merge with latest dev

.gitignore

@@ -15,4 +15,4 @@ docs/**bench**
 **/*/libposeidon-permute-c-mac.a
 **/*/go-iden3-crypto/
 node_modules
-.vscode
+.vscode

.gitmodules

@@ -1,3 +1,4 @@
 [submodule "depends/cryptography_cuda"]
 	path = depends/cryptography_cuda
 	url = git@github.com:okx/cryptography_cuda.git
+	branch =5c5c3ca7987125507c8eb17572ecb355ffed2256

field/Cargo.toml

@@ -36,8 +36,8 @@ proc-macro2 = "1"
 quote = "1"
 
 [features]
-default = ["no_cuda"]
-cuda = ["cryptography_cuda"]
+default = ["cryptography_cuda/no_cuda"]
+cuda = ["cryptography_cuda/cuda"]
 precompile = []
 no_cuda = ["cryptography_cuda/no_cuda"]
 

field/src/fft.rs

@@ -2,7 +2,7 @@ use alloc::vec::Vec;
 use core::cmp::{max, min};
 
 #[cfg(feature = "cuda")]
-use cryptography_cuda::{intt, ntt, types::NTTInputOutputOrder};
+use cryptography_cuda::{ntt, types::NTTInputOutputOrder};
 use plonky2_util::{log2_strict, reverse_index_bits_in_place};
 use unroll::unroll_for_loops;
 
@@ -34,6 +34,7 @@ pub fn fft_root_table<F: Field>(n: usize) -> FftRootTable<F> {
     root_table
 }
 
+#[allow(dead_code)]
 #[cfg(feature = "cuda")]
 fn fft_dispatch_gpu<F: Field>(
     input: &mut [F],

plonky2/Cargo.toml

@@ -12,16 +12,15 @@ keywords.workspace = true
 categories.workspace = true
 
 [features]
-default = ["gate_testing", "parallel", "rand_chacha", "std", "timing"]
+default = ["gate_testing", "parallel", "rand_chacha", "std", "timing", "cryptography_cuda/no_cuda"]
 gate_testing = []
 parallel = ["hashbrown/rayon", "plonky2_maybe_rayon/parallel"]
 std = ["anyhow/std", "rand/std", "itertools/use_std"]
 timing = ["std", "dep:web-time"]
-cuda =["cryptography_cuda"]
+cuda =["cryptography_cuda/cuda"]
 no_cuda = ["cryptography_cuda/no_cuda"]
 batch =[]
 
-
 [dependencies]
 ahash = { workspace = true }
 anyhow = { workspace = true }