On Mon, 5 May 2014, Michael Kerrisk (man-pages) wrote: > Date: Mon, 5 May 2014 21:46:22 +0200 > From: "Michael Kerrisk (man-pages)" <mtk.manpages@xxxxxxxxx> > To: Lukáš Czerner <lczerner@xxxxxxxxxx> > Cc: linux-man <linux-man@xxxxxxxxxxxxxxx> > Subject: Re: [PATCH v3] fallocate.2: Document FALLOC_FL_ZERO_RANGE > > Hi Lukáš, > > On Mon, May 5, 2014 at 3:06 PM, Lukáš Czerner <lczerner@xxxxxxxxxx> wrote: > > On Mon, 5 May 2014, Michael Kerrisk (man-pages) wrote: > > > >> Date: Mon, 05 May 2014 14:44:48 +0200 > >> From: "Michael Kerrisk (man-pages)" <mtk.manpages@xxxxxxxxx> > >> To: Lukas Czerner <lczerner@xxxxxxxxxx> > >> Cc: mtk.manpages@xxxxxxxxx, linux-man@xxxxxxxxxxxxxxx > >> Subject: Re: [PATCH v3] fallocate.2: Document FALLOC_FL_ZERO_RANGE > >> > >> > >> On 05/05/2014 02:04 PM, Lukas Czerner wrote: > >> > FALLOC_FL_ZERO_RANGE was added in Linux 3.14, > >> > for zeroing ranges in the allocated space in a file. > >> > > >> > Signed-off-by: Lukas Czerner <lczerner@xxxxxxxxxx> > >> > --- > >> > v2: Rebase and update the description > >> > v3: Ext4 zero range on extent based files > >> > >> Thank, Lukáš. > >> > >> In fact I'd already merged and tweaked your previous patch into > >> a branch, and the only new piece from this patch that I needed > >> was the piece about ext4 extent-based files. > >> > >> The text now reads: > >> > >> Zeroing file space > >> Specifying the FALLOC_FL_ZERO_RANGE flag (available since Linux > >> 3.14) in mode zeroes space in the byte range starting at offset > >> and continuing for len bytes. Within the specified range, > >> blocks are preallocated for the regions that span the holes in > >> the file. After a successful call, subsequent reads from this > >> range will return zeroes. > >> > >> Zeroing is done within the filesystem preferably by converting > >> the range into unwritten extents. This approach means that the > >> specified range will not be physically zeroed out on the > >> device, and I/O is required only to update file metadata. > > > > mostly for metadata as mentioned in my patch. Because we still need > > to manually zero out partial blocks. But that's not really all that > > important. > > Thanks for checking. Now I understand the "mostly"! How about we make the text: > > [[ > Zeroing is done within the filesystem preferably by converting the range into > unwritten extents. > This approach means that the specified range will not be physically zeroed > out on the device (except for partial blocks at the either end of the range), > and I/O is (otherwise) required only to update file metadata. > ]] Perfect. Just get rid of "file" in the "file metadata" and I think it's ok. The reason is that those are also file system metadata in the case you're actually allocating blocks. Thanks! -Lukas > > ? > > Cheers, > > Michael > > > > > > Thanks! > > -Lukas > > > >> > >> If the FALLOC_FL_KEEP_SIZE flag is additionally specified in > >> mode, the behavior of the call is similar, but the file size > >> will not be changed even if offset+len is greater than the file > >> size. This behaviour is the same as when preallocating space > >> with FALLOC_FL_KEEP_SIZE specified. > >> > >> Not all filesystems support FALLOC_FL_ZERO_RANGE; if a filesys‐ > >> tem doesn't support the operation, an error is returned. The > >> operation is supported on at least the following filesystems: > >> > >> * XFS (since Linux 3.14) > >> > >> * ext4, for extent-based files (since Linux 3.14) > >> ... > >> ERRORS > >> EINVAL mode is FALLOC_FL_COLLAPSE_RANGE or FAL‐ > >> LOC_FL_ZERO_RANGE, but the file referred to by fd is not > >> a regular file. > >> > >> Okay? > >> > >> Cheers, > >> > >> Michael > >> > > > >
