-
Notifications
You must be signed in to change notification settings - Fork 31
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.
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
Allocate pinned buffer for vectorized code #601
Merged
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bitfaster
commented
May 29, 2024
bitfaster
changed the title
Allocate aligned buffer for vectorized code
Allocate pinned buffer for vectorized code
Jun 1, 2024
The improvement on a newer AMD CPU is much less, almost noise.
Inc BaselineTabular
In PinnedTabular
Freq baselineTabular
Freq pinnedTabular
|
bitfaster
commented
Aug 31, 2024
bitfaster
commented
Nov 21, 2024
Pinned block is faster except for size=524288.
|
|
|
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Current AVX2 vectorized code doesn't have much of an advantage on .NET8 and .NET9. We can gain some speed by using the pinned object heap introduced in .NET5 and eliminating the fixed statement. With a fixed address, we can also do a trick to align to 64 bytes, such that each increment/frequency call is guaranteed to be on a single cache line.
Use of fixed results in a fixed local variable in IL, the runtime overhead comes from the JITted code. Explanation here.
This has little effect for smaller sizes, but is noticeable when the sketch is larger. Data captured from current generation Azure v5 VMs on both Intel and AMD CPUs.
AMD (Zen3 - Milan)
Intel (Sunny Cove - Ice Lake)
Alternative
It is possible to allocate aligned native memory, but this would mean that the Sketch/LFU are disposable.
https://github.com/dotnet/runtime/blob/main/docs/coding-guidelines/vectorization-guidelines.md#enforcing-memory-alignment
void* _pointer = NativeMemory.AlignedAlloc(byteCount: Size * sizeof(int), alignment: 32);