| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
The user can open multiple device FDs if it likes, however the open
function calls vfio_register_notifier() on device global state. Calling
vfio_register_notifier() twice will trigger a WARN_ON from
notifier_chain_register() and the first close will wrongly delete the
notifier and more.
Since these really want the new open/close_device() semantics just change
the function over.
Reviewed-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/13-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
The user can open multiple device FDs if it likes, however these open()
functions call vfio_register_notifier() on some device global
state. Calling vfio_register_notifier() twice in will trigger a WARN_ON
from notifier_chain_register() and the first close will wrongly delete the
notifier and more.
Since these really want the new open/close_device() semantics just change
the functions over.
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/12-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
mbochs_close() iterates over global device state and frees it. Currently
this is done every time a device FD is closed, but if multiple device FDs
are open this could corrupt other still active FDs.
Change this to use close_device() so it only runs on the last close.
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/11-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Like vfio_pci_dev_set_try_reset() this code wants to reset all of the
devices in the "reset group" which is the same membership as the device
set.
Instead of trying to reconstruct the device set from the PCI list go
directly from the device set's device list to execute the reset.
The same basic structure as vfio_pci_dev_set_try_reset() is used. The
'vfio_devices' struct is replaced with the device set linked list and we
simply sweep it multiple times under the lock.
This eliminates a memory allocation and get/put traffic and another
improperly locked test of pci_dev_driver().
Reviewed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/10-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
vfio_pci_try_bus_reset() is triggering a reset of the entire_dev set if
any device within it has accumulated a needs_reset. This reset can only be
done once all of the drivers operating the PCI devices to be reset are in
a known safe state.
Make this clearer by directly operating on the dev_set instead of the
vfio_pci_device. Rename the function to vfio_pci_dev_set_try_reset().
Use the device list inside the dev_set to check that all drivers are in a
safe state instead of working backwards from the pci_device.
The dev_set->lock directly prevents devices from joining/leaving the set,
or changing their state, which further implies the pci_device cannot
change drivers or that the vfio_device be freed, eliminating the need for
get/put's.
If a pci_device to be reset is not in the dev_set then the reset cannot be
used as we can't know what the state of that driver is. Directly measure
this by checking that every pci_device is in the dev_set - which
effectively proves that VFIO drivers are attached to everything.
Remove the odd interaction around vfio_pci_set_power_state() - have the
only caller avoid its redundant vfio_pci_set_power_state() instead of
avoiding it inside vfio_pci_dev_set_try_reset().
This restructuring corrects a call to pci_dev_driver() without holding the
device_lock() and removes a hard wiring to &vfio_pci_driver.
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/9-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
PCI wants to have the usual open/close_device() logic with the slight
twist that the open/close_device() must be done under a singelton lock
shared by all of the vfio_devices that are in the PCI "reset group".
The reset group, and thus the device set, is determined by what devices
pci_reset_bus() touches, which is either the entire bus or only the slot.
Rely on the core code to do everything reflck was doing and delete reflck
entirely.
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/8-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Platform simply wants to run some code when the device is first
opened/last closed. Use the core framework and locking for this. Aside
from removing a bit of code this narrows the locking scope from a global
lock.
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/7-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
FSL uses the internal reflck to implement the open_device() functionality,
conversion to the core code is straightforward.
The decision on which set to be part of is trivially based on the
is_fsl_mc_bus_dprc() and we use a 'struct device *' pointer as the set_id.
The dev_set lock is protecting the interrupts setup. The FSL MC devices
are using MSIs and only the DPRC device is allocating the MSIs from the
MSI domain. The other devices just take interrupts from a pool. The lock
is protecting the access to this pool.
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Tested-by: Diana Craciun OSS <diana.craciun@oss.nxp.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/6-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
The core code no longer requires these ops to be defined, so delete these
empty functions and leave the op as NULL. mtty's functions only log a
pointless message, delete that entirely.
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/5-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Currently the driver ops have an open/release pair that is called once
each time a device FD is opened or closed. Add an additional set of
open/close_device() ops which are called when the device FD is opened for
the first time and closed for the last time.
An analysis shows that all of the drivers require this semantic. Some are
open coding it as part of their reflck implementation, and some are just
buggy and miss it completely.
To retain the current semantics PCI and FSL depend on, introduce the idea
of a "device set" which is a grouping of vfio_device's that share the same
lock around opening.
The device set is established by providing a 'set_id' pointer. All
vfio_device's that provide the same pointer will be joined to the same
singleton memory and lock across the whole set. This effectively replaces
the oddly named reflck.
After conversion the set_id will be sourced from:
- A struct device from a fsl_mc_device (fsl)
- A struct pci_slot (pci)
- A struct pci_bus (pci)
- The struct vfio_device (everything)
The design ensures that the above pointers are live as long as the
vfio_device is registered, so they form reliable unique keys to group
vfio_devices into sets.
This implementation uses xarray instead of searching through the driver
core structures, which simplifies the somewhat tricky locking in this
area.
Following patches convert all the drivers.
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/4-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
This pairs with vfio_init_group_dev() and allows undoing any state that is
stored in the vfio_device unrelated to registration. Add appropriately
placed calls to all the drivers.
The following patch will use this to add pre-registration state for the
device set.
Signed-off-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/3-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Convert mbochs to use an atomic scheme for this like mtty was changed
into. The atomic fixes various race conditions with probing. Add the
missing error unwind. Also add the missing kfree of mdev_state->pages.
Fixes: 681c1615f891 ("vfio/mbochs: Convert to use vfio_register_group_dev()")
Reported-by: Cornelia Huck <cohuck@redhat.com>
Co-developed-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/2-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
The patch to move the get/put to core and the patch to convert the samples
to use vfio_device crossed in a way that this was missed. When both
patches are together the samples do not need their own get/put.
Fixes: 437e41368c01 ("vfio/mdpy: Convert to use vfio_register_group_dev()")
Fixes: 681c1615f891 ("vfio/mbochs: Convert to use vfio_register_group_dev()")
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Link: https://lore.kernel.org/r/1-v4-9ea22c5e6afb+1adf-vfio_reflck_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Only a single driver actually sets the ->request method, so don't print
a scary warning if it isn't.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20210726143524.155779-3-hch@lst.de
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
Without this setups with buŃ–lt-in mdev and mdev-drivers fail to
register like this:
[1.903149] Driver 'intel_vgpu_mdev' was unable to register with bus_type 'mdev' because the bus was not initialized.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20210726143524.155779-2-hch@lst.de
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
The only implementation of this in IGD returns a -ERRNO which is
implicitly cast through a size_t and then casted again and returned as a
ssize_t in vfio_pci_rw().
Fix the vfio_pci_regops->rw() return type to be ssize_t so all is
consistent.
Fixes: 28541d41c9e0 ("vfio/pci: Add infrastructure for additional device specific regions")
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Link: https://lore.kernel.org/r/0-v3-5db12d1bf576+c910-vfio_rw_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
VFIO_NOIOMMU is supposed to be an element in the VFIO menu, not start
a new menu. Correct this copy-paste mistake.
Fixes: 03a76b60f8ba ("vfio: Include No-IOMMU mode")
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Link: https://lore.kernel.org/r/0-v1-3f0b685c3679+478-vfio_menuconfig_jgg@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
|
|
|
|
Since commit 1b6b26ae7053 ("pipe: fix and clarify pipe write wakeup
logic") we have sanitized the pipe write logic, and would only try to
wake up readers if they needed it.
In particular, if the pipe already had data in it before the write,
there was no point in trying to wake up a reader, since any existing
readers must have been aware of the pre-existing data already. Doing
extraneous wakeups will only cause potential thundering herd problems.
However, it turns out that some Android libraries have misused the EPOLL
interface, and expected "edge triggered" be to "any new write will
trigger it". Even if there was no edge in sight.
Quoting Sandeep Patil:
"The commit 1b6b26ae7053 ('pipe: fix and clarify pipe write wakeup
logic') changed pipe write logic to wakeup readers only if the pipe
was empty at the time of write. However, there are libraries that
relied upon the older behavior for notification scheme similar to
what's described in [1]
One such library 'realm-core'[2] is used by numerous Android
applications. The library uses a similar notification mechanism as GNU
Make but it never drains the pipe until it is full. When Android moved
to v5.10 kernel, all applications using this library stopped working.
The library has since been fixed[3] but it will be a while before all
applications incorporate the updated library"
Our regression rule for the kernel is that if applications break from
new behavior, it's a regression, even if it was because the application
did something patently wrong. Also note the original report [4] by
Michal Kerrisk about a test for this epoll behavior - but at that point
we didn't know of any actual broken use case.
So add the extraneous wakeup, to approximate the old behavior.
[ I say "approximate", because the exact old behavior was to do a wakeup
not for each write(), but for each pipe buffer chunk that was filled
in. The behavior introduced by this change is not that - this is just
"every write will cause a wakeup, whether necessary or not", which
seems to be sufficient for the broken library use. ]
It's worth noting that this adds the extraneous wakeup only for the
write side, while the read side still considers the "edge" to be purely
about reading enough from the pipe to allow further writes.
See commit f467a6a66419 ("pipe: fix and clarify pipe read wakeup logic")
for the pipe read case, which remains that "only wake up if the pipe was
full, and we read something from it".
Link: https://lore.kernel.org/lkml/CAHk-=wjeG0q1vgzu4iJhW5juPkTsjTYmiqiMUYAebWW+0bam6w@mail.gmail.com/ [1]
Link: https://github.com/realm/realm-core [2]
Link: https://github.com/realm/realm-core/issues/4666 [3]
Link: https://lore.kernel.org/lkml/CAKgNAkjMBGeAwF=2MKK758BhxvW58wYTgYKB2V-gY1PwXxrH+Q@mail.gmail.com/ [4]
Link: https://lore.kernel.org/lkml/20210729222635.2937453-1-sspatil@android.com/
Reported-by: Sandeep Patil <sspatil@android.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
This makes 'perf top' abort in some cases, and the right fix will
involve surgery that is too much to do at this stage, so revert for now
and fix it in the next merge window.
This reverts commit 2d6b74baa7147251c30a46c4996e8cc224aa2dc5.
Cc: Riccardo Mancini <rickyman7@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Krister Johansen <kjlx@templeofstupid.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
When I use kfree_rcu() to free a large memory allocated by kmalloc_node(),
the following dump occurs.
BUG: kernel NULL pointer dereference, address: 0000000000000020
[...]
Oops: 0000 [#1] SMP
[...]
Workqueue: events kfree_rcu_work
RIP: 0010:__obj_to_index include/linux/slub_def.h:182 [inline]
RIP: 0010:obj_to_index include/linux/slub_def.h:191 [inline]
RIP: 0010:memcg_slab_free_hook+0x120/0x260 mm/slab.h:363
[...]
Call Trace:
kmem_cache_free_bulk+0x58/0x630 mm/slub.c:3293
kfree_bulk include/linux/slab.h:413 [inline]
kfree_rcu_work+0x1ab/0x200 kernel/rcu/tree.c:3300
process_one_work+0x207/0x530 kernel/workqueue.c:2276
worker_thread+0x320/0x610 kernel/workqueue.c:2422
kthread+0x13d/0x160 kernel/kthread.c:313
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:294
When kmalloc_node() a large memory, page is allocated, not slab, so when
freeing memory via kfree_rcu(), this large memory should not be used by
memcg_slab_free_hook(), because memcg_slab_free_hook() is is used for
slab.
Using page_objcgs_check() instead of page_objcgs() in
memcg_slab_free_hook() to fix this bug.
Link: https://lkml.kernel.org/r/20210728145655.274476-1-wanghai38@huawei.com
Fixes: 270c6a71460e ("mm: memcontrol/slab: Use helpers to access slab page's memcg_data")
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
SLUB uses page allocator for higher order allocations and update
unreclaimable slab stat for such allocations. At the moment, the bulk
free for SLUB does not share code with normal free code path for these
type of allocations and have missed the stat update. So, fix the stat
update by common code. The user visible impact of the bug is the
potential of inconsistent unreclaimable slab stat visible through
meminfo and vmstat.
Link: https://lkml.kernel.org/r/20210728155354.3440560-1-shakeelb@google.com
Fixes: 6a486c0ad4dc ("mm, sl[ou]b: improve memory accounting")
Signed-off-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Similar to commit 2da9f6305f30 ("mm/vmscan: fix NR_ISOLATED_FILE
corruption on 64-bit") avoid using unsigned int for nr_pages. With
unsigned int type the large unsigned int converts to a large positive
signed long.
Symptoms include CMA allocations hanging forever due to
alloc_contig_range->...->isolate_migratepages_block waiting forever in
"while (unlikely(too_many_isolated(pgdat)))".
Link: https://lkml.kernel.org/r/20210728042531.359409-1-aneesh.kumar@linux.ibm.com
Fixes: c5fc5c3ae0c8 ("mm: migrate: account THP NUMA migration counters correctly")
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Reported-by: Michael Ellerman <mpe@ellerman.id.au>
Reported-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Dan Carpenter reports:
The patch 2d146aa3aa84: "mm: memcontrol: switch to rstat" from Apr
29, 2021, leads to the following static checker warning:
kernel/cgroup/rstat.c:200 cgroup_rstat_flush()
warn: sleeping in atomic context
mm/memcontrol.c
3572 static unsigned long mem_cgroup_usage(struct mem_cgroup *memcg, bool swap)
3573 {
3574 unsigned long val;
3575
3576 if (mem_cgroup_is_root(memcg)) {
3577 cgroup_rstat_flush(memcg->css.cgroup);
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This is from static analysis and potentially a false positive. The
problem is that mem_cgroup_usage() is called from __mem_cgroup_threshold()
which holds an rcu_read_lock(). And the cgroup_rstat_flush() function
can sleep.
3578 val = memcg_page_state(memcg, NR_FILE_PAGES) +
3579 memcg_page_state(memcg, NR_ANON_MAPPED);
3580 if (swap)
3581 val += memcg_page_state(memcg, MEMCG_SWAP);
3582 } else {
3583 if (!swap)
3584 val = page_counter_read(&memcg->memory);
3585 else
3586 val = page_counter_read(&memcg->memsw);
3587 }
3588 return val;
3589 }
__mem_cgroup_threshold() indeed holds the rcu lock. In addition, the
thresholding code is invoked during stat changes, and those contexts
have irqs disabled as well. If the lock breaking occurs inside the
flush function, it will result in a sleep from an atomic context.
Use the irqsafe flushing variant in mem_cgroup_usage() to fix this.
Link: https://lkml.kernel.org/r/20210726150019.251820-1-hannes@cmpxchg.org
Fixes: 2d146aa3aa84 ("mm: memcontrol: switch to rstat")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Chris Down <chris@chrisdown.name>
Reviewed-by: Rik van Riel <riel@surriel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
For punch holes in EOF blocks, fallocate used buffer write to zero the
EOF blocks in last cluster. But since ->writepage will ignore EOF
pages, those zeros will not be flushed.
This "looks" ok as commit 6bba4471f0cc ("ocfs2: fix data corruption by
fallocate") will zero the EOF blocks when extend the file size, but it
isn't. The problem happened on those EOF pages, before writeback, those
pages had DIRTY flag set and all buffer_head in them also had DIRTY flag
set, when writeback run by write_cache_pages(), DIRTY flag on the page
was cleared, but DIRTY flag on the buffer_head not.
When next write happened to those EOF pages, since buffer_head already
had DIRTY flag set, it would not mark page DIRTY again. That made
writeback ignore them forever. That will cause data corruption. Even
directio write can't work because it will fail when trying to drop pages
caches before direct io, as it found the buffer_head for those pages
still had DIRTY flag set, then it will fall back to buffer io mode.
To make a summary of the issue, as writeback ingores EOF pages, once any
EOF page is generated, any write to it will only go to the page cache,
it will never be flushed to disk even file size extends and that page is
not EOF page any more. The fix is to avoid zero EOF blocks with buffer
write.
The following code snippet from qemu-img could trigger the corruption.
656 open("6b3711ae-3306-4bdd-823c-cf1c0060a095.conv.2", O_RDWR|O_DIRECT|O_CLOEXEC) = 11
...
660 fallocate(11, FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE, 2275868672, 327680 <unfinished ...>
660 fallocate(11, 0, 2275868672, 327680) = 0
658 pwrite64(11, "
Link: https://lkml.kernel.org/r/20210722054923.24389-2-junxiao.bi@oracle.com
Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
If append-dio feature is enabled, direct-io write and fallocate could
run in parallel to extend file size, fallocate used "orig_isize" to
record i_size before taking "ip_alloc_sem", when
ocfs2_zeroout_partial_cluster() zeroout EOF blocks, i_size maybe already
extended by ocfs2_dio_end_io_write(), that will cause valid data zeroed
out.
Link: https://lkml.kernel.org/r/20210722054923.24389-1-junxiao.bi@oracle.com
Fixes: 6bba4471f0cc ("ocfs2: fix data corruption by fallocate")
Signed-off-by: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Jun Piao <piaojun@huawei.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
STRING_SELFTEST is presented in the "Library routines" menu. Move it in
Kernel hacking > Kernel Testing and Coverage > Runtime Testing together
with other similar tests found in lib/
--- Runtime Testing
<*> Test functions located in the hexdump module at runtime
<*> Test string functions (NEW)
<*> Test functions located in the string_helpers module at runtime
<*> Test strscpy*() family of functions at runtime
<*> Test kstrto*() family of functions at runtime
<*> Test printf() family of functions at runtime
<*> Test scanf() family of functions at runtime
Link: https://lkml.kernel.org/r/20210719185158.190371-1-mcroce@linux.microsoft.com
Signed-off-by: Matteo Croce <mcroce@microsoft.com>
Cc: Peter Rosin <peda@axentia.se>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
The team maintaining the gve driver has undergone some changes,
this updates the MAINTAINERS file accordingly.
Signed-off-by: Catherine Sullivan <csully@google.com>
Signed-off-by: Jon Olson <jonolson@google.com>
Signed-off-by: David Awogbemila <awogbemila@google.com>
Signed-off-by: Jeroen de Borst <jeroendb@google.com>
Link: https://lore.kernel.org/r/20210729155258.442650-1-csully@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
|
|
The arch-specific Kconfig files use HAVE_IDE to indicate if IDE is
supported.
As IDE support and the HAVE_IDE config vanishes with commit b7fb14d3ac63
("ide: remove the legacy ide driver"), there is no need to mention
HAVE_IDE in all those arch-specific Kconfig files.
The issue was identified with ./scripts/checkkconfigsymbols.py.
Fixes: b7fb14d3ac63 ("ide: remove the legacy ide driver")
Suggested-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20210728182115.4401-1-lukas.bulwahn@gmail.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
In esd_usb2_setup_rx_urbs() MAX_RX_URBS coherent buffers are allocated
and there is nothing, that frees them:
1) In callback function the urb is resubmitted and that's all
2) In disconnect function urbs are simply killed, but URB_FREE_BUFFER
is not set (see esd_usb2_setup_rx_urbs) and this flag cannot be used
with coherent buffers.
So, all allocated buffers should be freed with usb_free_coherent()
explicitly.
Side note: This code looks like a copy-paste of other can drivers. The
same patch was applied to mcba_usb driver and it works nice with real
hardware. There is no change in functionality, only clean-up code for
coherent buffers.
Fixes: 96d8e90382dc ("can: Add driver for esd CAN-USB/2 device")
Link: https://lore.kernel.org/r/b31b096926dcb35998ad0271aac4b51770ca7cc8.1627404470.git.paskripkin@gmail.com
Cc: linux-stable <stable@vger.kernel.org>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
In ems_usb_start() MAX_RX_URBS coherent buffers are allocated and
there is nothing, that frees them:
1) In callback function the urb is resubmitted and that's all
2) In disconnect function urbs are simply killed, but URB_FREE_BUFFER
is not set (see ems_usb_start) and this flag cannot be used with
coherent buffers.
So, all allocated buffers should be freed with usb_free_coherent()
explicitly.
Side note: This code looks like a copy-paste of other can drivers. The
same patch was applied to mcba_usb driver and it works nice with real
hardware. There is no change in functionality, only clean-up code for
coherent buffers.
Fixes: 702171adeed3 ("ems_usb: Added support for EMS CPC-USB/ARM7 CAN/USB interface")
Link: https://lore.kernel.org/r/59aa9fbc9a8cbf9af2bbd2f61a659c480b415800.1627404470.git.paskripkin@gmail.com
Cc: linux-stable <stable@vger.kernel.org>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
In usb_8dev_start() MAX_RX_URBS coherent buffers are allocated and
there is nothing, that frees them:
1) In callback function the urb is resubmitted and that's all
2) In disconnect function urbs are simply killed, but URB_FREE_BUFFER
is not set (see usb_8dev_start) and this flag cannot be used with
coherent buffers.
So, all allocated buffers should be freed with usb_free_coherent()
explicitly.
Side note: This code looks like a copy-paste of other can drivers. The
same patch was applied to mcba_usb driver and it works nice with real
hardware. There is no change in functionality, only clean-up code for
coherent buffers.
Fixes: 0024d8ad1639 ("can: usb_8dev: Add support for USB2CAN interface from 8 devices")
Link: https://lore.kernel.org/r/d39b458cd425a1cf7f512f340224e6e9563b07bd.1627404470.git.paskripkin@gmail.com
Cc: linux-stable <stable@vger.kernel.org>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
Yasushi reported, that his Microchip CAN Analyzer stopped working
since commit 91c02557174b ("can: mcba_usb: fix memory leak in
mcba_usb"). The problem was in missing urb->transfer_dma
initialization.
In my previous patch to this driver I refactored mcba_usb_start() code
to avoid leaking usb coherent buffers. To archive it, I passed local
stack variable to usb_alloc_coherent() and then saved it to private
array to correctly free all coherent buffers on ->close() call. But I
forgot to initialize urb->transfer_dma with variable passed to
usb_alloc_coherent().
All of this was causing device to not work, since dma addr 0 is not
valid and following log can be found on bug report page, which points
exactly to problem described above.
| DMAR: [DMA Write] Request device [00:14.0] PASID ffffffff fault addr 0 [fault reason 05] PTE Write access is not set
Fixes: 91c02557174b ("can: mcba_usb: fix memory leak in mcba_usb")
Link: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=990850
Link: https://lore.kernel.org/r/20210725103630.23864-1-paskripkin@gmail.com
Cc: linux-stable <stable@vger.kernel.org>
Reported-by: Yasushi SHOJI <yasushi.shoji@gmail.com>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Tested-by: Yasushi SHOJI <yashi@spacecubics.com>
[mkl: fixed typos in commit message - thanks Yasushi SHOJI]
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
The hi3110_cmd() is supposed to return zero on success and negative
error codes on failure, but it was accidentally declared as a u8 when
it needs to be an int type.
Fixes: 57e83fb9b746 ("can: hi311x: Add Holt HI-311x CAN driver")
Link: https://lore.kernel.org/r/20210729141246.GA1267@kili
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
This patch adds Yasushi SHOJI as a reviewer for the Microchip CAN BUS
Analyzer Tool driver.
Link: https://lore.kernel.org/r/20210726111619.1023991-1-mkl@pengutronix.de
Acked-by: Yasushi SHOJI <yashi@spacecubics.com>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
|
|
Fix the following fallthrough warning (on ARM):
drivers/scsi/arm/fas216.c:1379:2: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]
default:
^
drivers/scsi/arm/fas216.c:1379:2: note: insert 'break;' to avoid fall-through
default:
^
break;
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/lkml/202107260355.bF00i5bi-lkp@intel.com/
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
|
|
Fix the following fallthrough warning (on ARM):
drivers/scsi/arm/acornscsi.c:2651:2: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]
case res_success:
^
drivers/scsi/arm/acornscsi.c:2651:2: note: insert '__attribute__((fallthrough));' to silence this warning
case res_success:
^
__attribute__((fallthrough));
drivers/scsi/arm/acornscsi.c:2651:2: note: insert 'break;' to avoid fall-through
case res_success:
^
break;
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/lkml/202107260355.bF00i5bi-lkp@intel.com/
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
|
|
Fix the following fallthrough warning:
arch/arm/mach-rpc/riscpc.c:52:2: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]
default:
^
arch/arm/mach-rpc/riscpc.c:52:2: note: insert 'break;' to avoid fall-through
default:
^
break;
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/lkml/202107260355.bF00i5bi-lkp@intel.com/
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
|
|
While reviewing the buffer item recovery code, the thought occurred to
me: in V5 filesystems we use log sequence number (LSN) tracking to avoid
replaying older metadata updates against newer log items. However, we
use the magic number of the ondisk buffer to find the LSN of the ondisk
metadata, which means that if an attacker can control the layout of the
realtime device precisely enough that the start of an rt bitmap block
matches the magic and UUID of some other kind of block, they can control
the purported LSN of that spoofed block and thereby break log replay.
Since realtime bitmap and summary blocks don't have headers at all, we
have no way to tell if a block really should be replayed. The best we
can do is replay unconditionally and hope for the best.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Carlos Maiolino <cmaiolino@redhat.com>
|
|
From the department of "generic/482 keeps on giving", we bring you
another tail update race condition:
iclog:
S1 C1
+-----------------------+-----------------------+
S2 EOIC
Two checkpoints in a single iclog. One is complete, the other just
contains the start record and overruns into a new iclog.
Timeline:
Before S1: Cache flush, log tail = X
At S1: Metadata stable, write start record and checkpoint
At C1: Write commit record, set NEED_FUA
Single iclog checkpoint, so no need for NEED_FLUSH
Log tail still = X, so no need for NEED_FLUSH
After C1,
Before S2: Cache flush, log tail = X
At S2: Metadata stable, write start record and checkpoint
After S2: Log tail moves to X+1
At EOIC: End of iclog, more journal data to write
Releases iclog
Not a commit iclog, so no need for NEED_FLUSH
Writes log tail X+1 into iclog.
At this point, the iclog has tail X+1 and NEED_FUA set. There has
been no cache flush for the metadata between X and X+1, and the
iclog writes the new tail permanently to the log. THis is sufficient
to violate on disk metadata/journal ordering.
We have two options here. The first is to detect this case in some
manner and ensure that the partial checkpoint write sets NEED_FLUSH
when the iclog is already marked NEED_FUA and the log tail changes.
This seems somewhat fragile and quite complex to get right, and it
doesn't actually make it obvious what underlying problem it is
actually addressing from reading the code.
The second option seems much cleaner to me, because it is derived
directly from the requirements of the C1 commit record in the iclog.
That is, when we write this commit record to the iclog, we've
guaranteed that the metadata/data ordering is correct for tail
update purposes. Hence if we only write the log tail into the iclog
for the *first* commit record rather than the log tail at the last
release, we guarantee that the log tail does not move past where the
the first commit record in the log expects it to be.
IOWs, taking the first option means that replay of C1 becomes
dependent on future operations doing the right thing, not just the
C1 checkpoint itself doing the right thing. This makes log recovery
almost impossible to reason about because now we have to take into
account what might or might not have happened in the future when
looking at checkpoints in the log rather than just having to
reconstruct the past...
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Because I cannot tell if the NEED_FLUSH flag is being set correctly
by the log force and CIL push machinery without it.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
From the department of "WTAF? How did we miss that!?"...
When we are recovering a buffer, the first thing we do is check the
buffer magic number and extract the LSN from the buffer. If the LSN
is older than the current LSN, we replay the modification to it. If
the metadata on disk is newer than the transaction in the log, we
skip it. This is a fundamental v5 filesystem metadata recovery
behaviour.
generic/482 failed with an attribute writeback failure during log
recovery. The write verifier caught the corruption before it got
written to disk, and the attr buffer dump looked like:
XFS (dm-3): Metadata corruption detected at xfs_attr3_leaf_verify+0x275/0x2e0, xfs_attr3_leaf block 0x19be8
XFS (dm-3): Unmount and run xfs_repair
XFS (dm-3): First 128 bytes of corrupted metadata buffer:
00000000: 00 00 00 00 00 00 00 00 3b ee 00 00 4d 2a 01 e1 ........;...M*..
00000010: 00 00 00 00 00 01 9b e8 00 00 00 01 00 00 05 38 ...............8
^^^^^^^^^^^^^^^^^^^^^^^
00000020: df 39 5e 51 58 ac 44 b6 8d c5 e7 10 44 09 bc 17 .9^QX.D.....D...
00000030: 00 00 00 00 00 02 00 83 00 03 00 cc 0f 24 01 00 .............$..
00000040: 00 68 0e bc 0f c8 00 10 00 00 00 00 00 00 00 00 .h..............
00000050: 00 00 3c 31 0f 24 01 00 00 00 3c 32 0f 88 01 00 ..<1.$....<2....
00000060: 00 00 3c 33 0f d8 01 00 00 00 00 00 00 00 00 00 ..<3............
00000070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
.....
The highlighted bytes are the LSN that was replayed into the
buffer: 0x100000538. This is cycle 1, block 0x538. Prior to replay,
that block on disk looks like this:
$ sudo xfs_db -c "fsb 0x417d" -c "type attr3" -c p /dev/mapper/thin-vol
hdr.info.hdr.forw = 0
hdr.info.hdr.back = 0
hdr.info.hdr.magic = 0x3bee
hdr.info.crc = 0xb5af0bc6 (correct)
hdr.info.bno = 105448
hdr.info.lsn = 0x100000900
^^^^^^^^^^^
hdr.info.uuid = df395e51-58ac-44b6-8dc5-e7104409bc17
hdr.info.owner = 131203
hdr.count = 2
hdr.usedbytes = 120
hdr.firstused = 3796
hdr.holes = 1
hdr.freemap[0-2] = [base,size]
Note the LSN stamped into the buffer on disk: 1/0x900. The version
on disk is much newer than the log transaction that was being
replayed. That's a bug, and should -never- happen.
So I immediately went to look at xlog_recover_get_buf_lsn() to check
that we handled the LSN correctly. I was wondering if there was a
similar "two commits with the same start LSN skips the second
replay" problem with buffers. I didn't get that far, because I found
a much more basic, rudimentary bug: xlog_recover_get_buf_lsn()
doesn't recognise buffers with XFS_ATTR3_LEAF_MAGIC set in them!!!
IOWs, attr3 leaf buffers fall through the magic number checks
unrecognised, so trigger the "recover immediately" behaviour instead
of undergoing an LSN check. IOWs, we incorrectly replay ATTR3 leaf
buffers and that causes silent on disk corruption of inode attribute
forks and potentially other things....
Git history shows this is *another* zero day bug, this time
introduced in commit 50d5c8d8e938 ("xfs: check LSN ordering for v5
superblocks during recovery") which failed to handle the attr3 leaf
buffers in recovery. And we've failed to handle them ever since...
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
When we log an inode, we format the "log inode" core and set an LSN
in that inode core. We do that via xfs_inode_item_format_core(),
which calls:
xfs_inode_to_log_dinode(ip, dic, ip->i_itemp->ili_item.li_lsn);
to format the log inode. It writes the LSN from the inode item into
the log inode, and if recovery decides the inode item needs to be
replayed, it recovers the log inode LSN field and writes it into the
on disk inode LSN field.
Now this might seem like a reasonable thing to do, but it is wrong
on multiple levels. Firstly, if the item is not yet in the AIL,
item->li_lsn is zero. i.e. the first time the inode it is logged and
formatted, the LSN we write into the log inode will be zero. If we
only log it once, recovery will run and can write this zero LSN into
the inode.
This means that the next time the inode is logged and log recovery
runs, it will *always* replay changes to the inode regardless of
whether the inode is newer on disk than the version in the log and
that violates the entire purpose of recording the LSN in the inode
at writeback time (i.e. to stop it going backwards in time on disk
during recovery).
Secondly, if we commit the CIL to the journal so the inode item
moves to the AIL, and then relog the inode, the LSN that gets
stamped into the log inode will be the LSN of the inode's current
location in the AIL, not it's age on disk. And it's not the LSN that
will be associated with the current change. That means when log
recovery replays this inode item, the LSN that ends up on disk is
the LSN for the previous changes in the log, not the current
changes being replayed. IOWs, after recovery the LSN on disk is not
in sync with the LSN of the modifications that were replayed into
the inode. This, again, violates the recovery ordering semantics
that on-disk writeback LSNs provide.
Hence the inode LSN in the log dinode is -always- invalid.
Thirdly, recovery actually has the LSN of the log transaction it is
replaying right at hand - it uses it to determine if it should
replay the inode by comparing it to the on-disk inode's LSN. But it
doesn't use that LSN to stamp the LSN into the inode which will be
written back when the transaction is fully replayed. It uses the one
in the log dinode, which we know is always going to be incorrect.
Looking back at the change history, the inode logging was broken by
commit 93f958f9c41f ("xfs: cull unnecessary icdinode fields") way
back in 2016 by a stupid idiot who thought he knew how this code
worked. i.e. me. That commit replaced an in memory di_lsn field that
was updated only at inode writeback time from the inode item.li_lsn
value - and hence always contained the same LSN that appeared in the
on-disk inode - with a read of the inode item LSN at inode format
time. CLearly these are not the same thing.
Before 93f958f9c41f, the log recovery behaviour was irrelevant,
because the LSN in the log inode always matched the on-disk LSN at
the time the inode was logged, hence recovery of the transaction
would never make the on-disk LSN in the inode go backwards or get
out of sync.
A symptom of the problem is this, caught from a failure of
generic/482. Before log recovery, the inode has been allocated but
never used:
xfs_db> inode 393388
xfs_db> p
core.magic = 0x494e
core.mode = 0
....
v3.crc = 0x99126961 (correct)
v3.change_count = 0
v3.lsn = 0
v3.flags2 = 0
v3.cowextsize = 0
v3.crtime.sec = Thu Jan 1 10:00:00 1970
v3.crtime.nsec = 0
After log recovery:
xfs_db> p
core.magic = 0x494e
core.mode = 020444
....
v3.crc = 0x23e68f23 (correct)
v3.change_count = 2
v3.lsn = 0
v3.flags2 = 0
v3.cowextsize = 0
v3.crtime.sec = Thu Jul 22 17:03:03 2021
v3.crtime.nsec = 751000000
...
You can see that the LSN of the on-disk inode is 0, even though it
clearly has been written to disk. I point out this inode, because
the generic/482 failure occurred because several adjacent inodes in
this specific inode cluster were not replayed correctly and still
appeared to be zero on disk when all the other metadata (inobt,
finobt, directories, etc) indicated they should be allocated and
written back.
The fix for this is two-fold. The first is that we need to either
revert the LSN changes in 93f958f9c41f or stop logging the inode LSN
altogether. If we do the former, log recovery does not need to
change but we add 8 bytes of memory per inode to store what is
largely a write-only inode field. If we do the latter, log recovery
needs to stamp the on-disk inode in the same manner that inode
writeback does.
I prefer the latter, because we shouldn't really be trying to log
and replay changes to the on disk LSN as the on-disk value is the
canonical source of the on-disk version of the inode. It also
matches the way we recover buffer items - we create a buf_log_item
that carries the current recovery transaction LSN that gets stamped
into the buffer by the write verifier when it gets written back
when the transaction is fully recovered.
However, this might break log recovery on older kernels even more,
so I'm going to simply ignore the logged value in recovery and stamp
the on-disk inode with the LSN of the transaction being recovered
that will trigger writeback on transaction recovery completion. This
will ensure that the on-disk inode LSN always reflects the LSN of
the last change that was written to disk, regardless of whether it
comes from log recovery or runtime writeback.
Fixes: 93f958f9c41f ("xfs: cull unnecessary icdinode fields")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Before waiting on a iclog in xfs_log_force_lsn(), we don't check to
see if the iclog has already been completed and the contents on
stable storage. We check for completed iclogs in xfs_log_force(), so
we should do the same thing for xfs_log_force_lsn().
This fixed some random up-to-30s pauses seen in unmounting
filesystems in some tests. A log force ends up waiting on completed
iclog, and that doesn't then get flushed (and hence the log force
get completed) until the background log worker issues a log force
that flushes the iclog in question. Then the unmount unblocks and
continues.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
After fixing the tail_lsn vs cache flush race, generic/482 continued
to fail in a similar way where cache flushes were missing before
iclog FUA writes. Tracing of iclog state changes during the fsstress
workload portion of the test (via xlog_iclog* events) indicated that
iclog writes were coming from two sources - CIL pushes and log
forces (due to fsync/O_SYNC operations). All of the cases where a
recovery problem was triggered indicated that the log force was the
source of the iclog write that was not preceeded by a cache flush.
This was an oversight in the modifications made in commit
eef983ffeae7 ("xfs: journal IO cache flush reductions"). Log forces
for fsync imply a data device cache flush has been issued if an
iclog was flushed to disk and is indicated to the caller via the
log_flushed parameter so they can elide the device cache flush if
the journal issued one.
The change in eef983ffeae7 results in iclogs only issuing a cache
flush if XLOG_ICL_NEED_FLUSH is set on the iclog, but this was not
added to the iclogs that the log force code flushes to disk. Hence
log forces are no longer guaranteeing that a cache flush is issued,
hence opening up a potential on-disk ordering failure.
Log forces should also set XLOG_ICL_NEED_FUA as well to ensure that
the actual iclogs it forces to the journal are also on stable
storage before it returns to the caller.
This patch introduces the xlog_force_iclog() helper function to
encapsulate the process of taking a reference to an iclog, switching
its state if WANT_SYNC and flushing it to stable storage correctly.
Both xfs_log_force() and xfs_log_force_lsn() are converted to use
it, as is xlog_unmount_write() which has an elaborate method of
doing exactly the same "write this iclog to stable storage"
operation.
Further, if the log force code needs to wait on a iclog in the
WANT_SYNC state, it needs to ensure that iclog also results in a
cache flush being issued. This covers the case where the iclog
contains the commit record of the CIL flush that the log force
triggered, but it hasn't been written yet because there is still an
active reference to the iclog.
Note: this whole cache flush whack-a-mole patch is a result of log
forces still being iclog state centric rather than being CIL
sequence centric. Most of this nasty code will go away in future
when log forces are converted to wait on CIL sequence push
completion rather than iclog completion. With the CIL push algorithm
guaranteeing that the CIL checkpoint is fully on stable storage when
it completes, we no longer need to iterate iclogs and push them to
ensure a CIL sequence push has completed and so all this nasty iclog
iteration and flushing code will go away.
Fixes: eef983ffeae7 ("xfs: journal IO cache flush reductions")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
We force iclogs in several places - we need them all to have the
same cache flush semantics, so start by factoring out the iclog
force into a common helper.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
There is a race between the new CIL async data device metadata IO
completion cache flush and the log tail in the iclog the flush
covers being updated. This can be seen by repeating generic/482 in a
loop and eventually log recovery fails with a failures such as this:
XFS (dm-3): Starting recovery (logdev: internal)
XFS (dm-3): bad inode magic/vsn daddr 228352 #0 (magic=0)
XFS (dm-3): Metadata corruption detected at xfs_inode_buf_verify+0x180/0x190, xfs_inode block 0x37c00 xfs_inode_buf_verify
XFS (dm-3): Unmount and run xfs_repair
XFS (dm-3): First 128 bytes of corrupted metadata buffer:
00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000040: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000060: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
XFS (dm-3): metadata I/O error in "xlog_recover_items_pass2+0x55/0xc0" at daddr 0x37c00 len 32 error 117
Analysis of the logwrite replay shows that there were no writes to
the data device between the FUA @ write 124 and the FUA at write @
125, but log recovery @ 125 failed. The difference was the one log
write @ 125 moved the tail of the log forwards from (1,8) to (1,32)
and so the inode create intent in (1,8) was not replayed and so the
inode cluster was zero on disk when replay of the first inode item
in (1,32) was attempted.
What this meant was that the journal write that occurred at @ 125
did not ensure that metadata completed before the iclog was written
was correctly on stable storage. The tail of the log moved forward,
so IO must have been completed between the two iclog writes. This
means that there is a race condition between the unconditional async
cache flush in the CIL push work and the tail LSN that is written to
the iclog. This happens like so:
CIL push work AIL push work
------------- -------------
Add to committing list
start async data dev cache flush
.....
<flush completes>
<all writes to old tail lsn are stable>
xlog_write
.... push inode create buffer
<start IO>
.....
xlog_write(commit record)
.... <IO completes>
log tail moves
xlog_assign_tail_lsn()
start_lsn == commit_lsn
<no iclog preflush!>
xlog_state_release_iclog
__xlog_state_release_iclog()
<writes *new* tail_lsn into iclog>
xlog_sync()
....
submit_bio()
<tail in log moves forward without flushing written metadata>
Essentially, this can only occur if the commit iclog is issued
without a cache flush. If the iclog bio is submitted with
REQ_PREFLUSH, then it will guarantee that all the completed IO is
one stable storage before the iclog bio with the new tail LSN in it
is written to the log.
IOWs, the tail lsn that is written to the iclog needs to be sampled
*before* we issue the cache flush that guarantees all IO up to that
LSN has been completed.
To fix this without giving up the performance advantage of the
flush/FUA optimisations (e.g. g/482 runtime halves with 5.14-rc1
compared to 5.13), we need to ensure that we always issue a cache
flush if the tail LSN changes between the initial async flush and
the commit record being written. THis requires sampling the tail_lsn
before we start the flush, and then passing the sampled tail LSN to
xlog_state_release_iclog() so it can determine if the the tail LSN
has changed while writing the checkpoint. If the tail LSN has
changed, then it needs to set the NEED_FLUSH flag on the iclog and
we'll issue another cache flush before writing the iclog.
Fixes: eef983ffeae7 ("xfs: journal IO cache flush reductions")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
Fold __xlog_state_release_iclog into its only caller to prepare
make an upcoming fix easier.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
[hch: split from a larger patch]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
The recent journal flush/FUA changes replaced the flushing of the
data device on every iclog write with an up-front async data device
cache flush. Unfortunately, the assumption of which this was based
on has been proven incorrect by the flush vs log tail update
ordering issue. As the fix for that issue uses the
XLOG_ICL_NEED_FLUSH flag to indicate that data device needs a cache
flush, we now need to (once again) ensure that an iclog write to
external logs that need a cache flush to be issued actually issue a
cache flush to the data device as well as the log device.
Fixes: eef983ffeae7 ("xfs: journal IO cache flush reductions")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
|
We incorrectly flush the log device instead of the data device when
trying to ensure metadata is correctly on disk before writing the
unmount record.
Fixes: eef983ffeae7 ("xfs: journal IO cache flush reductions")
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
|
Jason Gunthorpe
Yishai Hadas
Max Gurtovoy
Christoph Hellwig
Linus Torvalds
Arnaldo Carvalho de Melo
Wang Hai
Shakeel Butt
Aneesh Kumar K.V
Johannes Weiner
Junxiao Bi
Matteo Croce
Catherine Sullivan
Lukas Bulwahn
Pavel Skripkin
Dan Carpenter
Marc Kleine-Budde
Gustavo A. R. Silva
Darrick J. Wong
Dave Chinner