How can I avoid glitches in behavioural vhdl code simulations? [closed]

I've been trying to write VHDL code for a counter. Ultimately, I would like to take the output value, check it with a constant value, and see if the counter has reached the constant value. if it reaches there, I want to reset it. and if not I want to keep counting. What I wrote for the counter process is the following:

CNT: process(clk,reset)
begin
     if (reset='0') then
        cnt_tmp<= (others=>'0');
     elsif (rising_edge(clk)) then
        if (enable= '1') then
            if (match='1') then
                cnt_tmp <= (others => '0');
            else
                cnt_tmp <= cnt_tmp + 1;
            end if;
        else
            cnt_tmp <= cnt_tmp;
        end if;
    end if;
end process;
cnt<= cnt_tmp; 

So, when the match is 1, we synchronously reset the counter and it also has an enable pin.

This match signal is generated by bitwise xnor the constant with the counter output and "and" of the xnor outputs all together.

However, when I run the simulation, I see some glitches like this:

Does anyone have any idea how can I deal with this problem?

I am expecting to eliminate these glitches even though they don't affect the functionality in my simulations.