent models (#25278 )

2026-07-08 15:55:19 +08:00

llama-memory-hybrid-iswa.h

llama + spec: MTP Support (#22673 )

2026-05-16 20:06:23 +08:00

llama-memory-hybrid.cpp

llama: evict recurrent/SSM state on device release

2026-07-26 21:58:43 +02:00

llama-memory-hybrid.h

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-memory-recurrent.cpp

llama: evict recurrent/SSM state on device release

2026-07-26 21:58:43 +02:00

llama-memory-recurrent.h

llama: evict recurrent/SSM state on device release

2026-07-26 21:58:43 +02:00

llama-memory.cpp

memory : correctly handle failure in apply() (#14438 )

2025-06-30 18:03:03 +03:00

llama-memory.h

server: optionally evict the KV cache too (Phase 2 of VRAM sharing)

2026-07-26 21:58:43 +02:00

llama-mmap.cpp

llama : pin mmap-backed CPU weights for faster H2D uploads

2026-07-26 21:58:43 +02:00

llama-mmap.h

llama : pin mmap-backed CPU weights for faster H2D uploads

2026-07-26 21:58:43 +02:00

llama-model-loader.cpp

llama : pin mmap-backed CPU weights for faster H2D uploads

2026-07-26 21:58:43 +02:00

llama-model-loader.h

args: refactor mlock/mmap/directio into load-mode (#20834 )

2026-07-23 20:32:56 +08:00

llama-model-saver.cpp

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-model-saver.h

llama: fix llama-model-saver (#20503 )

2026-03-25 12:53:16 +02:00

llama-model.cpp

llama: guard buffer iteration against released device buffers

2026-07-26 21:58:43 +02:00

llama-model.h

server: on-demand VRAM sharing to time-share one GPU between models

2026-07-26 21:58:43 +02:00

llama-quant.cpp

llama: various bug fixes (#26051 )

2026-07-24 18:56:42 +02:00

llama-quant.h

llama : refactor src/llama.cpp (#10902 )

2025-01-03 10:18:53 +02:00

llama-sampler.cpp

llama: various bug fixes (#26051 )

2026-07-24 18:56:42 +02:00

llama-sampler.h

llama : rename llama-sampling to llama-sampler (#19363 )

2026-02-06 07:26:54 +01:00

llama-vocab.cpp

model: Add MiniMax-M3 (MSA: MiniMax Sparse Attention) support (#24908 )

2026-07-26 19:43:45 +02:00

llama-vocab.h

Add support for Laguna XS.2 & M.1 (#25165 )

2026-07-22 09:54:08 +08:00

llama.cpp

args: refactor mlock/mmap/directio into load-mode (#20834 )

2026-07-23 20:32:56 +08:00

unicode-data.cpp

server : better security control for public deployments (#9776 )

2024-10-08 13:27:04 +02:00

unicode-data.h

llama : reduce compile time and binary size (#9712 )

2024-10-02 15:49:55 +02:00

unicode.cpp

unicode,test: add Qwen3.5 non-backtracking tokenizer handler and regr… (#22110 )

2026-05-14 11:03:40 +02:00

unicode.h

vocab: fix Gemma4 tokenizer (#21343 )

2026-04-03 10:33:03 +02:00