Re: [PATCH v2] mm/memmap: Prevent double scanning of memmap by kmemleak

[Weikang Guo <guoweikang.kernel@gmail.com>]

Thansk， Catalin

Catalin Marinas <catalin.marinas@arm.com> wrote on Friday, 3 January
2025 at 22:28
>
> On Fri, Jan 03, 2025 at 07:01:50PM +0800, Guo Weikang wrote:
> > diff --git a/mm/mm_init.c b/mm/mm_init.c
> > index 24b68b425afb..71b58f5f2492 100644
> > --- a/mm/mm_init.c
> > +++ b/mm/mm_init.c
> > @@ -1580,6 +1580,10 @@ static void __init free_area_init_core(struct pglist_data *pgdat)
> >       }
> >  }
> >
> > +/*
> > + * Kmemleak will explicitly scan mem_map by traversing all valid `struct *page`,
> > + * so memblock does not need to be added to the scan list.
> > + */
> >  void __init *memmap_alloc(phys_addr_t size, phys_addr_t align,
> >                         phys_addr_t min_addr, int nid, bool exact_nid)
> >  {
>
> Nitpick: normally I'd place the comment in the code, before the 'if'
> statement. We keep the comments above functions for a description of
> the function.
>

Thank you for the suggestion! I agree that placing the comment before
the 'if' statement is
a better approach. I will update the patch accordingly.

> > @@ -1587,11 +1591,11 @@ void __init *memmap_alloc(phys_addr_t size, phys_addr_t align,
> >
> >       if (exact_nid)
> >               ptr = memblock_alloc_exact_nid_raw(size, align, min_addr,
> > -                                                MEMBLOCK_ALLOC_ACCESSIBLE,
> > +                                                MEMBLOCK_ALLOC_NOLEAKTRACE,
> >                                                  nid);
> >       else
> >               ptr = memblock_alloc_try_nid_raw(size, align, min_addr,
> > -                                              MEMBLOCK_ALLOC_ACCESSIBLE,
> > +                                              MEMBLOCK_ALLOC_NOLEAKTRACE,
> >                                                nid);
> >
> >       if (ptr && size > 0)
> > diff --git a/mm/sparse-vmemmap.c b/mm/sparse-vmemmap.c
> > index cec67c5f37d8..903a5422907b 100644
> > --- a/mm/sparse-vmemmap.c
> > +++ b/mm/sparse-vmemmap.c
> > @@ -27,7 +27,7 @@
> >  #include <linux/spinlock.h>
> >  #include <linux/vmalloc.h>
> >  #include <linux/sched.h>
> > -
> > +#include "internal.h"
> >  #include <asm/dma.h>
> >  #include <asm/pgalloc.h>
>
> Another nit: if the original authors of this file preferred the includes
> to be in blocks, I'd keep the internal.h include as a separate block
> after the asm/ includes.
>

Here is the order I actually refer to `mm/sparse.c`, but after looking at
other files, I think your suggestion is better.

> Apart from these minor things, the patch looks fine.
>

Tanks!

> Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
>
> Thanks.
>
> --
> Catalin


---
Guo