<p><br>
On Aug 20, 2011 5:52 AM, "Marco Stornelli" <<a href="mailto:marco.stornelli@gmail.com">marco.stornelli@gmail.com</a>> wrote:<br>
><br>
> Hi,<br>
><br>
><br>
> Il 28/06/2011 17:33, Josef Bacik ha scritto:<br>
>><br>
>> This just gets us ready to support the SEEK_HOLE and SEEK_DATA flags. Turns out<br>
>> using fiemap in things like cp cause more problems than it solves, so lets try<br>
>> and give userspace an interface that doesn't suck. We need to match solaris<br>
>> here, and the definitions are<br>
>><br>
>> *o* If /whence/ is SEEK_HOLE, the offset of the start of the<br>
>> next hole greater than or equal to the supplied offset<br>
>> is returned. The definition of a hole is provided near<br>
>> the end of the DESCRIPTION.<br>
>><br>
>> *o* If /whence/ is SEEK_DATA, the file pointer is set to the<br>
>> start of the next non-hole file region greater than or<br>
>> equal to the supplied offset.<br>
>><br>
><br>
> I'm implementing the SEEK_DATA/SEEK_HOLE management for pramfs and I've got some doubts about the right behavior:<br>
><br>
> 1) when we use SEEK_DATA/SEEK_HOLE, the offset used in lseek means always the offset from the start of the file, right?<br>
><br>
> 2) in case of a file with hole at the beginning and data at the end, if I do lseek(fd, 0, SEEK_HOLE) I should receive the end of the file because the idea is to search the *next* hole and we have always a virtual hole at the end of the file, right?<br>
><br>
> 3) about the last sentence of point 2), is it always true even if we have a case of block allocation beyond the end of file (fallocate with keep size option)?<br>
></p>
<p>Marco,</p>
<p>You may want to enable the xfstests test(s) for SEEK_HOLE and SEEK_DATA for pramfs. That should give you some confidence your implementing the api like other filesystems are.</p>
<p>Greg</p>