linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Wei Yang <richardw.yang@linux.intel.com>
To: David Rientjes <rientjes@google.com>
Cc: Wei Yang <richardw.yang@linux.intel.com>,
	akpm@linux-foundation.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/4] mm/page_alloc.c: extract commom part to check page
Date: Mon, 20 Jan 2020 08:33:59 +0800	[thread overview]
Message-ID: <20200120003359.GB26292@richard> (raw)
In-Reply-To: <alpine.DEB.2.21.2001191401240.43388@chino.kir.corp.google.com>

On Sun, Jan 19, 2020 at 02:06:18PM -0800, David Rientjes wrote:
>On Sun, 19 Jan 2020, Wei Yang wrote:
>
>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>> index d047bf7d8fd4..8cd06729169f 100644
>> --- a/mm/page_alloc.c
>> +++ b/mm/page_alloc.c
>> @@ -1025,13 +1025,9 @@ static inline bool page_expected_state(struct page *page,
>>  	return true;
>>  }
>>  
>> -static void free_pages_check_bad(struct page *page)
>> +static inline const char *__check_page(struct page *page)
>>  {
>> -	const char *bad_reason;
>> -	unsigned long bad_flags;
>> -
>> -	bad_reason = NULL;
>> -	bad_flags = 0;
>> +	const char *bad_reason = NULL;
>>  
>>  	if (unlikely(atomic_read(&page->_mapcount) != -1))
>>  		bad_reason = "nonzero mapcount";
>> @@ -1039,14 +1035,23 @@ static void free_pages_check_bad(struct page *page)
>>  		bad_reason = "non-NULL mapping";
>>  	if (unlikely(page_ref_count(page) != 0))
>>  		bad_reason = "nonzero _refcount";
>> -	if (unlikely(page->flags & PAGE_FLAGS_CHECK_AT_FREE)) {
>> -		bad_reason = "PAGE_FLAGS_CHECK_AT_FREE flag(s) set";
>> -		bad_flags = PAGE_FLAGS_CHECK_AT_FREE;
>> -	}
>>  #ifdef CONFIG_MEMCG
>>  	if (unlikely(page->mem_cgroup))
>>  		bad_reason = "page still charged to cgroup";
>>  #endif
>> +	return bad_reason;
>> +}
>> +
>> +static void free_pages_check_bad(struct page *page)
>> +{
>> +	const char *bad_reason = NULL;
>> +	unsigned long bad_flags = 0;
>> +
>> +	bad_reason = __check_page(page);
>> +	if (unlikely(page->flags & PAGE_FLAGS_CHECK_AT_FREE)) {
>> +		bad_reason = "PAGE_FLAGS_CHECK_AT_FREE flag(s) set";
>> +		bad_flags = PAGE_FLAGS_CHECK_AT_FREE;
>> +	}
>>  	bad_page(page, bad_reason, bad_flags);
>>  }
>>  
>> @@ -2044,12 +2049,7 @@ static void check_new_page_bad(struct page *page)
>>  	const char *bad_reason = NULL;
>>  	unsigned long bad_flags = 0;
>>  
>> -	if (unlikely(atomic_read(&page->_mapcount) != -1))
>> -		bad_reason = "nonzero mapcount";
>> -	if (unlikely(page->mapping != NULL))
>> -		bad_reason = "non-NULL mapping";
>> -	if (unlikely(page_ref_count(page) != 0))
>> -		bad_reason = "nonzero _refcount";
>> +	bad_reason = __check_page(page);
>>  	if (unlikely(page->flags & __PG_HWPOISON)) {
>>  		bad_reason = "HWPoisoned (hardware-corrupted)";
>>  		bad_flags = __PG_HWPOISON;
>> @@ -2061,10 +2061,6 @@ static void check_new_page_bad(struct page *page)
>>  		bad_reason = "PAGE_FLAGS_CHECK_AT_PREP flag set";
>>  		bad_flags = PAGE_FLAGS_CHECK_AT_PREP;
>>  	}
>> -#ifdef CONFIG_MEMCG
>> -	if (unlikely(page->mem_cgroup))
>> -		bad_reason = "page still charged to cgroup";
>> -#endif
>>  	bad_page(page, bad_reason, bad_flags);
>>  }
>>  
>
>I think this is compounding a previous problem in these functions: these 
>are all "if" clauses, not "else if" clauses so they are presumably ordered 
>based on least significant to most significant (we only see the last 
>bad_reason that we find).  For the page->mem_cgroup check, this leaves 
>bad_flags set but it doesn't match bad_reason.
>

I have thought about this. And curious about the order of those reasons.

>Could you instead fix the problem with these functions so that we actually 
>list *all* the problems with the page rather than only the last 
>conditional that is true?

Sure, thanks for the suggestion.

-- 
Wei Yang
Help you, Help me


  reply	other threads:[~2020-01-20  0:33 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-19 13:14 [PATCH 0/4] mm/page_alloc.c: cleanup on " Wei Yang
2020-01-19 13:14 ` [PATCH 1/4] mm/page_alloc.c: extract commom part to " Wei Yang
2020-01-19 22:06   ` David Rientjes
2020-01-20  0:33     ` Wei Yang [this message]
2020-01-19 13:14 ` [PATCH 2/4] mm/page_alloc.c: rename free_pages_check_bad() to check_free_page_bad() Wei Yang
2020-01-19 22:07   ` David Rientjes
2020-01-20  0:36     ` Wei Yang
2020-01-19 13:14 ` [PATCH 3/4] mm/page_alloc.c: rename free_pages_check() to check_free_page() Wei Yang
2020-01-19 13:14 ` [PATCH 4/4] mm/page_alloc.c: bad_[reason|flags] is not necessary when PageHWPoison Wei Yang
2020-01-19 22:07   ` David Rientjes

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200120003359.GB26292@richard \
    --to=richardw.yang@linux.intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=rientjes@google.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox