Re: [PATCH v3] hugetlb: simplify hugetlb handling in follow_page_mask

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi, Mike,

On Sun, Sep 18, 2022 at 07:13:48PM -0700, Mike Kravetz wrote:
> +struct page *hugetlb_follow_page_mask(struct vm_area_struct *vma,
> +				unsigned long address, unsigned int flags)
> +{
> +	struct hstate *h = hstate_vma(vma);
> +	struct mm_struct *mm = vma->vm_mm;
> +	unsigned long haddr = address & huge_page_mask(h);
> +	struct page *page = NULL;
> +	spinlock_t *ptl;
> +	pte_t *pte, entry;
> +
> +	/*
> +	 * FOLL_PIN is not supported for follow_page(). Ordinary GUP goes via
> +	 * follow_hugetlb_page().
> +	 */
> +	if (WARN_ON_ONCE(flags & FOLL_PIN))
> +		return NULL;
> +
> +retry:
> +	/*
> +	 * vma lock prevents racing with another thread doing a pmd unshare.
> +	 * This keeps pte as returned by huge_pte_offset valid.
> +	 */
> +	hugetlb_vma_lock_read(vma);

I'm not sure whether it's okay to take a rwsem here, as the code can be
called by e.g. FOLL_NOWAIT?

I'm wondering whether it's fine to just drop this anyway, just always walk
it lockless.  IIUC gup callers should be safe here because the worst case
is the caller will fetch a wrong page, but then it should be invalidated
very soon with mmu notifiers.  One thing worth mention is that pmd unshare
should never free a pgtable page.

IIUC it's also the same as fast-gup - afaiu we don't take the read vma lock
in fast-gup too but I also think it's safe.  But I hope I didn't miss
something.

-- 
Peter Xu





[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux