My interpretation of the preceeding is that there is agreement that
the functionality currently implemented in __setlease() should be
exported, even though the exported name may not be __setlease(). Is
this correct?
If so, that is just fine with me.
The question that I have now is when do you
On Thu, Jun 07, 2007 at 10:43:51AM -0400, Robert Rappaport wrote:
My interpretation of the preceeding is that there is agreement that
the functionality currently implemented in __setlease() should be
exported, even though the exported name may not be __setlease(). Is
this correct?
Yes.
If
Why isn't the existing setlease() export sufficient? The plan is to more
or less get rid of __setlease().
Trond
On Tue, 2007-06-05 at 18:04 -0400, Robert Rappaport wrote:
I have had some previous communications with Bruce on these topics,
and I am generally pleased with the proposed
Hi Bruce,
The file system does need to keep the local state up to date, like it does
with posix locks, so it might need to call __setlease(). The why we had it
before was that the call to the file system was done from outside of
setlease() and the file system was able to call setlease() which
On Tue, 2007-06-05 at 15:56 -0700, Marc Eshel wrote:
Hi Bruce,
The file system does need to keep the local state up to date, like it does
with posix locks, so it might need to call __setlease(). The why we had it
before was that the call to the file system was done from outside of
On Sat, Jun 02, 2007 at 02:21:22PM -0400, Trond Myklebust wrote:
Currently, the lease handling is done all in the VFS, and is done prior
to calling any filesystem operations. Bruce's break_lease() inode
operation allows the VFS to notify the filesystem that some operation is
going to be called
[EMAIL PROTECTED] wrote on 06/02/2007 10:39:10 AM:
On Fri, 2007-06-01 at 12:53 -0400, J. Bruce Fields wrote:
It also means that you're calling twice down
into the filesystem for every call to may_open() (once for
vfs_permission() and once for break_lease()) and 3 times in
On Sat, 2007-06-02 at 11:09 -0700, Marc Eshel wrote:
A cluster filesystem don't need the inode operation breake_lease. The fs
must be able to detect the conflict and break the lease using
__break_lease() call on all the nodes that hold a conflicting leases.
Marc.
Currently, the lease
On Sat, 2007-06-02 at 11:09 -0700, Marc Eshel wrote:
[EMAIL PROTECTED] wrote on 06/02/2007 10:39:10 AM:
^
BTW: your mailer is seriously broken.
Trond
-
To unsubscribe from this list: send the line unsubscribe linux-fsdevel in
the body of a message to [EMAIL
On Fri, Jun 01, 2007 at 09:14:53AM -0400, Peter Staubach wrote:
J. Bruce Fields wrote:
From: J. Bruce Fields [EMAIL PROTECTED]
Currently leases are only kept locally, so there's no way for a distributed
filesystem to enforce them against multiple clients. We're particularly
interested in
On Thu, May 31, 2007 at 06:34:09PM -0400, Trond Myklebust wrote:
On Thu, 2007-05-31 at 17:40 -0400, J. Bruce Fields wrote:
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7cf0c54..09aefb4 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1112,6 +1112,7 @@ struct
On Fri, Jun 01, 2007 at 12:44:16PM -0400, J. Bruce Fields wrote:
The only problem I'm aware of is that leases aren't broken on rename,
link, and unlink. This is kind of tricky to fix. David Richter (cc'd)
and I sketched out a few different approaches, and I think he has some
patches
On Fri, Jun 01, 2007 at 11:41:23AM -0600, Matthew Wilcox wrote:
Samba internally prohibits renaming or deleting an open file, to match
Windows semantics. So it won't notice the difference. At least, that's
what I remember from a discussion with Tridge when we were implementing
leases back in
On Thu, 2007-05-31 at 17:40 -0400, J. Bruce Fields wrote:
From: J. Bruce Fields [EMAIL PROTECTED]
Currently leases are only kept locally, so there's no way for a distributed
filesystem to enforce them against multiple clients. We're particularly
interested in the case of nfsd exporting a
14 matches
Mail list logo
