Re: [PATCH v4 6/6] binder: encapsulate individual alloc test cases

[Kees Cook <kees@kernel.org>]

On Wed, Jul 16, 2025 at 11:32:10PM -0700, Kees Cook wrote:
> This really screams for a struct-based way to in-place declare a
> seq_buf. The current macro only works on the stack. I think this
> will work; I'll send a patch once I get it tested:
> 
> #define DECLARE_SEQ_BUF(NAME, SIZE)		\
> 	char NAME##_buffer[size];		\
> 	struct seq_buf NAME = {			\
> 		.buffer = &NAME##_buffer,	\
> 		.size = SIZE,			\
> 	}

Oh, no, that wouldn't work in a struct -- there is a static initializer
there. And my memory is short: this would have been a revert of commit
7a8e9cdf9405 ("seq_buf: Make DECLARE_SEQ_BUF() usable").

I don't think it's possible to construct a seq_buf in a struct without a
runtime initialization. And that's true for anything that needs to
actually set a value (size) for the data type to be usable. Hmpf.

-- 
Kees Cook