Fix SIMD Fallback Import#232
Conversation
|
I also added some details to the references since I was looking through the papers anyway. (It follows the APA style used by Google Scholar.) |
762d335 to
a1cda64
Compare
b9affae to
fe9529e
Compare
src/literals.rs
Outdated
|
|
||
| use freqs::BYTE_FREQUENCIES; | ||
|
|
||
| #[cfg(feature = "simd-accel")] |
There was a problem hiding this comment.
I think you don't actually need this cfg switch, right? The cfg switch is done in lib.rs so that simd_fallback should never actually be explicitly used.
There was a problem hiding this comment.
You still import it as simd_fallback, though. Might be a good idea to provide a shim module in lib.rs depending on the feature flag (the shim in simd_accel only deals with the target arch feature).
There was a problem hiding this comment.
Oh, I see. I thought I was using path, but maybe it was somewhere else. Maybe this should be in lib.rs?
#[cfg(not(feature = "simd-accel"))]
#[path = "simd_fallback/mod.rs"]
mod simd_accel;I think that would work? Or something equivalent would be fine. I was just trying to stop the use of cfg switching everywhere.
There was a problem hiding this comment.
Yeah, that should work. Defining the shim in lib.rs is far better than using cfg here. I'll change this.
d9ad839 to
3ee97ee
Compare
|
Rebased. |
6f2bb0f to
203c509
Compare
Add SIMD accelerated multiple pattern search.
|
@killercup Could you rebase this against master? Sorry for the churn. Thanks! |
3ee97ee to
71cd186
Compare
|
@BurntSushi no problem! Rebased on master. |
|
Awesome, thanks so much! :D |
|
@killercup Hmm, it looks like this got merged into the |
|
Oh, Github didn't change the target branch :( just one second!
|
Fixes
cargo docand some other stuff.r? @BurntSushi