Re: [PATCH v2] arm64: ptrace: don't warn on missing siginfo in a group-stop

[Will Deacon]

On Mon, Oct 26, 2026 at 06:32:45PM +0000, Catalin Marinas wrote:
> On Fri, Oct 23, 2026 at 07:07:10PM +0800, Zhao Gongyi wrote:
> > PTRACE_ATTACH to a group-stopped child can enter do_jobctl_trap(),
> > which calls ptrace_stop() without siginfo. If the tracer writes
> > NT_ARM_SYSTEM_CALL, system_call_set() can change syscallno from
> > NO_SYSCALL before calling update_syscall_orig_x0_after_ptrace(). The
> > WARN_ON_ONCE(!info) then fires, although this is a valid ptrace stop.

I can't quite get my head around what it means to set NT_ARM_SYSTEM_CALL
in this context. Suddenly, in_syscall() will return true, which has
implications for things like signal handling that seem bizarre if we're
not actually in a syscall. Should we be avoiding this situation to start
with?

> > Remove the warning and guard the syscall-exit-stop check against a
> > missing siginfo. Keep recognizing both SIGTRAP and SIGTRAP | 0x80:
> > syscall-exit-stops use either value depending on whether the tracer
> > enabled PTRACE_O_TRACESYSGOOD. With no siginfo, allow the final
> > assignment to synchronize orig_x0 from x0.
> > 
> > Fixes: 88b839ce497c ("arm64: ptrace: Keep 'orig_x0' in-sync with x0 on syscall entry")
> > Cc: stable@xxxxxxxxxxxxxxx
> > Signed-off-by: Zhao Gongyi <zhaogongyi@xxxxxxxxxxxxx>
> > ---
> > Changes since v1:
> > - Remove the warning and early return, as suggested by Oleg.
> > - Keep recognizing syscall-exit-stops with or without PTRACE_O_TRACESYSGOOD.
> > 
> > v1: https://lore.kernel.org/r/20261016024536.18938-1-zhaogongyi@xxxxxxxxxxxxx
> > 
> > Tested on arm64 KVM at af32da41b032 with this v2 applied.
> > At a group-stop, PTRACE_GETSIGINFO returned -EINVAL and an
> > NT_ARM_SYSTEM_CALL write succeeded. Syscall-entry and seccomp
> > x0 rewrites worked. At syscall-exit-stops, NT_PRSTATUS writes succeeded
> > with and without PTRACE_O_TRACESYSGOOD (si_code 133 and 5,
> > respectively). No kernel warnings occurred.
> > 
> >  arch/arm64/kernel/ptrace.c | 6 +-----
> >  1 file changed, 1 insertion(+), 5 deletions(-)
> > 
> > diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
> > index f743cbec1..22fb9eeed 100644
> > --- a/arch/arm64/kernel/ptrace.c
> > +++ b/arch/arm64/kernel/ptrace.c
> > @@ -573,10 +573,6 @@ static void update_syscall_orig_x0_after_ptrace(struct task_struct *target)
> >  	if (regs->syscallno == NO_SYSCALL)
> >  		return;
> >  
> > -	/* We should only be called when target is in a ptrace stop */
> > -	if (WARN_ON_ONCE(!info))
> > -		return;
> 
> That's a slight semantic change since this function may now update
> orig_x0 even when !info. However, I think it makes more sense as the aim
> of commit 88b839ce497c ("arm64: ptrace: Keep 'orig_x0' in-sync with x0
> on syscall entry") was to update orig_x0 in all cases other than
> syscall-exit-stop. We can't have the latter with !info.
> 
> I'll wait for a couple of days to see if Will has any comments and queue
> it.

I was trying to limit the scope of this hack as much as possible, so I
think I'd prefer the early return tbh (i.e. just drop the WARN_ON_ONCE()
and update the comment).

Will