>
> Here it is, you're initializing 'prefix' here although it is being
> used in the previous patch. You should concentrate the whole prefix
> initialization and passing in one patch so that there are no breakages.
Yes I probably could have broken up the patches in a cleaner way. Thanks.
>
> Also, why is it exactly 44 chars long, is that some magic number
> that always works? At least this is what aer_print_error does
> too. I'm guessing someone chose it because it is large enough for
> dev_driver_string() and dev_name() and a couple more formatting
> characters. Oh well.
>
Yes, I was using the same code as in aer_print_error. This was brought up in a previous mail with Joe and we discussed cleaning all of this up in a later patch.
Lance
��.n��������+%������w��{.n�����{�����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
