Hi Geert,
thanks for reviewing this!
Am 23.06.2021 um 19:26 schrieb Geert Uytterhoeven:
Hi Michael,
On Thu, Jun 17, 2021 at 7:39 AM Michael Schmitz <schmitzmic@xxxxxxxxx> wrote:
Add secure_computing() call to syscall_trace_enter to actually
filter system calls.
Add necessary arch Kconfig options, define TIF_SECCOMP trace
flag and provide basic seccomp filter support in asm/syscall.h
syscall_get_nr currently uses the syscall nr stored in orig_d0
because we change d0 to a default return code before starting a
syscall trace. This may be inconsistent with syscall_rollback
copying orig_d0 to d0 (which we never check upon return from
trace). We use d0 for the return code from syscall_trace_enter
in entry.S currently, and could perhaps expand that to store
a new syscall number returned by the seccomp filter before
executing the syscall. This clearly needs some discussion.
Compiles (for Atari) and boots on ARAnyM, otherwise untested.
Signed-off-by: Michael Schmitz <schmitzmic@xxxxxxxxx>
Thanks for your patch!
--- a/arch/m68k/include/asm/syscall.h
+++ b/arch/m68k/include/asm/syscall.h
@@ -4,6 +4,39 @@
#include <uapi/linux/audit.h>
+#include <asm/unistd.h>
+
+extern const unsigned long sys_call_table[];
+
+static inline int syscall_get_nr(struct task_struct *task,
+ struct pt_regs *regs)
+{
+ return regs->orig_d0;
+}
+
+static inline void syscall_rollback(struct task_struct *task,
+ struct pt_regs *regs)
+{
+ regs->d0 = regs->orig_d0;
+}
+
+static inline void syscall_set_return_value(struct task_struct *task,
+ struct pt_regs *regs,
+ int error, long val)
+{
+ regs->d0 = (long) error ? error : val;
+}
+
+static inline void syscall_get_arguments(struct task_struct *task,
+ struct pt_regs *regs,
+ unsigned long *args)
+{
+ args[0] = regs->orig_d0;
+ args++;
+
+ memcpy(args, ®s->d0 + 1, 5 * sizeof(args[0]));
This doesn't look right to me: "®s->d0 + 1" is "®s->orig_d0"
again, and there are no registers after that.
Perhaps you meant "®s->d1"?
Well spotted - that's a copy&paste error I didn't catch.
I'll look at the patch you referenced in the other mail later - Adrian
was going to test my patch with his libseccomp version. If I can build a
simple test case myself, all the better.
Cheers,
Michael
Gr{oetje,eeting}s,
Geert