Re: [PATCH v6 06/20] mm/vma: Change munmap to use vma_munmap_struct() for accounting and surrounding vmas

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Tue, Aug 20, 2024 at 07:57:15PM GMT, Liam R. Howlett wrote:
> From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
>
> Clean up the code by changing the munmap operation to use a structure
> for the accounting and munmap variables.
>
> Since remove_mt() is only called in one location and the contents will
> be reduced to almost nothing.  The remains of the function can be added
> to vms_complete_munmap_vmas().
>
> Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
> Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> Reviewed-by: Suren Baghdasaryan <surenb@google.com>
> ---
>  mm/vma.c | 79 ++++++++++++++++++++++++++++----------------------------
>  mm/vma.h |  6 +++++
>  2 files changed, 46 insertions(+), 39 deletions(-)
>
> diff --git a/mm/vma.c b/mm/vma.c
> index e1aee43a3dc4..7b8b8b983399 100644
> --- a/mm/vma.c
> +++ b/mm/vma.c

[snip]

>  	mm = vms->mm;
> @@ -731,21 +708,26 @@ static void vms_complete_munmap_vmas(struct vma_munmap_struct *vms,
>  	if (vms->unlock)
>  		mmap_write_downgrade(mm);
>
> -	prev = vma_iter_prev_range(vms->vmi);
> -	next = vma_next(vms->vmi);
> -	if (next)
> -		vma_iter_prev_range(vms->vmi);
> -
>  	/*
>  	 * We can free page tables without write-locking mmap_lock because VMAs
>  	 * were isolated before we downgraded mmap_lock.
>  	 */
>  	mas_set(mas_detach, 1);
> -	unmap_region(mm, mas_detach, vms->vma, prev, next, vms->start, vms->end,
> -		     vms->vma_count, !vms->unlock);
> -	/* Statistics and freeing VMAs */
> +	unmap_region(mm, mas_detach, vms->vma, vms->prev, vms->next,
> +		     vms->start, vms->end, vms->vma_count, !vms->unlock);
> +	/* Update high watermark before we lower total_vm */
> +	update_hiwater_vm(mm);
> +	/* Stat accounting */
> +	WRITE_ONCE(mm->total_vm, READ_ONCE(mm->total_vm) - vms->nr_pages);
> +	mm->exec_vm -= vms->exec_vm;
> +	mm->stack_vm -= vms->stack_vm;
> +	mm->data_vm -= vms->data_vm;

See below, but I bisected a bug to this patch that manifested because of
miscalculated accounting. So I wonder whether it'd be a good idea to take
this opportunity, when updating mm->... stats to add some:

VM_WARN_ON(vms->exec_vm > mm->exec_vm);

etc. for each of the fields updated. This would help catch any accounting
issues like this with CONFIG_DEBUG_VM switched on.

[snip]

> @@ -824,6 +807,22 @@ static int vms_gather_munmap_vmas(struct vma_munmap_struct *vms,
>  			goto munmap_gather_failed;
>
>  		vma_mark_detached(next, true);
> +		nrpages = vma_pages(next);
> +
> +		vms->nr_pages += nrpages;
> +		if (next->vm_flags & VM_LOCKED)
> +			vms->locked_vm += nrpages;

So I bisected a self-test failure, memfd_secret specifically, to this
commit. This is because you are double-counting VM_LOCKED...

> +
> +		if (next->vm_flags & VM_ACCOUNT)
> +			vms->nr_accounted += nrpages;
> +
> +		if (is_exec_mapping(next->vm_flags))
> +			vms->exec_vm += nrpages;
> +		else if (is_stack_mapping(next->vm_flags))
> +			vms->stack_vm += nrpages;
> +		else if (is_data_mapping(next->vm_flags))
> +			vms->data_vm += nrpages;
> +
>  		if (next->vm_flags & VM_LOCKED)
>  			vms->locked_vm += vma_pages(next);

...the double counting being right here :) so I think we should drop the
above couple lines.

>
> @@ -847,7 +846,9 @@ static int vms_gather_munmap_vmas(struct vma_munmap_struct *vms,
>  		BUG_ON(next->vm_start < vms->start);
>  		BUG_ON(next->vm_start > vms->end);
>  #endif
> -	} for_each_vma_range(*(vms->vmi), next, vms->end);
> +	}
> +
> +	vms->next = vma_next(vms->vmi);
>
>  #if defined(CONFIG_DEBUG_VM_MAPLE_TREE)
>  	/* Make sure no VMAs are about to be lost. */
> diff --git a/mm/vma.h b/mm/vma.h
> index cb67acf59012..cbf55e0e0c4f 100644
> --- a/mm/vma.h
> +++ b/mm/vma.h
> @@ -33,12 +33,18 @@ struct vma_munmap_struct {
>  	struct vma_iterator *vmi;
>  	struct mm_struct *mm;
>  	struct vm_area_struct *vma;     /* The first vma to munmap */
> +	struct vm_area_struct *prev;    /* vma before the munmap area */
> +	struct vm_area_struct *next;    /* vma after the munmap area */
>  	struct list_head *uf;           /* Userfaultfd list_head */
>  	unsigned long start;            /* Aligned start addr (inclusive) */
>  	unsigned long end;              /* Aligned end addr (exclusive) */
>  	int vma_count;                  /* Number of vmas that will be removed */
>  	unsigned long nr_pages;         /* Number of pages being removed */
>  	unsigned long locked_vm;        /* Number of locked pages */
> +	unsigned long nr_accounted;     /* Number of VM_ACCOUNT pages */
> +	unsigned long exec_vm;
> +	unsigned long stack_vm;
> +	unsigned long data_vm;
>  	bool unlock;                    /* Unlock after the munmap */
>  };
>
> --
> 2.43.0
>