On Wed, 2 Oct 2024 13:39:14 +0530 Brahmajit Das <brahmajit.xyz@xxxxxxxxx> wrote:
> GCC 15 enables -Werror=unterminated-string-initialization by default.
> This results in the following build error/s
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> 917 | [0 ... (BITS_PER_LONG-1)] = "??",
> | ^~~~
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> fs/proc/task_mmu.c:917:49: error: initializer-string for array of ‘char’ is too long [-Werror=unterminate d-string-initialization]
> ...
>
> To fix this, the length of the second argument of arary mnemonics needs
"array"
> to be 3 instead of previously set 2 (i.e. from [BITS_PER_LONG][2] to
> [BITS_PER_LONG][3])
>
Yes, I'm not surprised that little party trick we used in there fools
gcc. And really it deserves to die.
> --- a/fs/proc/task_mmu.c
> +++ b/fs/proc/task_mmu.c
> @@ -909,8 +909,15 @@ static void show_smap_vma_flags(struct seq_file *m, struct vm_area_struct *vma)
> {
> /*
> * Don't forget to update Documentation/ on changes.
> + *
> + * The length of the second argument of arary mnemonics
"array". But really, just using "mnemonics[]" conveys all we need.
> + * needs to be 3 instead of previously set 2
> + * (i.e. from [BITS_PER_LONG][2] to [BITS_PER_LONG][3])
> + * to avoid spurious
> + * -Werror=unterminated-string-initialization warning
> + * with GCC 15
> */
> - static const char mnemonics[BITS_PER_LONG][2] = {
> + static const char mnemonics[BITS_PER_LONG][3] = {
> /*
> * In case if we meet a flag we don't know about.
> */
If we want to preserve the party trick then perhaps we can use
- [0 ... (BITS_PER_LONG-1)] = { "??" },
+ [0 ... (BITS_PER_LONG-1)] = { '?', '?' },
and this will shut gcc up?
If we do remove the party trick (as you have done) then this:
if (vma->vm_flags & (1UL << i)) {
seq_putc(m, mnemonics[i][0]);
seq_putc(m, mnemonics[i][1]);
seq_putc(m, ' ');
}
can be simplified (and probably sped up) with
if (vma->vm_flags & (1UL << i))
seq_printf(m, "%s ", mnemonics[i]);
yes?
