Incorrect zero gradient with no error thrown

I have am trying to take gradients of a loss function which is defined via this function below. I need to take gradients of tensor_conv w.r.t W and H, which are not mutated inside the function:

function tensor_conv!(est, W::AbstractArray, H::AbstractArray)
    K, N, L = size(W)
    T = size(H, 2)

    @. est = 0
    for lag = 0:(L-1)
        @views s_dot!(est[:, lag+1:T], W[:, :, lag+1]', H, lag, 1, 1)
    end
    
    return est
end

function s_dot!(B, Wl, H, lag, α, β)
    K, T = size(H)
    
    if lag < 0
        @views mul!(B, Wl, H[:, 1+lag:T], α, β) 
    else  # lag >= 0
        @views mul!(B, Wl, H[:, 1:T-lag], α, β)
    end

    return B
end

However, reverseDiff always gives me zero gradient when differentiating this function, it does not throw an error. Can someone explain what might be going wrong?

Start by reproducing the zero-gradient result with tensor_conv! and s_dot! using ReverseDiff, then isolate whether the mutation, @views, or mul! call affects differentiation. Compare the computed gradients with expected values and document the cause or a minimal failing case; no repository file or test is named in the issue.