On Sat, Sep 14, 2024 at 01:18:03PM -0700, Matthew Broomfield wrote: > On Thu, Aug 1, 2024 at 4:04 PM Matthew Broomfield > <mattysweeps@xxxxxxxxxx> wrote: > > > > When checking if SGID is cleared upon chmod / setfacl the write bit is > > flipped just to make sure permission changes are happening. This is > > orthogonal to testing SGID. > > > > Added this information to the test description. > > > > Signed-off-by: Matthew Broomfield <mattysweeps@xxxxxxxxxx> > > --- > > tests/generic/375 | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/tests/generic/375 b/tests/generic/375 > > index eb675144..44670350 100755 > > --- a/tests/generic/375 > > +++ b/tests/generic/375 > > @@ -7,6 +7,9 @@ > > # Check if SGID is cleared upon chmod / setfacl when the owner is not in the > > # owning group. > > # > > +# The test flips the write bit to easily observe that chmod / setfacl are > > +# successful. This is unrelated to testing SGID. > > +# > > . ./common/preamble > > _begin_fstest auto quick acl perms > > > > -- > > 2.46.0.rc2.264.g509ed76dc8-goog > > > > Hello, can someone review if the comment in this patch is accurate? > Thanks in advance for your time. I don't understand this new comment -- "flipping the write bit" is what the chmod does, and that's what clears setgid. --D
