From: "Kirill A. Shutemov" <kirill-oKw7cIdHH8eLwutG50LtGA@public.gmane.org>
To: linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Cc: "Kirill A. Shutemov"
<kirill.shutemov-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>,
Hugh Dickins <hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Andrea Arcangeli
<aarcange-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Andrew Morton
<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
Dave Hansen <dave.hansen-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
Vlastimil Babka <vbabka-AlSwsSmVLrQ@public.gmane.org>,
Christoph Lameter <cl-gkYfJU5Cukgdnm+yROfE0A@public.gmane.org>,
Naoya Horiguchi
<n-horiguchi-PaJj6Psr51x8UrSeD/g0lQ@public.gmane.org>,
Jerome Marchand
<jmarchan-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Yang Shi <yang.shi-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>,
Sasha Levin <sasha.levin-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org
Subject: THP-enabled filesystem vs. FALLOC_FL_PUNCH_HOLE
Date: Fri, 4 Mar 2016 14:26:03 +0300 [thread overview]
Message-ID: <20160304112603.GA9790@node.shutemov.name> (raw)
In-Reply-To: <1457023939-98083-1-git-send-email-kirill.shutemov-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
On Thu, Mar 03, 2016 at 07:51:50PM +0300, Kirill A. Shutemov wrote:
> Truncate and punch hole that only cover part of THP range is implemented
> by zero out this part of THP.
>
> This have visible effect on fallocate(FALLOC_FL_PUNCH_HOLE) behaviour.
> As we don't really create hole in this case, lseek(SEEK_HOLE) may have
> inconsistent results depending what pages happened to be allocated.
> Not sure if it should be considered ABI break or not.
Looks like this shouldn't be a problem. man 2 fallocate:
Within the specified range, partial filesystem blocks are zeroed,
and whole filesystem blocks are removed from the file. After a
successful call, subsequent reads from this range will return
zeroes.
It means we effectively have 2M filesystem block size.
And I don't see any guarantee about subsequent lseek(SEEK_HOLE) beheviour.
--
Kirill A. Shutemov
WARNING: multiple messages have this Message-ID (diff)
From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: linux-fsdevel@vger.kernel.org, linux-api@vger.kernel.org
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Hugh Dickins <hughd@google.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
Dave Hansen <dave.hansen@intel.com>,
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: THP-enabled filesystem vs. FALLOC_FL_PUNCH_HOLE
Date: Fri, 4 Mar 2016 14:26:03 +0300 [thread overview]
Message-ID: <20160304112603.GA9790@node.shutemov.name> (raw)
In-Reply-To: <1457023939-98083-1-git-send-email-kirill.shutemov@linux.intel.com>
On Thu, Mar 03, 2016 at 07:51:50PM +0300, Kirill A. Shutemov wrote:
> Truncate and punch hole that only cover part of THP range is implemented
> by zero out this part of THP.
>
> This have visible effect on fallocate(FALLOC_FL_PUNCH_HOLE) behaviour.
> As we don't really create hole in this case, lseek(SEEK_HOLE) may have
> inconsistent results depending what pages happened to be allocated.
> Not sure if it should be considered ABI break or not.
Looks like this shouldn't be a problem. man 2 fallocate:
Within the specified range, partial filesystem blocks are zeroed,
and whole filesystem blocks are removed from the file. After a
successful call, subsequent reads from this range will return
zeroes.
It means we effectively have 2M filesystem block size.
And I don't see any guarantee about subsequent lseek(SEEK_HOLE) beheviour.
--
Kirill A. Shutemov
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: "Kirill A. Shutemov" <kirill@shutemov.name>
To: linux-fsdevel@vger.kernel.org, linux-api@vger.kernel.org
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Hugh Dickins <hughd@google.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
Dave Hansen <dave.hansen@intel.com>,
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: THP-enabled filesystem vs. FALLOC_FL_PUNCH_HOLE
Date: Fri, 4 Mar 2016 14:26:03 +0300 [thread overview]
Message-ID: <20160304112603.GA9790@node.shutemov.name> (raw)
In-Reply-To: <1457023939-98083-1-git-send-email-kirill.shutemov@linux.intel.com>
On Thu, Mar 03, 2016 at 07:51:50PM +0300, Kirill A. Shutemov wrote:
> Truncate and punch hole that only cover part of THP range is implemented
> by zero out this part of THP.
>
> This have visible effect on fallocate(FALLOC_FL_PUNCH_HOLE) behaviour.
> As we don't really create hole in this case, lseek(SEEK_HOLE) may have
> inconsistent results depending what pages happened to be allocated.
> Not sure if it should be considered ABI break or not.
Looks like this shouldn't be a problem. man 2 fallocate:
Within the specified range, partial filesystem blocks are zeroed,
and whole filesystem blocks are removed from the file. After a
successful call, subsequent reads from this range will return
zeroes.
It means we effectively have 2M filesystem block size.
And I don't see any guarantee about subsequent lseek(SEEK_HOLE) beheviour.
--
Kirill A. Shutemov
next prev parent reply other threads:[~2016-03-04 11:26 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-03 16:51 [PATCHv3 00/29] huge tmpfs implementation using compound pages Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 01/29] rmap: introduce rmap_walk_locked() Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 02/29] rmap: extend try_to_unmap() to be usable by split_huge_page() Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 03/29] mm: make remove_migration_ptes() beyond mm/migration.c Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 04/29] thp: rewrite freeze_page()/unfreeze_page() with generic rmap walkers Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 05/29] mm: do not pass mm_struct into handle_mm_fault Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 06/29] mm: introduce fault_env Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 07/29] mm: postpone page table allocation until we have page to map Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 08/29] rmap: support file thp Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:51 ` [PATCHv3 09/29] mm: introduce do_set_pmd() Kirill A. Shutemov
2016-03-03 16:51 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 10/29] mm, rmap: account file thp pages Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 11/29] thp, vmstats: add counters for huge file pages Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 12/29] thp: support file pages in zap_huge_pmd() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 13/29] thp: handle file pages in split_huge_pmd() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 14/29] thp: handle file COW faults Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 15/29] thp: handle file pages in mremap() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 16/29] thp: skip file huge pmd on copy_huge_pmd() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 17/29] thp: prepare change_huge_pmd() for file thp Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 18/29] thp: run vma_adjust_trans_huge() outside i_mmap_rwsem Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 19/29] thp: file pages support for split_huge_page() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 20/29] thp, mlock: do not mlock PTE-mapped file huge pages Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 21/29] vmscan: split file huge pages before paging them out Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 22/29] page-flags: relax policy for PG_mappedtodisk and PG_reclaim Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 23/29] radix-tree: implement radix_tree_maybe_preload_order() Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 24/29] filemap: prepare find and delete operations for huge pages Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 25/29] truncate: handle file thp Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 26/29] shmem: prepare huge= mount option and sysfs knob Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 27/29] shmem: get_unmapped_area align huge page Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 28/29] shmem: add huge pages support Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-03 16:52 ` [PATCHv3 29/29] shmem, thp: respect MADV_{NO,}HUGEPAGE for file mappings Kirill A. Shutemov
2016-03-03 16:52 ` Kirill A. Shutemov
2016-03-04 4:20 ` [PATCHv3 00/29] huge tmpfs implementation using compound pages Sasha Levin
2016-03-04 4:20 ` Sasha Levin
2016-03-04 22:53 ` Kirill A. Shutemov
2016-03-04 22:53 ` Kirill A. Shutemov
[not found] ` <1457023939-98083-1-git-send-email-kirill.shutemov-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2016-03-04 11:26 ` Kirill A. Shutemov [this message]
2016-03-04 11:26 ` THP-enabled filesystem vs. FALLOC_FL_PUNCH_HOLE Kirill A. Shutemov
2016-03-04 11:26 ` Kirill A. Shutemov
2016-03-04 17:40 ` Dave Hansen
2016-03-04 17:40 ` Dave Hansen
[not found] ` <56D9C882.3040808-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2016-03-04 19:38 ` Hugh Dickins
2016-03-04 19:38 ` Hugh Dickins
2016-03-04 19:38 ` Hugh Dickins
2016-03-04 22:48 ` Kirill A. Shutemov
2016-03-04 22:48 ` Kirill A. Shutemov
[not found] ` <alpine.LSU.2.11.1603041100320.6011-fupSdm12i1nKWymIFiNcPA@public.gmane.org>
2016-03-04 23:05 ` Dave Chinner
2016-03-04 23:05 ` Dave Chinner
2016-03-04 23:05 ` Dave Chinner
2016-03-04 23:24 ` Kirill A. Shutemov
2016-03-04 23:24 ` Kirill A. Shutemov
2016-03-04 23:24 ` Kirill A. Shutemov
[not found] ` <20160304232412.GC12498-sVvlyX1904swdBt8bTSxpkEMvNT87kid@public.gmane.org>
2016-03-05 22:38 ` Dave Chinner
2016-03-05 22:38 ` Dave Chinner
2016-03-05 22:38 ` Dave Chinner
2016-03-06 0:30 ` Kirill A. Shutemov
2016-03-06 0:30 ` Kirill A. Shutemov
[not found] ` <20160306003034.GA13704-sVvlyX1904swdBt8bTSxpkEMvNT87kid@public.gmane.org>
2016-03-06 23:03 ` Dave Chinner
2016-03-06 23:03 ` Dave Chinner
2016-03-06 23:03 ` Dave Chinner
2016-03-06 23:33 ` Kirill A. Shutemov
2016-03-06 23:33 ` 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=20160304112603.GA9790@node.shutemov.name \
--to=kirill-okw7cidhh8elwutg50ltga@public.gmane.org \
--cc=aarcange-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
--cc=cl-gkYfJU5Cukgdnm+yROfE0A@public.gmane.org \
--cc=dave.hansen-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
--cc=hughd-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=jmarchan-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=kirill.shutemov-VuQAYsv1563Yd54FQh9/CA@public.gmane.org \
--cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=n-horiguchi-PaJj6Psr51x8UrSeD/g0lQ@public.gmane.org \
--cc=sasha.levin-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org \
--cc=vbabka-AlSwsSmVLrQ@public.gmane.org \
--cc=yang.shi-QSEj5FYQhm4dnm+yROfE0A@public.gmane.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.