From: David Hildenbrand <david@redhat.com>
To: Muchun Song <songmuchun@bytedance.com>,
corbet@lwn.net, mike.kravetz@oracle.com, tglx@linutronix.de,
mingo@redhat.com, bp@alien8.de, x86@kernel.org, hpa@zytor.com,
dave.hansen@linux.intel.com, luto@kernel.org,
peterz@infradead.org, viro@zeniv.linux.org.uk,
akpm@linux-foundation.org, paulmck@kernel.org,
mchehab+huawei@kernel.org, pawan.kumar.gupta@linux.intel.com,
rdunlap@infradead.org, oneukum@suse.com,
anshuman.khandual@arm.com, jroedel@suse.de,
almasrymina@google.com, rientjes@google.com, willy@infradead.org,
osalvador@suse.de, mhocko@suse.com, song.bao.hua@hisilicon.com,
naoya.horiguchi@nec.com
Cc: duanxiongchun@bytedance.com, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH v10 00/11] Free some vmemmap pages of HugeTLB page
Date: Thu, 17 Dec 2020 13:17:56 +0100 [thread overview]
Message-ID: <6fbd36f0-4864-89ff-15ef-9750059defab@redhat.com> (raw)
In-Reply-To: <20201217121303.13386-1-songmuchun@bytedance.com>
On 17.12.20 13:12, Muchun Song wrote:
> Hi all,
>
> This patch series will free some vmemmap pages(struct page structures)
> associated with each hugetlbpage when preallocated to save memory.
>
> In order to reduce the difficulty of the first version of code review.
> From this version, we disable PMD/huge page mapping of vmemmap if this
> feature was enabled. This accutualy eliminate a bunch of the complex code
> doing page table manipulation. When this patch series is solid, we cam add
> the code of vmemmap page table manipulation in the future.
>
> The struct page structures (page structs) are used to describe a physical
> page frame. By default, there is a one-to-one mapping from a page frame to
> it's corresponding page struct.
>
> The HugeTLB pages consist of multiple base page size pages and is supported
> by many architectures. See hugetlbpage.rst in the Documentation directory
> for more details. On the x86 architecture, HugeTLB pages of size 2MB and 1GB
> are currently supported. Since the base page size on x86 is 4KB, a 2MB
> HugeTLB page consists of 512 base pages and a 1GB HugeTLB page consists of
> 4096 base pages. For each base page, there is a corresponding page struct.
>
> Within the HugeTLB subsystem, only the first 4 page structs are used to
> contain unique information about a HugeTLB page. HUGETLB_CGROUP_MIN_ORDER
> provides this upper limit. The only 'useful' information in the remaining
> page structs is the compound_head field, and this field is the same for all
> tail pages.
>
> By removing redundant page structs for HugeTLB pages, memory can returned to
> the buddy allocator for other uses.
>
> When the system boot up, every 2M HugeTLB has 512 struct page structs which
> size is 8 pages(sizeof(struct page) * 512 / PAGE_SIZE).
>
> HugeTLB struct pages(8 pages) page frame(8 pages)
> +-----------+ ---virt_to_page---> +-----------+ mapping to +-----------+
> | | | 0 | -------------> | 0 |
> | | +-----------+ +-----------+
> | | | 1 | -------------> | 1 |
> | | +-----------+ +-----------+
> | | | 2 | -------------> | 2 |
> | | +-----------+ +-----------+
> | | | 3 | -------------> | 3 |
> | | +-----------+ +-----------+
> | | | 4 | -------------> | 4 |
> | 2MB | +-----------+ +-----------+
> | | | 5 | -------------> | 5 |
> | | +-----------+ +-----------+
> | | | 6 | -------------> | 6 |
> | | +-----------+ +-----------+
> | | | 7 | -------------> | 7 |
> | | +-----------+ +-----------+
> | |
> | |
> | |
> +-----------+
>
> The value of page->compound_head is the same for all tail pages. The first
> page of page structs (page 0) associated with the HugeTLB page contains the 4
> page structs necessary to describe the HugeTLB. The only use of the remaining
> pages of page structs (page 1 to page 7) is to point to page->compound_head.
> Therefore, we can remap pages 2 to 7 to page 1. Only 2 pages of page structs
> will be used for each HugeTLB page. This will allow us to free the remaining
> 6 pages to the buddy allocator.
>
> Here is how things look after remapping.
>
> HugeTLB struct pages(8 pages) page frame(8 pages)
> +-----------+ ---virt_to_page---> +-----------+ mapping to +-----------+
> | | | 0 | -------------> | 0 |
> | | +-----------+ +-----------+
> | | | 1 | -------------> | 1 |
> | | +-----------+ +-----------+
> | | | 2 | ----------------^ ^ ^ ^ ^ ^
> | | +-----------+ | | | | |
> | | | 3 | ------------------+ | | | |
> | | +-----------+ | | | |
> | | | 4 | --------------------+ | | |
> | 2MB | +-----------+ | | |
> | | | 5 | ----------------------+ | |
> | | +-----------+ | |
> | | | 6 | ------------------------+ |
> | | +-----------+ |
> | | | 7 | --------------------------+
> | | +-----------+
> | |
> | |
> | |
> +-----------+
>
> When a HugeTLB is freed to the buddy system, we should allocate 6 pages for
> vmemmap pages and restore the previous mapping relationship.
>
> Apart from 2MB HugeTLB page, we also have 1GB HugeTLB page. It is similar
> to the 2MB HugeTLB page. We also can use this approach to free the vmemmap
> pages.
>
> In this case, for the 1GB HugeTLB page, we can save 4088 pages(There are
> 4096 pages for struct page structs, we reserve 2 pages for vmemmap and 8
> pages for page tables. So we can save 4088 pages). This is a very substantial
> gain. On our server, run some SPDK/QEMU applications which will use 1024GB
> hugetlbpage. With this feature enabled, we can save ~16GB(1G hugepage)/~11GB
> (2MB hugepage, the worst case is 10GB while the best is 12GB) memory.
>
> Because there are vmemmap page tables reconstruction on the freeing/allocating
> path, it increases some overhead. Here are some overhead analysis.
>
> 1) Allocating 10240 2MB hugetlb pages.
>
> a) With this patch series applied:
> # time echo 10240 > /proc/sys/vm/nr_hugepages
>
> real 0m0.166s
> user 0m0.000s
> sys 0m0.166s
>
> # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
> Attaching 2 probes...
>
> @latency:
> [8K, 16K) 8360 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
> [16K, 32K) 1868 |@@@@@@@@@@@ |
> [32K, 64K) 10 | |
> [64K, 128K) 2 | |
>
> b) Without this patch series:
> # time echo 10240 > /proc/sys/vm/nr_hugepages
>
> real 0m0.066s
> user 0m0.000s
> sys 0m0.066s
>
> # bpftrace -e 'kprobe:alloc_fresh_huge_page { @start[tid] = nsecs; } kretprobe:alloc_fresh_huge_page /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
> Attaching 2 probes...
>
> @latency:
> [4K, 8K) 10176 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
> [8K, 16K) 62 | |
> [16K, 32K) 2 | |
>
> Summarize: this feature is about ~2x slower than before.
>
> 2) Freeing 10240 2MB hugetlb pages.
>
> a) With this patch series applied:
> # time echo 0 > /proc/sys/vm/nr_hugepages
>
> real 0m0.004s
> user 0m0.000s
> sys 0m0.002s
>
> # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
> Attaching 2 probes...
>
> @latency:
> [16K, 32K) 10240 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
>
> b) Without this patch series:
> # time echo 0 > /proc/sys/vm/nr_hugepages
>
> real 0m0.077s
> user 0m0.001s
> sys 0m0.075s
>
> # bpftrace -e 'kprobe:__free_hugepage { @start[tid] = nsecs; } kretprobe:__free_hugepage /@start[tid]/ { @latency = hist(nsecs - @start[tid]); delete(@start[tid]); }'
> Attaching 2 probes...
>
> @latency:
> [4K, 8K) 9950 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@|
> [8K, 16K) 287 |@ |
> [16K, 32K) 3 | |
>
> Summarize: The overhead of __free_hugepage is about ~2-4x slower than before.
> But according to the allocation test above, I think that here is
> also ~2x slower than before.
>
> But why the 'real' time of patched is smaller than before? Because
> In this patch series, the freeing hugetlb is asynchronous(through
> kwoker).
>
> Although the overhead has increased, the overhead is not significant. Like Mike
> said, "However, remember that the majority of use cases create hugetlb pages at
> or shortly after boot time and add them to the pool. So, additional overhead is
> at pool creation time. There is no change to 'normal run time' operations of
> getting a page from or returning a page to the pool (think page fault/unmap)".
>
Just FYI, I'll be offline until first week of January. I'm planning on
reviewing when I'm back.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2020-12-17 12:19 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-17 12:12 [PATCH v10 00/11] Free some vmemmap pages of HugeTLB page Muchun Song
2020-12-17 12:12 ` [PATCH v10 01/11] mm/memory_hotplug: Factor out bootmem core functions to bootmem_info.c Muchun Song
2020-12-17 12:12 ` [PATCH v10 02/11] mm/hugetlb: Introduce a new config HUGETLB_PAGE_FREE_VMEMMAP Muchun Song
2020-12-18 15:41 ` Oscar Salvador
2020-12-21 23:56 ` Mike Kravetz
2020-12-17 12:12 ` [PATCH v10 03/11] mm/hugetlb: Free the vmemmap pages associated with each HugeTLB page Muchun Song
2020-12-21 9:11 ` Oscar Salvador
2020-12-21 11:25 ` [External] " Muchun Song
2020-12-21 11:25 ` Muchun Song
2020-12-21 13:43 ` Oscar Salvador
2020-12-21 15:52 ` Muchun Song
2020-12-21 15:52 ` Muchun Song
2020-12-21 18:00 ` Oscar Salvador
2020-12-22 1:03 ` Mike Kravetz
2020-12-22 2:49 ` Muchun Song
2020-12-22 2:49 ` Muchun Song
2020-12-17 12:12 ` [PATCH v10 04/11] mm/hugetlb: Defer freeing of HugeTLB pages Muchun Song
2020-12-21 10:27 ` Oscar Salvador
2020-12-21 11:07 ` [External] " Muchun Song
2020-12-21 11:07 ` Muchun Song
2020-12-21 14:14 ` Oscar Salvador
2020-12-21 15:18 ` Muchun Song
2020-12-21 15:18 ` Muchun Song
2020-12-17 12:12 ` [PATCH v10 05/11] mm/hugetlb: Allocate the vmemmap pages associated with each HugeTLB page Muchun Song
2020-12-17 12:12 ` [PATCH v10 06/11] mm/hugetlb: Set the PageHWPoison to the raw error page Muchun Song
2020-12-17 12:12 ` [PATCH v10 07/11] mm/hugetlb: Flush work when dissolving hugetlb page Muchun Song
2020-12-21 10:40 ` Oscar Salvador
2020-12-21 11:07 ` [External] " Muchun Song
2020-12-21 11:07 ` Muchun Song
2020-12-17 12:13 ` [PATCH v10 08/11] mm/hugetlb: Add a kernel parameter hugetlb_free_vmemmap Muchun Song
2020-12-17 12:13 ` [PATCH v10 09/11] mm/hugetlb: Introduce nr_free_vmemmap_pages in the struct hstate Muchun Song
2020-12-21 8:16 ` Oscar Salvador
2020-12-21 9:33 ` [External] " Muchun Song
2020-12-21 9:33 ` Muchun Song
2020-12-17 12:13 ` [PATCH v10 10/11] mm/hugetlb: Gather discrete indexes of tail page Muchun Song
2020-12-18 9:06 ` Oscar Salvador
2020-12-18 9:41 ` [External] " Muchun Song
2020-12-18 9:41 ` Muchun Song
2020-12-17 12:13 ` [PATCH v10 11/11] mm/hugetlb: Optimize the code with the help of the compiler Muchun Song
2020-12-17 12:17 ` David Hildenbrand [this message]
2020-12-17 14:59 ` [PATCH v10 00/11] Free some vmemmap pages of HugeTLB page Oscar Salvador
2020-12-17 15:52 ` [External] " Muchun Song
2020-12-17 15:52 ` Muchun Song
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=6fbd36f0-4864-89ff-15ef-9750059defab@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=almasrymina@google.com \
--cc=anshuman.khandual@arm.com \
--cc=bp@alien8.de \
--cc=corbet@lwn.net \
--cc=dave.hansen@linux.intel.com \
--cc=duanxiongchun@bytedance.com \
--cc=hpa@zytor.com \
--cc=jroedel@suse.de \
--cc=linux-doc@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=mchehab+huawei@kernel.org \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=mingo@redhat.com \
--cc=naoya.horiguchi@nec.com \
--cc=oneukum@suse.com \
--cc=osalvador@suse.de \
--cc=paulmck@kernel.org \
--cc=pawan.kumar.gupta@linux.intel.com \
--cc=peterz@infradead.org \
--cc=rdunlap@infradead.org \
--cc=rientjes@google.com \
--cc=song.bao.hua@hisilicon.com \
--cc=songmuchun@bytedance.com \
--cc=tglx@linutronix.de \
--cc=viro@zeniv.linux.org.uk \
--cc=willy@infradead.org \
--cc=x86@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.