On Thu, Nov 18, 2021 at 04:04:58PM -0600, Eric W. Biederman wrote: > > Recently to prevent issues with SECCOMP_RET_KILL and similar signals > being changed before they are delivered SA_IMMUTABLE was added. > > Unfortunately this broke debuggers[1][2] which reasonably expect to be > able to trap synchronous SIGTRAP and SIGSEGV even when the target > process is not configured to handle those signals. > > Update force_sig_to_task to support both the case when we can > allow the debugger to intercept and possibly ignore the > signal and the case when it is not safe to let userspace > known about the signal until the process has exited. > > Reported-by: Kyle Huey <me@xxxxxxxxxxxx> > Reported-by: kernel test robot <oliver.sang@xxxxxxxxx> > Cc: stable@xxxxxxxxxxxxxxx > [1] https://lkml.kernel.org/r/CAP045AoMY4xf8aC_4QU_-j7obuEPYgTcnQQP3Yxk=2X90jtpjw@xxxxxxxxxxxxxx > [2] https://lkml.kernel.org/r/20211117150258.GB5403@xsang-OptiPlex-902 > Fixes: 00b06da29cf9 ("signal: Add SA_IMMUTABLE to ensure forced siganls do not get changed") > Signed-off-by: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx> > --- > kernel/signal.c | 23 ++++++++++++++++------- > 1 file changed, 16 insertions(+), 7 deletions(-) > > diff --git a/kernel/signal.c b/kernel/signal.c > index 7c4b7ae714d4..02058c983bd6 100644 > --- a/kernel/signal.c > +++ b/kernel/signal.c > @@ -1298,6 +1298,12 @@ int do_send_sig_info(int sig, struct kernel_siginfo *info, struct task_struct *p > return ret; > } > > +enum sig_handler { > + HANDLER_CURRENT, /* If reachable use the current handler */ > + HANDLER_SIG_DFL, /* Always use SIG_DFL handler semantics */ > + HANDLER_EXIT, /* Only visible as the proces exit code */ Oh, I just noticed this typo "proces" -> "process" -Kees -- Kees Cook
