rewrite hello-neon as a generic vectorization sample #1011
Comments
|
After this merges and before I delete the old one, https://developer.android.com/ndk/guides/cpu-arm-neon will need to be updated since it points to hello-neon. |
DanAlbert
added a commit
to DanAlbert/ndk-samples
that referenced
this issue
Jun 12, 2024
This adds a new generic vectorization sample to replace hello-neon. Most importantly, it covers non-Neon options for SIMD. One of the useful things it shows, for example, is that there's actually no reason to write SIMD code the way that hello-neon does any more. This also solves a much simpler problem (small matrix multiplication), which makes it easier to see how to deal with the SIMD features rather than figuring out what a FIR filter is. Finally, this sample benchmarks each of the implementations so it's obvious what is and isn't worth doing. I was sort of surprised that auto-vectorization didn't do better, and was pleased to learn that there's no reason at all to write Neon intrinsics. I'll delete hello-neon after this merges and I've fixed up the doc links. android#1011
DanAlbert
added a commit
to DanAlbert/ndk-samples
that referenced
this issue
Jun 12, 2024
This adds a new generic vectorization sample to replace hello-neon. Most importantly, it covers non-Neon options for SIMD. One of the useful things it shows, for example, is that there's actually no reason to write SIMD code the way that hello-neon does any more. This also solves a much simpler problem (small matrix multiplication), which makes it easier to see how to deal with the SIMD features rather than figuring out what a FIR filter is. Finally, this sample benchmarks each of the implementations so it's obvious what is and isn't worth doing. I was sort of surprised that auto-vectorization didn't do better, and was pleased to learn that there's no reason at all to write Neon intrinsics. I'll delete hello-neon after this merges and I've fixed up the doc links. android#1011
DanAlbert
added a commit
to DanAlbert/ndk-samples
that referenced
this issue
Jun 12, 2024
This adds a new generic vectorization sample to replace hello-neon. Most importantly, it covers non-Neon options for SIMD. One of the useful things it shows, for example, is that there's actually no reason to write SIMD code the way that hello-neon does any more. This also solves a much simpler problem (small matrix multiplication), which makes it easier to see how to deal with the SIMD features rather than figuring out what a FIR filter is. Finally, this sample benchmarks each of the implementations so it's obvious what is and isn't worth doing. I was sort of surprised that auto-vectorization didn't do better, and was pleased to learn that there's no reason at all to write Neon intrinsics. I'll delete hello-neon after this merges and I've fixed up the doc links. android#1011
DanAlbert
added a commit
to DanAlbert/ndk-samples
that referenced
this issue
Jun 17, 2024
This adds a new generic vectorization sample to replace hello-neon. Most importantly, it covers non-Neon options for SIMD. One of the useful things it shows, for example, is that there's actually no reason to write SIMD code the way that hello-neon does any more. This also solves a much simpler problem (small matrix multiplication), which makes it easier to see how to deal with the SIMD features rather than figuring out what a FIR filter is. Finally, this sample benchmarks each of the implementations so it's obvious what is and isn't worth doing. I was sort of surprised that auto-vectorization didn't do better, and was pleased to learn that there's no reason at all to write Neon intrinsics. I'll delete hello-neon after this merges and I've fixed up the doc links. android#1011
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
This sample probably made more sense back when it was written, but now it's not really guiding people toward doing the right thing.
Much more useful, I think would be rewriting it as a general purpose vectorization demo. Pick some sort of vectorization problem (probably something simple like a matrix multiply), and build an implementation for each of:
<algorithm>stuff, if the problem we pick is suitable for themAnd then the UI would show the benchmark results of each so people can see the differences.
The text was updated successfully, but these errors were encountered: