linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Dave Hansen <dave.hansen@intel.com>
To: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Vlastimil Babka <vbabka@suse.cz>,
	Christoph Lameter <cl@gentwo.org>,
	Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>,
	Jerome Marchand <jmarchan@redhat.com>,
	Yang Shi <yang.shi@linaro.org>,
	Sasha Levin <sasha.levin@oracle.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCHv2 17/28] thp: skip file huge pmd on copy_huge_pmd()
Date: Tue, 16 Feb 2016 07:46:37 -0800	[thread overview]
Message-ID: <56C3445D.3040305@intel.com> (raw)
In-Reply-To: <20160216101450.GE46557@black.fi.intel.com>

On 02/16/2016 02:14 AM, Kirill A. Shutemov wrote:
> On Fri, Feb 12, 2016 at 10:42:09AM -0800, Dave Hansen wrote:
>> On 02/11/2016 06:21 AM, Kirill A. Shutemov wrote:
>>> File pmds can be safely skip on copy_huge_pmd(), we can re-fault them
>>> later. COW for file mappings handled on pte level.
>>
>> Is this different from 4k pages?  I figured we might skip copying
>> file-backed ptes on fork, but I couldn't find the code.
> 
> Currently, we only filter out on per-VMA basis. See first comment in
> copy_page_range().
> 
> Here we handle PMD mapped file pages in COW mapping. File THP can be
> mapped into COW mapping as result of read page fault.

OK...  So, copy_page_range() has a check for "Don't copy ptes where a
page fault will fill them correctly."  Seems sane enough, but the check
is implemented using a check for the VMA having !vma->anon_vma, which is
a head-scratcher for a moment.  Why does that apply to huge tmpfs?

Ahh, MAP_PRIVATE.  MAP_PRIVATE vmas have ->anon_vma because they have
essentially-anonymous pages for when they do a COW, so they don't hit
that check and they go through the copy_*() functions, including
copy_huge_pmd().

We don't handle 2M COW operations yet so we simply decline to copy these
pages.  Might cost us page faults down the road, but it makes things
easier to implement for now.

Did I get that right?

Any chance we could get a bit of that into the patch descriptions so
that the next hapless reviewer can spend their time looking at your code
instead of relearning the fork() handling for MAP_PRIVATE?

  reply	other threads:[~2016-02-16 15:46 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-11 14:21 [PATCHv2 00/28] huge tmpfs implementation using compound pages Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 01/28] thp, dax: do not try to withdraw pgtable from non-anon VMA Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 02/28] rmap: introduce rmap_walk_locked() Kirill A. Shutemov
2016-02-11 18:52   ` Andi Kleen
2016-02-16  9:36     ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 03/28] rmap: extend try_to_unmap() to be usable by split_huge_page() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 04/28] mm: make remove_migration_ptes() beyond mm/migration.c Kirill A. Shutemov
2016-02-12 16:54   ` Dave Hansen
2016-02-16  9:54     ` Kirill A. Shutemov
2016-02-16 15:29       ` Dave Hansen
2016-02-11 14:21 ` [PATCHv2 05/28] thp: rewrite freeze_page()/unfreeze_page() with generic rmap walkers Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 06/28] mm: do not pass mm_struct into handle_mm_fault Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 07/28] mm: introduce fault_env Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 08/28] mm: postpone page table allocation until do_set_pte() Kirill A. Shutemov
2016-02-12 17:44   ` Dave Hansen
2016-02-16 14:26     ` Kirill A. Shutemov
2016-02-16 17:17       ` Dave Hansen
2016-02-23 13:05         ` Kirill A. Shutemov
2016-02-16 17:38       ` Dave Hansen
2016-02-23 22:58         ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 09/28] rmap: support file thp Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 10/28] mm: introduce do_set_pmd() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 11/28] mm, rmap: account file thp pages Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 12/28] thp, vmstats: add counters for huge file pages Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 13/28] thp: support file pages in zap_huge_pmd() Kirill A. Shutemov
2016-02-12 18:33   ` Dave Hansen
2016-02-16 10:00     ` Kirill A. Shutemov
2016-02-16 15:31       ` Dave Hansen
2016-02-18 12:19         ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 14/28] thp: handle file pages in split_huge_pmd() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 15/28] thp: handle file COW faults Kirill A. Shutemov
2016-02-12 18:36   ` Dave Hansen
2016-02-16 10:08     ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 16/28] thp: handle file pages in mremap() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 17/28] thp: skip file huge pmd on copy_huge_pmd() Kirill A. Shutemov
2016-02-12 18:42   ` Dave Hansen
2016-02-16 10:14     ` Kirill A. Shutemov
2016-02-16 15:46       ` Dave Hansen [this message]
2016-02-18 12:41         ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 18/28] thp: prepare change_huge_pmd() for file thp Kirill A. Shutemov
2016-02-12 18:48   ` Dave Hansen
2016-02-16 10:15     ` Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 19/28] thp: run vma_adjust_trans_huge() outside i_mmap_rwsem Kirill A. Shutemov
2016-02-12 18:50   ` Dave Hansen
2016-02-16 10:16     ` Kirill A. Shutemov
2016-02-16 15:49   ` Dave Hansen
2016-02-11 14:21 ` [PATCHv2 20/28] thp: file pages support for split_huge_page() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 21/28] vmscan: split file huge pages before paging them out Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 22/28] page-flags: relax policy for PG_mappedtodisk and PG_reclaim Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 23/28] radix-tree: implement radix_tree_maybe_preload_order() Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 24/28] filemap: prepare find and delete operations for huge pages Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 25/28] truncate: handle file thp Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 26/28] shmem: prepare huge=N mount option and /proc/sys/vm/shmem_huge Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 27/28] shmem: get_unmapped_area align huge page Kirill A. Shutemov
2016-02-11 14:21 ` [PATCHv2 28/28] shmem: add huge pages support Kirill A. Shutemov

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=56C3445D.3040305@intel.com \
    --to=dave.hansen@intel.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=cl@gentwo.org \
    --cc=hughd@google.com \
    --cc=jmarchan@redhat.com \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=n-horiguchi@ah.jp.nec.com \
    --cc=sasha.levin@oracle.com \
    --cc=vbabka@suse.cz \
    --cc=yang.shi@linaro.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).