def read_data():
    H, W = map(int, input().split())
    s = [input() for i in range(H)]
    return H, W, s

def parse_board(H, W, s):
    reds = [False] * (H * W)
    start = -1
    goal = -1
    for h in range(H):
        sh = s[h]
        for w in range(W):
            c = sh[w]
            if c == 'S':
                start = h * W + w
            elif c == 'G':
                goal = h * W + w
            elif c == 'R':
                reds[h * W + w] = True
    return reds, start, goal

def solve(H, W, s):
    next_move = [(), ()]
    next_move[0] = ((-2, -1), (-2, 1), (-1, -2), (-1, 2), (1, -2), (1, 2), (2, -1), (2, 1))
    next_move[1] = ((-1, -1), (-1, 1), (1, -1), (1, 1))
    reds, start, goal = parse_board(H, W, s)
    return wfs(start, goal, reds, H, W, next_move)

def wfs(start, goal, reds, H, W, next_move):
    HW = H * W
    dist = [-1] * (HW * 2)
    goals = [goal, goal + HW]
    steps = 0
    state = 0
    frontiers = [start]
    dist[start] = 0
    while frontiers:
        steps += 1
        new_frontiers = []
        for spos in frontiers:
            state, pos = divmod(spos, HW)
            h, w = divmod(pos, W)
            for dh, dw in next_move[state]:
                nh = h + dh
                nw = w + dw
                if 0 <= nh < H and 0 <= nw < W:
                    npos = nh * W + nw
                    new_state = 1 - state if reds[npos] else state
                    nspos = new_state * HW + npos
                    if dist[nspos] == -1:
                        dist[nspos] = steps
                        new_frontiers.append(nspos)
                        if nspos in goals:
                            return steps
        frontiers = new_frontiers
    return -1

if __name__ == '__main__':
    pars = read_data()
    print(solve(*pars))