On giovedì 2 marzo 2023 10:59:31 CET Jan Kara wrote: > On Wed 01-03-23 14:14:16, Al Viro wrote: > > On Wed, Mar 01, 2023 at 02:00:18PM +0100, Jan Kara wrote: > > > On Wed 01-03-23 12:20:56, Fabio M. De Francesco wrote: > > > > On venerdì 24 febbraio 2023 04:26:57 CET Al Viro wrote: > > > > > Fabio's "switch to kmap_local_page()" patchset (originally after the > > > > > > > > > > ext2 counterpart, with a lot of cleaning up done to it; as the matter > > > > > of > > > > > fact, ext2 side is in need of similar cleanups - calling conventions > > > > > there > > > > > are bloody awful). > > > > > > > > If nobody else is already working on these cleanups in ext2 following > > > > your > > > > suggestion, I'd be happy to work on this by the end of this week. I only > > > > need > > > > a confirmation because I'd hate to duplicate someone else work. > > > > > > > > > Plus the equivalents of minix stuff... > > > > > > > > I don't know this other filesystem but I could take a look and see > > > > whether it > > > > resembles somehow sysv and ext2 (if so, this work would be pretty simple > > > > too, > > > > thanks to your kind suggestions when I worked on sysv and ufs). > > > > > > > > I'm adding Jan to the Cc list to hear whether he is aware of anybody > > > > else > > > > working on this changes for ext2. I'm waiting for a reply from you (@Al) > > > > or > > > > Jan to avoid duplication (as said above). > > > > > > I'm not sure what exactly Al doesn't like about how ext2 handles pages and > > > mapping but if you have some cleanups in mind, sure go ahead. I don't have > > > any plans on working on that code in the near term. > > > > I think I've pushed a demo patchset to vfs.git at some point back in > > January... Yep - see #work.ext2 in there; completely untested, though. > > OK, I think your changes to ext2_rename() in PATCH 1 leak a reference and > mapping of old_page but otherwise I like the patches. So Fabio, if you can > pick them up and push this to completion, it would be nice. Thanks! > @Jan, I was sure you would have liked them :-) I'm happy to pick them up and push them to completion. But... when yesterday Al showed his demo patchset I probably interpreted his reply the wrong way and thought that since he spent time for the demo he wanted to put this to completion on his own. Now I see that you are interpreting his message as an invite to use them to shorten the time... Furthermore I'm not sure about how I should credit him. Should I merely add a "Suggested-by:" tag or more consistent "Co-authored-by: Al Viro <...>"? Since he did so much I'd rather the second but I need his permission. @Al, Can I really proceed with *your* work? What should the better suited tag be to credit you for the patches? If you can reply today or at least by Friday, I'll pick your demo patchset, put it to completion, make the patches and test them with (x)fstests on a QEMU/KVM x86_32 bit VM, with 6GB RAM, running an HIGHMEM64GB enabled kernel. Thanks, Fabio
