ggml-webgpu: move from parameter buffer pool to single buffer with offsets (llama/21278)

Browse Source

* Work towards removing bitcast

* Move rest of existing types over

* Add timeout back to wait and remove synchronous set_tensor/memset_tensor

* move to unpackf16 for wider compatibility

* cleanup

* Remove deadlock condition in free_bufs

* Start work on removing parameter buffer pools

* Simplify and optimize further

* simplify profile futures

* Fix stride

* Try using a single command buffer per batch

* formatting

...

This commit is contained in:

Reese Levine

2026-04-30 11:29:01 +03:00

committed by Georgi Gerganov

parent 3f5117610b

commit d6cfdc669c

2 changed files with 373 additions and 416 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

ggml/src/ggml-webgpu/ggml-webgpu.cpp

+349 -397

File diff suppressed because it is too large Load Diff