shm: Respect PROT_MAX when creating private mappings

We were previously unconditionally adding PROT_WRITE to the maxprot of
private mapping (because a private mapping can be written even if the
fd i

shm: Respect PROT_MAX when creating private mappings

We were previously unconditionally adding PROT_WRITE to the maxprot of
private mapping (because a private mapping can be written even if the
fd is read-only), but this might violate the user's PROT_MAX request.

While here, rename cap_maxprot to max_maxprot. This is the intersection
of the maximum protections imposed by capsicum rights on the fd (not
really relevant for private mappings) and the user-required maximum
protections (which were not being obeyed). In particular, cap_maxprot
is a misnomer after the introduction of PROT_MAX.

Add some regression test cases. mmap__maxprot_shm fails without this
patch.

Note: Capsicum's CAP_MMAP_W is a bit ambiguous. Should it be required
in order to create writeable private mappings? Currently it is, even
though such mappings don't permit writes to the object referenced by the
fd.

Reported by: brooks
Reviewed by: brooks
MFC after: 1 month
Fixes: c7841c6b8e41 ("Relax restrictions on private mappings of POSIX shm objects.")
Differential Revision: https://reviews.freebsd.org/D46741

show more ...

Remove $FreeBSD$: two-line .h pattern

Remove /^\s*\*\n \*\s+\$FreeBSD\$$\n/

mmap map_at_zero test: handle W^X

Use kern.elfXX.allow_wx to decide whether to map W+X or W-only memory.

Future work could expand this test to add an "allow_wx" axis to the
test matrix, but I would

mmap map_at_zero test: handle W^X

Use kern.elfXX.allow_wx to decide whether to map W+X or W-only memory.

Future work could expand this test to add an "allow_wx" axis to the
test matrix, but I would argue that a separate test should be written,
since that's orthogonal to map_at_zero.

MFC after: 1 week
Sponsored by: Dell EMC Isilon

show more ...

Fix write only mappings on arm64

When trapping on a wrote access to a buffer the kernel has mapped as write
only we should only pass the VM_PROT_WRITE flag. Previously the call to
vm_fault_trap as t

Fix write only mappings on arm64

When trapping on a wrote access to a buffer the kernel has mapped as write
only we should only pass the VM_PROT_WRITE flag. Previously the call to
vm_fault_trap as the VM_PROT_READ flag was unexpected.

Reported by: manu
Sponsored by: Innovate UK

show more ...

Merge ^/head r313644 through r313895.

Revert r313565 -- :mmap__bad_arguments passes again after r313655

PR: 216976
Sponsored by: Dell EMC Isilon

Merge ^/head r313301 through r313643.

Expect :mmap__bad_arguments to fail

Some recent changes to vm related to mmap(2) have broken the prot checks that
would result with an EINVAL with this case

I suspect r313352 is the root-cause the

Expect :mmap__bad_arguments to fail

Some recent changes to vm related to mmap(2) have broken the prot checks that
would result with an EINVAL with this case

I suspect r313352 is the root-cause the issue

PR: 216976
Sponsored by: Dell EMC Isilon

show more ...

Merge ^/head r311940 through r312200.

Fix -Wsign-compare warnings

The loop index (i) doesn't need to be size_t as its comparison is signed

MFC after: 1 week
Sponsored by: Dell EMC Isilon

Fix leaks and test for getpagesize() returning == -1

- close file descriptors after use.
- Always munmap memory regions after mmap'ing them.
- Make sure getpagesize() returns a value greater than 0

Fix leaks and test for getpagesize() returning == -1

- close file descriptors after use.
- Always munmap memory regions after mmap'ing them.
- Make sure getpagesize() returns a value greater than 0 and use a
cached value instead of always calling getpagesize(3).

CID: 1331374-1331377, 1331653-1331662
Differential Revision: https://reviews.freebsd.org/D6011
MFC after: 2 weeks
Reported by: Coverity
Reviewed by: cem
Sponsored by: EMC / Isilon Storage Division

show more ...

Merge from head

Sponsored by: Gandi.net

Merge from head

Merge from head

Merge r286744-r287584 from head.

Merge ^/head r286858 through r287489.

Add more mmap tests related to character devices.
- Add cdev-related tests for bad args.
- Add two simple tests cases for mapping /dev/zero that test for
MAP_ANON-like behavior.

Reviewed by: alc,

Add more mmap tests related to character devices.
- Add cdev-related tests for bad args.
- Add two simple tests cases for mapping /dev/zero that test for
MAP_ANON-like behavior.

Reviewed by: alc, kib
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D3323

show more ...

Catch up with head, primarily for the 1.14.4.0 firmware.

Merge ^/head r285924 through r286421.

Add various tests to ensure that invalid arguments passed to mmap()
trigger failures.

Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D3269

Convert the map_at_zero test case to ATF. In particular, this will
facilitate adding more mmap() tests.

MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D3268

Catch up with HEAD (r280229-r284686).

Merge sync of head

Merge from head

MFH: r281855-r282312

Sponsored by: The FreeBSD Foundation