On Thu, Apr 4, 2024, at 20:17, Andrew Morton wrote:
> On Thu, 4 Apr 2024 14:44:30 +0200 Arnd Bergmann <arnd@xxxxxxxxxx> wrote:
>
>> From: Arnd Bergmann <arnd@xxxxxxxx>
>>
>> This header is no longer included implicitly and instead needs to be
>> pulled in directly:
>>
>> mm/kasan/hw_tags.c: In function 'unpoison_vmalloc_pages':
>> mm/kasan/hw_tags.c:280:16: error: implicit declaration of function 'find_vm_area'; did you mean 'find_vma_prev'? [-Werror=implicit-function-declaration]
>> 280 | area = find_vm_area((void *)addr);
>> | ^~~~~~~~~~~~
>> | find_vma_prev
>> mm/kasan/hw_tags.c:280:14: error: assignment to 'struct vm_struct *' from 'int' makes pointer from integer without a cast [-Werror=int-conversion]
>> 280 | area = find_vm_area((void *)addr);
>> | ^
>> mm/kasan/hw_tags.c:284:29: error: invalid use of undefined type 'struct vm_struct'
>> 284 | for (i = 0; i < area->nr_pages; i++) {
>> | ^~
>> mm/kasan/hw_tags.c:285:41: error: invalid use of undefined type 'struct vm_struct'
>> 285 | struct page *page = area->pages[i];
>> | ^~
>
> Thanks, but I'd like to know which patch this patch is fixing, please.
> Is it mainline or linux-next? I'm suspecting it might be a fix for
> fix-missing-vmalloch-includes.patch but without knowing how to
> reproduce this I can't determine anything.
It only showed up yesterday in linux-next. I thought about
bisecting it but ended up not doing it as it seemed simple
enough.
fix-missing-vmalloch-includes.patch looks like the right
place to me, given both the timing and contents, so please
fold my change into that.
Arnd
