Re: [git pull] vfs pile 3

[Al Viro]

On Fri, Dec 23, 2016 at 12:03:21AM +, Al Viro wrote:
>   Assorted cleanups and fixes all over the place.

Fix for aio compat in !CONFIG_AIO builds folded in, commit message on
seq_file patch unmangled.  Other than that, identical to the previous.

The following changes since commit e93b1cc8a8965da137ffea0b88e5f62fa1d2a9e6:

  Merge branch 'for_linus' of 
git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs (2016-12-19 
08:23:53 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus

for you to fetch changes up to faf0dcebd7b387187f29ff811d47df465ea4c9f9:

  Merge branch 'work.namespace' into for-linus (2016-12-22 23:04:31 -0500)


Al Viro (7):
  remove a bogus claim about namespace_sem being held by callers of 
mnt_alloc_id()
  clone_private_mount() doesn't need to touch namespace_sem
  reorganize do_make_slave()
  move aio compat to fs/aio.c
  [iov_iter] fix iterate_all_kinds() on empty iterators
  sg_write()/bsg_write() is not fit to be called under KERNEL_DS
  Merge branch 'work.namespace' into for-linus

Aleksa Sarai (1):
  fs: exec: apply CLOEXEC before changing dumpable task flags

Darrick J. Wong (1):
  vfs: fix isize/pos/len checks for reflink & dedupe

Jeff Layton (1):
  ufs: fix function declaration for ufs_truncate_blocks

Tomasz Majchrzak (1):
  seq_file: reset iterator to first record for zero offset

 block/bsg.c |  3 ++
 drivers/scsi/sg.c   |  3 ++
 fs/aio.c| 97 -
 fs/compat.c | 75 --
 fs/exec.c   | 10 -
 fs/namespace.c  |  8 +---
 fs/ocfs2/refcounttree.c |  2 +-
 fs/pnode.c  | 74 ++---
 fs/read_write.c | 18 +
 fs/seq_file.c   |  7 
 fs/ufs/inode.c  |  2 +-
 fs/xfs/xfs_reflink.c|  2 +-
 include/linux/aio.h |  5 ---
 kernel/sys_ni.c |  3 ++
 lib/iov_iter.c  | 55 +---
 15 files changed, 197 insertions(+), 167 deletions(-)

Re: [git pull] vfs pile 3

[Al Viro]

On Fri, Dec 23, 2016 at 12:03:21AM +, Al Viro wrote:
>   Assorted cleanups and fixes all over the place.

Fix for aio compat in !CONFIG_AIO builds folded in, commit message on
seq_file patch unmangled.  Other than that, identical to the previous.

The following changes since commit e93b1cc8a8965da137ffea0b88e5f62fa1d2a9e6:

  Merge branch 'for_linus' of 
git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs (2016-12-19 
08:23:53 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus

for you to fetch changes up to faf0dcebd7b387187f29ff811d47df465ea4c9f9:

  Merge branch 'work.namespace' into for-linus (2016-12-22 23:04:31 -0500)


Al Viro (7):
  remove a bogus claim about namespace_sem being held by callers of 
mnt_alloc_id()
  clone_private_mount() doesn't need to touch namespace_sem
  reorganize do_make_slave()
  move aio compat to fs/aio.c
  [iov_iter] fix iterate_all_kinds() on empty iterators
  sg_write()/bsg_write() is not fit to be called under KERNEL_DS
  Merge branch 'work.namespace' into for-linus

Aleksa Sarai (1):
  fs: exec: apply CLOEXEC before changing dumpable task flags

Darrick J. Wong (1):
  vfs: fix isize/pos/len checks for reflink & dedupe

Jeff Layton (1):
  ufs: fix function declaration for ufs_truncate_blocks

Tomasz Majchrzak (1):
  seq_file: reset iterator to first record for zero offset

 block/bsg.c |  3 ++
 drivers/scsi/sg.c   |  3 ++
 fs/aio.c| 97 -
 fs/compat.c | 75 --
 fs/exec.c   | 10 -
 fs/namespace.c  |  8 +---
 fs/ocfs2/refcounttree.c |  2 +-
 fs/pnode.c  | 74 ++---
 fs/read_write.c | 18 +
 fs/seq_file.c   |  7 
 fs/ufs/inode.c  |  2 +-
 fs/xfs/xfs_reflink.c|  2 +-
 include/linux/aio.h |  5 ---
 kernel/sys_ni.c |  3 ++
 lib/iov_iter.c  | 55 +---
 15 files changed, 197 insertions(+), 167 deletions(-)

Re: [git pull] vfs pile 3

[Linus Torvalds]

On Sat, Aug 6, 2016 at 11:46 PM, Al Viro  wrote:
>
> I went for the second variant (backmerge), but if you prefer the third one,
> just pull #for-linus-2 and I'll send a separate pull request for the last
> commit.  Or just cherry-pick that last commit from #for-linus after having
> pulled #for-linus-2.

Please don't do back-merges to avoid conflicts. I'd *much* rather just
get the conflict. Especially when it's that trivial.

   Linus

Re: [git pull] vfs pile 3

[Linus Torvalds]

On Sat, Aug 6, 2016 at 11:46 PM, Al Viro  wrote:
>
> I went for the second variant (backmerge), but if you prefer the third one,
> just pull #for-linus-2 and I'll send a separate pull request for the last
> commit.  Or just cherry-pick that last commit from #for-linus after having
> pulled #for-linus-2.

Please don't do back-merges to avoid conflicts. I'd *much* rather just
get the conflict. Especially when it's that trivial.

   Linus

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 19:07, Al Viro ha scritto:

On Sat, Oct 13, 2012 at 12:04:55PM -0400, Christoph Hellwig wrote:

On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:

You know, I'm in the middle of dealing with one such TODO.  Yours, as it
were.  From six years ago.  kernel_thread() unexporting.  TODO comments
of any form are routinely shat upon and ignored, especially when shuffled
away into less read parts of the tree... ;-/

I'd rather see it done fs-by-fs.  Starting with something reasonably easy
to test - minixfs would do nicely.  Don't get me wrong - I'm all for
burying ->truncate(); what I'm worried about is that we'll end up burying
the warning about the reasons why vmtruncate() was a bad idea, leaving the
functionality exactly as it used to be...


As mentioned I agree with the concern in principle.  Let's start by
taking Marco's patches for filesystems that use vmtruncate but don't
actually implement ->truncate.  There's a few I remember offhand, e.g.
procfs and ufs right now.  Then we can do the actual work required ones
piece by piece.


Umm... That would be what, procfs?  Frankly, I'm not sure that ATTR_SIZE for
procfs actually should not be silently ignored.  ->i_size there is completely
synthetic - it's not as if truncation would actually change the contents.

And ufs situation is quite different - there vmtruncate() is used only on the
->write_begin() side.  ->setattr() is already vmtruncate-free.  What's needed
there is an analog of e.g. ext2_write_failed().



I'm open to change the series and give any help. My original idea was to 
do a cleanup patch and after that give to each fs maintainer the 
possibility to do ad-hoc fix. Each fs  maintainer has got a deep 
knowledge of its fs so it was a safe approach from testing point of view 
and so on. However if you tell me that in this case another approach is 
better is ok for me. I'll fix the patches according to the comments of 
Christoph.


Regards,

Marco
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 19:07, Al Viro ha scritto:

On Sat, Oct 13, 2012 at 12:04:55PM -0400, Christoph Hellwig wrote:

On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:

You know, I'm in the middle of dealing with one such TODO.  Yours, as it
were.  From six years ago.  kernel_thread() unexporting.  TODO comments
of any form are routinely shat upon and ignored, especially when shuffled
away into less read parts of the tree... ;-/

I'd rather see it done fs-by-fs.  Starting with something reasonably easy
to test - minixfs would do nicely.  Don't get me wrong - I'm all for
burying -truncate(); what I'm worried about is that we'll end up burying
the warning about the reasons why vmtruncate() was a bad idea, leaving the
functionality exactly as it used to be...


As mentioned I agree with the concern in principle.  Let's start by
taking Marco's patches for filesystems that use vmtruncate but don't
actually implement -truncate.  There's a few I remember offhand, e.g.
procfs and ufs right now.  Then we can do the actual work required ones
piece by piece.


Umm... That would be what, procfs?  Frankly, I'm not sure that ATTR_SIZE for
procfs actually should not be silently ignored.  -i_size there is completely
synthetic - it's not as if truncation would actually change the contents.

And ufs situation is quite different - there vmtruncate() is used only on the
-write_begin() side.  -setattr() is already vmtruncate-free.  What's needed
there is an analog of e.g. ext2_write_failed().



I'm open to change the series and give any help. My original idea was to 
do a cleanup patch and after that give to each fs maintainer the 
possibility to do ad-hoc fix. Each fs  maintainer has got a deep 
knowledge of its fs so it was a safe approach from testing point of view 
and so on. However if you tell me that in this case another approach is 
better is ok for me. I'll fix the patches according to the comments of 
Christoph.


Regards,

Marco
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 12:04:55PM -0400, Christoph Hellwig wrote:
> On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:
> > You know, I'm in the middle of dealing with one such TODO.  Yours, as it
> > were.  From six years ago.  kernel_thread() unexporting.  TODO comments
> > of any form are routinely shat upon and ignored, especially when shuffled
> > away into less read parts of the tree... ;-/
> > 
> > I'd rather see it done fs-by-fs.  Starting with something reasonably easy
> > to test - minixfs would do nicely.  Don't get me wrong - I'm all for
> > burying ->truncate(); what I'm worried about is that we'll end up burying
> > the warning about the reasons why vmtruncate() was a bad idea, leaving the
> > functionality exactly as it used to be...
> 
> As mentioned I agree with the concern in principle.  Let's start by
> taking Marco's patches for filesystems that use vmtruncate but don't 
> actually implement ->truncate.  There's a few I remember offhand, e.g.
> procfs and ufs right now.  Then we can do the actual work required ones
> piece by piece.

Umm... That would be what, procfs?  Frankly, I'm not sure that ATTR_SIZE for
procfs actually should not be silently ignored.  ->i_size there is completely
synthetic - it's not as if truncation would actually change the contents.

And ufs situation is quite different - there vmtruncate() is used only on the
->write_begin() side.  ->setattr() is already vmtruncate-free.  What's needed
there is an analog of e.g. ext2_write_failed().
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Christoph Hellwig]

On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:
> You know, I'm in the middle of dealing with one such TODO.  Yours, as it
> were.  From six years ago.  kernel_thread() unexporting.  TODO comments
> of any form are routinely shat upon and ignored, especially when shuffled
> away into less read parts of the tree... ;-/
> 
> I'd rather see it done fs-by-fs.  Starting with something reasonably easy
> to test - minixfs would do nicely.  Don't get me wrong - I'm all for
> burying ->truncate(); what I'm worried about is that we'll end up burying
> the warning about the reasons why vmtruncate() was a bad idea, leaving the
> functionality exactly as it used to be...

As mentioned I agree with the concern in principle.  Let's start by
taking Marco's patches for filesystems that use vmtruncate but don't 
actually implement ->truncate.  There's a few I remember offhand, e.g.
procfs and ufs right now.  Then we can do the actual work required ones
piece by piece.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 11:48:10AM -0400, Christoph Hellwig wrote:
> On Sat, Oct 13, 2012 at 08:51:28AM +0100, Al Viro wrote:
> > It's somewhat pointless on its own...  If you were doing something with
> > the callers afterwards - sure, it would be make sense, but as it is...
> 
> I'd really like to see ->truncate and vmtruncate done, so from that side
> I'm absolutely in favour of this series.  What I'm a bit concerned about
> is that it just does the trivial 1:1 conversion and not actually
> converts the sequence of operations to the proper form, which was one
> of the two big reasons of moving away from ->truncate to start with.
> 
> I'd love to see the full conversion, but without adequate test coverage
> for all the fringe filesystems that might be a bit too much to expect
> from Marco.
> 
> I think just doing the easy conversions he did, and putting a TODO
> comment explaining how it should be taken further at each of the sites
> would be valueable on its own.

You know, I'm in the middle of dealing with one such TODO.  Yours, as it
were.  From six years ago.  kernel_thread() unexporting.  TODO comments
of any form are routinely shat upon and ignored, especially when shuffled
away into less read parts of the tree... ;-/

I'd rather see it done fs-by-fs.  Starting with something reasonably easy
to test - minixfs would do nicely.  Don't get me wrong - I'm all for
burying ->truncate(); what I'm worried about is that we'll end up burying
the warning about the reasons why vmtruncate() was a bad idea, leaving the
functionality exactly as it used to be...
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Christoph Hellwig]

On Sat, Oct 13, 2012 at 08:51:28AM +0100, Al Viro wrote:
> It's somewhat pointless on its own...  If you were doing something with
> the callers afterwards - sure, it would be make sense, but as it is...

I'd really like to see ->truncate and vmtruncate done, so from that side
I'm absolutely in favour of this series.  What I'm a bit concerned about
is that it just does the trivial 1:1 conversion and not actually
converts the sequence of operations to the proper form, which was one
of the two big reasons of moving away from ->truncate to start with.

I'd love to see the full conversion, but without adequate test coverage
for all the fringe filesystems that might be a bit too much to expect
from Marco.

I think just doing the easy conversions he did, and putting a TODO
comment explaining how it should be taken further at each of the sites
would be valueable on its own.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 09:51, Al Viro ha scritto:

On Sat, Oct 13, 2012 at 09:20:45AM +0200, Marco Stornelli wrote:

Il 13/10/2012 02:20, Al Viro ha scritto:

Stuff from Jeff Layton, mostly.  Sanitizing interplay between
audit and namei, removing a lot of insanity from audit_inode() mess
and getting things ready for his ESTALE patchset.  Please, pull from
the usual place -
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus



Al,

do you see any problem to apply the patch series "drop vmtruncate"?


It's somewhat pointless on its own...  If you were doing something with
the callers afterwards - sure, it would be make sense, but as it is...



The goal of the patch was to remove a deprecated function trying to 
improve the code, removing the inode operation and doing a general 
cleanup. It's pointless for me to have "dead"/"old" code, however thanks 
for you comment.


Marco
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 09:20:45AM +0200, Marco Stornelli wrote:
> Il 13/10/2012 02:20, Al Viro ha scritto:
> > Stuff from Jeff Layton, mostly.  Sanitizing interplay between
> >audit and namei, removing a lot of insanity from audit_inode() mess
> >and getting things ready for his ESTALE patchset.  Please, pull from
> >the usual place -
> >git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus
> >
> 
> Al,
> 
> do you see any problem to apply the patch series "drop vmtruncate"?

It's somewhat pointless on its own...  If you were doing something with
the callers afterwards - sure, it would be make sense, but as it is...
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 02:20, Al Viro ha scritto:

Stuff from Jeff Layton, mostly.  Sanitizing interplay between
audit and namei, removing a lot of insanity from audit_inode() mess
and getting things ready for his ESTALE patchset.  Please, pull from
the usual place -
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus



Al,

do you see any problem to apply the patch series "drop vmtruncate"? Do 
you think it's better to push it for 3.8?


Thanks,

Marco
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 02:20, Al Viro ha scritto:

Stuff from Jeff Layton, mostly.  Sanitizing interplay between
audit and namei, removing a lot of insanity from audit_inode() mess
and getting things ready for his ESTALE patchset.  Please, pull from
the usual place -
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus



Al,

do you see any problem to apply the patch series drop vmtruncate? Do 
you think it's better to push it for 3.8?


Thanks,

Marco
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 09:20:45AM +0200, Marco Stornelli wrote:
 Il 13/10/2012 02:20, Al Viro ha scritto:
  Stuff from Jeff Layton, mostly.  Sanitizing interplay between
 audit and namei, removing a lot of insanity from audit_inode() mess
 and getting things ready for his ESTALE patchset.  Please, pull from
 the usual place -
 git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus
 
 
 Al,
 
 do you see any problem to apply the patch series drop vmtruncate?

It's somewhat pointless on its own...  If you were doing something with
the callers afterwards - sure, it would be make sense, but as it is...
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Marco Stornelli]

Il 13/10/2012 09:51, Al Viro ha scritto:

On Sat, Oct 13, 2012 at 09:20:45AM +0200, Marco Stornelli wrote:

Il 13/10/2012 02:20, Al Viro ha scritto:

Stuff from Jeff Layton, mostly.  Sanitizing interplay between
audit and namei, removing a lot of insanity from audit_inode() mess
and getting things ready for his ESTALE patchset.  Please, pull from
the usual place -
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git for-linus



Al,

do you see any problem to apply the patch series drop vmtruncate?


It's somewhat pointless on its own...  If you were doing something with
the callers afterwards - sure, it would be make sense, but as it is...



The goal of the patch was to remove a deprecated function trying to 
improve the code, removing the inode operation and doing a general 
cleanup. It's pointless for me to have dead/old code, however thanks 
for you comment.


Marco
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Christoph Hellwig]

On Sat, Oct 13, 2012 at 08:51:28AM +0100, Al Viro wrote:
 It's somewhat pointless on its own...  If you were doing something with
 the callers afterwards - sure, it would be make sense, but as it is...

I'd really like to see -truncate and vmtruncate done, so from that side
I'm absolutely in favour of this series.  What I'm a bit concerned about
is that it just does the trivial 1:1 conversion and not actually
converts the sequence of operations to the proper form, which was one
of the two big reasons of moving away from -truncate to start with.

I'd love to see the full conversion, but without adequate test coverage
for all the fringe filesystems that might be a bit too much to expect
from Marco.

I think just doing the easy conversions he did, and putting a TODO
comment explaining how it should be taken further at each of the sites
would be valueable on its own.

--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 11:48:10AM -0400, Christoph Hellwig wrote:
 On Sat, Oct 13, 2012 at 08:51:28AM +0100, Al Viro wrote:
  It's somewhat pointless on its own...  If you were doing something with
  the callers afterwards - sure, it would be make sense, but as it is...
 
 I'd really like to see -truncate and vmtruncate done, so from that side
 I'm absolutely in favour of this series.  What I'm a bit concerned about
 is that it just does the trivial 1:1 conversion and not actually
 converts the sequence of operations to the proper form, which was one
 of the two big reasons of moving away from -truncate to start with.
 
 I'd love to see the full conversion, but without adequate test coverage
 for all the fringe filesystems that might be a bit too much to expect
 from Marco.
 
 I think just doing the easy conversions he did, and putting a TODO
 comment explaining how it should be taken further at each of the sites
 would be valueable on its own.

You know, I'm in the middle of dealing with one such TODO.  Yours, as it
were.  From six years ago.  kernel_thread() unexporting.  TODO comments
of any form are routinely shat upon and ignored, especially when shuffled
away into less read parts of the tree... ;-/

I'd rather see it done fs-by-fs.  Starting with something reasonably easy
to test - minixfs would do nicely.  Don't get me wrong - I'm all for
burying -truncate(); what I'm worried about is that we'll end up burying
the warning about the reasons why vmtruncate() was a bad idea, leaving the
functionality exactly as it used to be...
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Christoph Hellwig]

On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:
 You know, I'm in the middle of dealing with one such TODO.  Yours, as it
 were.  From six years ago.  kernel_thread() unexporting.  TODO comments
 of any form are routinely shat upon and ignored, especially when shuffled
 away into less read parts of the tree... ;-/
 
 I'd rather see it done fs-by-fs.  Starting with something reasonably easy
 to test - minixfs would do nicely.  Don't get me wrong - I'm all for
 burying -truncate(); what I'm worried about is that we'll end up burying
 the warning about the reasons why vmtruncate() was a bad idea, leaving the
 functionality exactly as it used to be...

As mentioned I agree with the concern in principle.  Let's start by
taking Marco's patches for filesystems that use vmtruncate but don't 
actually implement -truncate.  There's a few I remember offhand, e.g.
procfs and ufs right now.  Then we can do the actual work required ones
piece by piece.
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [git pull] vfs pile 3

[Al Viro]

On Sat, Oct 13, 2012 at 12:04:55PM -0400, Christoph Hellwig wrote:
 On Sat, Oct 13, 2012 at 05:01:15PM +0100, Al Viro wrote:
  You know, I'm in the middle of dealing with one such TODO.  Yours, as it
  were.  From six years ago.  kernel_thread() unexporting.  TODO comments
  of any form are routinely shat upon and ignored, especially when shuffled
  away into less read parts of the tree... ;-/
  
  I'd rather see it done fs-by-fs.  Starting with something reasonably easy
  to test - minixfs would do nicely.  Don't get me wrong - I'm all for
  burying -truncate(); what I'm worried about is that we'll end up burying
  the warning about the reasons why vmtruncate() was a bad idea, leaving the
  functionality exactly as it used to be...
 
 As mentioned I agree with the concern in principle.  Let's start by
 taking Marco's patches for filesystems that use vmtruncate but don't 
 actually implement -truncate.  There's a few I remember offhand, e.g.
 procfs and ufs right now.  Then we can do the actual work required ones
 piece by piece.

Umm... That would be what, procfs?  Frankly, I'm not sure that ATTR_SIZE for
procfs actually should not be silently ignored.  -i_size there is completely
synthetic - it's not as if truncation would actually change the contents.

And ufs situation is quite different - there vmtruncate() is used only on the
-write_begin() side.  -setattr() is already vmtruncate-free.  What's needed
there is an analog of e.g. ext2_write_failed().
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/