Skip to content

Releases: Vithulep/llama.cpp

b5631

11 Jun 09:39
1f7d50b
Compare
Choose a tag to compare
vulkan: Track descriptor pools/sets per-context (#14109)

Use the same descriptor set layout for all pipelines (MAX_PARAMETER_COUNT == 8)
and move it to the vk_device. Move all the descriptor pool and set tracking to
the context - none of it is specific to pipelines anymore. It has a single vector
of pools and vector of sets, and a single counter to track requests and a single
counter to track use.

b5618

10 Jun 06:51
1f63e75
Compare
Choose a tag to compare
metal : use less stack memory in FA kernel (#14088)

* metal : use less stack memory in FA kernel

ggml-ci

* cont : fix BF16 variant