extern/whisper.cpp
1
0
Fork 1
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2024-11-07 08:34:37 +01:00
Code Issues Packages Projects Releases Wiki Activity

metal : fix asserts for setThreadgroupMemoryLength (close #1435)

Browse Source
This commit is contained in:
Georgi Gerganov 2023-11-07 11:02:16 +02:00
parent 435a6b74e3
commit 3bd7d48f51
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ggml-metal.m
View File

@ -1030,7 +1030,7 @@ void ggml_metal_graph_compute(
[encoder setBytes:&ne00 length:sizeof(ne00) atIndex:2]; [encoder setBytes:&ne00 length:sizeof(ne00) atIndex:2];
[encoder setBytes:&ne01 length:sizeof(ne01) atIndex:3]; [encoder setBytes:&ne01 length:sizeof(ne01) atIndex:3];
[encoder setBytes:&ne02 length:sizeof(ne02) atIndex:4]; [encoder setBytes:&ne02 length:sizeof(ne02) atIndex:4];
[encoder setThreadgroupMemoryLength:nth/32*sizeof(float) atIndex:0]; [encoder setThreadgroupMemoryLength:GGML_PAD(nth/32*sizeof(float), 16) atIndex:0];
[encoder dispatchThreadgroups:MTLSizeMake(ne01*ne02*ne03, 1, 1) threadsPerThreadgroup:MTLSizeMake(nth, 1, 1)]; [encoder dispatchThreadgroups:MTLSizeMake(ne01*ne02*ne03, 1, 1) threadsPerThreadgroup:MTLSizeMake(nth, 1, 1)];
} break; } break;
@ -1342,7 +1342,7 @@ void ggml_metal_graph_compute(
[encoder setBytes:&ne00 length:sizeof( int64_t) atIndex:2]; [encoder setBytes:&ne00 length:sizeof( int64_t) atIndex:2];
[encoder setBytes:&nb01 length:sizeof(uint64_t) atIndex:3]; [encoder setBytes:&nb01 length:sizeof(uint64_t) atIndex:3];
[encoder setBytes:&eps length:sizeof( float) atIndex:4]; [encoder setBytes:&eps length:sizeof( float) atIndex:4];
[encoder setThreadgroupMemoryLength:nth/32*sizeof(float) atIndex:0]; [encoder setThreadgroupMemoryLength:GGML_PAD(nth/32*sizeof(float), 16) atIndex:0];
const int64_t nrows = ggml_nrows(src0); const int64_t nrows = ggml_nrows(src0);
@ -1361,7 +1361,7 @@ void ggml_metal_graph_compute(
[encoder setBytes:&ne00 length:sizeof( int64_t) atIndex:2]; [encoder setBytes:&ne00 length:sizeof( int64_t) atIndex:2];
[encoder setBytes:&nb01 length:sizeof(uint64_t) atIndex:3]; [encoder setBytes:&nb01 length:sizeof(uint64_t) atIndex:3];
[encoder setBytes:&eps length:sizeof( float) atIndex:4]; [encoder setBytes:&eps length:sizeof( float) atIndex:4];
[encoder setThreadgroupMemoryLength:nth*sizeof(float) atIndex:0]; [encoder setThreadgroupMemoryLength:GGML_PAD(nth*sizeof(float), 16) atIndex:0];
const int64_t nrows = ggml_nrows(src0); const int64_t nrows = ggml_nrows(src0);