linux-kernel - Re: [PATCH 5/10] mm/fs: route MADV_REMOVE to FALLOC_FL_PUNCH

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Mon, 21 May 2012 17:08:29 -0700
From:	john stultz <johnstul@...ibm.com>
To:	Hugh Dickins <hughd@...gle.com>,
	Arve Hjønnevåg <arve@...roid.com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Christoph Hellwig <hch@...radead.org>,
	Cong Wang <amwang@...hat.com>,
	Al Viro <viro@...iv.linux.org.uk>,
	Colin Cross <ccross@...roid.com>,
	John Stulz <john.stulz@...aro.org>,
	Greg Kroah-Hartman <gregkh@...ux-foundation.org>,
	"Theodore Ts'o" <tytso@....edu>,
	Andreas Dilger <adilger@...ger.ca>,
	Mark Fasheh <mfasheh@...e.de>,
	Joel Becker <jlbec@...lplan.org>,
	Dave Chinner <david@...morbit.com>, Ben Myers <bpm@....com>,
	Michael Kerrisk <mtk.manpages@...il.com>, linux-mm@...ck.org,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 5/10] mm/fs: route MADV_REMOVE to FALLOC_FL_PUNCH_HOLE

On Sat, May 12, 2012 at 5:13 AM, Hugh Dickins <hughd@...gle.com> wrote:
> Now tmpfs supports hole-punching via fallocate(), switch madvise_remove()
> to use do_fallocate() instead of vmtruncate_range(): which extends
> madvise(,,MADV_REMOVE) support from tmpfs to ext4, ocfs2 and xfs.
>
> There is one more user of vmtruncate_range() in our tree, staging/android's
> ashmem_shrink(): convert it to use do_fallocate() too (but if its unpinned
> areas are already unmapped - I don't know - then it would do better to use
> shmem_truncate_range() directly).

I suspect shmem_truncate_range directly would be the right approach,
but am not totally sure.
Arve: Any thoughts?

Hugh: Do you have a git tree with this set available somewhere?  I was
working on my own tmpfs support for FALLOC_FL_PUNCH_HOLE, along with
my volatile range work, so I'd like to rebase on top of your work
here.

thanks
-john


>
> Based-on-patch-by: Cong Wang <amwang@...hat.com>
> Signed-off-by: Hugh Dickins <hughd@...gle.com>
> ---
>  drivers/staging/android/ashmem.c |    8 +++++---
>  mm/madvise.c                     |   15 +++++++--------
>  2 files changed, 12 insertions(+), 11 deletions(-)
>
> --- 3045N.orig/drivers/staging/android/ashmem.c 2012-05-05 10:42:33.564056626 -0700
> +++ 3045N/drivers/staging/android/ashmem.c      2012-05-05 10:46:25.692062478 -0700
> @@ -19,6 +19,7 @@
>  #include <linux/module.h>
>  #include <linux/file.h>
>  #include <linux/fs.h>
> +#include <linux/falloc.h>
>  #include <linux/miscdevice.h>
>  #include <linux/security.h>
>  #include <linux/mm.h>
> @@ -363,11 +364,12 @@ static int ashmem_shrink(struct shrinker
>
>        mutex_lock(&ashmem_mutex);
>        list_for_each_entry_safe(range, next, &ashmem_lru_list, lru) {
> -               struct inode *inode = range->asma->file->f_dentry->d_inode;
>                loff_t start = range->pgstart * PAGE_SIZE;
> -               loff_t end = (range->pgend + 1) * PAGE_SIZE - 1;
> +               loff_t end = (range->pgend + 1) * PAGE_SIZE;
>
> -               vmtruncate_range(inode, start, end);
> +               do_fallocate(range->asma->file,
> +                               FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
> +                               start, end - start);
>                range->purged = ASHMEM_WAS_PURGED;
>                lru_del(range);
>
> --- 3045N.orig/mm/madvise.c     2012-05-05 10:42:33.572056784 -0700
> +++ 3045N/mm/madvise.c  2012-05-05 10:46:25.692062478 -0700
> @@ -11,8 +11,10 @@
>  #include <linux/mempolicy.h>
>  #include <linux/page-isolation.h>
>  #include <linux/hugetlb.h>
> +#include <linux/falloc.h>
>  #include <linux/sched.h>
>  #include <linux/ksm.h>
> +#include <linux/fs.h>
>
>  /*
>  * Any behaviour which results in changes to the vma->vm_flags needs to
> @@ -200,8 +202,7 @@ static long madvise_remove(struct vm_are
>                                struct vm_area_struct **prev,
>                                unsigned long start, unsigned long end)
>  {
> -       struct address_space *mapping;
> -       loff_t offset, endoff;
> +       loff_t offset;
>        int error;
>
>        *prev = NULL;   /* tell sys_madvise we drop mmap_sem */
> @@ -217,16 +218,14 @@ static long madvise_remove(struct vm_are
>        if ((vma->vm_flags & (VM_SHARED|VM_WRITE)) != (VM_SHARED|VM_WRITE))
>                return -EACCES;
>
> -       mapping = vma->vm_file->f_mapping;
> -
>        offset = (loff_t)(start - vma->vm_start)
>                        + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
> -       endoff = (loff_t)(end - vma->vm_start - 1)
> -                       + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
>
> -       /* vmtruncate_range needs to take i_mutex */
> +       /* filesystem's fallocate may need to take i_mutex */
>        up_read(&current->mm->mmap_sem);
> -       error = vmtruncate_range(mapping->host, offset, endoff);
> +       error = do_fallocate(vma->vm_file,
> +                               FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
> +                               offset, end - start);
>        down_read(&current->mm->mmap_sem);
>        return error;
>  }
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/