chfast
commented
2 years ago div<uint256, udivrem>/64 -0.0036 -0.0036 23 23 23 23
div<uint256, udivrem>/128 +0.0026 +0.0026 29 29 29 29
div<uint256, udivrem>/192 -0.0414 -0.0414 34 33 34 33
div<uint256, udivrem>/256 -0.0468 -0.0468 30 28 30 28
div<uint512, udivrem>/64 -0.0025 -0.0025 37 37 37 37
div<uint512, udivrem>/128 -0.0144 -0.0145 49 48 49 48
div<uint512, udivrem>/192 -0.0440 -0.0440 72 68 72 68
div<uint512, udivrem>/256 -0.0480 -0.0480 69 66 69 66
mod<addmod>/64 -0.0115 -0.0115 30 29 30 29
mod<addmod>/128 -0.0041 -0.0041 33 33 33 33
mod<addmod>/192 -0.0133 -0.0133 41 40 41 40
mod<addmod>/256 -0.0111 -0.0111 35 35 35 35
mod<addmod_public>/64 -0.0054 -0.0054 33 33 33 33
mod<addmod_public>/128 -0.0044 -0.0044 37 36 37 36
mod<addmod_public>/192 -0.0403 -0.0403 44 42 44 42
mod<addmod_public>/256 -0.0279 -0.0279 37 36 37 36
mod<addmod_simple>/64 +0.0046 +0.0046 32 32 32 32
mod<addmod_simple>/128 +0.0131 +0.0131 34 35 34 35
mod<addmod_simple>/192 -0.0250 -0.0250 41 40 41 40
mod<addmod_simple>/256 -0.0008 -0.0008 35 35 35 35
mod<addmod_prenormalize>/64 -0.0137 -0.0137 66 65 66 65
mod<addmod_prenormalize>/128 +0.0170 +0.0170 69 70 69 70
mod<addmod_prenormalize>/192 -0.0308 -0.0308 81 79 81 79
mod<addmod_prenormalize>/256 -0.0205 -0.0205 62 61 62 61
mod<addmod_daosvik>/64 -0.0034 -0.0034 33 32 33 32
mod<addmod_daosvik>/128 -0.0102 -0.0102 36 36 36 36
mod<addmod_daosvik>/192 -0.0448 -0.0448 43 41 43 41
mod<addmod_daosvik>/256 -0.0248 -0.0248 36 35 36 35
mod<mulmod>/64 +0.0123 +0.0123 49 49 49 49
mod<mulmod>/128 +0.0191 +0.0191 59 60 59 60
mod<mulmod>/192 -0.0710 -0.0710 85 79 85 79
mod<mulmod>/256 -0.0648 -0.0648 82 77 82 77
ecmod<addmod_public> +0.0164 +0.0163 19 19 19 19
ecmod<addmod_simple> -0.0050 -0.0050 26 26 26 26
ecmod<addmod_prenormalize> -0.0016 -0.0016 10 10 10 10
ecmod<addmod_daosvik> +0.0006 +0.0006 18 18 18 18
ecmod<mulmod> -0.0508 -0.0508 79 75 79 75 
codecov-commenter
Inline subc() in submul() and use borrow flag idiom which is recognizable by GCC.