On 2018年02月01日 02:26, David Sterba wrote: > On Fri, Jan 19, 2018 at 01:37:15PM +0800, Qu Wenruo wrote: >> The long planned cmds-check re-construction is finally here. >> >> As the original cmds-check.c is getting larger and larger (already over >> 15K lines), it's always a good idea to split it into its own check/ >> directory. >> >> This patchset do the following work: >> 1) Move cmds-check.c to check/main.c >> 2) Put codes shared by both original and lowmem mode into >> check/common.[ch] >> 3) Put lowmem code into check/lowmem.[ch] >> With minor renaming to get rid of unnecessary _v2 suffix. >> >> The modification looks scary, but no functional change at all. >> >> And considering how much the file structure changed, it's a good idea to >> put PART1 as quick as possible, and there will be less pressure to >> rebase new incoming fsck related codes. > > I agree with the way you split check, my attempts were going along the > same lines. The question whether to split first and then add fixes shall > be resolved as I'm going to merge the split first. This would need > refreshing the lowmem fixes but the other way around would not avoid the > merge conflicts anyway. > > The patch 15 does not apply cleanly to current devel, the conflict did > not seem big, but I did not try to resolve it and merge the last patch. > So, 1-14 goes to devel, please refresh 15 and 16.
No problem. Update under way. > >> The real move work happens in the 15th patch, which due to its size >> (500KB+), it may not be able to reach mail list. >> So please fetch the whole patchset from github: >> https://github.com/adam900710/btrfs-progs/tree/split_check >> >> There will be a part 2, mostly moving original mode to its own >> check/original.[ch], along with extra comment explaining how the two >> different modes work. > > I'm going to do a few file renames, eg. original.c -> mode-original.c > and similar. > It would be better if you send part 2 after that. Of course. It would be a while for part 2. > I'm > expecting more patches that would do no functional change so we can > settle down with the new check/ structure. > > As 'check' is the 1st level command, the cmds-check.c will need to be > restored at some point, but it will be only a simple wrapper around the > commandline handling. OK, I think it would be part of PART 2 to restore it. Thanks, Qu > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majord...@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html >
signature.asc
Description: OpenPGP digital signature
