Re: [PATCH 2/2] FIEMAP ioctl for ext4

2007-11-12 Thread David Chinner
On Tue, Nov 13, 2007 at 02:30:06AM +0530, Kalpak Shah wrote: > Recently there was discussion about an "FIle Extent MAP"(FIEMAP) ioctl for > efficiently mapping the extents and holes of a file. This will be many times > more efficient than FIBMAP by cutting down the number of ioctls. > > This pat

[ANNOUNCE] Ceph distributed file system

2007-11-12 Thread Sage Weil
Hi everyone, Ceph is a distributed network file system designed to provide excellent performance, reliability, and scalability with POSIX semantics. I periodically see frustration on this list with the lack of a scalable GPL distributed file system with sufficiently robust replication and fail

Re: [rfc][patch 3/5] afs: new aops

2007-11-12 Thread Nick Piggin
On Tue, Nov 13, 2007 at 12:30:05AM +, David Howells wrote: > Nick Piggin <[EMAIL PROTECTED]> wrote: > > > PAGE_CACHE_SIZE should be used to address the pagecache. > > Perhaps, but the function being called from there takes pages not page cache > slots. If I have to allow for PAGE_CACHE_SIZE

Re: [rfc][patch 3/5] afs: new aops

2007-11-12 Thread David Howells
Nick Piggin <[EMAIL PROTECTED]> wrote: > PAGE_CACHE_SIZE should be used to address the pagecache. Perhaps, but the function being called from there takes pages not page cache slots. If I have to allow for PAGE_CACHE_SIZE > PAGE_SIZE then I need to modify my code, if not then the assertion needs

Re: [rfc][patch 3/5] afs: new aops

2007-11-12 Thread Nick Piggin
On Mon, Nov 12, 2007 at 03:29:14PM +, David Howells wrote: > Nick Piggin <[EMAIL PROTECTED]> wrote: > > > - ASSERTCMP(start + len, <=, PAGE_SIZE); > > + ASSERTCMP(len, <=, PAGE_CACHE_SIZE); > > Do you guarantee this will work if PAGE_CACHE_SIZE != PAGE_SIZE? If not, you > can't make this

Re: [PATCH 1/2] Add FIEMAP header file

2007-11-12 Thread Mark Fasheh
On Tue, Nov 13, 2007 at 02:30:02AM +0530, Kalpak Shah wrote: > Recently there was discussion about an "FIle Extent MAP"(FIEMAP) ioctl for > efficiently mapping the extents and holes of a file. This will be many > times more efficient than FIBMAP by cutting down the number of ioctls. Thanks for sen

[PATCH 1/2] Add FIEMAP header file

2007-11-12 Thread Kalpak Shah
Recently there was discussion about an "FIle Extent MAP"(FIEMAP) ioctl for efficiently mapping the extents and holes of a file. This will be many times more efficient than FIBMAP by cutting down the number of ioctls. This patch adds the FIEMAP header file in include/linux. Signed-off-by: Andrea

[PATCH 2/2] FIEMAP ioctl for ext4

2007-11-12 Thread Kalpak Shah
Recently there was discussion about an "FIle Extent MAP"(FIEMAP) ioctl for efficiently mapping the extents and holes of a file. This will be many times more efficient than FIBMAP by cutting down the number of ioctls. This patch adds the FIEMAP ioctl for ext4. The spec for the FIEMAP ioctl was p

Re: msync(2) bug(?), returns AOP_WRITEPAGE_ACTIVATE to userland

2007-11-12 Thread Hugh Dickins
On Fri, 9 Nov 2007, Erez Zadok wrote: > In message <[EMAIL PROTECTED]>, Hugh Dickins writes: > > > Three, I believe you need to add a flush_dcache_page(lower_page) > > after the copy_highpage(lower_page): some architectures will need > > that to see the new data if they have lower_page mapped (tho

Re: [rfc][patch 3/5] afs: new aops

2007-11-12 Thread David Howells
Nick Piggin <[EMAIL PROTECTED]> wrote: > - ASSERTCMP(start + len, <=, PAGE_SIZE); > + ASSERTCMP(len, <=, PAGE_CACHE_SIZE); Do you guarantee this will work if PAGE_CACHE_SIZE != PAGE_SIZE? If not, you can't make this particular change. Do we ever intend to have PAGE_CACHE_SIZE != PAGE_SI