From mboxrd@z Thu Jan  1 00:00:00 1970
From: baruch@tkos.co.il (Baruch Siach)
Date: Fri, 22 Oct 2010 08:28:26 +0200
Subject: [PATCH RESEND] ARM: fix spinlock recursion in adjust_pte()
In-Reply-To: <1287680982-22971-1-git-send-email-mika.westerberg@iki.fi>
References: <20101021164433.GH4772@gw.healthdatacare.com>
	<1287680982-22971-1-git-send-email-mika.westerberg@iki.fi>
Message-ID: <20101022062825.GH6225@tarshish>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

Hi Mika,

On Thu, Oct 21, 2010 at 08:09:42PM +0300, Mika Westerberg wrote:
> When running following code in a machine which has VIVT caches and
> USE_SPLIT_PTLOCKS is not defined:
> 
>   fd = open("/etc/passwd", O_RDONLY);
>   addr = mmap(NULL, 4096, PROT_READ, MAP_SHARED, fd, 0);
>   addr2 = mmap(NULL, 4096, PROT_READ, MAP_SHARED, fd, 0);
> 
>   v = *((int *)addr);
> 
> we will hang in spinlock recursion in the page fault handler:
> 
>   BUG: spinlock recursion on CPU#0, mmap_test/717

[snip]

Do you have any idea when was this bug introduced? Does it affect already 
release kernels other than .36?

baruch

> Same thing can be achieved by running:
> 
>   # useradd dummy
> 
> This comes from the fact that when USE_SPLIT_PTLOCKS is not defined,
> the only lock protecting the page tables is mm->page_table_lock
> which is already locked before update_mmu_cache() is called.
> 
> Signed-off-by: Mika Westerberg <mika.westerberg@iki.fi>
> ---
>  arch/arm/mm/fault-armv.c |   28 ++++++++++++++++++++++++++--
>  1 files changed, 26 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
> index 9b906de..56036ff 100644
> --- a/arch/arm/mm/fault-armv.c
> +++ b/arch/arm/mm/fault-armv.c
> @@ -65,6 +65,30 @@ static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
>  	return ret;
>  }
>  
> +#if USE_SPLIT_PTLOCKS
> +/*
> + * If we are using split PTE locks, then we need to take the page
> + * lock here.  Otherwise we are using shared mm->page_table_lock
> + * which is already locked, thus cannot take it.
> + */
> +static inline void do_pte_lock(spinlock_t *ptl)
> +{
> +	/*
> +	 * Use nested version here to indicate that we are already
> +	 * holding one similar spinlock.
> +	 */
> +	spin_lock_nested(ptl, SINGLE_DEPTH_NESTING);
> +}
> +
> +static inline void do_pte_unlock(spinlock_t *ptl)
> +{
> +	spin_unlock(ptl);
> +}
> +#else /* !USE_SPLIT_PTLOCKS */
> +static inline void do_pte_lock(spinlock_t *ptl) {}
> +static inline void do_pte_unlock(spinlock_t *ptl) {}
> +#endif /* USE_SPLIT_PTLOCKS */
> +
>  static int adjust_pte(struct vm_area_struct *vma, unsigned long address,
>  	unsigned long pfn)
>  {
> @@ -89,11 +113,11 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address,
>  	 */
>  	ptl = pte_lockptr(vma->vm_mm, pmd);
>  	pte = pte_offset_map_nested(pmd, address);
> -	spin_lock(ptl);
> +	do_pte_lock(ptl);
>  
>  	ret = do_adjust_pte(vma, address, pfn, pte);
>  
> -	spin_unlock(ptl);
> +	do_pte_unlock(ptl);
>  	pte_unmap_nested(pte);
>  
>  	return ret;
> -- 
> 1.5.6.5

-- 
                                                     ~. .~   Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
   - baruch at tkos.co.il - tel: +972.2.679.5364, http://www.tkos.co.il -