Messages in this thread | | | Date | Wed, 17 Apr 2024 13:23:18 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 00/19] Enable -Wshadow=local for kernel/sched |
| |
* Matthew Wilcox <willy@infradead.org> wrote:
> On Tue, Apr 16, 2024 at 05:29:02PM -0700, Linus Torvalds wrote: > > On Tue, 16 Apr 2024 at 14:15, Kees Cook <keescook@chromium.org> wrote: > > > > > > I was looking at -Wshadow=local again, and remembered this series. It > > > sounded like things were close, but a tweak was needed. What would be > > > next to get this working? > > > > So what is the solution to > > > > #define MAX(a,b) ({ \ > > typeof(a) __a = (a); \ > > typeof(b) __b = (b); \ > > __a > __b ? __a : __b; \ > > }) > > #define __MAX(a, __a, b, __b) ({ \ > typeof(a) __a = (a); \ > typeof(b) __b = (b); \ > __a > __b ? __a : __b; \ > }) > > #define MAX(a, b) __MAX(a, UNIQUE_ID(a), b, UNIQUE_ID(b)) > > At least, I think that was the plan. This was two years ago and I've > mostly forgotten.
I think as long as we can keep any additional complexity inside macros it would be acceptable, at least from the scheduler's POV. A UNIQUE_ID() layer of indirection for names doesn't sound look a too high price.
I had good reasults with -Wshadow in user-space projects: once the false positives got ironed out, the vast percentage of new warnings was for genuinely problematic new code. But they rarely used block-nested macros like the kernel does.
Thanks,
Ingo
| |