linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: John Hubbard <jhubbard@nvidia.com>
To: jglisse@redhat.com, linux-mm@kvack.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org,
	Ralph Campbell <rcampbell@nvidia.com>,
	stable@vger.kernel.org, Evgeny Baskakov <ebaskakov@nvidia.com>,
	Mark Hairgrove <mhairgrove@nvidia.com>
Subject: Re: [PATCH 03/14] mm/hmm: HMM should have a callback before MM is destroyed v2
Date: Fri, 16 Mar 2018 21:39:00 -0700	[thread overview]
Message-ID: <b1406b15-858f-00a2-e2c1-a950d190f0e1@nvidia.com> (raw)
In-Reply-To: <b0cd570b-dfe4-4b42-18bb-967d1dbddcb3@nvidia.com>

On 03/16/2018 08:47 PM, John Hubbard wrote:
> On 03/16/2018 07:36 PM, John Hubbard wrote:
>> On 03/16/2018 12:14 PM, jglisse@redhat.com wrote:
>>> From: Ralph Campbell <rcampbell@nvidia.com>
>>>
>>
>> <snip>
>>
>>> +static void hmm_release(struct mmu_notifier *mn, struct mm_struct *mm)
>>> +{
>>> +	struct hmm *hmm = mm->hmm;
>>> +	struct hmm_mirror *mirror;
>>> +	struct hmm_mirror *mirror_next;
>>> +
>>> +	down_write(&hmm->mirrors_sem);
>>> +	list_for_each_entry_safe(mirror, mirror_next, &hmm->mirrors, list) {
>>> +		list_del_init(&mirror->list);
>>> +		if (mirror->ops->release)
>>> +			mirror->ops->release(mirror);
>>> +	}
>>> +	up_write(&hmm->mirrors_sem);
>>> +}
>>> +
>>
>> OK, as for actual code review:
>>
>> This part of the locking looks good. However, I think it can race against
>> hmm_mirror_register(), because hmm_mirror_register() will just add a new 
>> mirror regardless.
>>
>> So:
>>
>> thread 1                                      thread 2
>> --------------                                -----------------
>> hmm_release                                   hmm_mirror_register 
>>     down_write(&hmm->mirrors_sem);                <blocked: waiting for sem>
>>         // deletes all list items
>>     up_write
>>                                                   unblocked: adds new mirror
>>                                               
>>

Mark Hairgrove just pointed out some more fun facts:

1. Because hmm_mirror_register() needs to be called with an mm that has a non-zero
refcount, you generally cannot get an hmm_release callback, so the above race should
not happen.

2. We looked around, and the code is missing a call to mmu_notifier_unregister().
That means that it is going to leak memory and not let the mm get released either.

Maybe having each mirror have its own mmu notifier callback is a possible way
to solve this.

thanks,
-- 
John Hubbard
NVIDIA

  reply	other threads:[~2018-03-17  4:39 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-16 19:14 [PATCH 0/4] hmm: fixes and documentations v2 jglisse
2018-03-16 19:14 ` [PATCH 01/14] mm/hmm: documentation editorial update to HMM documentation jglisse
2018-03-16 19:14 ` [PATCH 02/14] mm/hmm: fix header file if/else/endif maze jglisse
2018-03-16 21:09   ` Andrew Morton
2018-03-16 21:18     ` Jerome Glisse
2018-03-16 21:35       ` Andrew Morton
2018-03-16 21:40         ` John Hubbard
2018-03-17  1:20   ` [PATCH 02/14] mm/hmm: fix header file if/else/endif maze v2 jglisse
2018-03-16 19:14 ` [PATCH 03/14] mm/hmm: HMM should have a callback before MM is destroyed v2 jglisse
2018-03-16 21:12   ` Andrew Morton
2018-03-16 21:26     ` Jerome Glisse
2018-03-16 21:37       ` John Hubbard
2018-03-17  2:36   ` John Hubbard
2018-03-17  3:47     ` John Hubbard
2018-03-17  4:39       ` John Hubbard [this message]
2018-03-16 19:14 ` [PATCH 04/14] mm/hmm: hmm_pfns_bad() was accessing wrong struct jglisse
2018-03-17  2:04   ` John Hubbard
2018-03-16 19:14 ` [PATCH 05/14] mm/hmm: use struct for hmm_vma_fault(), hmm_vma_get_pfns() parameters jglisse
2018-03-17  3:08   ` John Hubbard
2018-03-16 19:14 ` [PATCH 06/14] mm/hmm: remove HMM_PFN_READ flag and ignore peculiar architecture jglisse
2018-03-17  3:30   ` John Hubbard
2018-03-16 19:14 ` [PATCH 07/14] mm/hmm: use uint64_t for HMM pfn instead of defining hmm_pfn_t to ulong jglisse
2018-03-17  3:59   ` John Hubbard
2018-03-16 19:14 ` [PATCH 08/14] mm/hmm: cleanup special vma handling (VM_SPECIAL) jglisse
2018-03-17  4:35   ` John Hubbard
2018-03-16 19:14 ` [PATCH 09/14] mm/hmm: do not differentiate between empty entry or missing directory jglisse
2018-03-19 23:06   ` John Hubbard
2018-03-20  2:08     ` Jerome Glisse

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=b1406b15-858f-00a2-e2c1-a950d190f0e1@nvidia.com \
    --to=jhubbard@nvidia.com \
    --cc=akpm@linux-foundation.org \
    --cc=ebaskakov@nvidia.com \
    --cc=jglisse@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mhairgrove@nvidia.com \
    --cc=rcampbell@nvidia.com \
    --cc=stable@vger.kernel.org \
    /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;
as well as URLs for NNTP newsgroup(s).