Re: [PATCH v9 08/69] mm: start tracking VMAs with maple tree

[Qian Cai <quic_qiancai@quicinc.com>]

On Wed, May 04, 2022 at 01:07:52AM +0000, Liam Howlett wrote:
> From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
> 
> Start tracking the VMAs with the new maple tree structure in parallel with
> the rb_tree.  Add debug and trace events for maple tree operations and
> duplicate the rb_tree that is created on forks into the maple tree.
> 
> The maple tree is added to the mm_struct including the mm_init struct,
> added support in required mm/mmap functions, added tracking in kernel/fork
> for process forking, and used to find the unmapped_area and checked
> against what the rbtree finds.
> 
> This also moves the mmap_lock() in exit_mmap() since the oom reaper call
> does walk the VMAs.  Otherwise lockdep will be unhappy if oom happens.
> 
> Signed-off-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
...
> @@ -2479,9 +2693,14 @@ int expand_downwards(struct vm_area_struct *vma,
>  			return -ENOMEM;
>  	}
>  
> +	if (mas_preallocate(&mas, vma, GFP_KERNEL))

We have thousands of leaks reports here from kmemleak while running some
syscall fuzzers on today's linux-next tree.

unreferenced object 0xffff0803c0063d00 (size 256):
  comm "trinity-c25", pid 177739, jiffies 4297781293 (age 354.988s)
  hex dump (first 32 bytes):
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
     slab_post_alloc_hook
     kmem_cache_alloc_bulk
     mas_alloc_nodes
     mas_preallocate
     mas_preallocate at lib/maple_tree.c:5579
     expand_downwards
     expand_downwards at mm/mmap.c:2094
     expand_stack
     do_page_fault
     do_translation_fault
     do_mem_abort
     el1_abort
     el1h_64_sync_handler
     el1h_64_sync
     handle_futex_death
     exit_robust_list
     futex_exit_release
     exit_mm_release

> +		return -ENOMEM;
> +
>  	/* We must make sure the anon_vma is allocated. */
> -	if (unlikely(anon_vma_prepare(vma)))
> +	if (unlikely(anon_vma_prepare(vma))) {
> +		mas_destroy(&mas);
>  		return -ENOMEM;
> +	}
>  
>  	/*
>  	 * vma->vm_start/vm_end cannot change under us because the caller