Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

llama_decode lock #595

Merged
merged 2 commits into from
Mar 13, 2024
Merged

llama_decode lock #595

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
011d019
Added a lock object into `SafeLlamaModelHandle` which all calls to `l…
martindevans Mar 12, 2024
7e1f472
Modified the lock to be global over _all_ inferences. This seems to b…
martindevans Mar 12, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 15 additions & 2 deletions LLama/Native/SafeLLamaContextHandle.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -192,6 +192,18 @@ public uint TokenToSpan(LLamaToken token, Span<byte> dest)
#endregion

#region infer
/// <summary>
/// This object exists to ensure there is only ever 1 inference running at a time. This is a workaround for thread safety issues in llama.cpp itself.
/// Most notably CUDA, which seems to use some global singleton resources and will crash if multiple inferences are run (even against different models).
///
/// For more information see these issues:
/// - https://github.com/SciSharp/LLamaSharp/issues/596
/// - https://github.com/ggerganov/llama.cpp/issues/3960
///
/// If these are ever resolved this lock can probably be removed.
/// </summary>
private static readonly object GlobalInferenceLock = new();

/// <summary>
/// </summary>
/// <param name="batch"></param>
Expand All @@ -202,8 +214,9 @@ public uint TokenToSpan(LLamaToken token, Span<byte> dest)
/// </returns>
public DecodeResult Decode(LLamaBatch batch)
{
using (batch.ToNativeBatch(out var nb))
return (DecodeResult)NativeApi.llama_decode(this, nb);
lock (GlobalInferenceLock)
using (batch.ToNativeBatch(out var nb))
return (DecodeResult)NativeApi.llama_decode(this, nb);
}

/// <summary>
Expand Down
Loading