From: Mike Kravetz <mike.kravetz@oracle.com>
To: Roman Gushchin <guro@fb.com>, Minchan Kim <minchan@kernel.org>
Cc: David Hildenbrand <david@redhat.com>,
Michal Hocko <mhocko@suse.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Shakeel Butt <shakeelb@google.com>,
Oscar Salvador <osalvador@suse.de>,
Muchun Song <songmuchun@bytedance.com>,
David Rientjes <rientjes@google.com>,
Miaohe Lin <linmiaohe@huawei.com>,
Peter Zijlstra <peterz@infradead.org>,
Matthew Wilcox <willy@infradead.org>,
HORIGUCHI NAOYA <naoya.horiguchi@nec.com>,
"Aneesh Kumar K . V" <aneesh.kumar@linux.ibm.com>,
Waiman Long <longman@redhat.com>, Peter Xu <peterx@redhat.com>,
Mina Almasry <almasrymina@google.com>,
Hillf Danton <hdanton@sina.com>,
Andrew Morton <akpm@linux-foundation.org>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>
Subject: Re: [PATCH 1/8] mm: cma: introduce cma_release_nowait()
Date: Fri, 26 Mar 2021 14:32:01 -0700 [thread overview]
Message-ID: <0f7f3c4e-530a-5cd2-2719-259e509366e4@oracle.com> (raw)
In-Reply-To: <d1e712f3-c2f2-dcad-85c0-dc152bb8eecb@oracle.com>
On 3/25/21 4:49 PM, Mike Kravetz wrote:
> On 3/25/21 4:19 PM, Roman Gushchin wrote:
>> On Thu, Mar 25, 2021 at 01:12:51PM -0700, Minchan Kim wrote:
>>> On Thu, Mar 25, 2021 at 06:15:11PM +0100, David Hildenbrand wrote:
>>>> On 25.03.21 17:56, Mike Kravetz wrote:
>>>>> On 3/25/21 3:22 AM, Michal Hocko wrote:
>>>>>> On Thu 25-03-21 10:56:38, David Hildenbrand wrote:
>>>>>>> On 25.03.21 01:28, Mike Kravetz wrote:
>>>>>>>> From: Roman Gushchin <guro@fb.com>
>>>>>>>>
>>>>>>>> cma_release() has to lock the cma_lock mutex to clear the cma bitmap.
>>>>>>>> It makes it a blocking function, which complicates its usage from
>>>>>>>> non-blocking contexts. For instance, hugetlbfs code is temporarily
>>>>>>>> dropping the hugetlb_lock spinlock to call cma_release().
>>>>>>>>
>>>>>>>> This patch introduces a non-blocking cma_release_nowait(), which
>>>>>>>> postpones the cma bitmap clearance. It's done later from a work
>>>>>>>> context. The first page in the cma allocation is used to store
>>>>>>>> the work struct. Because CMA allocations and de-allocations are
>>>>>>>> usually not that frequent, a single global workqueue is used.
>>>>>>>>
>>>>>>>> To make sure that subsequent cma_alloc() call will pass, cma_alloc()
>>>>>>>> flushes the cma_release_wq workqueue. To avoid a performance
>>>>>>>> regression in the case when only cma_release() is used, gate it
>>>>>>>> by a per-cma area flag, which is set by the first call
>>>>>>>> of cma_release_nowait().
>>>>>>>>
>>>>>>>> Signed-off-by: Roman Gushchin <guro@fb.com>
>>>>>>>> [mike.kravetz@oracle.com: rebased to v5.12-rc3-mmotm-2021-03-17-22-24]
>>>>>>>> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
>>>>>>>> ---
>>>>>>>
>>>>>>>
>>>>>>> 1. Is there a real reason this is a mutex and not a spin lock? It seems to
>>>>>>> only protect the bitmap. Are bitmaps that huge that we spend a significant
>>>>>>> amount of time in there?
>>>>>>
>>>>>> Good question. Looking at the code it doesn't seem that there is any
>>>>>> blockable operation or any heavy lifting done under the lock.
>>>>>> 7ee793a62fa8 ("cma: Remove potential deadlock situation") has introduced
>>>>>> the lock and there was a simple bitmat protection back then. I suspect
>>>>>> the patch just followed the cma_mutex lead and used the same type of the
>>>>>> lock. cma_mutex used to protect alloc_contig_range which is sleepable.
>>>>>>
>>>>>> This all suggests that there is no real reason to use a sleepable lock
>>>>>> at all and we do not need all this heavy lifting.
>>>>>>
>>>>>
>>>>> When Roman first proposed these patches, I brought up the same issue:
>>>>>
>>>>> https://lore.kernel.org/linux-mm/20201022023352.GC300658@carbon.dhcp.thefacebook.com/
>>>>>
>>>>> Previously, Roman proposed replacing the mutex with a spinlock but
>>>>> Joonsoo was opposed.
>>>>>
>>>>> Adding Joonsoo on Cc:
>>>>>
>>>>
>>>> There has to be a good reason not to. And if there is a good reason,
>>>> lockless clearing might be one feasible alternative.
>>>
>>> I also don't think nowait variant is good idea. If the scanning of
>>> bitmap is *really* significant, it might be signal that we need to
>>> introduce different technique or data structure not bitmap rather
>>> than a new API variant.
>>
>> I'd also prefer to just replace the mutex with a spinlock rather than fiddling
>> with a delayed release.
>>
>
> I hope Joonsoo or someone else brings up specific concerns. I do not
> know enough about all CMA use cases. Certainly, in this specific use
> case converting to a spinlock would not be an issue. Do note that we
> would want to convert to an irq safe spinlock and disable irqs if that
> makes any difference in the discussion.
>
Suggestions on how to move forward would be appreciated. I can think of
the following options.
- Use the the cma_release_nowait() routine as defined in this patch.
- Just change the mutex to an irq safe spinlock. AFAICT, the potential
downsides could be:
- Interrupts disabled during long bitmap scans
- Wasted cpu cycles (spinning) if there is much contention on lock
Both of these would be more of an issue on small/embedded systems. I
took a quick look at the callers of cma_alloc/cma_release and nothing
stood out that could lead to high degrees of contention. However, I
could have missed something.
- Another idea I had was to allow the user to specify the locking type
when creating a cma area. In this way, cma areas which may have
release calls from atomic context would be set up with an irq safe
spinlock. Others, would use the mutex. I admit this is a hackish
way to address the issue, but perhaps not much worse than the separate
cma_release_nowait interface?
- Change the CMA bitmap to some other data structure and algorithm.
This would obviously take more work.
Thanks,
--
Mike Kravetz
next prev parent reply other threads:[~2021-03-26 21:33 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-25 0:28 [PATCH 0/8] make hugetlb put_page safe for all calling contexts Mike Kravetz
2021-03-25 0:28 ` [PATCH 1/8] mm: cma: introduce cma_release_nowait() Mike Kravetz
2021-03-25 9:39 ` Oscar Salvador
2021-03-25 9:45 ` Michal Hocko
2021-03-25 9:54 ` Oscar Salvador
2021-03-25 10:10 ` Michal Hocko
2021-03-25 10:11 ` Michal Hocko
2021-03-25 10:13 ` David Hildenbrand
2021-03-25 10:17 ` Oscar Salvador
2021-03-25 10:24 ` Michal Hocko
2021-03-25 9:56 ` David Hildenbrand
2021-03-25 10:22 ` Michal Hocko
2021-03-25 16:56 ` Mike Kravetz
2021-03-25 17:15 ` David Hildenbrand
2021-03-25 20:12 ` Minchan Kim
2021-03-25 23:19 ` Roman Gushchin
2021-03-25 23:49 ` Mike Kravetz
2021-03-26 21:32 ` Mike Kravetz [this message]
2021-03-29 7:46 ` Michal Hocko
2021-03-29 22:27 ` Mike Kravetz
2021-03-25 0:28 ` [PATCH 2/8] mm: hugetlb: don't drop hugetlb_lock around cma_release() call Mike Kravetz
2021-03-25 0:28 ` [PATCH 3/8] hugetlb: add per-hstate mutex to synchronize user adjustments Mike Kravetz
2021-03-25 10:47 ` Michal Hocko
2021-03-25 12:29 ` Oscar Salvador
2021-03-26 1:52 ` Miaohe Lin
2021-03-25 0:28 ` [PATCH 4/8] hugetlb: create remove_hugetlb_page() to separate functionality Mike Kravetz
2021-03-25 10:49 ` Michal Hocko
2021-03-26 2:10 ` Miaohe Lin
2021-03-26 19:57 ` Mike Kravetz
2021-03-27 1:40 ` Miaohe Lin
2021-03-27 6:36 ` [External] " Muchun Song
2021-03-25 0:28 ` [PATCH 5/8] hugetlb: call update_and_free_page without hugetlb_lock Mike Kravetz
2021-03-25 10:55 ` Michal Hocko
2021-03-25 17:12 ` Mike Kravetz
2021-03-25 19:39 ` Michal Hocko
2021-03-25 20:33 ` Mike Kravetz
2021-03-27 6:54 ` [External] " Muchun Song
2021-03-28 21:40 ` Mike Kravetz
2021-03-25 0:28 ` [PATCH 6/8] hugetlb: change free_pool_huge_page to remove_pool_huge_page Mike Kravetz
2021-03-25 11:06 ` Michal Hocko
2021-03-25 17:29 ` Mike Kravetz
2021-03-25 0:28 ` [PATCH 7/8] hugetlb: make free_huge_page irq safe Mike Kravetz
2021-03-25 11:21 ` Michal Hocko
2021-03-25 17:32 ` Mike Kravetz
2021-03-27 7:06 ` [External] " Muchun Song
2021-03-29 7:49 ` Michal Hocko
2021-03-29 22:44 ` Mike Kravetz
2021-03-25 0:28 ` [PATCH 8/8] hugetlb: add lockdep_assert_held() calls for hugetlb_lock Mike Kravetz
2021-03-25 11:22 ` Michal Hocko
2021-03-26 2:12 ` Miaohe Lin
2021-03-27 8:14 ` [External] " Muchun Song
2021-03-26 1:42 ` [PATCH 0/8] make hugetlb put_page safe for all calling contexts Miaohe Lin
2021-03-26 20:00 ` Mike Kravetz
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=0f7f3c4e-530a-5cd2-2719-259e509366e4@oracle.com \
--to=mike.kravetz@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=almasrymina@google.com \
--cc=aneesh.kumar@linux.ibm.com \
--cc=david@redhat.com \
--cc=guro@fb.com \
--cc=hdanton@sina.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=linmiaohe@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=longman@redhat.com \
--cc=mhocko@suse.com \
--cc=minchan@kernel.org \
--cc=naoya.horiguchi@nec.com \
--cc=osalvador@suse.de \
--cc=peterx@redhat.com \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--cc=shakeelb@google.com \
--cc=songmuchun@bytedance.com \
--cc=willy@infradead.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).