From: Qi Zheng <zhengqi.arch@bytedance.com>
To: Jason Gunthorpe <jgg@nvidia.com>, David Hildenbrand <david@redhat.com>
Cc: akpm@linux-foundation.org, tglx@linutronix.de,
hannes@cmpxchg.org, mhocko@kernel.org, vdavydov.dev@gmail.com,
kirill.shutemov@linux.intel.com, mika.penttila@nextfour.com,
linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, songmuchun@bytedance.com
Subject: Re: [PATCH v2 6/9] mm: free user PTE page table pages
Date: Thu, 2 Sep 2021 14:53:58 +0800 [thread overview]
Message-ID: <47c7d6fd-056b-0e40-ae30-e2f099a97a95@bytedance.com> (raw)
In-Reply-To: <20210901153247.GJ1721383@nvidia.com>
On 2021/9/1 PM11:32, Jason Gunthorpe wrote:
>
>>> Also, I don't really understand how this scheme works with
>>> get_user_pages_fast.
>>
>> With the RCU change it in #8 it should work just fine, because RCU
>> synchronize has to wait either until all other CPUs have left the RCU read
>> section, or re-enabled interrupts.
>
> So at this point in the series fast gup is broken, that does mean the
> series presentation really needs to be reworked. The better
> presentation is to add the API changes, with a
> no-functional-difference implementation, push the new API in well
> split patches to all the consumption sites, then change the API to
> have the new semantics.
>
> RCU and refcount to free the page levels seems like a reasonable
> approach, but I have to say I haven't thought it through fully - are
> all the contexts that have the pte deref safe to do call_rcu?
See Documentation/RCU/rcubarrier.rst:
"Since call_rcu() never blocks, this code can safely be used from within
IRQ context."
So I think call_rcu() can be safely run in any context.
Thinks,
Qi
>
> Jason
>
>
next prev parent reply other threads:[~2021-09-02 6:54 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-19 3:18 [PATCH v2 0/9] Free user PTE page table pages Qi Zheng
2021-08-19 3:18 ` [PATCH v2 1/9] mm: introduce pmd_install() helper Qi Zheng
2021-08-24 16:26 ` David Hildenbrand
2021-08-25 16:20 ` Qi Zheng
2021-08-25 16:32 ` David Hildenbrand
2021-08-26 3:04 ` Qi Zheng
2021-08-19 3:18 ` [PATCH v2 2/9] mm: remove redundant smp_wmb() Qi Zheng
2021-08-19 3:18 ` [PATCH v2 3/9] mm: rework the parameter of lock_page_or_retry() Qi Zheng
2021-08-19 3:18 ` [PATCH v2 4/9] mm: move pte_alloc{,_map,_map_lock}() to a separate file Qi Zheng
2021-08-19 3:18 ` [PATCH v2 5/9] mm: pte_refcount infrastructure Qi Zheng
2021-08-19 3:18 ` [PATCH v2 6/9] mm: free user PTE page table pages Qi Zheng
2021-08-19 7:01 ` David Hildenbrand
2021-08-19 10:18 ` [External] " Qi Zheng
2021-09-01 13:53 ` Jason Gunthorpe
2021-09-01 13:57 ` David Hildenbrand
2021-09-01 15:32 ` Jason Gunthorpe
2021-09-01 16:13 ` David Hildenbrand
2021-09-01 16:16 ` Jason Gunthorpe
2021-09-01 16:19 ` David Hildenbrand
2021-09-01 17:10 ` Jason Gunthorpe
2021-09-01 17:49 ` David Hildenbrand
2021-09-01 17:55 ` Jason Gunthorpe
2021-09-01 17:58 ` David Hildenbrand
2021-09-01 18:09 ` Jason Gunthorpe
2021-09-01 18:10 ` David Hildenbrand
2021-09-02 7:04 ` Qi Zheng
2021-09-02 6:53 ` Qi Zheng [this message]
2021-08-19 3:18 ` [PATCH v2 7/9] mm: add THP support for pte_ref Qi Zheng
2021-08-19 3:18 ` [PATCH v2 8/9] mm: free PTE page table by using rcu mechanism Qi Zheng
2021-08-19 3:18 ` [PATCH v2 9/9] mm: use mmu_gather to free PTE page table Qi Zheng
2021-09-01 12:32 ` [PATCH v2 0/9] Free user PTE page table pages David Hildenbrand
2021-09-01 16:07 ` Jason Gunthorpe
2021-09-01 16:10 ` David Hildenbrand
2021-09-02 3:37 ` Qi Zheng
2021-09-15 14:52 ` Qi Zheng
2021-09-15 14:59 ` Jason Gunthorpe
2021-09-16 5:32 ` Qi Zheng
2021-09-16 8:30 ` David Hildenbrand
2021-09-16 8:41 ` Qi Zheng
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=47c7d6fd-056b-0e40-ae30-e2f099a97a95@bytedance.com \
--to=zhengqi.arch@bytedance.com \
--cc=akpm@linux-foundation.org \
--cc=david@redhat.com \
--cc=hannes@cmpxchg.org \
--cc=jgg@nvidia.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=mika.penttila@nextfour.com \
--cc=songmuchun@bytedance.com \
--cc=tglx@linutronix.de \
--cc=vdavydov.dev@gmail.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;
as well as URLs for NNTP newsgroup(s).