On 25-Mär 13:07, Kees Cook wrote:

On Wed, Mar 25, 2020 at 08:39:56PM +0100, KP Singh wrote:

quoted

On 25-Mär 12:28, Kees Cook wrote:

quoted

On Wed, Mar 25, 2020 at 04:26:24PM +0100, KP Singh wrote:

quoted

+noinline __weak RET bpf_lsm_##NAME(__VA_ARGS__)	\

I don't think the __weak is needed any more here?

This was suggested in:

 https://lore.kernel.org/bpf/20200221022537.wbmhdfkdbfvw2pww@ast-mbp/ (local)

"I think I saw cases when gcc ignored 'noinline' when function is
defined in the same file and still performed inlining while keeping
the function body.  To be safe I think __weak is necessary. That will
guarantee noinline."

It happened to work nicely with the previous approach for the special
hooks but the actual reason for adding the __weak was to guarrantee
that these functions don't get inlined.

Oh, hrm. Well, okay. That rationale would imply that the "noinline"
macro needs adjustment instead, but that can be separate, something like:

include/linux/compiler_attributes.h

-#define noinline __attribute__((__noinline__))
+#define noinline __attribute__((__noinline__)) __attribute__((__weak__))

With a comment, etc...

Sounds reasonable, I will drop the __weak from this and send a
separate patch for this.

- KP

-Kees

quoted

quoted

quoted

+{						\
+	return DEFAULT;				\

I'm impressed that LSM_RET_VOID actually works. :)

All the credit goes to Andrii :)

- KP

quoted

-Kees

quoted

+}
+
+#include <linux/lsm_hook_defs.h>
+#undef LSM_HOOK
 
 const struct bpf_prog_ops lsm_prog_ops = {
 };
-- 
2.20.1

-- 
Kees Cook

-- 
Kees Cook