Skip to content

GitLab

Unverified Commit 63269668 authored by frob's avatar frob Committed by GitHub
Browse files

Prevent underflow when FreeMemory < overhead (#8014) 


Co-authored-by: default avatarRichard Lyons <frob@cloudstaff.com>
parent 900f64e6
Hide whitespace changes
Inline Side-by-side
Showing with 3 additions and 3 deletions
llm/memory.go
View file @ 63269668
......@@ -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++
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment