Messages in this thread | | | Date | Thu, 20 Feb 2014 07:56:21 +0900 | Subject | Re: [PATCH v5 3/10] ext4: Add support FALLOC_FL_COLLAPSE_RANGE for fallocate | From | Namjae Jeon <> |
| |
2014-02-20 6:58 GMT+09:00, Dongsu Park <dongsu.park@profitbricks.com>: > Hi Namjae, > see below: Hi Dongsu. > > On 19.02.2014 01:38, Namjae Jeon wrote: >> This patch implements fallocate's FALLOC_FL_COLLAPSE_RANGE for Ext4. > ...<snip>... >> + /* >> + * Don't start shifting extents until we make sure the hole is big >> + * enough to accomodate the shift. >> + */ >> + path = ext4_ext_find_extent(inode, start - 1, NULL, 0); >> + depth = path->p_depth; >> + extent = path[depth].p_ext; >> + ex_start = extent->ee_block; >> + ex_end = extent->ee_block + ext4_ext_get_actual_len(extent); >> + ext4_ext_drop_refs(path); >> + kfree(path); >> + >> + if ((start == ex_start && shift > ex_start) || >> + (shift > start - ex_end)) { > > This line causes a compile error. So the fix would be like that: Oops.. I made a stupid mistake when creating the patch at dawn.. Thanks for your pointing. > > - (shift > start - ex_end)) { > + (shift > start - ex_end)) > >> + return -EINVAL; >> + >> + /* Its safe to start updating extents */ Yes, Right. > ...<snip>... > > Apart from that, the whole patchset seems to work without error, > both on xfs and on ext4. > Of course I had to remove Lukas' patches before testing collapse-range. > So it would be excellent if two patch series could get somehow merged. > > Tested-by: Dongsu Park <dongsu.park@profitbricks.com> Thank you so much for your testing! > > Regards, > Dongsu >
| |