Re: [PATCH -mm] mm: Fix races between swapoff and flush dcache
Andrew Morton writes:
> On Fri, 2 Mar 2018 16:04:26 +0800 "Huang, Ying" wrote:
>
>> From: Huang Ying
>>
>> >From commit 4b3ef9daa4fc ("mm/swap: split swap cache into 64MB
>> trunks") on, after swapoff, the address_space associated with the swap
>> device will be freed. So page_mapping() users which may touch the
>> address_space need some kind of mechanism to prevent the address_space
>> from being freed during accessing.
>>
>> The dcache flushing functions (flush_dcache_page(), etc) in
>> architecture specific code may access the address_space of swap device
>> for anonymous pages in swap cache via page_mapping() function. But in
>> some cases there are no mechanisms to prevent the swap device from
>> being swapoff, for example,
>>
>> CPU1 CPU2
>> __get_user_pages() swapoff()
>> flush_dcache_page()
>> mapping = page_mapping()
>> ... exit_swap_address_space()
>> ...kvfree(spaces)
>> mapping_mapped(mapping)
>>
>> The address space may be accessed after being freed.
>>
>> But from cachetlb.txt and Russell King, flush_dcache_page() only care
>> about file cache pages, for anonymous pages, flush_anon_page() should
>> be used. The implementation of flush_dcache_page() in all
>> architectures follows this too. They will check whether
>> page_mapping() is NULL and whether mapping_mapped() is true to
>> determine whether to flush the dcache immediately. And they will use
>> interval tree (mapping->i_mmap) to find all user space mappings.
>> While mapping_mapped() and mapping->i_mmap isn't used by anonymous
>> pages in swap cache at all.
>>
>> So, to fix the race between swapoff and flush dcache, __page_mapping()
>> is add to return the address_space for file cache pages and NULL
>> otherwise. All page_mapping() invoking in flush dcache functions are
>> replaced with __page_mapping().
>>
>> The patch is only build tested, because I have no machine with
>> architecture other than x86.
>>
>> ...
>>
>> +/*
>> + * For file cache pages, return the address_space, otherwise return NULL
>> + */
>> +struct address_space *__page_mapping(struct page *page)
>> +{
>> +struct address_space *mapping;
>> +
>> +page = compound_head(page);
>> +
>> +/* This happens if someone calls flush_dcache_page on slab page */
>> +if (unlikely(PageSlab(page)))
>> +return NULL;
>> +
>> +mapping = page->mapping;
>> +if ((unsigned long)mapping & PAGE_MAPPING_ANON)
>> +return NULL;
>> +
>> +return (void *)((unsigned long)mapping & ~PAGE_MAPPING_FLAGS);
>> +}
>> +
>
> I think page_mapping_file() would be a better name.
Thanks! I will use that name.
> And do we really need to duplicate page_mapping()? Could it be
>
> struct address_space *page_mapping_file(struct page *page)
> {
> if (PageSwapCache(page))
> return NULL;
> return page_mapping(page);
> }
Yes. This looks better.
> (We don't need to run compound_head() here, do we?)
Yes. I think so.
Best Regards,
Huang, Ying
Re: [PATCH -mm] mm: Fix races between swapoff and flush dcache
On Fri, 2 Mar 2018 16:04:26 +0800 "Huang, Ying" wrote:
> From: Huang Ying
>
> >From commit 4b3ef9daa4fc ("mm/swap: split swap cache into 64MB
> trunks") on, after swapoff, the address_space associated with the swap
> device will be freed. So page_mapping() users which may touch the
> address_space need some kind of mechanism to prevent the address_space
> from being freed during accessing.
>
> The dcache flushing functions (flush_dcache_page(), etc) in
> architecture specific code may access the address_space of swap device
> for anonymous pages in swap cache via page_mapping() function. But in
> some cases there are no mechanisms to prevent the swap device from
> being swapoff, for example,
>
> CPU1 CPU2
> __get_user_pages()swapoff()
> flush_dcache_page()
> mapping = page_mapping()
> ... exit_swap_address_space()
> ... kvfree(spaces)
> mapping_mapped(mapping)
>
> The address space may be accessed after being freed.
>
> But from cachetlb.txt and Russell King, flush_dcache_page() only care
> about file cache pages, for anonymous pages, flush_anon_page() should
> be used. The implementation of flush_dcache_page() in all
> architectures follows this too. They will check whether
> page_mapping() is NULL and whether mapping_mapped() is true to
> determine whether to flush the dcache immediately. And they will use
> interval tree (mapping->i_mmap) to find all user space mappings.
> While mapping_mapped() and mapping->i_mmap isn't used by anonymous
> pages in swap cache at all.
>
> So, to fix the race between swapoff and flush dcache, __page_mapping()
> is add to return the address_space for file cache pages and NULL
> otherwise. All page_mapping() invoking in flush dcache functions are
> replaced with __page_mapping().
>
> The patch is only build tested, because I have no machine with
> architecture other than x86.
>
> ...
>
> +/*
> + * For file cache pages, return the address_space, otherwise return NULL
> + */
> +struct address_space *__page_mapping(struct page *page)
> +{
> + struct address_space *mapping;
> +
> + page = compound_head(page);
> +
> + /* This happens if someone calls flush_dcache_page on slab page */
> + if (unlikely(PageSlab(page)))
> + return NULL;
> +
> + mapping = page->mapping;
> + if ((unsigned long)mapping & PAGE_MAPPING_ANON)
> + return NULL;
> +
> + return (void *)((unsigned long)mapping & ~PAGE_MAPPING_FLAGS);
> +}
> +
I think page_mapping_file() would be a better name.
And do we really need to duplicate page_mapping()? Could it be
struct address_space *page_mapping_file(struct page *page)
{
if (PageSwapCache(page))
return NULL;
return page_mapping(page);
}
(We don't need to run compound_head() here, do we?)
