search

sarah-quinones / gemm

MIT License
76 stars 11 forks source link

Compilation error when compiling to aarch64-apple-ios #25

Open santiagomed opened 6 months ago

santiagomed commented 6 months ago

Hi. I am trying to compile a project that uses the Candle ML framework into aarch64-apple-ios. Candle uses gemm as a dependency, so I get the following compilation error:

error: instruction requires: fullfp16
    --> /Users/santiagomedina/.cargo/registry/src/index.crates.io-6f17d22bba15001f/gemm-common-0.17.1/src/simd.rs:1982:18
     |
1982 |                 "fmla {0:v}.8h, {1:v}.8h, {2:v}.h[0]",
     |                  ^
     |
note: instantiated into assembly here
    --> <inline asm>:1:2
     |
1    |     fmla v0.8h, v1.8h, v2.h[0]
     |     ^

Saw this Candle issue so I though I'd open an issue here as well to check on any updates.

akashicMarga commented 5 months ago

@santiagomed check this discussion. i tried compiling candle for android. this rust flag helped remove gemm issue.

https://github.com/huggingface/candle/discussions/2081#discussioncomment-9217656

Pox-here commented 4 months ago

Experiencing the same. Tried to follow the thread further but found currently no solution. Was this issue solved?

    --> /home/[user]/.cargo/registry/src/index.crates.io-6f17d22bba15001f/gemm-common-0.17.1/src/simd.rs:2024:18
     |
2024 |                 "fmla {0:v}.8h, {1:v}.8h, {2:v}.h[7]",
     |                  ^
     |
note: instantiated into assembly here
    --> <inline asm>:1:2
     |
1    |     fmla v0.8h, v1.8h, v2.h[7]

I tried adding the flag rustflags = [ "-C", "target-feature=+fp16", ]

But it did not resolve the issue