On Tue, May 12, 2020 at 06:35:04PM -0500, Tyler Hicks wrote: > On 2020-05-06 14:15:21, Kees Cook wrote: > > @@ -954,7 +965,11 @@ static void __init ramoops_register_dummy(void) > > pdata.console_size = ramoops_console_size; > > pdata.ftrace_size = ramoops_ftrace_size; > > pdata.pmsg_size = ramoops_pmsg_size; > > - pdata.dump_oops = dump_oops; > > + /* Parse deprecated module param "dump_oops" into "max_reason". */ > > + if (ramoops_dump_oops != -1) > > + pdata.max_reason = ramoops_dump_oops ? KMSG_DUMP_OOPS > > + : KMSG_DUMP_PANIC; > > + pdata.max_reason = ramoops_max_reason; > > This isn't quite right. We're conditionally assigning pdata.max_reason > and then immediately re-assigning it. > > IIUC, we're just missing an else block and it should look like this: > > /* Parse deprecated module param "dump_oops" into "max_reason". */ > if (ramoops_dump_oops != -1) > pdata.max_reason = ramoops_dump_oops ? KMSG_DUMP_OOPS > : KMSG_DUMP_PANIC; > else > pdata.max_reason = ramoops_max_reason; Whoops, I forgot to CC you Tyler! This should be fixed now here: https://lore.kernel.org/lkml/20200515184434.8470-6-keescook@xxxxxxxxxxxx/ -- Kees Cook
