export/fuse: Allow other users access to the export
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v2 Latest6 patches 3 repliesIn review
- v14 patches 9 repliesSuperseded
Patches in v2 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/6 | export/fuse: Pass default_permissions for mountblock/ tests/ tests/ | +9 | 0 replies | |
| 2/6 | export/fuse: Add allow-other optionblock/ qapi/ tests/ +1 more | +65 | 1 reply | |
| 3/6 | export/fuse: Give SET_ATTR_SIZE its own branchblock/ | +11 | 0 replies | |
| 4/6 | export/fuse: Let permissions be adjustableblock/ | +62 | 0 replies | |
| 5/6 | iotests/308: Test +w on read-only FUSE exportstests/ tests/ | +15 | 0 replies | |
| 6/6 | iotests/fuse-allow-other: Test allow-othertests/ tests/ | +263 | 1 reply |
Reviews trailers per patch
1 of 6 patches has a Reviewed-by or Acked-by.
| Patch | Kevin Wolf |
|---|---|
| 1/6 | |
| 2/6 | |
| 3/6 | R-b |
| 4/6 | |
| 5/6 | |
| 6/6 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH v2 0/6] export/fuse: Allow other users access to the export
Hi,
The v1 cover letter is here:
https://lists.nongnu.org/archive/html/qemu-block/2021-06/msg00730.html
In v2, I changed the following:
- default_permissions is now passed always. This is the right thing to do regardless of whether allow_other is active or not.
- allow_other is no longer a bool, but an off/on/auto enum. auto is the default, in which case we will try to mount the export with allow_other first, and then fall back to mounting it without.
- Changing the file mode is now possible even without allow_other (because default_permissions is always active now), but only for the user/owner. Giving the group or others any permissions only makes sense with allow_other, the same applies to changing the UID or GID. Giving a read-only export +w makes no sense and hence yields an EROFS error now.
- I decided just testing some default_permission quirks is boring. So the new fuse-allow-other iotest does rely on sudo -n -u nobody working now, and actually tests what allow_other is supposed to do. (Also, it is skipped if allow_other does not work.)
git-backport-diff against v1:
Key:
[----] : patches are identical
[####] : number of functional differences between upstream/downstream patch [down] : patch is downstream-only
The flags [FC] indicate (F)unctional and (C)ontextual differences, respectively
001/6:[down] 'export/fuse: Pass default_permissions for mount' 002/6:[0089] [FC] 'export/fuse: Add allow-other option' 003/6:[----] [--] 'export/fuse: Give SET_ATTR_SIZE its own branch' 004/6:[0039] [FC] 'export/fuse: Let permissions be adjustable' 005/6:[down] 'iotests/308: Test +w on read-only FUSE exports' 006/6:[down] 'iotests/fuse-allow-other: Test allow-other'
Max Reitz (6):
export/fuse: Pass default_permissions for mount
export/fuse: Add allow-other option
export/fuse: Give SET_ATTR_SIZE its own branch
export/fuse: Let permissions be adjustable
iotests/308: Test +w on read-only FUSE exports
iotests/fuse-allow-other: Test allow-other
qapi/block-export.json | 33 +++- block/export/fuse.c | 121 +++++++++--- tests/qemu-iotests/308 | 20 +- tests/qemu-iotests/308.out | 6 +- tests/qemu-iotests/common.rc | 6 +- tests/qemu-iotests/tests/fuse-allow-other | 175 ++++++++++++++++++ tests/qemu-iotests/tests/fuse-allow-other.out | 88 +++++++++ 7 files changed, 421 insertions(+), 28 deletions(-) create mode 100755 tests/qemu-iotests/tests/fuse-allow-other create mode 100644 tests/qemu-iotests/tests/fuse-allow-other.out
Signature · 1 line
2.31.1