[PATCH] Bind Mount Extensions 0.05
Herbert Poetzl
Greetings!
The following patch extends the 'noatime', 'nodiratime' and
last but not least the 'ro' (read only) mount option to the
vfs --bind mounts, allowing them to behave like any other
mount, by honoring those mount flags (which are silently
ignored by the current implementation in 2.4.x and 2.6.x)
I don't want to pollute your mailbox with useless patches,
so for those who are interested in this stuff, get them
here (for 2.4.27 and 2.6.8.1)
http://www.13thfloor.at/patches/
many thanks to Willy Tarreau for spotting the bug in the
previous bme0.04 for linux 2.4.x.
enjoy,
Herbert
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majo...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Jonathan Sambrook
>
> Greetings!
>
> The following patch extends the 'noatime', 'nodiratime' and
> last but not least the 'ro' (read only) mount option to the
> vfs --bind mounts, allowing them to behave like any other
> mount, by honoring those mount flags (which are silently
> ignored by the current implementation in 2.4.x and 2.6.x)
>
What's the likelyhood/timeframe for this getting into mainline kernels?
Cheers,
Jonathan
--
Jonathan Sambrook
Software Developer
Designer Servers
Christoph Hellwig
> What's the likelyhood/timeframe for this getting into mainline kernels?
This patch? not at all. THe feature? Maybe this year, but as it'll need
interface changes I'm not sure it's a good idea to do it in 2.6.
Herbert Poetzl
>
> Greetings!
>
> The following patch extends the 'noatime', 'nodiratime' and
> last but not least the 'ro' (read only) mount option to the
> vfs --bind mounts, allowing them to behave like any other
> mount, by honoring those mount flags (which are silently
> ignored by the current implementation in 2.4.x and 2.6.x)
>
> I don't want to pollute your mailbox with useless patches,
> so for those who are interested in this stuff, get them
> here (for 2.4.27 and 2.6.8.1)
patch for 2.6.8.1 is broken, but it will be updated soon
please don't use it for now ...
TIA,
Herbert
Herbert Poetzl
> On Wed, Aug 18, 2004 at 02:49:59PM +0100, Jonathan Sambrook wrote:
> > What's the likelyhood/timeframe for this getting into mainline kernels?
>
> This patch? not at all. THe feature?
> Maybe this year, but as it'll need interface changes
what interface changes are needed?
TIA,
Herbert
Herbert Poetzl
> On Wed, Aug 18, 2004 at 02:51:04PM +0200, Herbert Poetzl wrote:
> >
> > Greetings!
> >
> > The following patch extends the 'noatime', 'nodiratime' and
> > last but not least the 'ro' (read only) mount option to the
> > vfs --bind mounts, allowing them to behave like any other
> > mount, by honoring those mount flags (which are silently
> > ignored by the current implementation in 2.4.x and 2.6.x)
> >
> > I don't want to pollute your mailbox with useless patches,
> > so for those who are interested in this stuff, get them
> > here (for 2.4.27 and 2.6.8.1)
>
> patch for 2.6.8.1 is broken, but it will be updated soon
> please don't use it for now ...
okay, the patch for 2.6.8.1 was updated and should be fine
now, please let me know if you encounter any issues, as the
version for 2.6.8.1 is relatively new ...
http://www.13thfloor.at/patches/patch-2.6.8.1-bme0.05.1.diff
best,
Serge E. Hallyn
> now, please let me know if you encounter any issues, as the
> version for 2.6.8.1 is relatively new ...
>
> http://www.13thfloor.at/patches/patch-2.6.8.1-bme0.05.1.diff
This patch works wonderfully for me (with ext3).
Have you investigated which cases cause MNT_IS_RDONLY to be called
with a NULL vfsmount, and whether any of these cases are meaningful?
Quoting Christoph Hellwig (h...@infradead.org):
> On Wed, Aug 18, 2004 at 02:49:59PM +0100, Jonathan Sambrook wrote:
> > What's the likelyhood/timeframe for this getting into mainline kernels?
>
> This patch? not at all. THe feature? Maybe this year, but as it'll need
> interface changes I'm not sure it's a good idea to do it in 2.6.
What are the interface changes which will improve this patch?
Is waiting until 2.7 unnecessary given the new kernel development model?
Or are the interface changes just that drastic? :)
thanks,
-serge