On Wed, 14 Oct 2020 23:16:14 +0800
Qiujun Huang <hqjag...@gmail.com> wrote:
> It may be better to check each page is aligned by 4 bytes. The 2
> least significant bits of the address will be used as flags.
>
> Signed-off-by: Qiujun Huang <hqjag...@gmail.com>
> ---
> kernel/trace/ring_buffer.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)
>
> diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c
> index 93ef0ab6ea20..9dec7d58b177 100644
> --- a/kernel/trace/ring_buffer.c
> +++ b/kernel/trace/ring_buffer.c
> @@ -1420,7 +1420,8 @@ static int rb_check_pages(struct ring_buffer_per_cpu
> *cpu_buffer)
> return 0;
> }
>
> -static int __rb_allocate_pages(long nr_pages, struct list_head *pages, int
> cpu)
> +static int __rb_allocate_pages(struct ring_buffer_per_cpu *cpu_buffer,
> + long nr_pages, struct list_head *pages, int cpu)
> {
> struct buffer_page *bpage, *tmp;
> bool user_thread = current->mm != NULL;
> @@ -1464,6 +1465,8 @@ static int __rb_allocate_pages(long nr_pages, struct
> list_head *pages, int cpu)
> if (!bpage)
> goto free_pages;
>
> + rb_check_bpage(cpu_buffer, bpage);
> +
>
Why add it here, and not just add this check to the scan in
rb_check_pages()?
-- Steve
