Messages in this thread Patch in this message | | | From | Michael Ellerman <> | Subject | Re: [REGRESSION] x86/entry: Tracer no longer has opportunity to change the syscall number at entry via orig_ax | Date | Sun, 13 Sep 2020 17:44:09 +1000 |
| |
Kees Cook <keescook@chromium.org> writes: > On Wed, Sep 09, 2020 at 11:53:42PM +1000, Michael Ellerman wrote: >> I can observe the difference between v5.8 and mainline, using the >> raw_syscall trace event and running the seccomp_bpf selftest which turns >> a getpid (39) into a getppid (110). >> >> With v5.8 we see getppid on entry and exit: >> >> seccomp_bpf-1307 [000] .... 22974.874393: sys_enter: NR 110 (7ffff22c46e0, 40a350, 4, fffffffffffff7ab, 7fa6ee0d4010, 0) >> seccomp_bpf-1307 [000] .N.. 22974.874401: sys_exit: NR 110 = 1304 >> >> Whereas on mainline we see an enter for getpid and an exit for getppid: >> >> seccomp_bpf-1030 [000] .... 21.806766: sys_enter: NR 39 (7ffe2f6d1ad0, 40a350, 7ffe2f6d1ad0, 0, 0, 407299) >> seccomp_bpf-1030 [000] .... 21.806767: sys_exit: NR 110 = 1027 > > For my own notes, this is how I reproduced it: > > # ./perf-$VER record -e raw_syscalls:sys_enter -e raw_syscalls:sys_exit & > # ./seccomp_bpf > # fg > ctrl-c > # ./perf-$VER script | grep seccomp_bpf | awk '{print $7}' | sort | uniq -c > $VER.log > *repeat* > # diff -u old.log new.log > ... > > (Is there an easier way to get those results?)
I did more or less the same thing, except I ran the trace event manually (via debugfs), which is no better really.
I think the right way to test it would be to have a test that modifies the syscall via seccomp and also monitors the trace event using perf events. But that wouldn't be easier :)
> I will go see if I can figure out the best way to correct this.
I think this works?
diff --git a/kernel/entry/common.c b/kernel/entry/common.c index 18683598edbc..901361e2f8ea 100644 --- a/kernel/entry/common.c +++ b/kernel/entry/common.c @@ -60,13 +60,15 @@ static long syscall_trace_enter(struct pt_regs *regs, long syscall, return ret; } + syscall = syscall_get_nr(current, regs); + if (unlikely(ti_work & _TIF_SYSCALL_TRACEPOINT)) trace_sys_enter(regs, syscall); syscall_enter_audit(regs, syscall); /* The above might have changed the syscall number */ - return ret ? : syscall_get_nr(current, regs); + return ret ? : syscall; } static __always_inline long
cheers
| |