From: Anshuman Khandual <anshuman.khandual@arm.com>
To: Mike Kravetz <mike.kravetz@oracle.com>,
Muchun Song <songmuchun@bytedance.com>,
Peter Xu <peterx@redhat.com>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Michal Hocko <mhocko@suse.com>,
Naoya Horiguchi <naoya.horiguchi@linux.dev>,
James Houghton <jthoughton@google.com>,
Mina Almasry <almasrymina@google.com>,
"Aneesh Kumar K . V" <aneesh.kumar@linux.vnet.ibm.com>,
Paul Walmsley <paul.walmsley@sifive.com>,
Christian Borntraeger <borntraeger@linux.ibm.com>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [RFC PATCH 1/3] hugetlb: skip to end of PT page mapping when pte not present
Date: Wed, 1 Jun 2022 12:28:44 +0530 [thread overview]
Message-ID: <294314f1-d306-0eb9-059c-f10cb79b9e76@arm.com> (raw)
In-Reply-To: <95bd9375-36d2-7e34-83a8-d9eedfe70956@oracle.com>
On 5/31/22 22:35, Mike Kravetz wrote:
> On 5/30/22 19:04, Muchun Song wrote:
>> On Mon, May 30, 2022 at 03:56:43PM -0400, Peter Xu wrote:
>>> Hi, Mike,
>>>
>>> On Fri, May 27, 2022 at 03:58:47PM -0700, Mike Kravetz wrote:
>>>> +unsigned long hugetlb_mask_last_hp(struct hstate *h)
>>>> +{
>>>> + unsigned long hp_size = huge_page_size(h);
>>>> +
>>>> + if (hp_size == P4D_SIZE)
>>>> + return PGDIR_SIZE - P4D_SIZE;
>>>> + else if (hp_size == PUD_SIZE)
>>>> + return P4D_SIZE - PUD_SIZE;
>>>> + else if (hp_size == PMD_SIZE)
>>>> + return PUD_SIZE - PMD_SIZE;
>>>> +
>>>> + return ~(0);
>>>> +}
>>>
>>> How about:
>>>
>>> unsigned long hugetlb_mask_last_hp(struct hstate *h)
>>> {
>>> unsigned long hp_size = huge_page_size(h);
>>>
>>> return hp_size * (PTRS_PER_PTE - 1);
>>> }
>>>
>>
>> +1
>>
>
>
> I like this as well, but I wonder if we should put something like the
> following in just to be safe.
>
> BUILD_BUG_ON(PTRS_PER_PTE != PTRS_PER_PMD);
> BUILD_BUG_ON(PTRS_PER_PTE != PTRS_PER_PUD);
> BUILD_BUG_ON(PTRS_PER_PTE != PTRS_PER_P4D);
Exactly, assuming that PTRS_PER_PTE entries are present per page table page for
all supported HugeTLB levels might be bit risky, particularly for the top level.
Hence rather than all these additional BUILD_BUG_ON() checks for using standard
page table page entries i.e PTRS_PER_PTE, it might be better to just stick with
the original check either via 'if else' as proposed or better with a switch case.
next prev parent reply other threads:[~2022-06-01 6:59 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-27 22:58 [RFC PATCH 0/3] hugetlb: speed up linear address scanning Mike Kravetz
2022-05-27 22:58 ` [RFC PATCH 1/3] hugetlb: skip to end of PT page mapping when pte not present Mike Kravetz
2022-05-30 10:10 ` Baolin Wang
2022-05-31 16:56 ` Mike Kravetz
2022-06-15 21:22 ` Mike Kravetz
2022-06-16 3:48 ` Baolin Wang
2022-05-30 19:56 ` Peter Xu
2022-05-31 2:04 ` Muchun Song
2022-05-31 17:05 ` Mike Kravetz
2022-06-01 6:58 ` Anshuman Khandual [this message]
2022-05-31 17:00 ` Mike Kravetz
2022-06-15 17:27 ` Mike Kravetz
2022-06-15 17:51 ` Peter Xu
2022-05-27 22:58 ` [RFC PATCH 2/3] hugetlb: do not update address in huge_pmd_unshare Mike Kravetz
2022-05-30 10:14 ` Baolin Wang
2022-05-30 15:36 ` Muchun Song
2022-05-31 17:06 ` Mike Kravetz
2022-05-27 22:58 ` [RFC PATCH 3/3] hugetlb: Lazy page table copies in fork() Mike Kravetz
2022-05-31 17:25 ` David Hildenbrand
2022-06-01 5:20 ` 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=294314f1-d306-0eb9-059c-f10cb79b9e76@arm.com \
--to=anshuman.khandual@arm.com \
--cc=akpm@linux-foundation.org \
--cc=almasrymina@google.com \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=borntraeger@linux.ibm.com \
--cc=jthoughton@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=mike.kravetz@oracle.com \
--cc=naoya.horiguchi@linux.dev \
--cc=paul.walmsley@sifive.com \
--cc=peterx@redhat.com \
--cc=songmuchun@bytedance.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).