Message-ID: <b1cada27-33b0-f53a-4059-07c54d9f1bc4@redhat.com>
Date:   Tue, 14 Feb 2023 18:39:21 +0100
From:   David Hildenbrand <david@...hat.com>
To:     Yang Shi <shy828301@...il.com>
Cc:     Chih-En Lin <shiyn.lin@...il.com>,
        Pasha Tatashin <pasha.tatashin@...een.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Qi Zheng <zhengqi.arch@...edance.com>,
        "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        Christophe Leroy <christophe.leroy@...roup.eu>,
        John Hubbard <jhubbard@...dia.com>,
        Nadav Amit <namit@...are.com>, Barry Song <baohua@...nel.org>,
        Steven Rostedt <rostedt@...dmis.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...hat.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Mark Rutland <mark.rutland@....com>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Jiri Olsa <jolsa@...nel.org>,
        Namhyung Kim <namhyung@...nel.org>,
        Peter Xu <peterx@...hat.com>, Vlastimil Babka <vbabka@...e.cz>,
        Zach O'Keefe <zokeefe@...gle.com>,
        Yun Zhou <yun.zhou@...driver.com>,
        Hugh Dickins <hughd@...gle.com>,
        Suren Baghdasaryan <surenb@...gle.com>,
        Yu Zhao <yuzhao@...gle.com>, Juergen Gross <jgross@...e.com>,
        Tong Tiangen <tongtiangen@...wei.com>,
        Liu Shixin <liushixin2@...wei.com>,
        Anshuman Khandual <anshuman.khandual@....com>,
        Li kunyu <kunyu@...china.com>,
        Minchan Kim <minchan@...nel.org>,
        Miaohe Lin <linmiaohe@...wei.com>,
        Gautam Menghani <gautammenghani201@...il.com>,
        Catalin Marinas <catalin.marinas@....com>,
        Mark Brown <broonie@...nel.org>, Will Deacon <will@...nel.org>,
        Vincenzo Frascino <Vincenzo.Frascino@....com>,
        Thomas Gleixner <tglx@...utronix.de>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Andy Lutomirski <luto@...nel.org>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        "Liam R. Howlett" <Liam.Howlett@...cle.com>,
        Fenghua Yu <fenghua.yu@...el.com>,
        Andrei Vagin <avagin@...il.com>,
        Barret Rhoden <brho@...gle.com>,
        Michal Hocko <mhocko@...e.com>,
        "Jason A. Donenfeld" <Jason@...c4.com>,
        Alexey Gladkov <legion@...nel.org>,
        linux-kernel@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        linux-mm@...ck.org, linux-trace-kernel@...r.kernel.org,
        linux-perf-users@...r.kernel.org,
        Dinglan Peng <peng301@...due.edu>,
        Pedro Fonseca <pfonseca@...due.edu>,
        Jim Huang <jserv@...s.ncku.edu.tw>,
        Huichun Feng <foxhoundsk.tw@...il.com>
Subject: Re: [PATCH v4 00/14] Introduce Copy-On-Write to Page Table

On 14.02.23 18:23, Yang Shi wrote:
> On Tue, Feb 14, 2023 at 1:58 AM David Hildenbrand <david@...hat.com> wrote:
>>
>> On 10.02.23 18:20, Chih-En Lin wrote:
>>> On Fri, Feb 10, 2023 at 11:21:16AM -0500, Pasha Tatashin wrote:
>>>>>>> Currently, copy-on-write is only used for the mapped memory; the child
>>>>>>> process still needs to copy the entire page table from the parent
>>>>>>> process during forking. The parent process might take a lot of time and
>>>>>>> memory to copy the page table when the parent has a big page table
>>>>>>> allocated. For example, the memory usage of a process after forking with
>>>>>>> 1 GB mapped memory is as follows:
>>>>>>
>>>>>> For some reason, I was not able to reproduce performance improvements
>>>>>> with a simple fork() performance measurement program. The results that
>>>>>> I saw are the following:
>>>>>>
>>>>>> Base:
>>>>>> Fork latency per gigabyte: 0.004416 seconds
>>>>>> Fork latency per gigabyte: 0.004382 seconds
>>>>>> Fork latency per gigabyte: 0.004442 seconds
>>>>>> COW kernel:
>>>>>> Fork latency per gigabyte: 0.004524 seconds
>>>>>> Fork latency per gigabyte: 0.004764 seconds
>>>>>> Fork latency per gigabyte: 0.004547 seconds
>>>>>>
>>>>>> AMD EPYC 7B12 64-Core Processor
>>>>>> Base:
>>>>>> Fork latency per gigabyte: 0.003923 seconds
>>>>>> Fork latency per gigabyte: 0.003909 seconds
>>>>>> Fork latency per gigabyte: 0.003955 seconds
>>>>>> COW kernel:
>>>>>> Fork latency per gigabyte: 0.004221 seconds
>>>>>> Fork latency per gigabyte: 0.003882 seconds
>>>>>> Fork latency per gigabyte: 0.003854 seconds
>>>>>>
>>>>>> Given, that page table for child is not copied, I was expecting the
>>>>>> performance to be better with COW kernel, and also not to depend on
>>>>>> the size of the parent.
>>>>>
>>>>> Yes, the child won't duplicate the page table, but fork will still
>>>>> traverse all the page table entries to do the accounting.
>>>>> And, since this patch expends the COW to the PTE table level, it's not
>>>>> the mapped page (page table entry) grained anymore, so we have to
>>>>> guarantee that all the mapped page is available to do COW mapping in
>>>>> the such page table.
>>>>> This kind of checking also costs some time.
>>>>> As a result, since the accounting and the checking, the COW PTE fork
>>>>> still depends on the size of the parent so the improvement might not
>>>>> be significant.
>>>>
>>>> The current version of the series does not provide any performance
>>>> improvements for fork(). I would recommend removing claims from the
>>>> cover letter about better fork() performance, as this may be
>>>> misleading for those looking for a way to speed up forking. In my
>>>
>>>   From v3 to v4, I changed the implementation of the COW fork() part to do
>>> the accounting and checking. At the time, I also removed most of the
>>> descriptions about the better fork() performance. Maybe it's not enough
>>> and still has some misleading. I will fix this in the next version.
>>> Thanks.
>>>
>>>> case, I was looking to speed up Redis OSS, which relies on fork() to
>>>> create consistent snapshots for driving replicates/backups. The O(N)
>>>> per-page operation causes fork() to be slow, so I was hoping that this
>>>> series, which does not duplicate the VA during fork(), would make the
>>>> operation much quicker.
>>>
>>> Indeed, at first, I tried to avoid the O(N) per-page operation by
>>> deferring the accounting and the swap stuff to the page fault. But,
>>> as I mentioned, it's not suitable for the mainline.
>>>
>>> Honestly, for improving the fork(), I have an idea to skip the per-page
>>> operation without breaking the logic. However, this will introduce the
>>> complicated mechanism and may has the overhead for other features. It
>>> might not be worth it. It's hard to strike a balance between the
>>> over-complicated mechanism with (probably) better performance and data
>>> consistency with the page status. So, I would focus on the safety and
>>> stable approach at first.
>>
>> Yes, it is most probably possible, but complexity, robustness and
>> maintainability have to be considered as well.
>>
>> Thanks for implementing this approach (only deduplication without other
>> optimizations) and evaluating it accordingly. It's certainly "cleaner",
>> such that we only have to mess with unsharing and not with other
>> accounting/pinning/mapcount thingies. But it also highlights how
>> intrusive even this basic deduplication approach already is -- and that
>> most benefits of the original approach requires even more complexity on top.
>>
>> I am not quite sure if the benefit is worth the price (I am not to
>> decide and I would like to hear other options).
>>
>> My quick thoughts after skimming over the core parts of this series
>>
>> (1) forgetting to break COW on a PTE in some pgtable walker feels quite
>>       likely (meaning that it might be fairly error-prone) and forgetting
>>       to break COW on a PTE table, accidentally modifying the shared
>>       table.
>> (2) break_cow_pte() can fail, which means that we can fail some
>>       operations (possibly silently halfway through) now. For example,
>>       looking at your change_pte_range() change, I suspect it's wrong.
>> (3) handle_cow_pte_fault() looks quite complicated and needs quite some
>>       double-checking: we temporarily clear the PMD, to reset it
>>       afterwards. I am not sure if that is correct. For example, what
>>       stops another page fault stumbling over that pmd_none() and
>>       allocating an empty page table? Maybe there are some locking details
>>       missing or they are very subtle such that we better document them. I
>>      recall that THP played quite some tricks to make such cases work ...
>>
>>>
>>>>> Actually, at the RFC v1 and v2, we proposed the version of skipping
>>>>> those works, and we got a significant improvement. You can see the
>>>>> number from RFC v2 cover letter [1]:
>>>>> "In short, with 512 MB mapped memory, COW PTE decreases latency by 93%
>>>>> for normal fork"
>>>>
>>>> I suspect the 93% improvement (when the mapcount was not updated) was
>>>> only for VAs with 4K pages. With 2M mappings this series did not
>>>> provide any benefit is this correct?
>>>
>>> Yes. In this case, the COW PTE performance is similar to the normal
>>> fork().
>>
>>
>> The thing with THP is, that during fork(), we always allocate a backup
>> PTE table, to be able to PTE-map the THP whenever we have to. Otherwise
>> we'd have to eventually fail some operations we don't want to fail --
>> similar to the case where break_cow_pte() could fail now due to -ENOMEM
>> although we really don't want to fail (e.g., change_pte_range() ).
>>
>> I always considered that wasteful, because in many scenarios, we'll
>> never ever split a THP and possibly waste memory.
> 
> When you say "split THP", do you mean split the compound page to base
> pages? IIUC the backup PTE table page is used to guarantee the PMD
> split (just convert pmd mapped THP to PTE-mapped but not split the
> compound page) succeed. You may already notice there is no return
> value for PMD split.

Yes, as I raised in my other reply.

> 
> The PMD split may be called quite often, for example, MADV_DONTNEED,
> mbind, mlock, and even in memory reclamation context  (THP swap).

Yes, but with a single MADV_DONTNEED call you cannot PTE-map more than 2 
THP (all other overlapped THP will get zapped). Same with most other 
operations.

There are corner cases, though. I recall that s390x/kvm wants to break 
all THP in a given VMA range. But that operation could safely fail if we 
can't do that.

Certainly needs some investigation, that's most probably why it hasn't 
been done yet.

> 
>>
>> Optimizing that for THP (e.g., don't always allocate backup THP, have
>> some global allocation backup pool for splits + refill when
>> close-to-empty) might provide similar fork() improvements, both in speed
>> and memory consumption when it comes to anonymous memory.
> 
> It might work. But may be much more complicated than what you thought
> when handling multiple parallel PMD splits.


I consider the whole PTE-table linking to THPs complicated enough to 
eventually replace it by something differently complicated that wastes 
less memory ;)

-- 
Thanks,

David / dhildenb

Powered by blists - more mailing lists