Re: [RFC PATCH] mm, memory_hotplug: fix off-by-one in is_pageblock_removable

[Mike Rapoport <rppt@linux.ibm.com>]

On Mon, Feb 18, 2019 at 07:15:44PM +0100, Michal Hocko wrote:
> From: Michal Hocko <mhocko@suse.com>
> 
> Rong Chen has reported the following boot crash
> [   40.305212] PGD 0 P4D 0
> [   40.308255] Oops: 0000 [#1] PREEMPT SMP PTI
> [   40.313055] CPU: 1 PID: 239 Comm: udevd Not tainted 5.0.0-rc4-00149-gefad4e4 #1
> [   40.321348] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014
> [   40.330813] RIP: 0010:page_mapping+0x12/0x80
> [   40.335709] Code: 5d c3 48 89 df e8 0e ad 02 00 85 c0 75 da 89 e8 5b 5d c3 0f 1f 44 00 00 53 48 89 fb 48 8b 43 08 48 8d 50 ff a8 01 48 0f 45 da <48> 8b 53 08 48 8d 42 ff 83 e2 01 48 0f 44 c3 48 83 38 ff 74 2f 48
> [   40.356704] RSP: 0018:ffff88801fa87cd8 EFLAGS: 00010202
> [   40.362714] RAX: ffffffffffffffff RBX: fffffffffffffffe RCX: 000000000000000a
> [   40.370798] RDX: fffffffffffffffe RSI: ffffffff820b9a20 RDI: ffff88801e5c0000
> [   40.378830] RBP: 6db6db6db6db6db7 R08: ffff88801e8bb000 R09: 0000000001b64d13
> [   40.386902] R10: ffff88801fa87cf8 R11: 0000000000000001 R12: ffff88801e640000
> [   40.395033] R13: ffffffff820b9a20 R14: ffff88801f145258 R15: 0000000000000001
> [   40.403138] FS:  00007fb2079817c0(0000) GS:ffff88801dd00000(0000) knlGS:0000000000000000
> [   40.412243] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [   40.418846] CR2: 0000000000000006 CR3: 000000001fa82000 CR4: 00000000000006a0
> [   40.426951] Call Trace:
> [   40.429843]  __dump_page+0x14/0x2c0
> [   40.433947]  is_mem_section_removable+0x24c/0x2c0
> [   40.439327]  removable_show+0x87/0xa0
> [   40.443613]  dev_attr_show+0x25/0x60
> [   40.447763]  sysfs_kf_seq_show+0xba/0x110
> [   40.452363]  seq_read+0x196/0x3f0
> [   40.456282]  __vfs_read+0x34/0x180
> [   40.460233]  ? lock_acquire+0xb6/0x1e0
> [   40.464610]  vfs_read+0xa0/0x150
> [   40.468372]  ksys_read+0x44/0xb0
> [   40.472129]  ? do_syscall_64+0x1f/0x4a0
> [   40.476593]  do_syscall_64+0x5e/0x4a0
> [   40.480809]  ? trace_hardirqs_off_thunk+0x1a/0x1c
> [   40.486195]  entry_SYSCALL_64_after_hwframe+0x49/0xbe
> 
> and bisected it down to efad4e475c31 ("mm, memory_hotplug:
> is_mem_section_removable do not pass the end of a zone"). The reason for
> the crash is that the mapping is garbage for poisoned (uninitialized) page.
> This shouldn't happen as all pages in the zone's boundary should be
> initialized. Later debugging revealed that the actual problem is an
> off-by-one when evaluating the end_page. start_pfn + nr_pages resp.
> zone_end_pfn refers to a pfn after the range and as such it might belong
> to a differen memory section. This along with CONFIG_SPARSEMEM then
> makes the loop condition completely bogus because a pointer arithmetic
> doesn't work for pages from two different sections in that memory model.
> 
> Fix the issue by reworking is_pageblock_removable to be pfn based and
> only use struct page where necessary. This makes the code slightly
> easier to follow and we will remove the problematic pointer arithmetic
> completely.
> 
> Fixes: efad4e475c31 ("mm, memory_hotplug: is_mem_section_removable do not pass the end of a zone")
> Reported-by: <rong.a.chen@intel.com>
> Signed-off-by: Michal Hocko <mhocko@suse.com>

Acked-by: Mike Rapoport <rppt@linux.ibm.com>

> ---
>  mm/memory_hotplug.c | 27 +++++++++++++++------------
>  1 file changed, 15 insertions(+), 12 deletions(-)
> 
> diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
> index 124e794867c5..1ad28323fb9f 100644
> --- a/mm/memory_hotplug.c
> +++ b/mm/memory_hotplug.c
> @@ -1188,11 +1188,13 @@ static inline int pageblock_free(struct page *page)
>  	return PageBuddy(page) && page_order(page) >= pageblock_order;
>  }
>  
> -/* Return the start of the next active pageblock after a given page */
> -static struct page *next_active_pageblock(struct page *page)
> +/* Return the pfn of the start of the next active pageblock after a given pfn */
> +static unsigned long next_active_pageblock(unsigned long pfn)
>  {
> +	struct page *page = pfn_to_page(pfn);
> +
>  	/* Ensure the starting page is pageblock-aligned */
> -	BUG_ON(page_to_pfn(page) & (pageblock_nr_pages - 1));
> +	BUG_ON(pfn & (pageblock_nr_pages - 1));
>  
>  	/* If the entire pageblock is free, move to the end of free page */
>  	if (pageblock_free(page)) {
> @@ -1200,16 +1202,16 @@ static struct page *next_active_pageblock(struct page *page)
>  		/* be careful. we don't have locks, page_order can be changed.*/
>  		order = page_order(page);
>  		if ((order < MAX_ORDER) && (order >= pageblock_order))
> -			return page + (1 << order);
> +			return pfn + (1 << order);
>  	}
>  
> -	return page + pageblock_nr_pages;
> +	return pfn + pageblock_nr_pages;
>  }
>  
> -static bool is_pageblock_removable_nolock(struct page *page)
> +static bool is_pageblock_removable_nolock(unsigned long pfn)
>  {
> +	struct page *page = pfn_to_page(pfn);
>  	struct zone *zone;
> -	unsigned long pfn;
>  
>  	/*
>  	 * We have to be careful here because we are iterating over memory
> @@ -1232,13 +1234,14 @@ static bool is_pageblock_removable_nolock(struct page *page)
>  /* Checks if this range of memory is likely to be hot-removable. */
>  bool is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages)
>  {
> -	struct page *page = pfn_to_page(start_pfn);
> -	unsigned long end_pfn = min(start_pfn + nr_pages, zone_end_pfn(page_zone(page)));
> -	struct page *end_page = pfn_to_page(end_pfn);
> +	unsigned long end_pfn, pfn;
> +
> +	end_pfn = min(start_pfn + nr_pages,
> +			zone_end_pfn(page_zone(pfn_to_page(start_pfn))));
>  
>  	/* Check the starting page of each pageblock within the range */
> -	for (; page < end_page; page = next_active_pageblock(page)) {
> -		if (!is_pageblock_removable_nolock(page))
> +	for (pfn = start_pfn; pfn < end_pfn; pfn = next_active_pageblock(pfn)) {
> +		if (!is_pageblock_removable_nolock(pfn))
>  			return false;
>  		cond_resched();
>  	}
> -- 
> 2.20.1
> 

-- 
Sincerely yours,
Mike.