mirrors/ollama-for-amd
1
0
Fork 0
mirror of https://github.com/likelovewant/ollama-for-amd.git synced 2025-12-23 23:18:26 +00:00
Code Issues Packages Projects Releases Wiki Activity

merge upstream and fix conflicts

Browse Source
This commit is contained in:
likelovewant
2024-12-12 13:17:55 +08:00
parent a0caaa2bc8
commit a64347c6d4
383 changed files with 59815 additions and 43978 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
llm/memory.go
View File

@@ -182,7 +182,7 @@ func EstimateGPULayers(gpus []discover.GpuInfo, ggml *GGML, projectors []string,
gzo = gpuZeroOverhead
}
// Only include GPUs that can fit the graph, gpu minimum, the layer buffer and at least more layer
if (gpus[i].FreeMemory - overhead) < gzo+max(graphPartialOffload, graphFullOffload)+gpus[i].MinimumMemory+2*layerSize {
if gpus[i].FreeMemory < overhead+gzo+max(graphPartialOffload, graphFullOffload)+gpus[i].MinimumMemory+2*layerSize {
slog.Debug("gpu has too little memory to allocate any layers",
"id", gpus[i].ID,
"library", gpus[i].Library,
@@ -228,7 +228,7 @@ func EstimateGPULayers(gpus []discover.GpuInfo, ggml *GGML, projectors []string,
for j := len(gpusWithSpace); j > 0; j-- {
g := gpusWithSpace[i%j]
used := gpuAllocations[g.i] + max(graphPartialOffload, graphFullOffload)
if (g.g.FreeMemory - overhead) > used+layerSize {
if g.g.FreeMemory > overhead+used+layerSize {
gpuAllocations[g.i] += layerSize
layerCounts[g.i]++
layerCount++
@@ -251,7 +251,7 @@ func EstimateGPULayers(gpus []discover.GpuInfo, ggml *GGML, projectors []string,
for j := len(gpusWithSpace); j > 0; j-- {
g := gpusWithSpace[layerCount%j]
used := gpuAllocations[g.i] + max(graphPartialOffload, graphFullOffload)
if (g.g.FreeMemory - overhead) > used+memoryLayerOutput {
if g.g.FreeMemory > overhead+used+memoryLayerOutput {
gpuAllocations[g.i] += memoryLayerOutput
layerCounts[g.i]++
layerCount++