At boot time, KVM stashes the host MDCR_EL2 value, but only does this
when the kernel is not running in hyp mode (i.e. is non-VHE). In these
cases, the stashed value of MDCR_EL2.HPMN happens to be zero, which can
lead to CONSTRAINED UNPREDICTABLE behaviour.
Since we use this value to derive the MDCR_EL2 value when switching
to/from a guest, after a guest have been run, the performance counters
do not behave as expected. This has been observed to result in accesses
via PMXEVTYPER_EL0 and PMXEVCNTR_EL0 not affecting the relevant
counters, resulting in events not being counted. In these cases, only
the fixed-purpose cycle counter appears to work as expected.
Fix this by always stashing the host MDCR_EL2 value, regardless of VHE.
Cc: Christopher Dall <christoffer.dall@arm.com> Cc: James Morse <james.morse@arm.com> Cc: Will Deacon <will.deacon@arm.com> Cc: stable@vger.kernel.org Fixes: 1e947bad0b63b351 ("arm64: KVM: Skip HYP setup when already running in HYP") Tested-by: Robin Murphy <robin.murphy@arm.com> Signed-off-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ensure that the writes into the context image are completed prior to the
register mmio to trigger execution. Although previously we were assured
by the SDM that all writes are flushed before an uncached memory
transaction (our mmio write to submit the context to HW for execution),
we have empirical evidence to believe that this is not actually the
case.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108656
References: https://bugs.freedesktop.org/show_bug.cgi?id=108315
References: https://bugs.freedesktop.org/show_bug.cgi?id=106887 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20181108081740.25615-1-chris@chris-wilson.co.uk Cc: stable@vger.kernel.org
(cherry picked from commit 987abd5c62f92ee4970b45aa077f47949974e615) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unfortunately drm_dp_get_mst_branch_device which is called from both
drm_dp_mst_handle_down_rep and drm_dp_mst_handle_up_rep seem to rely
on that mgr->mst_primary is not NULL, which seem to be wrong as it can be
cleared with simultaneous mode set, if probing fails or in other case.
mgr->lock mutex doesn't protect against that as it might just get
assigned to NULL right before, not simultaneously.
There are currently bugs 107738, 108616 bugs which crash in
drm_dp_get_mst_branch_device, caused by this issue.
v2: Refactored the code, as it was nicely noticed.
Fixed Bugzilla bug numbers(second was 108616, but not 108816)
and added links.
[changed title and added stable cc] Signed-off-by: Lyude Paul <lyude@redhat.com> Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com> Cc: stable@vger.kernel.org
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108616
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107738 Link: https://patchwork.freedesktop.org/patch/msgid/20181109090012.24438-1-stanislav.lisovskiy@intel.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Leaving the DRM driver enabled on reboot or kexec has the annoying
effect of leaving the display generating transactions whilst the
IOMMU has been shut down.
In turn, the IOMMU driver (which shares its interrupt line with
the VOP) starts warning either on shutdown or when entering the
secondary kernel in the kexec case (nothing is expected on that
front).
A cheap way of ensuring that things are nicely shut down is to
register a shutdown callback in the platform driver.
The page migration code employs try_to_unmap() to try and unmap the source
page. This is accomplished by using rmap_walk to find all vmas where the
page is mapped. This search stops when page mapcount is zero. For shared
PMD huge pages, the page map count is always 1 no matter the number of
mappings. Shared mappings are tracked via the reference count of the PMD
page. Therefore, try_to_unmap stops prematurely and does not completely
unmap all mappings of the source page.
This problem can result is data corruption as writes to the original
source page can happen after contents of the page are copied to the target
page. Hence, data is lost.
This problem was originally seen as DB corruption of shared global areas
after a huge page was soft offlined due to ECC memory errors. DB
developers noticed they could reproduce the issue by (hotplug) offlining
memory used to back huge pages. A simple testcase can reproduce the
problem by creating a shared PMD mapping (note that this must be at least
PUD_SIZE in size and PUD_SIZE aligned (1GB on x86)), and using
migrate_pages() to migrate process pages between nodes while continually
writing to the huge pages being migrated.
To fix, have the try_to_unmap_one routine check for huge PMD sharing by
calling huge_pmd_unshare for hugetlbfs huge pages. If it is a shared
mapping it will be 'unshared' which removes the page table entry and drops
the reference on the PMD page. After this, flush caches and TLB.
mmu notifiers are called before locking page tables, but we can not be
sure of PMD sharing until page tables are locked. Therefore, check for
the possibility of PMD sharing before locking so that notifiers can
prepare for the worst possible case.
Link: http://lkml.kernel.org/r/20180823205917.16297-2-mike.kravetz@oracle.com
[mike.kravetz@oracle.com: make _range_in_vma() a static inline] Link: http://lkml.kernel.org/r/6063f215-a5c8-2f0c-465a-2c515ddc952d@oracle.com Fixes: 39dde65c9940 ("shared page table for hugetlb page") Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com> Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: Davidlohr Bueso <dave@stgolabs.net> Cc: Jerome Glisse <jglisse@redhat.com> Cc: Mike Kravetz <mike.kravetz@oracle.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com> Reviewed-by: Jérôme Glisse <jglisse@redhat.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This bug has been experienced several times by the Oracle DB team. The
BUG is in remove_inode_hugepages() as follows:
/*
* If page is mapped, it was faulted in after being
* unmapped in caller. Unmap (again) now after taking
* the fault mutex. The mutex will prevent faults
* until we finish removing the page.
*
* This race can only happen in the hole punch case.
* Getting here in a truncate operation is a bug.
*/
if (unlikely(page_mapped(page))) {
BUG_ON(truncate_op);
In this case, the elevated map count is not the result of a race.
Rather it was incorrectly incremented as the result of a bug in the huge
pmd sharing code. Consider the following:
- Process A maps a hugetlbfs file of sufficient size and alignment
(PUD_SIZE) that a pmd page could be shared.
- Process B maps the same hugetlbfs file with the same size and
alignment such that a pmd page is shared.
- Process B then calls mprotect() to change protections for the mapping
with the shared pmd. As a result, the pmd is 'unshared'.
- Process B then calls mprotect() again to chage protections for the
mapping back to their original value. pmd remains unshared.
- Process B then forks and process C is created. During the fork
process, we do dup_mm -> dup_mmap -> copy_page_range to copy page
tables. Copying page tables for hugetlb mappings is done in the
routine copy_hugetlb_page_range.
In copy_hugetlb_page_range(), the destination pte is obtained by:
dst_pte = huge_pte_alloc(dst, addr, sz);
If pmd sharing is possible, the returned pointer will be to a pte in an
existing page table. In the situation above, process C could share with
either process A or process B. Since process A is first in the list,
the returned pte is a pointer to a pte in process A's page table.
However, the check for pmd sharing in copy_hugetlb_page_range is:
/* If the pagetables are shared don't copy or take references */
if (dst_pte == src_pte)
continue;
Since process C is sharing with process A instead of process B, the
above test fails. The code in copy_hugetlb_page_range which follows
assumes dst_pte points to a huge_pte_none pte. It copies the pte entry
from src_pte to dst_pte and increments this map count of the associated
page. This is how we end up with an elevated map count.
To solve, check the dst_pte entry for huge_pte_none. If !none, this
implies PMD sharing so do not copy.
Link: http://lkml.kernel.org/r/20181105212315.14125-1-mike.kravetz@oracle.com Fixes: c5c99429fa57 ("fix hugepages leak due to pagetable page sharing") Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com> Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Michal Hocko <mhocko@kernel.org> Cc: Hugh Dickins <hughd@google.com> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Davidlohr Bueso <dave@stgolabs.net> Cc: Prakash Sangappa <prakash.sangappa@oracle.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
gcc-8 complains about the prototype for this function:
lib/ubsan.c:432:1: error: ignoring attribute 'noreturn' in declaration of a built-in function '__ubsan_handle_builtin_unreachable' because it conflicts with attribute 'const' [-Werror=attributes]
This is actually a GCC's bug. In GCC internals
__ubsan_handle_builtin_unreachable() declared with both 'noreturn' and
'const' attributes instead of only 'noreturn':
fs/configfs/symlink.c:67:3: warning:
'strncpy' output truncated before terminating nul copying as many
bytes from a string as its length
fs/configfs/symlink.c: In function 'configfs_get_link':
fs/configfs/symlink.c:63:13: note: length computed here
Using strncpy() is indeed less than perfect since the length of data to
be copied has already been determined with strlen(). Replace strncpy()
with memcpy() to address the warning and optimize the code a little.
fuse_request_send_notify_reply() may fail if the connection was reset for
some reason (e.g. fs was unmounted). Don't leak request reference in this
case. Besides leaking memory, this resulted in fc->num_waiting not being
decremented and hence fuse_wait_aborted() left in a hanging and unkillable
state.
Fixes: 2d45ba381a74 ("fuse: add retrieve request") Fixes: b8f95e5d13f5 ("fuse: umount should wait for all requests") Reported-and-tested-by: syzbot+6339eda9cb4ebbc4c37b@syzkaller.appspotmail.com Signed-off-by: Miklos Szeredi <mszeredi@redhat.com> Cc: <stable@vger.kernel.org> #v2.6.36 Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In async IO blocking case the additional reference to the io is taken for
it to survive fuse_aio_complete(). In non blocking case this additional
reference is not needed, however we still reference io to figure out
whether to wait for completion or not. This is wrong and will lead to
use-after-free. Fix it by storing blocking information in separate
variable.
This was spotted by KASAN when running generic/208 fstest.
Fix an issue with the 32-bit range error path in `rtc_hctosys' where no
error code is set and consequently the successful preceding call result
from `rtc_read_time' is propagated to `rtc_hctosys_ret'. This in turn
makes any subsequent call to `hctosys_show' incorrectly report in sysfs
that the system time has been set from this RTC while it has not.
Set the error to ERANGE then if we can't express the result due to an
overflow.
Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org> Fixes: b3a5ac42ab18 ("rtc: hctosys: Ensure system time doesn't overflow time_t") Cc: stable@vger.kernel.org # 4.17+ Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Make sure we have a saved filehandle, otherwise we'll oops with a null
pointer dereference in nfs4_preprocess_stateid_op().
Signed-off-by: Scott Mayhew <smayhew@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When truncating the encode buffer, the page_ptr is getting
advanced, causing the next page to be skipped while encoding.
The page is still included in the response, so the response
contains a page of bogus data.
We need to adjust the page_ptr backwards to ensure we encode
the next page into the correct place.
We saw this triggered when concurrent directory modifications caused
nfsd4_encode_direct_fattr() to return nfserr_noent, and the resulting
call to xdr_truncate_encode() corrupted the READDIR reply.
Signed-off-by: Frank Sorenson <sorenson@redhat.com> Cc: stable@vger.kernel.org Signed-off-by: J. Bruce Fields <bfields@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Timothy Baldwin <timbaldwin@fastmail.co.uk> wrote:
> As per mount_namespaces(7) unprivileged users should not be able to look under mount points:
>
> Mounts that come as a single unit from more privileged mount are locked
> together and may not be separated in a less privileged mount namespace.
>
> However they can:
>
> 1. Create a mount namespace.
> 2. In the mount namespace open a file descriptor to the parent of a mount point.
> 3. Destroy the mount namespace.
> 4. Use the file descriptor to look under the mount point.
>
> I have reproduced this with Linux 4.16.18 and Linux 4.18-rc8.
>
> The setup:
>
> $ sudo sysctl kernel.unprivileged_userns_clone=1
> kernel.unprivileged_userns_clone = 1
> $ mkdir -p A/B/Secret
> $ sudo mount -t tmpfs hide A/B
>
>
> "Secret" is indeed hidden as expected:
>
> $ ls -lR A
> A:
> total 0
> drwxrwxrwt 2 root root 40 Feb 12 21:08 B
>
> A/B:
> total 0
>
>
> The attack revealing "Secret":
>
> $ unshare -Umr sh -c "exec unshare -m ls -lR /proc/self/fd/4/ 4<A"
> /proc/self/fd/4/:
> total 0
> drwxr-xr-x 3 root root 60 Feb 12 21:08 B
>
> /proc/self/fd/4/B:
> total 0
> drwxr-xr-x 2 root root 40 Feb 12 21:08 Secret
>
> /proc/self/fd/4/B/Secret:
> total 0
I tracked this down to put_mnt_ns running passing UMOUNT_SYNC and
disconnecting all of the mounts in a mount namespace. Fix this by
factoring drop_mounts out of drop_collected_mounts and passing
0 instead of UMOUNT_SYNC.
There are two possible behavior differences that result from this.
- No longer setting UMOUNT_SYNC will no longer set MNT_SYNC_UMOUNT on
the vfsmounts being unmounted. This effects the lazy rcu walk by
kicking the walk out of rcu mode and forcing it to be a non-lazy
walk.
- No longer disconnecting locked mounts will keep some mounts around
longer as they stay because the are locked to other mounts.
There are only two users of drop_collected mounts: audit_tree.c and
put_mnt_ns.
In audit_tree.c the mounts are private and there are no rcu lazy walks
only calls to iterate_mounts. So the changes should have no effect
except for a small timing effect as the connected mounts are disconnected.
In put_mnt_ns there may be references from process outside the mount
namespace to the mounts. So the mounts remaining connected will
be the bug fix that is needed. That rcu walks are allowed to continue
appears not to be a problem especially as the rcu walk change was about
an implementation detail not about semantics.
Cc: stable@vger.kernel.org Fixes: 5ff9d8a65ce8 ("vfs: Lock in place mounts from more privileged users") Reported-by: Timothy Baldwin <timbaldwin@fastmail.co.uk> Tested-by: Timothy Baldwin <timbaldwin@fastmail.co.uk> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Jonathan Calmels from NVIDIA reported that he's able to bypass the
mount visibility security check in place in the Linux kernel by using
a combination of the unbindable property along with the private mount
propagation option to allow a unprivileged user to see a path which
was purposefully hidden by the root user.
Reproducer:
# Hide a path to all users using a tmpfs
root@castiana:~# mount -t tmpfs tmpfs /sys/devices/
root@castiana:~#
# As an unprivileged user, unshare user namespace and mount namespace
stgraber@castiana:~$ unshare -U -m -r
# Confirm the path is still not accessible
root@castiana:~# ls /sys/devices/
# Make /sys recursively unbindable and private
root@castiana:~# mount --make-runbindable /sys
root@castiana:~# mount --make-private /sys
# Recursively bind-mount the rest of /sys over to /mnnt
root@castiana:~# mount --rbind /sys/ /mnt
# Access our hidden /sys/device as an unprivileged user
root@castiana:~# ls /mnt/devices/
breakpoint cpu cstate_core cstate_pkg i915 intel_pt isa kprobe
LNXSYSTM:00 msr pci0000:00 platform pnp0 power software system
tracepoint uncore_arb uncore_cbox_0 uncore_cbox_1 uprobe virtual
Solve this by teaching copy_tree to fail if a mount turns out to be
both unbindable and locked.
Cc: stable@vger.kernel.org Fixes: 5ff9d8a65ce8 ("vfs: Lock in place mounts from more privileged users") Reported-by: Jonathan Calmels <jcalmels@nvidia.com> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It was recently pointed out that the one instance of testing MNT_LOCKED
outside of the namespace_sem is in ksys_umount.
Fix that by adding a test inside of do_umount with namespace_sem and
the mount_lock held. As it helps to fail fails the existing test is
maintained with an additional comment pointing out that it may be racy
because the locks are not held.
Cc: stable@vger.kernel.org Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Fixes: 5ff9d8a65ce8 ("vfs: Lock in place mounts from more privileged users") Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently bh is set to NULL only during first iteration of for cycle,
then this pointer is not cleared after end of using.
Therefore rollback after errors can lead to extra brelse(bh) call,
decrements bh counter and later trigger an unexpected warning in __brelse()
Patch moves brelse() calls in body of cycle to exclude requirement of
brelse() call in rollback.
Fixes: 33afdcc5402d ("ext4: add a function which sets up group blocks ...") Signed-off-by: Vasily Averin <vvs@virtuozzo.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Cc: stable@kernel.org # 3.3+ Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add support for platforms where pit_shutdown() doesn't work because of a
quirk in the PIT emulation. On these platforms setting the counter register
to zero causes the PIT to start running again, negating the shutdown.
Provide a global variable that controls whether the counter register is
zero'ed, which platform specific code can override.
We currently allow cloning a range from a file which includes the last
block of the file even if the file's size is not aligned to the block
size. This is fine and useful when the destination file has the same size,
but when it does not and the range ends somewhere in the middle of the
destination file, it leads to corruption because the bytes between the EOF
and the end of the block have undefined data (when there is support for
discard/trimming they have a value of 0x00).
The bytes in the range from 786532 (512Kb + 256Kb + 100 bytes) to 790527
(512Kb + 256Kb + 4Kb - 1) got corrupted, having now a value of 0x00 instead
of 0xb5.
This is similar to the problem we had for deduplication that got recently
fixed by commit de02b9f6bb65 ("Btrfs: fix data corruption when
deduplicating between different files").
Fix this by not allowing such operations to be performed and return the
errno -EINVAL to user space. This is what XFS is doing as well at the VFS
level. This change however now makes us return -EINVAL instead of
-EOPNOTSUPP for cases where the source range maps to an inline extent and
the destination range's end is smaller then the destination file's size,
since the detection of inline extents is done during the actual process of
dropping file extent items (at __btrfs_drop_extents()). Returning the
-EINVAL error is done early on and solely based on the input parameters
(offsets and length) and destination file's size. This makes us consistent
with XFS and anyone else supporting cloning since this case is now checked
at a higher level in the VFS and is where the -EINVAL will be returned
from starting with kernel 4.20 (the VFS changed was introduced in 4.20-rc1
by commit 07d19dc9fbe9 ("vfs: avoid problematic remapping requests into
partial EOF block"). So this change is more geared towards stable kernels,
as it's unlikely the new VFS checks get removed intentionally.
A test case for fstests follows soon, as well as an update to filter
existing tests that expect -EOPNOTSUPP to accept -EINVAL as well.
CC: <stable@vger.kernel.org> # 4.4+ Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the cow_file_range fails, the related resources are unlocked
according to the range [start..end), so the unlock cannot be repeated in
run_delalloc_nocow.
In some cases (e.g. cur_offset <= end && cow_start != -1), cur_offset is
not updated correctly, so move the cur_offset update before
cow_file_range.
Alpha has had c_ispeed and c_ospeed, but still set speeds in c_cflags
using arbitrary flags. Because BOTHER is not defined, the general
Linux code doesn't allow setting arbitrary baud rates, and because
CBAUDEX == 0, we can have an array overrun of the baud_rate[] table in
drivers/tty/tty_baudrate.c if (c_cflags & CBAUD) == 037.
Resolve both problems by #defining BOTHER to 037 on Alpha.
However, userspace still needs to know if setting BOTHER is actually
safe given legacy kernels (does anyone actually care about that on
Alpha anymore?), so enable the TCGETS2/TCSETS*2 ioctls on Alpha, even
though they use the same structure. Define struct termios2 just for
compatibility; it is the exact same structure as struct termios. In a
future patchset, this will be cleaned up so the uapi headers are
usable from libc.
Signed-off-by: H. Peter Anvin (Intel) <hpa@zytor.com> Cc: Jiri Slaby <jslaby@suse.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Kate Stewart <kstewart@linuxfoundation.org> Cc: Philippe Ombredanne <pombredanne@nexb.com> Cc: Eugene Syromiatnikov <esyr@redhat.com> Cc: <linux-alpha@vger.kernel.org> Cc: <linux-serial@vger.kernel.org> Cc: Johan Hovold <johan@kernel.org> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On architectures with CBAUDEX == 0 (Alpha and PowerPC), the code in tty_baudrate.c does
not do any limit checking on the tty_baudrate[] array, and in fact a
buffer overrun is possible on both architectures. Add a limit check to
prevent that situation.
This will be followed by a much bigger cleanup/simplification patch.
Signed-off-by: H. Peter Anvin (Intel) <hpa@zytor.com> Requested-by: Cc: Johan Hovold <johan@kernel.org> Cc: Jiri Slaby <jslaby@suse.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Kate Stewart <kstewart@linuxfoundation.org> Cc: Philippe Ombredanne <pombredanne@nexb.com> Cc: Eugene Syromiatnikov <esyr@redhat.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: stable <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently the NUMA distance map parsing does not validate the distance
table for the distance-matrix rules 1-2 in [1].
However the arch NUMA code may enforce some of these rules, but not all.
Such is the case for the arm64 port, which does not enforce the rule that
the distance between separates nodes cannot equal LOCAL_DISTANCE.
The patch adds the following rules validation:
- distance of node to self equals LOCAL_DISTANCE
- distance of separate nodes > LOCAL_DISTANCE
This change avoids a yet-unresolved crash reported in [2].
A note on dealing with symmetrical distances between nodes:
Validating symmetrical distances between nodes is difficult. If it were
mandated in the bindings that every distance must be recorded in the
table, then it would be easy. However, it isn't.
In addition to this, it is also possible to record [b, a] distance only
(and not [a, b]). So, when processing the table for [b, a], we cannot
assert that current distance of [a, b] != [b, a] as invalid, as [a, b]
distance may not be present in the table and current distance would be
default at REMOTE_DISTANCE.
As such, we maintain the policy that we overwrite distance [a, b] = [b, a]
for b > a. This policy is different to kernel ACPI SLIT validation, which
allows non-symmetrical distances (ACPI spec SLIT rules allow it). However,
the distance debug message is dropped as it may be misleading (for a distance
which is later overwritten).
Some final notes on semantics:
- It is implied that it is the responsibility of the arch NUMA code to
reset the NUMA distance map for an error in distance map parsing.
- It is the responsibility of the FW NUMA topology parsing (whether OF or
ACPI) to enforce NUMA distance rules, and not arch NUMA code.
I noticed during the creation of another bugfix that the BCH_CONST_PARAMS
option that is set by DOCG3 breaks setting variable parameters for any
other users of the BCH library code.
The only other user we have today is the MTD_NAND software BCH
implementation (most flash controllers use hardware BCH these days
and are not affected). I considered removing BCH_CONST_PARAMS entirely
because of the inherent conflict, but according to the description in
lib/bch.c there is a significant performance benefit in keeping it.
To avoid the immediate problem of the conflict between MTD_NAND_BCH
and DOCG3, this only sets the constant parameters if MTD_NAND_BCH
is disabled, which should fix the problem for all cases that
are affected. This should also work for all stable kernels.
Note that there is only one machine that actually seems to use the
DOCG3 driver (arch/arm/mach-pxa/mioa701.c), so most users should have
the driver disabled, but it almost certainly shows up if we wanted
to test random kernels on machines that use software BCH in MTD.
If there's no entry to drop in bucket that corresponds to the hash,
early_drop() should look for it in other buckets. But since it increments
hash instead of bucket number, it actually looks in the same bucket 8
times: hsize is 16k by default (14 bits) and hash is 32-bit value, so
reciprocal_scale(hash, hsize) returns the same value for hash..hash+7 in
most cases.
Fix it by increasing bucket number instead of hash and rename _hash
to bucket to avoid future confusion.
THP allocation might be really disruptive when allocated on NUMA system
with the local node full or hard to reclaim. Stefan has posted an
allocation stall report on 4.12 based SLES kernel which suggests the
same issue:
The defrag mode is "madvise" and from the above report it is clear that
the THP has been allocated for MADV_HUGEPAGA vma.
Andrea has identified that the main source of the problem is
__GFP_THISNODE usage:
: The problem is that direct compaction combined with the NUMA
: __GFP_THISNODE logic in mempolicy.c is telling reclaim to swap very
: hard the local node, instead of failing the allocation if there's no
: THP available in the local node.
:
: Such logic was ok until __GFP_THISNODE was added to the THP allocation
: path even with MPOL_DEFAULT.
:
: The idea behind the __GFP_THISNODE addition, is that it is better to
: provide local memory in PAGE_SIZE units than to use remote NUMA THP
: backed memory. That largely depends on the remote latency though, on
: threadrippers for example the overhead is relatively low in my
: experience.
:
: The combination of __GFP_THISNODE and __GFP_DIRECT_RECLAIM results in
: extremely slow qemu startup with vfio, if the VM is larger than the
: size of one host NUMA node. This is because it will try very hard to
: unsuccessfully swapout get_user_pages pinned pages as result of the
: __GFP_THISNODE being set, instead of falling back to PAGE_SIZE
: allocations and instead of trying to allocate THP on other nodes (it
: would be even worse without vfio type1 GUP pins of course, except it'd
: be swapping heavily instead).
Fix this by removing __GFP_THISNODE for THP requests which are
requesting the direct reclaim. This effectivelly reverts 5265047ac301
on the grounds that the zone/node reclaim was known to be disruptive due
to premature reclaim when there was memory free. While it made sense at
the time for HPC workloads without NUMA awareness on rare machines, it
was ultimately harmful in the majority of cases. The existing behaviour
is similar, if not as widespare as it applies to a corner case but
crucially, it cannot be tuned around like zone_reclaim_mode can. The
default behaviour should always be to cause the least harm for the
common case.
If there are specialised use cases out there that want zone_reclaim_mode
in specific cases, then it can be built on top. Longterm we should
consider a memory policy which allows for the node reclaim like behavior
for the specific memory ranges which would allow a
: Both patches look correct to me but I'm responding to this one because
: it's the fix. The change makes sense and moves further away from the
: severe stalling behaviour we used to see with both THP and zone reclaim
: mode.
:
: I put together a basic experiment with usemem configured to reference a
: buffer multiple times that is 80% the size of main memory on a 2-socket
: box with symmetric node sizes and defrag set to "always". The defrag
: setting is not the default but it would be functionally similar to
: accessing a buffer with madvise(MADV_HUGEPAGE). Usemem is configured to
: reference the buffer multiple times and while it's not an interesting
: workload, it would be expected to complete reasonably quickly as it fits
: within memory. The results were;
:
: usemem
: vanilla noreclaim-v1
: Amean Elapsd-1 42.78 ( 0.00%) 26.87 ( 37.18%)
: Amean Elapsd-3 27.55 ( 0.00%) 7.44 ( 73.00%)
: Amean Elapsd-4 5.72 ( 0.00%) 5.69 ( 0.45%)
:
: This shows the elapsed time in seconds for 1 thread, 3 threads and 4
: threads referencing buffers 80% the size of memory. With the patches
: applied, it's 37.18% faster for the single thread and 73% faster with two
: threads. Note that 4 threads showing little difference does not indicate
: the problem is related to thread counts. It's simply the case that 4
: threads gets spread so their workload mostly fits in one node.
:
: The overall view from /proc/vmstats is more startling
:
: 4.19.0-rc1 4.19.0-rc1
: vanillanoreclaim-v1r1
: Minor Faults 35593425 708164
: Major Faults 484088 36
: Swap Ins 3772837 0
: Swap Outs 3932295 0
:
: Massive amounts of swap in/out without the patch
:
: Direct pages scanned 6013214 0
: Kswapd pages scanned 0 0
: Kswapd pages reclaimed 0 0
: Direct pages reclaimed 4033009 0
:
: Lots of reclaim activity without the patch
:
: Kswapd efficiency 100% 100%
: Kswapd velocity 0.000 0.000
: Direct efficiency 67% 100%
: Direct velocity 11191.956 0.000
:
: Mostly from direct reclaim context as you'd expect without the patch.
:
: Page writes by reclaim 3932314.000 0.000
: Page writes file 19 0
: Page writes anon 3932295 0
: Page reclaim immediate 42336 0
:
: Writes from reclaim context is never good but the patch eliminates it.
:
: We should never have default behaviour to thrash the system for such a
: basic workload. If zone reclaim mode behaviour is ever desired but on a
: single task instead of a global basis then the sensible option is to build
: a mempolicy that enforces that behaviour.
This was a severe regression compared to previous kernels that made
important workloads unusable and it starts when __GFP_THISNODE was
added to THP allocations under MADV_HUGEPAGE. It is not a significant
risk to go to the previous behavior before __GFP_THISNODE was added, it
worked like that for years.
This was simply an optimization to some lucky workloads that can fit in
a single node, but it ended up breaking the VM for others that can't
possibly fit in a single node, so going back is safe.
[mhocko@suse.com: rewrote the changelog based on the one from Andrea] Link: http://lkml.kernel.org/r/20180925120326.24392-2-mhocko@kernel.org Fixes: 5265047ac301 ("mm, thp: really limit transparent hugepage allocation to local node") Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: Stefan Priebe <s.priebe@profihost.ag> Debugged-by: Andrea Arcangeli <aarcange@redhat.com> Reported-by: Alex Williamson <alex.williamson@redhat.com> Reviewed-by: Mel Gorman <mgorman@techsingularity.net> Tested-by: Mel Gorman <mgorman@techsingularity.net> Cc: Zi Yan <zi.yan@cs.rutgers.edu> Cc: Vlastimil Babka <vbabka@suse.cz> Cc: David Rientjes <rientjes@google.com> Cc: "Kirill A. Shutemov" <kirill@shutemov.name> Cc: <stable@vger.kernel.org> [4.1+] Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Somehow, file system metadata was corrupted, which causes
ocfs2_check_dir_entry() to fail in function ocfs2_dir_foreach_blk_el().
According to the original design intention, if above happens we should
skip the problematic block and continue to retrieve dir entry. But
there is obviouse misuse of brelse around related code.
After failure of ocfs2_check_dir_entry(), current code just moves to
next position and uses the problematic buffer head again and again
during which the problematic buffer head is released for multiple times.
I suppose, this a serious issue which is long-lived in ocfs2. This may
cause other file systems which is also used in a the same host insane.
So we should also consider about bakcporting this patch into linux
-stable.
Link: http://lkml.kernel.org/r/HK2PR06MB045211675B43EED794E597B6D56E0@HK2PR06MB0452.apcprd06.prod.outlook.com Signed-off-by: Changwei Ge <ge.changwei@h3c.com> Suggested-by: Changkuo Shi <shi.changkuo@h3c.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Mark Fasheh <mark@fasheh.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Junxiao Bi <junxiao.bi@oracle.com> Cc: Joseph Qi <jiangqi903@gmail.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commands with protection information included were not truncating the
protection iov_iter to the number of protection bytes in the command.
This resulted in vhost_scsi mis-calculating the size of the protection
SGL in vhost_scsi_calc_sgls(), and including both the protection and
data SG entries in the protection SGL.
Fixes: 09b13fa8c1a1 ("vhost/scsi: Add ANY_LAYOUT support in vhost_scsi_handle_vq") Signed-off-by: Greg Edwards <gedwards@ddn.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Fixes: 09b13fa8c1a1093e9458549ac8bb203a7c65c62a Cc: stable@vger.kernel.org Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is a potential execution path in which function
platform_get_resource() returns NULL. If this happens,
we will end up having a NULL pointer dereference.
Fix this by replacing devm_ioremap with devm_ioremap_resource,
which has the NULL check and the memory region request.
This code was detected with the help of Coccinelle.
Cc: stable@vger.kernel.org Fixes: 97b7129cd2af ("reset: hisilicon: change the definition of hisi_reset_init") Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reading the registers without waiting for engine idle returns
unpredictable values. These unpredictable values result in display
corruption - if atyfb_imageblit reads the content of DP_PIX_WIDTH with the
bit DP_HOST_TRIPLE_EN set (from previous invocation), the driver would
never ever clear the bit, resulting in display corruption.
We don't want to wait for idle because it would degrade performance, so
this patch modifies the driver so that it never reads accelerator
registers.
HOST_CNTL doesn't have to be read, we can just write it with
HOST_BYTE_ALIGN because no other part of the driver cares if
HOST_BYTE_ALIGN is set.
DP_PIX_WIDTH is written in the functions atyfb_copyarea and atyfb_fillrect
with the default value and in atyfb_imageblit with the value set according
to the source image data.
The code for manual bit triple is not endian-clean. It builds the variable
"hostdword" using byte accesses, therefore we must read the variable with
"le32_to_cpu".
The patch also enables (hardware or software) bit triple only if the image
is monochrome (image->depth). If we want to blit full-color image, we
shouldn't use the triple code.
splice_dentry() is used by three places. For two places, req->r_dentry
is passed to splice_dentry(). In the case of error, req->r_dentry does
not get updated. So splice_dentry() should not drop reference.
Fixes the signedness bug returning '(-22)' on the return type by removing the
sanity checker in rockchip_ddrclk_get_parent(). The function should return
and unsigned value only and it's safe to remove the sanity checker as the
core functions that call get_parent like clk_core_get_parent_by_index already
ensures the validity of the clk index returned (index >= core->num_parents).
Fixes: a4f182bf81f18 ("clk: rockchip: add new clock-type for the ddrclk") Cc: stable@vger.kernel.org Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com> Reviewed-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Heiko Stuebner <heiko@sntech.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit a982e45dc150 ("clk: at91: PLL recalc_rate() now using cached MUL
and DIV values") removed a check that prevents a division by zero. This
now causes a stacktrace when booting the kernel on a at91 platform if
the PLL DIV register contains zero. This commit reintroduces this check.
Fixes: a982e45dc150 ("clk: at91: PLL recalc_rate() now using cached...") Cc: <stable@vger.kernel.org> Signed-off-by: Ronald Wahl <rwahl@gmx.de> Acked-by: Ludovic Desroches <ludovic.desroches@microchip.com> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When driver is built as module and DT node contains clocks compatible
(e.g. "samsung,s2mps11-clk"), the module will not be autoloaded because
module aliases won't match.
The modalias from uevent: of:NclocksT<NULL>Csamsung,s2mps11-clk
The modalias from driver: platform:s2mps11-clk
The devices are instantiated by parent's MFD. However both Device Tree
bindings and parent define the compatible for clocks devices. In case
of module matching this DT compatible will be used.
The issue will not happen if this is a built-in (no need for module
matching) or when clocks DT node does not contain compatible (not
correct from bindings perspective but working for driver).
Note when backporting to stable kernels: adjust the list of device ID
entries.
Cc: <stable@vger.kernel.org> Fixes: 53c31b3437a6 ("mfd: sec-core: Add of_compatible strings for clock MFD cells") Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org> Acked-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The bootloader may pass physical address of the boot parameters structure
to the MMUv3 kernel in the register a2. Code in the _SetupMMU block in
the arch/xtensa/kernel/head.S is supposed to map that physical address to
the virtual address in the configured virtual memory layout.
This code haven't been updated when additional 256+256 and 512+512
memory layouts were introduced and it may produce wrong addresses when
used with these layouts.
Cc: stable@vger.kernel.org Signed-off-by: Max Filippov <jcmvbkbc@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Xtensa ABI requires stack alignment to be at least 16. In noMMU
configuration ARCH_SLAB_MINALIGN is used to align stack. Make it at
least 16.
This fixes the following runtime error in noMMU configuration, caused by
interaction between insufficiently aligned stack and alloca function,
that results in corruption of on-stack variable in the libc function
glob:
Caught unhandled exception in 'sh' (pid = 47, pc = 0x02d05d65)
- should not happen
EXCCAUSE is 15
Cc: stable@vger.kernel.org Signed-off-by: Max Filippov <jcmvbkbc@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This section collects all source .note.* sections together in the
vmlinux image. Without it .note.Linux section may be placed at address
0, while the rest of the kernel is at its normal address, resulting in a
huge vmlinux.bin image that may not be linked into the xtensa Image.elf.
Cc: stable@vger.kernel.org Signed-off-by: Max Filippov <jcmvbkbc@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After commit e509bd7da149dc349160 ("genirq: Allow migration of chained
interrupts by installing default action") Loongson-3 fails at here:
setup_irq(LOONGSON_HT1_IRQ, &cascade_irqaction);
This is because both chained_action and cascade_irqaction don't have
IRQF_SHARED flag. This will cause Loongson-3 resume fails because HPET
timer interrupt can't be delivered during S3. So we set the irqchip of
the chained irq to loongson_irq_chip which doesn't disable the chained
irq in CP0.Status.
Masking/unmasking the CPU UART irq in CP0_Status (and redirecting it to
other CPUs) may cause interrupts be lost, especially in multi-package
machines (Package-0's UART irq cannot be delivered to others). So make
mask_loongson_irq() and unmask_loongson_irq() be no-ops.
The original problem (UART IRQ may deliver to any core) is also because
of masking/unmasking the CPU UART irq in CP0_Status. So it is safe to
remove all of the stuff.
In the C-code we need to put the physical address of the hpmc handler in
the interrupt vector table (IVA) in order to get HPMCs working. Since
on parisc64 function pointers are indirect (in fact they are function
descriptors) we instead export the address as variable and not as
function.
This reverts a small part of commit f39cce654f9a ("parisc: Add
cfi_startproc and cfi_endproc to assembly code").
Make sure that the HPMC (High Priority Machine Check) handler is 16-byte
aligned and that it's length in the IVT is a multiple of 16 bytes.
Otherwise PDC may decide not to call the HPMC crash handler.
The os_hpmc_size variable sometimes wasn't aligned at word boundary and thus
triggered the unaligned fault handler at startup.
Fix it by aligning it properly.
Using memcpy() from a string that is shorter than the length copied means
the destination buffer is being filled with arbitrary data from the kernel
rodata segment. Instead, use strncpy() which will fill the trailing bytes
with zeros.
This was found with the future CONFIG_FORTIFY_SOURCE feature.
Cc: Daniel Micay <danielmicay@gmail.com> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
This patch fixes a race condition that can result into the interface being
up and carrier on, but with transmits disabled in the hardware.
The bug may show up by repeatedly IFF_DOWN+IFF_UP the interface, which
allows e1000_watchdog() interleave with e1000_down().
CPU x CPU y
--------------------------------------------------------------------
e1000_down():
netif_carrier_off()
e1000_watchdog():
if (carrier == off) {
netif_carrier_on();
enable_hw_transmit();
}
disable_hw_transmit();
e1000_watchdog():
/* carrier on, do nothing */
Signed-off-by: Vincenzo Maffione <v.maffione@gmail.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
Currently if the stat type is invalid then data[i] is being set
either by dereferencing a null pointer p, or it is reading from
an incorrect previous location if we had a valid stat type
previously. Fix this by skipping over the read of p on an invalid
stat type.
Detected by CoverityScan, CID#113385 ("Explicit null dereferenced")
Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: Alexander Duyck <alexander.h.duyck@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
The reason is that the length of the new brk is not page aligned when we
try to populate the it. There is no reason to bug on that though.
do_brk_flags already aligns the length properly so the mapping is
expanded as it should. All we need is to tell mm_populate about it.
Besides that there is absolutely no reason to to bug_on in the first
place. The worst thing that could happen is that the last page wouldn't
get populated and that is far from putting system into an inconsistent
state.
Fix the issue by moving the length sanitization code from do_brk_flags
up to vm_brk_flags. The only other caller of do_brk_flags is brk
syscall entry and it makes sure to provide the proper length so t here
is no need for sanitation and so we can use do_brk_flags without it.
Also remove the bogus BUG_ONs.
[osalvador@techadventures.net: fix up vm_brk_flags s@request@len@] Link: http://lkml.kernel.org/r/20180706090217.GI32658@dhcp22.suse.cz Signed-off-by: Michal Hocko <mhocko@suse.com> Reported-by: syzbot <syzbot+5dcb560fe12aa5091c06@syzkaller.appspotmail.com> Tested-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Reviewed-by: Oscar Salvador <osalvador@suse.de> Cc: Zi Yan <zi.yan@cs.rutgers.edu> Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: "Huang, Ying" <ying.huang@intel.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[bwh: Backported to 4.9:
- There is no do_brk_flags() function; update do_brk()
- Adjust context] Signed-off-by: Ben Hutchings <ben.hutchings@codethink.co.uk> Signed-off-by: Sasha Levin <sashal@kernel.org>
Otherwise fuse_dev_do_write() could come in and finish off the request, and
the set_bit(FR_SENT, ...) could trigger the WARN_ON(test_bit(FR_SENT, ...))
in request_end().
Using waitqueue_active() is racy. Make sure we issue a wake_up()
unconditionally after storing into fc->blocked. After that it's okay to
optimize with waitqueue_active() since the first wake up provides the
necessary barrier for all waiters, not the just the woken one.
There is another cast from unsigned long to int which causes
a bounds check to fail with specially crafted input. The value is
then used as an index in the slot array in cdrom_slot_status().
This issue is similar to CVE-2018-16658 and CVE-2018-10940.
p9stat_free is more of a cleanup function than a 'free' function as it
only frees the content of the struct; there are chances of use-after-free
if it is improperly used (e.g. p9stat_free called twice as it used to be
possible to)
Clearing dangling pointers makes the function idempotent and safer to use.
the 9p client code overwrites our glock.client_id pointing to a static
buffer by an allocated string holding the network provided value which
we do not care about; free and reset the value as appropriate.
This is almost identical to the leak in v9fs_file_getlock() fixed by
Al Viro in commit ce85dd58ad5a6 ("9p: we are leaking glock.client_id
in v9fs_file_getlock()"), which was returned as an error by a coverity
false positive -- while we are here attempt to make the code slightly
more robust to future change of the net/9p/client code and hopefully
more clear to coverity that there is no problem.
The driver ignored the width alignment which exists due to the UYVY
colorspace format. Fix the width alignment and make use of the the
provided v4l2 helper function to set the width, height and all
alignments in one.
The SC16IS752 is a dual-channel device. The two channels are largely
independent, but the IRQ signals are wired together as an open-drain,
active low signal which will be driven low while either of the
channels requires attention, which can be for significant periods of
time until operations complete and the interrupt can be acknowledged.
In that respect it is should be treated as a true level-sensitive IRQ.
The kernel, however, needs to be able to exit interrupt context in
order to use I2C or SPI to access the device registers (which may
involve sleeping). Therefore the interrupt needs to be masked out or
paused in some way.
The usual way to manage sleeping from within an interrupt handler
is to use a threaded interrupt handler - a regular interrupt routine
does the minimum amount of work needed to triage the interrupt before
waking the interrupt service thread. If the threaded IRQ is marked as
IRQF_ONESHOT the kernel will automatically mask out the interrupt
until the thread runs to completion. The sc16is7xx driver used to
use a threaded IRQ, but a patch switched to using a kthread_worker
in order to set realtime priorities on the handler thread and for
other optimisations. The end result is non-threaded IRQ that
schedules some work then returns IRQ_HANDLED, making the kernel
think that all IRQ processing has completed.
The work-around to prevent a constant stream of interrupts is to
mark the interrupt as edge-sensitive rather than level-sensitive,
but interpreting an active-low source as a falling-edge source
requires care to prevent a total cessation of interrupts. Whereas
an edge-triggering source will generate a new edge for every interrupt
condition a level-triggering source will keep the signal at the
interrupting level until it no longer requires attention; in other
words, the host won't see another edge until all interrupt conditions
are cleared. It is therefore vital that the interrupt handler does not
exit with an outstanding interrupt condition, otherwise the kernel
will not receive another interrupt unless some other operation causes
the interrupt state on the device to be cleared.
The existing sc16is7xx driver has a very simple interrupt "thread"
(kthread_work job) that processes interrupts on each channel in turn
until there are no more. If both channels are active and the first
channel starts interrupting while the handler for the second channel
is running then it will not be detected and an IRQ stall ensues. This
could be handled easily if there was a shared IRQ status register, or
a convenient way to determine if the IRQ had been deasserted for any
length of time, but both appear to be lacking.
Avoid this problem (or at least make it much less likely to happen)
by reducing the granularity of per-channel interrupt processing
to one condition per iteration, only exiting the overall loop when
both channels are no longer interrupting.
Call pcie_bus_configure_settings() on MIPS, like for other platforms.
The function pcie_bus_configure_settings() makes sure the MPS (Max
Payload Size) across the bus is uniform and provides the ability to
tune the MRSS (Max Read Request Size) and MPS (Max Payload Size) to
higher performance values. Some devices will not operate properly if
these aren't set correctly because the firmware doesn't always do it.
ld: arch/powerpc/boot/wrapper.a(crt0.o): in function '_zimage_start':
(.text+0x58): multiple definition of '_zimage_start';
arch/powerpc/boot/pseries-head.o:(.text+0x0): first defined here
Clang requires the .weak directive to appear after the symbol is
declared. The binutils manual says:
This directive sets the weak attribute on the comma separated list of
symbol names. If the symbols do not already exist, they will be
created.
So it appears this is different with clang. The only reference I could
see for this was an OpenBSD mailing list post[1].
Changing it to be after the declaration fixes building with Clang, and
still works with GCC.
Signed-off-by: Joel Stanley <joel@jms.id.au> Reviewed-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
After changing CPU online status, it will not be sent any IPIs such as in
__flush_cache_all() on software coherency systems. Do this before disabling
local IRQ.
altera_hw_filt_init() which calls append_internal() assumes
that the node was successfully linked in while in fact it can
silently fail. So the call-site needs to set return to -ENOMEM
on append_internal() returning NULL and exit through the err path.
Fixes: 349bcf02e361 ("[media] Altera FPGA based CI driver module") Signed-off-by: Nicholas Mc Guire <hofrat@osadl.org> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A DMM timeout "timed out waiting for done" has been observed on DRA7
devices. The timeout happens rarely, and only when the system is under
heavy load.
Debugging showed that the timeout can be made to happen much more
frequently by optimizing the DMM driver, so that there's almost no code
between writing the last DMM descriptors to RAM, and writing to DMM
register which starts the DMM transaction.
The current theory is that a wmb() does not properly ensure that the
data written to RAM is observable by all the components in the system.
This DMM timeout has caused interesting (and rare) bugs as the error
handling was not functioning properly (the error handling has been fixed
in previous commits):
* If a DMM timeout happened when a GEM buffer was being pinned for
display on the screen, a timeout error would be shown, but the driver
would continue programming DSS HW with broken buffer, leading to
SYNCLOST floods and possible crashes.
* If a DMM timeout happened when other user (say, video decoder) was
pinning a GEM buffer, a timeout would be shown but if the user
handled the error properly, no other issues followed.
* If a DMM timeout happened when a GEM buffer was being released, the
driver does not even notice the error, leading to crashes or hang
later.
This patch adds wmb() and readl() calls after the last bit is written to
RAM, which should ensure that the execution proceeds only after the data
is actually in RAM, and thus observable by DMM.
The read-back should not be needed. Further study is required to understand
if DMM is somehow special case and read-back is ok, or if DRA7's memory
barriers do not work correctly.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When enumerating page size definitions to check hardware support,
we construct a constant which is (1U << (def->shift - 10)).
However, the array of page size definitions is only initalised for
various MMU_PAGE_* constants, so it contains a number of 0-initialised
elements with def->shift == 0. This means we end up shifting by a
very large number, which gives the following UBSan splat:
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[220]: Failed to apply ACL on /dev/v4l-subdev0: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[224]: Failed to apply ACL on /dev/v4l-subdev1: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[215]: Failed to apply ACL on /dev/v4l-subdev10: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[228]: Failed to apply ACL on /dev/v4l-subdev2: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[232]: Failed to apply ACL on /dev/v4l-subdev5: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[217]: Failed to apply ACL on /dev/v4l-subdev11: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[214]: Failed to apply ACL on /dev/dri/card1: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[216]: Failed to apply ACL on /dev/v4l-subdev8: Operation not supported
| Sep 04 09:05:10 imx6qdl-variscite-som systemd-udevd[226]: Failed to apply ACL on /dev/v4l-subdev9: Operation not supported
and nasty follow-ups: Starting weston from sddm as unpriviledged user fails
with some hints on missing access rights."
Select the CONFIG_TMPFS_POSIX_ACL option to fix these issues.
The issue is found by a fuzzing test.
If tty_find_polling_driver() recevies an incorrect input such as
',,' or '0b', the len becomes 0 and strncmp() always return 0.
In this case, a null p->ops->poll_init() is called and it causes a kernel
panic.
Fix this by checking name length against zero in tty_find_polling_driver().
If an error occurs during an unplug operation, it's possible for
eeh_dump_dev_log() to be called when edev->pdn is null, which
currently leads to dereferencing a null pointer.
Handle this by skipping the error log for those devices.
Signed-off-by: Sam Bobroff <sbobroff@linux.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Sasha Levin <sashal@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit d595567dc4f0 (MD: fix invalid stored role for a disk) broke linear
hotadd. Let's only fix the role for disks in raid1/10.
Based on Guoqing's original patch.
Reported-by: kernel test robot <rong.a.chen@intel.com> Cc: Gioh Kim <gi-oh.kim@profitbricks.com> Cc: Guoqing Jiang <gqjiang@suse.com> Signed-off-by: Shaohua Li <shli@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We can't use entry->bytes if our entry is a bitmap entry, we need to use
entry->max_extent_size in that case. Fix up all the logic to make this
consistent.
CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
At inode.c:compress_file_range(), under the "free_pages_out" label, we can
end up dereferencing the "pages" pointer when it has a NULL value. This
case happens when "start" has a value of 0 and we fail to allocate memory
for the "pages" pointer. When that happens we jump to the "cont" label and
then enter the "if (start == 0)" branch where we immediately call the
cow_file_range_inline() function. If that function returns 0 (success
creating an inline extent) or an error (like -ENOMEM for example) we jump
to the "free_pages_out" label and then access "pages[i]" leading to a NULL
pointer dereference, since "nr_pages" has a value greater than zero at
that point.
Fix this by setting "nr_pages" to 0 when we fail to allocate memory for
the "pages" pointer.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=201119 Fixes: 771ed689d2cd ("Btrfs: Optimize compressed writeback and reads") CC: stable@vger.kernel.org # 4.4+ Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[CAUSE]
Before rescan we call qgroup_rescan_zero_tracking() to zero out all
qgroups' accounting numbers.
However we don't mark all qgroups dirty, but rely on rescan to do so.
If we have any high level qgroup without children, it won't be marked
dirty during rescan, since we cannot reach that qgroup.
This will cause QGROUP_INFO items of childless qgroups never get updated
in the quota tree, thus their numbers will stay the same in "btrfs
qgroup show" output.
[FIX]
Just mark all qgroups dirty in qgroup_rescan_zero_tracking(), so even if
we have childless qgroups, their QGROUP_INFO items will still get
updated during rescan.
After replaying the log we end up with file bar having 2 hard links, both
with the name 'bar' and one in the directory with inode number 258 and the
other in the directory with inode number 259. Also, we end up with the
directory inode 259 still existing and with the directory inode 258 still
named as 'A', instead of 'B'. In this scenario, file 'bar' should only
have one hard link, located at directory inode 258, the directory inode
259 should not exist anymore and the name for directory inode 258 should
be 'B'.
This incorrect behaviour happens because when attempting to log the old
parents of an inode, we skip any parents that no longer exist. Fix this
by forcing a full commit if an old parent no longer exists.
A test case for fstests follows soon.
CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Allocating new chunks modifies both the extent and chunk tree, which can
trigger new chunk allocations. So instead of doing list_for_each_safe,
just do while (!list_empty()) so we make sure we don't exit with other
pending bg's still on our list.
CC: stable@vger.kernel.org # 4.4+ Reviewed-by: Omar Sandoval <osandov@fb.com> Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com> Signed-off-by: Josef Bacik <josef@toxicpanda.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We need to clear the max_extent_size when we clear bits from a bitmap
since it could have been from the range that contains the
max_extent_size.
CC: stable@vger.kernel.org # 4.4+ Reviewed-by: Liu Bo <bo.liu@linux.alibaba.com> Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
While testing my backport I noticed there was a panic if I ran
generic/416 generic/417 generic/418 all in a row. This just happened to
uncover a race where we had outstanding IO after we destroy all of our
workqueues, and then we'd go to queue the endio work on those free'd
workqueues.
This is because we aren't waiting for the caching threads to be done
before freeing everything up, so to fix this make sure we wait on any
outstanding caching that's being done before we free up the block group,
so we're sure to be done with all IO by the time we get to
btrfs_stop_all_workers(). This fixes the panic I was seeing
consistently in testing.
We check whether any device the file system is using supports discard in
the ioctl call, but then we attempt to trim free extents on every device
regardless of whether discard is supported. Due to the way we mask off
EOPNOTSUPP, we can end up issuing the trim operations on each free range
on devices that don't support it, just wasting time.
Fixes: 499f377f49f08 ("btrfs: iterate over unused chunk space in FITRIM") CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Jeff Mahoney <jeffm@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
btrfs_trim_fs iterates over the fs_devices->alloc_list while holding the
device_list_mutex. The problem is that ->alloc_list is protected by the
chunk mutex. We don't want to hold the chunk mutex over the trim of the
entire file system. Fortunately, the ->dev_list list is protected by
the dev_list mutex and while it will give us all devices, including
read-only devices, we already just skip the read-only devices. Then we
can continue to take and release the chunk mutex while scanning each
device.
Fixes: 499f377f49f ("btrfs: iterate over unused chunk space in FITRIM") CC: stable@vger.kernel.org # 4.4+ Signed-off-by: Jeff Mahoney <jeffm@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes: 681ae50917df9 ("Btrfs: cleanup reserved space when freeing tree log on error") CC: <stable@vger.kernel.org> # v3.13 Signed-off-by: Jeff Mahoney <jeffm@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[BUG]
For certain crafted image, whose csum root leaf has missing backref, if
we try to trigger write with data csum, it could cause deadlock with the
following kernel WARN_ON():
[CAUSE]
That crafted image has missing backref for csum tree root leaf. And
when we try to allocate new tree block, since there is no
EXTENT/METADATA_ITEM for csum tree root, btrfs consider it's free slot
and use it.
The extent tree of the image looks like:
Normal image | This fuzzed image
----------------------------------+--------------------------------
BG 29360128 | BG 29360128
One empty slot | One empty slot 29364224: backref to UUID tree | 29364224: backref to UUID tree
Two empty slots | Two empty slots 29376512: backref to CSUM tree | One empty slot (bad type) <<< 29380608: backref to D_RELOC tree | 29380608: backref to D_RELOC tree
... | ...
Since bytenr 29376512 has no METADATA/EXTENT_ITEM, when btrfs try to
alloc tree block, it's an valid slot for btrfs.
And for finish_ordered_write, when we need to insert csum, we try to CoW
csum tree root.
By accident, empty slots at bytenr BG_OFFSET, BG_OFFSET + 8K,
BG_OFFSET + 12K is already used by tree block COW for other trees, the
next empty slot is BG_OFFSET + 16K, which should be the backref for CSUM
tree.
But due to the bad type, btrfs can recognize it and still consider it as
an empty slot, and will try to use it for csum tree CoW.
Then in the following call trace, we will try to lock the new tree
block, which turns out to be the old csum tree root which is already
locked:
btrfs_search_slot() called on csum tree root, which is at 29376512
|- btrfs_cow_block()
|- btrfs_set_lock_block()
| |- Now locks tree block 29376512 (old csum tree root)
|- __btrfs_cow_block()
|- btrfs_alloc_tree_block()
|- btrfs_reserve_extent()
| Now it returns tree block 29376512, which extent tree
| shows its empty slot, but it's already hold by csum tree
|- btrfs_init_new_buffer()
|- btrfs_tree_lock()
| Triggers WARN_ON(eb->lock_owner == current->pid)
|- wait_event()
Wait lock owner to release the lock, but it's
locked by ourself, so it will deadlock
[FIX]
This patch will do the lock_owner and current->pid check at
btrfs_init_new_buffer().
So above deadlock can be avoided.
Since such problem can only happen in crafted image, we will still
trigger kernel warning for later aborted transaction, but with a little
more meaningful warning message.
[CAUSE]
Extent tree corruption. In this particular case, reloc tree root's
owner is DATA_RELOC_TREE (should be TREE_RELOC), thus its backref is
corrupted and we failed the owner check in walk_up_tree().
[FIX]
It's pretty hard to take care of every extent tree corruption, but at
least we can remove such BUG_ON() and exit more gracefully.
And since in this particular image, DATA_RELOC_TREE and TREE_RELOC share
the same root (which is obviously invalid), we needs to make
__del_reloc_root() more robust to detect such invalid sharing to avoid
possible NULL dereference as root->node can be NULL in this case.
Fix child-node lookup during probe, which ended up searching the whole
device tree depth-first starting at the parent rather than just matching
on its children.
To make things worse, the parent pmc node could end up being prematurely
freed as of_find_node_by_name() drops a reference to its first argument.
Fixes: 3568df3d31d6 ("soc: tegra: Add thermal reset (thermtrip) support to PMC") Cc: stable <stable@vger.kernel.org> # 4.0 Cc: Mikko Perttunen <mperttunen@nvidia.com> Signed-off-by: Johan Hovold <johan@kernel.org> Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com> Signed-off-by: Thierry Reding <treding@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It is possible for corrupted filesystem images to produce very large
block offsets that may wrap when a length is added, and wrongly pass
the buffer size test.