Skip to content

fs: Support for LOOKUP_NONBLOCK / RESOLVE_NONBLOCK

Jens Axboe linux-fsdevel 3 versions – In review

Get series with b4
b4 am [email protected]

Versions newest first · linked by author and title

  1. v3 Latest4 patches 41 repliesIn review
  2. v2 RFC5 patches 10 repliesSuperseded
  3. v12 patches 25 repliesSuperseded

Changes in v3 from the cover letter

- Simplify the LOOKUP_NONBLOCK to _just_ cover LOOKUP_RCU and lookup_fast(), as per Linus's suggestion. I verified fast path operations are indeed just that, and it avoids having to mess with the inode lock and mnt_want_write() completely.

- Make O_TMPFILE return -EAGAIN for LOOKUP_NONBLOCK.

- Improve/add a few comments.

- Folded what was left of the open part of LOOKUP_NONBLOCK into the main patch.

Patches in v3 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/4fs: make unlazy_walk() error handling consistentfs/namei.c+17 -260 replies
2/4fs: add support for LOOKUP_NONBLOCKfs/namei.c include/linux/namei.h+27 -116 replies
3/4fs: expose LOOKUP_NONBLOCK through openat2() RESOLVE_NONBLOCKfs/open.c include/linux/fcntl.h include/uapi/linux/openat2.h+11 -15 replies
4/4io_uring: enable LOOKUP_NONBLOCK path resolution for filename lookupsfs/io_uring.c+24 -200 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v3 yet.

Read the v3 thread

Cover letter [PATCHSET v3 0/4] fs: Support for LOOKUP_NONBLOCK / RESOLVE_NONBLOCK

Hi,

Wanted to throw out what the current state of this is, as we keep getting closer to something palatable.

This time I've included the io_uring change too. I've tested this through both openat2, and through io_uring as well.

I'm pretty happy with this at this point. The core change is very simple, and the users end up being trivial too.

Also available here:

https://git.kernel.dk/cgit/linux-block/log/?h=nonblock-path-lookup

Changes since v2:

- Simplify the LOOKUP_NONBLOCK to _just_ cover LOOKUP_RCU and lookup_fast(), as per Linus's suggestion. I verified fast path operations are indeed just that, and it avoids having to mess with the inode lock and mnt_want_write() completely.

- Make O_TMPFILE return -EAGAIN for LOOKUP_NONBLOCK.

- Improve/add a few comments.

- Folded what was left of the open part of LOOKUP_NONBLOCK into the main patch.

Signature · 1 line
Jens Axboe