feat: remove redundant additions (#6)

Co-authored-by: Khashayar Barooti <khashayar.baroot@epfl.ch>

Author: zac-williamson

README.md

@@ -21,6 +21,15 @@ with the following command.
 
 The benchmark will be generated at `./gates_report.json`.
 
+Current benchmarks as of 1 Mar 2025
+
+| num blocks hashed | num bytes hashed | gates for `sha512::hash` | gates for `sha512::digest_var` |
+| --- | --- | --- | --- |
+| 1 block | 111 | 39,476 | 41,261 |
+| 2 blocks | 239 ||66,927 | 69,816 |
+| 3 blocks | 367 | 94,377 | 98,355 |
+| 4 blocks | 495 | 121,826 | 126,914 |
+
 ## Installation
 
 In your _Nargo.toml_ file, add the version of this library you would like to install under dependency:

src/benchmarks/mod.nr

@@ -1,30 +1,52 @@
-use crate::permutation::digest;
-
-// compression benchmark is 41924
-// NOTE: does this align with expectations? Seems high
-// #[export]
-// fn bench_compression(input: [u8; 128]) {
-//     let foo = msg_u8_to_u64(input);
-//     let r = sha512_compression(foo, [0; 8]);
-//     println(f"{r}");
-// }
-// cost of 2 hashes = 115,221
-// cost of 1 hash = 63,938
-// extra hash = 51k, 13k gate setup? seems appropriate
-// note: a lot of cost overhead vs compression algorithm. A lot of this will be witness extension, but maybe there are overheads we can cut?
-#[export]
-fn bench_digest(input: [u8; 128]) {
-    let result = digest::<_, 1>(input);
-
-    let expected: [u8; 64] = [
-        0xcf, 0x83, 0xe1, 0x35, 0x7e, 0xef, 0xb8, 0xbd, 0xf1, 0x54, 0x28, 0x50, 0xd6, 0x6d, 0x80,
-        0x07, 0xd6, 0x20, 0xe4, 0x05, 0x0b, 0x57, 0x15, 0xdc, 0x83, 0xf4, 0xa9, 0x21, 0xd3, 0x6c,
-        0xe9, 0xce, 0x47, 0xd0, 0xd1, 0x3c, 0x5d, 0x85, 0xf2, 0xb0, 0xff, 0x83, 0x18, 0xd2, 0x87,
-        0x7e, 0xec, 0x2f, 0x63, 0xb9, 0x31, 0xbd, 0x47, 0x41, 0x7a, 0x81, 0xa5, 0x38, 0x32, 0x7a,
-        0xf9, 0x27, 0xda, 0x3e,
-    ];
-
-    println(f"result {result}");
-    println(f"expected {expected}");
-    assert_eq(result, expected);
+use crate::formatting::msg_u8_to_u64;
+
+global MAX_BYTES_1_BLOCK: u32 = 111;
+global MAX_BYTES_2_BLOCKS: u32 = MAX_BYTES_1_BLOCK + 128;
+global MAX_BYTES_3_BLOCKS: u32 = MAX_BYTES_2_BLOCKS + 128;
+global MAX_BYTES_4_BLOCKS: u32 = MAX_BYTES_3_BLOCKS + 128;
+
+#[export]
+fn bench_sha512_compression(input: [u8; 128], h: [u64; 8]) -> [u64; 8] {
+    let r = crate::sha512_compression(msg_u8_to_u64(input), h);
+    r
+}
+
+#[export]
+fn bench_sha512_1_block(input: [u8; MAX_BYTES_1_BLOCK]) -> [u8; 64] {
+    crate::sha512::digest(input)
+}
+
+#[export]
+fn bench_sha512_2_blocks(input: [u8; MAX_BYTES_2_BLOCKS]) -> [u8; 64] {
+    crate::sha512::digest(input)
+}
+
+#[export]
+fn bench_sha512_3_blocks(input: [u8; MAX_BYTES_3_BLOCKS]) -> [u8; 64] {
+    crate::sha512::digest(input)
+}
+
+#[export]
+fn bench_sha512_4_blocks(input: [u8; MAX_BYTES_4_BLOCKS]) -> [u8; 64] {
+    crate::sha512::digest(input)
+}
+
+#[export]
+fn bench_sha512_var_1_block(input: BoundedVec<u8, MAX_BYTES_1_BLOCK>) -> [u8; 64] {
+    crate::sha512::sha512_var(input)
+}
+
+#[export]
+fn bench_sha512_var_2_blocks(input: BoundedVec<u8, MAX_BYTES_2_BLOCKS>) -> [u8; 64] {
+    crate::sha512::sha512_var(input)
+}
+
+#[export]
+fn bench_sha512_var_3_blocks(input: BoundedVec<u8, MAX_BYTES_3_BLOCKS>) -> [u8; 64] {
+    crate::sha512::sha512_var(input)
+}
+
+#[export]
+fn bench_sha512_var_4_blocks(input: BoundedVec<u8, MAX_BYTES_4_BLOCKS>) -> [u8; 64] {
+    crate::sha512::sha512_var(input)
 }