Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

hexagon: make vmem and buffer-size configurable (#22487)

Browse Source 
* hexagon: allow host to set max vmem size

We use a sane default but it's helpful to allow for an override if needed.

* hexagon: add support for measuring vmem space and move pinned mmaping management to host

* hexagon: update vmem checks to use uint64

* hexagon: bump op buffers to 16 (matches max mmaps)

* hexagon: bump default vmem to 3.2GB

* hexagon: add support for autodetecting vmem space and some logging cleanup in that area

* hexagon: fix whitespace warnings

* Update scripts/snapdragon/adb/run-cli.sh

Co-authored-by: Pascal <admin@serveurperso.com>

* hex-adb: fix run-completion script

---------

Co-authored-by: Pascal <admin@serveurperso.com>
This commit is contained in:
Max Krasnyansky
2026-04-29 11:51:21 -07:00
committed by GitHub
co-authored by Pascal
parent 098705a29e
commit 41a63be28e
7 changed files with 180 additions and 121 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-hexagon/htp/htp-ops.h
+2 -6
View File
@@ -90,15 +90,11 @@ enum htp_op_code {
#define HTP_OP_MAX_INPUTS 6 // aka GGML_MAX_SRCS
#define HTP_OP_MAX_PARAMS 16 // aka GGML_MAX_OP_PARAMS
#define HTP_OP_MAX_BUFS 8
#define HTP_OP_MAX_BUFS 16
#define HTP_OP_MAX_REQS 256
#define HTP_OP_MAX_TENSORS (HTP_OP_MAX_REQS * HTP_OP_MAX_INPUTS + HTP_OP_MAX_REQS)
#if __HVX_ARCH__ < 75
#define HTP_OP_MAX_VMEM (3167538380u)
#else
#define HTP_OP_MAX_VMEM (3221225472u)
#endif
#define HTP_OP_MAX_VMEM_DEFAULT (3355443200u)
#define HTP_MMAP_MAX_VMEM (2147483648u)