commit a1a7e3a36e01ca6e67014f8cf673cb8e47be5550 upstream.
Without doing verify_sec_ctx_len() check in xfrm_add_acquire(), it may be
out-of-bounds to access uctx->ctx_str with uctx->ctx_len, as noticed by
syz:
BUG: KASAN: slab-out-of-bounds in selinux_xfrm_alloc_user+0x237/0x430
Read of size 768 at addr ffff8880123be9b4 by task syz-executor.1/11650
Call Trace:
dump_stack+0xe8/0x16e
print_address_description.cold.3+0x9/0x23b
kasan_report.cold.4+0x64/0x95
memcpy+0x1f/0x50
selinux_xfrm_alloc_user+0x237/0x430
security_xfrm_policy_alloc+0x5c/0xb0
xfrm_policy_construct+0x2b1/0x650
xfrm_add_acquire+0x21d/0xa10
xfrm_user_rcv_msg+0x431/0x6f0
netlink_rcv_skb+0x15a/0x410
xfrm_netlink_rcv+0x6d/0x90
netlink_unicast+0x50e/0x6a0
netlink_sendmsg+0x8ae/0xd40
sock_sendmsg+0x133/0x170
___sys_sendmsg+0x834/0x9a0
__sys_sendmsg+0x100/0x1e0
do_syscall_64+0xe5/0x660
entry_SYSCALL_64_after_hwframe+0x6a/0xdf
So fix it by adding the missing verify_sec_ctx_len check there.
Fixes: 980ebd2579 ("[IPSEC]: Sync series - acquire insert")
Reported-by: Hangbin Liu <[email protected]>
Signed-off-by: Xin Long <[email protected]>
Signed-off-by: Steffen Klassert <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 171d449a028573b2f0acdc7f31ecbb045391b320 upstream.
It's not sufficient to do 'uctx->len != (sizeof(struct xfrm_user_sec_ctx) +
uctx->ctx_len)' check only, as uctx->len may be greater than nla_len(rt),
in which case it will cause slab-out-of-bounds when accessing uctx->ctx_str
later.
This patch is to fix it by return -EINVAL when uctx->len > nla_len(rt).
Fixes: df71837d50 ("[LSM-IPSec]: Security association restriction.")
Signed-off-by: Xin Long <[email protected]>
Signed-off-by: Steffen Klassert <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit ba80013fba656b9830ef45cd40a6a1e44707f47a upstream.
It has been discovered that this feature can globally block the RX port,
so it should be allowed for highly privileged users only.
Fixes: 03404e8ae652("IB/mlx5: Add support to dropless RQ")
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Maor Gottlieb <[email protected]>
Signed-off-by: Leon Romanovsky <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit df81dfcfd6991d547653d46c051bac195cd182c1 upstream.
The handling of notify->work did not properly maintain notify->kref in two
cases:
1) where the work was already scheduled, another irq_set_affinity_locked()
would get the ref and (no-op-ly) schedule the work. Thus when
irq_affinity_notify() ran, it would drop the original ref but not the
additional one.
2) when cancelling the (old) work in irq_set_affinity_notifier(), if there
was outstanding work a ref had been got for it but was never put.
Fix both by checking the return values of the work handling functions
(schedule_work() for (1) and cancel_work_sync() for (2)) and put the
extra ref if the return value indicates preexisting work.
Fixes: cd7eab44e9 ("genirq: Add IRQ affinity notifiers")
Fixes: 59c39840f5 ("genirq: Prevent use-after-free and work list corruption")
Signed-off-by: Edward Cree <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Ben Hutchings <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit dde9f095583b3f375ba23979045ee10dfcebec2f upstream.
When an AFS service handler function aborts a call, AF_RXRPC marks the call
as complete - which means that it's not going to get any more packets from
the receiver. This is a problem because reception of the final ACK is what
triggers afs_deliver_to_call() to drop the final ref on the afs_call
object.
Instead, aborted AFS service calls may then just sit around waiting for
ever or until they're displaced by a new call on the same connection
channel or a connection-level abort.
Fix this by calling afs_set_call_complete() to finalise the afs_call struct
representing the call.
However, we then need to drop the ref that stops the call from being
deallocated. We can do this in afs_set_call_complete(), as the work queue
is holding a separate ref of its own, but then we shouldn't do it in
afs_process_async_call() and afs_delete_async_call().
call->drop_ref is set to indicate that a ref needs dropping for a call and
this is dealt with when we transition a call to AFS_CALL_COMPLETE.
But then we also need to get rid of the ref that pins an asynchronous
client call. We can do this by the same mechanism, setting call->drop_ref
for an async client call too.
We can also get rid of call->incoming since nothing ever sets it and only
one thing ever checks it (futilely).
A trace of the rxrpc_call and afs_call struct ref counting looks like:
<idle>-0 [001] ..s5 164.764892: rxrpc_call: c=00000002 SEE u=3 sp=rxrpc_new_incoming_call+0x473/0xb34 a=00000000442095b5
<idle>-0 [001] .Ns5 164.766001: rxrpc_call: c=00000002 QUE u=4 sp=rxrpc_propose_ACK+0xbe/0x551 a=00000000442095b5
<idle>-0 [001] .Ns4 164.766005: rxrpc_call: c=00000002 PUT u=3 sp=rxrpc_new_incoming_call+0xa3f/0xb34 a=00000000442095b5
<idle>-0 [001] .Ns7 164.766433: afs_call: c=00000002 WAKE u=2 o=11 sp=rxrpc_notify_socket+0x196/0x33c
kworker/1:2-1810 [001] ...1 164.768409: rxrpc_call: c=00000002 SEE u=3 sp=rxrpc_process_call+0x25/0x7ae a=00000000442095b5
kworker/1:2-1810 [001] ...1 164.769439: rxrpc_tx_packet: c=00000002 e9f1a7a8:95786a88:00000008:09c5 00000001 00000000 02 22 ACK CallAck
kworker/1:2-1810 [001] ...1 164.769459: rxrpc_call: c=00000002 PUT u=2 sp=rxrpc_process_call+0x74f/0x7ae a=00000000442095b5
kworker/1:2-1810 [001] ...1 164.770794: afs_call: c=00000002 QUEUE u=3 o=12 sp=afs_deliver_to_call+0x449/0x72c
kworker/1:2-1810 [001] ...1 164.770829: afs_call: c=00000002 PUT u=2 o=12 sp=afs_process_async_call+0xdb/0x11e
kworker/1:2-1810 [001] ...2 164.771084: rxrpc_abort: c=00000002 95786a88:00000008 s=0 a=1 e=1 K-1
kworker/1:2-1810 [001] ...1 164.771461: rxrpc_tx_packet: c=00000002 e9f1a7a8:95786a88:00000008:09c5 00000002 00000000 04 00 ABORT CallAbort
kworker/1:2-1810 [001] ...1 164.771466: afs_call: c=00000002 PUT u=1 o=12 sp=SRXAFSCB_ProbeUuid+0xc1/0x106
The abort generated in SRXAFSCB_ProbeUuid(), labelled "K-1", indicates that
the local filesystem/cache manager didn't recognise the UUID as its own.
Fixes: 2067b2b3f4 ("afs: Fix the CB.ProbeUuid service handler to reply correctly")
Signed-off-by: David Howells <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 2d47fbacf2725a67869f4d3634c2415e7dfab2f4 upstream.
The following modify sequence (loosely based on ipoib) will lose a pkey
modifcation:
- Modify (pkey index, port)
- Modify (new pkey index, NO port)
After the first modify, the qp_pps list will have saved the pkey and the
unit on the main list.
During the second modify, get_new_pps() will fetch the port from qp_pps
and read the new pkey index from qp_attr->pkey_index. The state will
still be zero, or IB_PORT_PKEY_NOT_VALID. Because of the invalid state,
the new values will never replace the one in the qp pps list, losing the
new pkey.
This happens because the following if statements will never correct the
state because the first term will be false. If the code had been executed,
it would incorrectly overwrite valid values.
if ((qp_attr_mask & IB_QP_PKEY_INDEX) && (qp_attr_mask & IB_QP_PORT))
new_pps->main.state = IB_PORT_PKEY_VALID;
if (!(qp_attr_mask & (IB_QP_PKEY_INDEX | IB_QP_PORT)) && qp_pps) {
new_pps->main.port_num = qp_pps->main.port_num;
new_pps->main.pkey_index = qp_pps->main.pkey_index;
if (qp_pps->main.state != IB_PORT_PKEY_NOT_VALID)
new_pps->main.state = IB_PORT_PKEY_VALID;
}
Fix by joining the two if statements with an or test to see if qp_pps is
non-NULL and in the correct state.
Fixes: 1dd017882e01 ("RDMA/core: Fix protection fault in get_pkey_idx_qp_list")
Link: https://lore.kernel.org/r/[email protected]
Reviewed-by: Kaike Wan <[email protected]>
Signed-off-by: Mike Marciniszyn <[email protected]>
Reviewed-by: Leon Romanovsky <[email protected]>
Signed-off-by: Jason Gunthorpe <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 1d8006abaab4cb90f81add86e8d1bf9411add05a upstream.
There is no compensating cgroup_bpf_put() for each ancestor cgroup in
cgroup_bpf_inherit(). If compute_effective_progs returns error, those cgroups
won't be freed ever. Fix it by putting them in cleanup code path.
Fixes: e10360f815ca ("bpf: cgroup: prevent out-of-order release of cgroup bpf")
Signed-off-by: Andrii Nakryiko <[email protected]>
Signed-off-by: Alexei Starovoitov <[email protected]>
Acked-by: Roman Gushchin <[email protected]>
Link: https://lore.kernel.org/bpf/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 0e91506ba00730f088961a8d39f8693b0f8e3fea upstream.
Commit aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option +
quirk mechanism") was added to deal with spurious wakeups on one specific
model of the HP x2 10 series. In the mean time I have learned that there
are at least 3 different HP x2 10 models:
Bay Trail SoC + AXP288 PMIC
Cherry Trail SoC + AXP288 PMIC
Cherry Trail SoC + TI PMIC
And the original quirk is only correct for (and only matches the)
Cherry Trail SoC + TI PMIC model.
The Bay Trail SoC + AXP288 PMIC model has different DMI strings, has
the external EC interrupt on a different GPIO pin and only needs to ignore
wakeups on the EC interrupt, the INT0002 device works fine on this model.
This commit adds an extra DMI based quirk for the HP x2 10 BYT + AXP288
model, ignoring wakeups for ACPI GPIO events on the EC interrupt pin
on this model. This fixes spurious wakeups from suspend on this model.
Fixes: aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option + quirk mechanism")
Signed-off-by: Hans de Goede <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Mika Westerberg <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 2ccb21f5516afef5e251184eeefbf36db90206d7 upstream.
Commit aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option +
quirk mechanism") was added to deal with spurious wakeups on one specific
model of the HP x2 10 series.
The approach taken there was to add a bool controlling wakeup support for
all ACPI GPIO events. This was sufficient for the specific HP x2 10 model
the commit was trying to fix, but in the mean time other models have
turned up which need a similar workaround to avoid spurious wakeups from
suspend, but only for one of the pins on which the ACPI tables request
ACPI GPIO events.
Since the honor_wakeup option was added to be able to ignore wake events,
the name was perhaps not the best, this commit renames it to ignore_wake
and changes it to a string with the following format:
gpiolib_acpi.ignore_wake=controller@pin[,controller@pin[,...]]
This allows working around spurious wakeup issues on a per pin basis.
This commit also reworks the existing quirk for the HP x2 10 so that
it functions as before.
Note:
-This removes the honor_wakeup parameter. This has only been upstream for
a short time and to the best of my knowledge there are no users using
this module parameter.
-The controller@pin[,controller@pin[,...]] syntax is based on an existing
kernel module parameter using the same controller@pin format. That version
uses ';' as separator, but in practice that is problematic because grub2
cannot handle this without taking special care to escape the ';', so here
we are using a ',' as separator instead which does not have this issue.
Fixes: aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option + quirk mechanism")
Signed-off-by: Hans de Goede <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Mika Westerberg <[email protected]>
Reviewed-by: Andy Shevchenko <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit efaa87fa0947d525cf7c075316adde4e3ac7720b upstream.
Commit aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option +
quirk mechanism") added a quirk for some models of the HP x2 10 series.
There are 2 issues with the comment describing the quirk:
1) The comment claims the DMI quirk applies to all Cherry Trail based HP x2
10 models. In the mean time I have learned that there are at least 3
models of the HP x2 10 models:
Bay Trail SoC + AXP288 PMIC
Cherry Trail SoC + AXP288 PMIC
Cherry Trail SoC + TI PMIC
And this quirk's DMI matches only match the Cherry Trail SoC + TI PMIC
SoC, which is good because we want a slightly different quirk for the
others. This commit updates the comment to make it clear that the quirk
is only for the Cherry Trail SoC + TI PMIC models.
2) The comment says that it is ok to disable wakeup on all ACPI GPIO event
handlers, because there is only the one for the embedded-controller
events. This is not true, there also is a handler for the special
INT0002 device which is related to USB wakeups. We need to also disable
wakeups on that one because the device turns of the USB-keyboard built
into the dock when closing the lid. The XHCI controller takes a while
to notice this, so it only notices it when already suspended, causing
a spurious wakeup because of this. So disabling wakeup on all handlers
is the right thing to do, but not because there only is the one handler
for the EC events. This commit updates the comment to correctly reflect
this.
Fixes: aa23ca3d98f7 ("gpiolib: acpi: Add honor_wakeup module-option + quirk mechanism")
Signed-off-by: Hans de Goede <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Mika Westerberg <[email protected]>
Signed-off-by: Linus Walleij <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 8380ce479010f2f779587b462a9b4681934297c3 upstream.
Depending on CONFIG_VMAP_STACK and the THREAD_SIZE / PAGE_SIZE ratio the
space for task stacks can be allocated using __vmalloc_node_range(),
alloc_pages_node() and kmem_cache_alloc_node().
In the first and the second cases page->mem_cgroup pointer is set, but
in the third it's not: memcg membership of a slab page should be
determined using the memcg_from_slab_page() function, which looks at
page->slab_cache->memcg_params.memcg . In this case, using
mod_memcg_page_state() (as in account_kernel_stack()) is incorrect:
page->mem_cgroup pointer is NULL even for pages charged to a non-root
memory cgroup.
It can lead to kernel_stack per-memcg counters permanently showing 0 on
some architectures (depending on the configuration).
In order to fix it, let's introduce a mod_memcg_obj_state() helper,
which takes a pointer to a kernel object as a first argument, uses
mem_cgroup_from_obj() to get a RCU-protected memcg pointer and calls
mod_memcg_state(). It allows to handle all possible configurations
(CONFIG_VMAP_STACK and various THREAD_SIZE/PAGE_SIZE values) without
spilling any memcg/kmem specifics into fork.c .
Note: This is a special version of the patch created for stable
backports. It contains code from the following two patches:
- mm: memcg/slab: introduce mem_cgroup_from_obj()
- mm: fork: fix kernel_stack memcg stats for various stack implementations
[[email protected]: introduce mem_cgroup_from_obj()]
Link: http://lkml.kernel.org/r/[email protected]
Fixes: 4d96ba3530 ("mm: memcg/slab: stop setting page->mem_cgroup pointer for slab pages")
Signed-off-by: Roman Gushchin <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Shakeel Butt <[email protected]>
Acked-by: Johannes Weiner <[email protected]>
Cc: Michal Hocko <[email protected]>
Cc: Bharata B Rao <[email protected]>
Cc: Shakeel Butt <[email protected]>
Cc: <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit b943f045a9af9fd02f923e43fe8d7517e9961701 upstream.
Fix the crash like this:
BUG: Kernel NULL pointer dereference on read at 0x00000000
Faulting instruction address: 0xc000000000c3447c
Oops: Kernel access of bad area, sig: 11 [#1]
LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
CPU: 11 PID: 7519 Comm: lt-ndctl Not tainted 5.6.0-rc7-autotest #1
...
NIP [c000000000c3447c] vmemmap_populated+0x98/0xc0
LR [c000000000088354] vmemmap_free+0x144/0x320
Call Trace:
section_deactivate+0x220/0x240
__remove_pages+0x118/0x170
arch_remove_memory+0x3c/0x150
memunmap_pages+0x1cc/0x2f0
devm_action_release+0x30/0x50
release_nodes+0x2f8/0x3e0
device_release_driver_internal+0x168/0x270
unbind_store+0x130/0x170
drv_attr_store+0x44/0x60
sysfs_kf_write+0x68/0x80
kernfs_fop_write+0x100/0x290
__vfs_write+0x3c/0x70
vfs_write+0xcc/0x240
ksys_write+0x7c/0x140
system_call+0x5c/0x68
The crash is due to NULL dereference at
test_bit(idx, ms->usage->subsection_map);
due to ms->usage = NULL in pfn_section_valid()
With commit d41e2f3bd546 ("mm/hotplug: fix hot remove failure in
SPARSEMEM|!VMEMMAP case") section_mem_map is set to NULL after
depopulate_section_mem(). This was done so that pfn_page() can work
correctly with kernel config that disables SPARSEMEM_VMEMMAP. With that
config pfn_to_page does
__section_mem_map_addr(__sec) + __pfn;
where
static inline struct page *__section_mem_map_addr(struct mem_section *section)
{
unsigned long map = section->section_mem_map;
map &= SECTION_MAP_MASK;
return (struct page *)map;
}
Now with SPASEMEM_VMEMAP enabled, mem_section->usage->subsection_map is
used to check the pfn validity (pfn_valid()). Since section_deactivate
release mem_section->usage if a section is fully deactivated,
pfn_valid() check after a subsection_deactivate cause a kernel crash.
static inline int pfn_valid(unsigned long pfn)
{
...
return early_section(ms) || pfn_section_valid(ms, pfn);
}
where
static inline int pfn_section_valid(struct mem_section *ms, unsigned long pfn)
{
int idx = subsection_map_index(pfn);
return test_bit(idx, ms->usage->subsection_map);
}
Avoid this by clearing SECTION_HAS_MEM_MAP when mem_section->usage is
freed. For architectures like ppc64 where large pages are used for
vmmemap mapping (16MB), a specific vmemmap mapping can cover multiple
sections. Hence before a vmemmap mapping page can be freed, the kernel
needs to make sure there are no valid sections within that mapping.
Clearing the section valid bit before depopulate_section_memap enables
this.
[[email protected]: add comment]
Link: http://lkml.kernel.org/r/[email protected]: http://lkml.kernel.org/r/[email protected]
Fixes: d41e2f3bd546 ("mm/hotplug: fix hot remove failure in SPARSEMEM|!VMEMMAP case")
Reported-by: Sachin Sant <[email protected]>
Signed-off-by: Aneesh Kumar K.V <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Tested-by: Sachin Sant <[email protected]>
Reviewed-by: Baoquan He <[email protected]>
Reviewed-by: Wei Yang <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Acked-by: Pankaj Gupta <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: David Hildenbrand <[email protected]>
Cc: Oscar Salvador <[email protected]>
Cc: Mike Rapoport <[email protected]>
Cc: <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 53cdc1cb29e87ce5a61de5bb393eb08925d14ede upstream.
We see multiple issues with the implementation/interface to compute
whether a memory block can be offlined (exposed via
/sys/devices/system/memory/memoryX/removable) and would like to simplify
it (remove the implementation).
1. It runs basically lockless. While this might be good for performance,
we see possible races with memory offlining that will require at
least some sort of locking to fix.
2. Nowadays, more false positives are possible. No arch-specific checks
are performed that validate if memory offlining will not be denied
right away (and such check will require locking). For example, arm64
won't allow to offline any memory block that was added during boot -
which will imply a very high error rate. Other archs have other
constraints.
3. The interface is inherently racy. E.g., if a memory block is detected
to be removable (and was not a false positive at that time), there is
still no guarantee that offlining will actually succeed. So any
caller already has to deal with false positives.
4. It is unclear which performance benefit this interface actually
provides. The introducing commit 5c755e9fd8 ("memory-hotplug: add
sysfs removable attribute for hotplug memory remove") mentioned
"A user-level agent must be able to identify which sections
of memory are likely to be removable before attempting the
potentially expensive operation."
However, no actual performance comparison was included.
Known users:
- lsmem: Will group memory blocks based on the "removable" property. [1]
- chmem: Indirect user. It has a RANGE mode where one can specify
removable ranges identified via lsmem to be offlined. However,
it also has a "SIZE" mode, which allows a sysadmin to skip the
manual "identify removable blocks" step. [2]
- powerpc-utils: Uses the "removable" attribute to skip some memory
blocks right away when trying to find some to offline+remove.
However, with ballooning enabled, it already skips this
information completely (because it once resulted in many false
negatives). Therefore, the implementation can deal with false
positives properly already. [3]
According to Nathan Fontenot, DLPAR on powerpc is nowadays no longer
driven from userspace via the drmgr command (powerpc-utils). Nowadays
it's managed in the kernel - including onlining/offlining of memory
blocks - triggered by drmgr writing to /sys/kernel/dlpar. So the
affected legacy userspace handling is only active on old kernels. Only
very old versions of drmgr on a new kernel (unlikely) might execute
slower - totally acceptable.
With CONFIG_MEMORY_HOTREMOVE, always indicating "removable" should not
break any user space tool. We implement a very bad heuristic now.
Without CONFIG_MEMORY_HOTREMOVE we cannot offline anything, so report
"not removable" as before.
Original discussion can be found in [4] ("[PATCH RFC v1] mm:
is_mem_section_removable() overhaul").
Other users of is_mem_section_removable() will be removed next, so that
we can remove is_mem_section_removable() completely.
[1] http://man7.org/linux/man-pages/man1/lsmem.1.html
[2] http://man7.org/linux/man-pages/man8/chmem.8.html
[3] https://github.com/ibm-power-utilities/powerpc-utils
[4] https://lkml.kernel.org/r/[email protected]
Also, this patch probably fixes a crash reported by Steve.
http://lkml.kernel.org/r/CAPcyv4jpdaNvJ67SkjyUJLBnBnXXQv686BiVW042g03FUmWLXw@mail.gmail.com
Reported-by: "Scargall, Steve" <[email protected]>
Suggested-by: Michal Hocko <[email protected]>
Signed-off-by: David Hildenbrand <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Reviewed-by: Nathan Fontenot <[email protected]>
Acked-by: Michal Hocko <[email protected]>
Cc: Dan Williams <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: "Rafael J. Wysocki" <[email protected]>
Cc: Badari Pulavarty <[email protected]>
Cc: Robert Jennings <[email protected]>
Cc: Heiko Carstens <[email protected]>
Cc: Karel Zak <[email protected]>
Cc: <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit d795a90e2ba024dbf2f22107ae89c210b98b08b8 upstream.
claim_swapfile() currently keeps the inode locked when it is successful,
or the file is already swapfile (with -EBUSY). And, on the other error
cases, it does not lock the inode.
This inconsistency of the lock state and return value is quite confusing
and actually causing a bad unlock balance as below in the "bad_swap"
section of __do_sys_swapon().
This commit fixes this issue by moving the inode_lock() and IS_SWAPFILE
check out of claim_swapfile(). The inode is unlocked in
"bad_swap_unlock_inode" section, so that the inode is ensured to be
unlocked at "bad_swap". Thus, error handling codes after the locking now
jumps to "bad_swap_unlock_inode" instead of "bad_swap".
=====================================
WARNING: bad unlock balance detected!
5.5.0-rc7+ #176 Not tainted
-------------------------------------
swapon/4294 is trying to release lock (&sb->s_type->i_mutex_key) at: __do_sys_swapon+0x94b/0x3550
but there are no more locks to release!
other info that might help us debug this:
no locks held by swapon/4294.
stack backtrace:
CPU: 5 PID: 4294 Comm: swapon Not tainted 5.5.0-rc7-BTRFS-ZNS+ #176
Hardware name: ASUS All Series/H87-PRO, BIOS 2102 07/29/2014
Call Trace:
dump_stack+0xa1/0xea
print_unlock_imbalance_bug.cold+0x114/0x123
lock_release+0x562/0xed0
up_write+0x2d/0x490
__do_sys_swapon+0x94b/0x3550
__x64_sys_swapon+0x54/0x80
do_syscall_64+0xa4/0x4b0
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x7f15da0a0dc7
Fixes: 1638045c36 ("mm: set S_SWAPFILE on blockdev swap devices")
Signed-off-by: Naohiro Aota <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Tested-by: Qais Youef <[email protected]>
Reviewed-by: Andrew Morton <[email protected]>
Reviewed-by: Darrick J. Wong <[email protected]>
Cc: Christoph Hellwig <[email protected]>
Cc: <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Linus Torvalds <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit a0761a301746ec2d92d7fcb82af69c0a6a4339aa upstream.
If we know that we have an encrypted link (based on having had
a key configured for TX in the past) then drop all data frames
in the key selection handler if there's no key anymore.
This fixes an issue with mac80211 internal TXQs - there we can
buffer frames for an encrypted link, but then if the key is no
longer there when they're dequeued, the frames are sent without
encryption. This happens if a station is disconnected while the
frames are still on the TXQ.
Detecting that a link should be encrypted based on a first key
having been configured for TX is fine as there are no use cases
for a connection going from with encryption to no encryption.
With extended key IDs, however, there is a case of having a key
configured for only decryption, so we can't just trigger this
behaviour on a key being configured.
Cc: [email protected]
Reported-by: Jouni Malinen <[email protected]>
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Link: https://lore.kernel.org/r/iwlwifi.20200326150855.6865c7f28a14.I9fb1d911b064262d33e33dfba730cdeef83926ca@changeid
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit ea697a8bf5a4161e59806fab14f6e4a46dc7dcb0 upstream.
Some USB bridge devices will return a default set of characteristics during
initialization. And then, once an attached drive has spun up, substitute
the actual parameters reported by the drive. According to the SCSI spec,
the device should return a UNIT ATTENTION in case any reported parameters
change. But in this case the change is made silently after a small window
where default values are reported.
Commit a83da8a450 ("scsi: sd: Optimal I/O size should be a multiple of
physical block size") validated the reported optimal I/O size against the
physical block size to overcome problems with devices reporting nonsensical
transfer sizes. However, this validation did not account for the fact that
aforementioned devices will return default values during a brief window
during spin-up. The subsequent change in reported characteristics would
invalidate the checking that had previously been performed.
Unset a previously configured optimal I/O size should the sanity checking
fail on subsequent revalidate attempts.
Link: https://lore.kernel.org/r/[email protected]
Cc: Bryan Gurney <[email protected]>
Cc: <[email protected]>
Reported-by: Bernhard Sulzer <[email protected]>
Tested-by: Bernhard Sulzer <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit e33a814e772cdc36436c8c188d8c42d019fda639 upstream.
gcc 10 will default to -fno-common, which causes this error at link
time:
(.text+0x0): multiple definition of `yylloc'; dtc-lexer.lex.o (symbol from plugin):(.text+0x0): first defined here
This is because both dtc-lexer as well as dtc-parser define the same
global symbol yyloc. Before with -fcommon those were merged into one
defintion. The proper solution would be to to mark this as "extern",
however that leads to:
dtc-lexer.l:26:16: error: redundant redeclaration of 'yylloc' [-Werror=redundant-decls]
26 | extern YYLTYPE yylloc;
| ^~~~~~
In file included from dtc-lexer.l:24:
dtc-parser.tab.h:127:16: note: previous declaration of 'yylloc' was here
127 | extern YYLTYPE yylloc;
| ^~~~~~
cc1: all warnings being treated as errors
which means the declaration is completely redundant and can just be
dropped.
Signed-off-by: Dirk Mueller <[email protected]>
Signed-off-by: David Gibson <[email protected]>
[robh: cherry-pick from upstream]
Cc: [email protected]
Signed-off-by: Rob Herring <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit be40920fbf1003c38ccdc02b571e01a75d890c82 upstream.
When I tried to compile tools/perf from the top directory with the -C
option, the O= option didn't work correctly if I passed a relative path:
$ make O=BUILD -C tools/perf/
make: Entering directory '/home/mhiramat/ksrc/linux/tools/perf'
BUILD: Doing 'make -j8' parallel build
../scripts/Makefile.include:4: *** O=/home/mhiramat/ksrc/linux/tools/perf/BUILD does not exist. Stop.
make: *** [Makefile:70: all] Error 2
make: Leaving directory '/home/mhiramat/ksrc/linux/tools/perf'
The O= directory existence check failed because the check script ran in
the build target directory instead of the directory where I ran the make
command.
To fix that, once change directory to $(PWD) and check O= directory,
since the PWD is set to where the make command runs.
Fixes: c883122acc ("perf tools: Let O= makes handle relative paths")
Reported-by: Randy Dunlap <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Borislav Petkov <[email protected]>
Cc: Geert Uytterhoeven <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Masahiro Yamada <[email protected]>
Cc: Michal Marek <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sasha Levin <[email protected]>
Cc: Steven Rostedt (VMware) <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/158351957799.3363.15269768530697526765.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit c80b18cbb04b7b101af9bd14550f13d9866c646a upstream.
For some unexplained reason, commit d1d1a96bdb ("rtlwifi: rtl8188ee:
Remove local configuration variable") broke at least one system. As
the only net effect of the change was to remove 2 bytes from the start
of struct phy_status_rpt, this patch adds 2 bytes of padding at the
beginning of the struct.
Fixes: d1d1a96bdb ("rtlwifi: rtl8188ee: Remove local configuration variable")
Cc: Stable <[email protected]> # V5.4+
Reported-by: Ashish <[email protected]>
Tested-by: Ashish <[email protected]>
Signed-off-by: Larry Finger <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 1efde2754275dbd9d11c6e0132a4f09facf297ab upstream.
Do not depend on dwfl_module_addrsym() because it can fail on user-space
shared libraries.
Actually, same bug was fixed by commit 664fee3dc3 ("perf probe: Do not
use dwfl_module_addrsym if dwarf_diename finds symbol name"), but commit
07d369857808 ("perf probe: Fix wrong address verification) reverted to
get actual symbol address from symtab.
This fixes it again by getting symbol address from DIE, and only if the
DIE has only address range, it uses dwfl_module_addrsym().
Fixes: 07d369857808 ("perf probe: Fix wrong address verification)
Reported-by: Alexandre Ghiti <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Tested-by: Alexandre Ghiti <[email protected]>
Cc: Alexander Shishkin <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Sasha Levin <[email protected]>
Link: http://lore.kernel.org/lkml/158281812176.476.14164573830975116234.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 6b8d68f1ce9266b05a55e93c62923ff51daae4c1 upstream.
When we put an event with multiple probes, perf-probe fails to delete
with filters. This comes from a failure to list up the event name
because of overwrapping its name.
To fix this issue, skip to list up the event which has same name.
Without this patch:
# perf probe -l \*
probe_perf:map__map_ip (on perf_sample__fprintf_brstackoff:21@
probe_perf:map__map_ip (on perf_sample__fprintf_brstackoff:25@
probe_perf:map__map_ip (on append_inlines:12@util/machine.c in
probe_perf:map__map_ip (on unwind_entry:19@util/machine.c in /
probe_perf:map__map_ip (on map__map_ip@util/map.h in /home/mhi
probe_perf:map__map_ip (on map__map_ip@util/map.h in /home/mhi
# perf probe -d \*
"*" does not hit any event.
Error: Failed to delete events. Reason: No such file or directory (Code: -2)
With it:
# perf probe -d \*
Removed event: probe_perf:map__map_ip
#
Fixes: 72363540c009 ("perf probe: Support multiprobe event")
Reported-by: Arnaldo Carvalho de Melo <[email protected]>
Reported-by: He Zhe <[email protected]>
Signed-off-by: Masami Hiramatsu <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Adrian Hunter <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Cc: [email protected]
Link: http://lore.kernel.org/lkml/158287666197.16697.7514373548551863562.stgit@devnote2
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 870b4333a62e45b0b2000d14b301b7b8b8cad9da upstream.
In order to use efi_mem_type(), one needs CONFIG_EFI enabled. Otherwise
that function is undefined. Use IS_ENABLED() to check and avoid the
ifdeffery as the compiler optimizes away the following unreachable code
then.
Fixes: 985e537a4082 ("x86/ioremap: Map EFI runtime services data as encrypted for SEV")
Reported-by: Randy Dunlap <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Randy Dunlap <[email protected]>
Cc: Tom Lendacky <[email protected]>
Cc: <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit dfa7ea303f56a3a8b1ed3b91ef35af2da67ca4ee upstream.
The L3 interconnect's memory map is from 0x0 to
0xffffffff. Out of this, System memory (SDRAM) can be
accessed from 0x80000000 to 0xffffffff (2GB)
OMAP5 does support 4GB of SDRAM but upper 2GB can only be
accessed by the MPU subsystem.
Add the dma-ranges property to reflect the physical address limit
of the L3 bus.
Cc: [email protected]
Signed-off-by: Roger Quadros <[email protected]>
Signed-off-by: Tony Lindgren <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit cfb5d65f25959f724081bae8445a0241db606af6 upstream.
The L3 interconnect's memory map is from 0x0 to
0xffffffff. Out of this, System memory (SDRAM) can be
accessed from 0x80000000 to 0xffffffff (2GB)
DRA7 does support 4GB of SDRAM but upper 2GB can only be
accessed by the MPU subsystem.
Add the dma-ranges property to reflect the physical address limit
of the L3 bus.
Issues ere observed only with SATA on DRA7-EVM with 4GB RAM
and CONFIG_ARM_LPAE enabled. This is because the controller
supports 64-bit DMA and its driver sets the dma_mask to 64-bit
thus resulting in DMA accesses beyond L3 limit of 2G.
Setting the correct bus_dma_limit fixes the issue.
Signed-off-by: Roger Quadros <[email protected]>
Cc: [email protected]
Signed-off-by: Tony Lindgren <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 7614209736fbc4927584d4387faade4f31444fce upstream.
CEPH_OSDMAP_FULL/NEARFULL aren't set since mimic, so we need to consult
per-pool flags as well. Unfortunately the backwards compatibility here
is lacking:
- the change that deprecated OSDMAP_FULL/NEARFULL went into mimic, but
was guarded by require_osd_release >= RELEASE_LUMINOUS
- it was subsequently backported to luminous in v12.2.2, but that makes
no difference to clients that only check OSDMAP_FULL/NEARFULL because
require_osd_release is not client-facing -- it is for OSDs
Since all kernels are affected, the best we can do here is just start
checking both map flags and pool flags and send that to stable.
These checks are best effort, so take osdc->lock and look up pool flags
just once. Remove the FIXME, since filesystem quotas are checked above
and RADOS quotas are reflected in POOL_FLAG_FULL: when the pool reaches
its quota, both POOL_FLAG_FULL and POOL_FLAG_FULL_QUOTA are set.
Cc: [email protected]
Reported-by: Yanhu Cao <[email protected]>
Signed-off-by: Ilya Dryomov <[email protected]>
Reviewed-by: Jeff Layton <[email protected]>
Acked-by: Sage Weil <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 8959b304c7062889b1276092cc8590dc1ba98f65 upstream.
The implementation if .irq_disable() which kicks in between
the gpiolib and the driver is not properly mimicking the
expected semantics of the irqchip core: the irqchip will
call .irq_disable() if that exists, else it will call
mask_irq() which first checks if .irq_mask() is defined
before calling it.
Since we are calling it unconditionally, we get this bug
from drivers/pinctrl/qcom/pinctrl-ssbi-gpio.c, as it only
defines .irq_mask_ack and not .irq_mask:
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = (ptrval)
(...)
PC is at 0x0
LR is at gpiochip_irq_disable+0x20/0x30
Fix this by only calling .irq_mask() if it exists.
Cc: Brian Masney <[email protected]>
Cc: Hans Verkuil <[email protected]>
Cc: [email protected]
Reviewed-by: Bartosz Golaszewski <[email protected]>
Fixes: 461c1a7d47 ("gpiolib: override irq_enable/disable")
Signed-off-by: Linus Walleij <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 52afa505a03d914081f40cb869a3248567a57573 upstream.
The commit 19ba1eb15a ("Input: psmouse - add a custom serio protocol
to send extra information") introduced usage of the BIT() macro
for SERIO_* flags; this macro is not provided in UAPI headers.
Replace if with similarly defined _BITUL() macro defined
in <linux/const.h>.
Fixes: 19ba1eb15a ("Input: psmouse - add a custom serio protocol to send extra information")
Signed-off-by: Eugene Syromiatnikov <[email protected]>
Cc: <[email protected]> # v5.0+
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Dmitry Torokhov <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 1369d0abe469fb4cdea8a5bce219d38cb857a658 upstream.
This laptop (and perhaps other variants of the same model) reports an
SMBus-capable Synaptics touchpad. Everything (including suspend and
resume) works fine when RMI is enabled via the kernel command line, so
let's add it to the whitelist.
Signed-off-by: Yussuf Khalil <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Cc: [email protected]
Signed-off-by: Dmitry Torokhov <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 32cf3a610c35cb21e3157f4bbf29d89960e30a36 upstream.
These functions are supposed to return negative error codes but instead
it returns true on failure and false on success. The error codes are
eventually propagated back to user space.
Fixes: 48a2b78348 ("Input: add Raydium I2C touchscreen driver")
Signed-off-by: Dan Carpenter <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Cc: [email protected]
Signed-off-by: Dmitry Torokhov <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit e1b9f99ff8c40bba6e59de9ad4a659447b1e4112 upstream.
The driver forgets to disable and unprepare clk when remove.
Add a call to clk_disable_unprepare to fix it.
Signed-off-by: Chuhong Yuan <[email protected]>
Signed-off-by: Wolfram Sang <[email protected]>
Cc: [email protected]
Signed-off-by: Greg Kroah-Hartman <[email protected]>
commit 7937fd3227055892e169f4b34d21157e57d919e2 upstream.
The code now compiles without ACPI, but there's a warning since
iwl_mvm_get_ppag_table() isn't used, and iwl_mvm_ppag_init() must
not unconditionally fail but return success instead.
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Luca Coelho <[email protected]>
Signed-off-by: Kalle Valo <[email protected]>
[Drop hunk removing iwl_mvm_get_ppag_table() since it doesn't exist in
5.4]
Signed-off-by: Jason Andryuk <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
[ Upstream commit ba3b01d7a6f4ab9f8a0557044c9a7678f64ae070 ]
Commit 6825d3ea6c ("iommu/vt-d: Add debugfs support to show register
contents") dumps the register contents for all IOMMU devices.
Currently, a 64 bit read(dmar_readq) is done for all the IOMMU registers,
even though some of the registers are 32 bits, which is incorrect.
Use the correct read function variant (dmar_readl/dmar_readq) while
reading the contents of 32/64 bit registers respectively.
Signed-off-by: Megha Dey <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Acked-by: Lu Baolu <[email protected]>
Signed-off-by: Joerg Roedel <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
[ Upstream commit 5eb01ddfcfb25e6ebc404a41deae946bde776731 ]
The HNS3 driver supports to configure TC numbers and TC to priority
map via "tc" tool. But when delete the rule, will fail, because
the HNS3 driver needs at least one TC, but the "tc" tool sets TC
number to zero when delete.
This patch makes sure that the TC number is at least one.
Fixes: 30d240dfa2 ("net: hns3: Add mqprio hardware offload support in hns3 driver")
Signed-off-by: Yonglong Liu <[email protected]>
Signed-off-by: Huazhong Tan <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
[ Upstream commit f3cc008bf6d59b8d93b4190e01d3e557b0040e15 ]
This patch fixes an off-by-one error in strncpy size argument in
drivers/net/ethernet/samsung/sxgbe/sxgbe_main.c. The issue is that in:
strncmp(opt, "eee_timer:", 6)
the passed string literal: "eee_timer:" has 10 bytes (without the NULL
byte) and the passed size argument is 6. As a result, the logic will
also accept other, malformed strings, e.g. "eee_tiXXX:".
This bug doesn't seem to have any security impact since its present in
module's cmdline parsing code.
Signed-off-by: Dominik Czarnota <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
[ Upstream commit 7395f62d95aafacdb9bd4996ec2f95b4a655d7e6 ]
Clang warns:
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:2860:9: warning:
converting the result of '?:' with integer constants to a boolean always
evaluates to 'true' [-Wtautological-constant-compare]
return DPAA_FD_DATA_ALIGNMENT ? ALIGN(headroom,
^
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c:131:34: note: expanded
from macro 'DPAA_FD_DATA_ALIGNMENT'
\#define DPAA_FD_DATA_ALIGNMENT (fman_has_errata_a050385() ? 64 : 16)
^
1 warning generated.
This was exposed by commit 3c68b8fffb48 ("dpaa_eth: FMan erratum A050385
workaround") even though it appears to have been an issue since the
introductory commit 9ad1a37493 ("dpaa_eth: add support for DPAA
Ethernet") since DPAA_FD_DATA_ALIGNMENT has never been able to be zero.
Just replace the whole boolean expression with the true branch, as it is
always been true.
Link: https://github.com/ClangBuiltLinux/linux/issues/928
Signed-off-by: Nathan Chancellor <[email protected]>
Reviewed-by: Madalin Bucur <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
[ Upstream commit ba32679cac50c38fdf488296f96b1f3175532b8e ]
When trying to transmit to an unknown destination, the mesh code would
unconditionally transmit a HWMP PREQ even if HWMP is not the current
path selection algorithm.
Signed-off-by: Nicolas Cavallari <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Johannes Berg <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
[ Upstream commit 394b61711f3ce33f75bf70a3e22938464a13b3ee ]
When trying to rescan disks in petitboot shell, we hit the following
softlockup stacktrace:
Kernel panic - not syncing: System is deadlocked on memory
[ 241.223394] CPU: 32 PID: 693 Comm: sh Not tainted 5.4.16-openpower1 #1
[ 241.223406] Call Trace:
[ 241.223415] [c0000003f07c3180] [c000000000493fc4] dump_stack+0xa4/0xd8 (unreliable)
[ 241.223432] [c0000003f07c31c0] [c00000000007d4ac] panic+0x148/0x3cc
[ 241.223446] [c0000003f07c3260] [c000000000114b10] out_of_memory+0x468/0x4c4
[ 241.223461] [c0000003f07c3300] [c0000000001472b0] __alloc_pages_slowpath+0x594/0x6d8
[ 241.223476] [c0000003f07c3420] [c00000000014757c] __alloc_pages_nodemask+0x188/0x1a4
[ 241.223492] [c0000003f07c34a0] [c000000000153e10] alloc_pages_current+0xcc/0xd8
[ 241.223508] [c0000003f07c34e0] [c0000000001577ac] alloc_slab_page+0x30/0x98
[ 241.223524] [c0000003f07c3520] [c0000000001597fc] new_slab+0x138/0x40c
[ 241.223538] [c0000003f07c35f0] [c00000000015b204] ___slab_alloc+0x1e4/0x404
[ 241.223552] [c0000003f07c36c0] [c00000000015b450] __slab_alloc+0x2c/0x48
[ 241.223566] [c0000003f07c36f0] [c00000000015b754] kmem_cache_alloc_node+0x9c/0x1b4
[ 241.223582] [c0000003f07c3760] [c000000000218c48] blk_alloc_queue_node+0x34/0x270
[ 241.223599] [c0000003f07c37b0] [c000000000226574] blk_mq_init_queue+0x2c/0x78
[ 241.223615] [c0000003f07c37e0] [c0000000002ff710] scsi_mq_alloc_queue+0x28/0x70
[ 241.223631] [c0000003f07c3810] [c0000000003005b8] scsi_alloc_sdev+0x184/0x264
[ 241.223647] [c0000003f07c38a0] [c000000000300ba0] scsi_probe_and_add_lun+0x288/0xa3c
[ 241.223663] [c0000003f07c3a00] [c000000000301768] __scsi_scan_target+0xcc/0x478
[ 241.223679] [c0000003f07c3b20] [c000000000301c64] scsi_scan_channel.part.9+0x74/0x7c
[ 241.223696] [c0000003f07c3b70] [c000000000301df4] scsi_scan_host_selected+0xe0/0x158
[ 241.223712] [c0000003f07c3bd0] [c000000000303f04] store_scan+0x104/0x114
[ 241.223727] [c0000003f07c3cb0] [c0000000002d5ac4] dev_attr_store+0x30/0x4c
[ 241.223741] [c0000003f07c3cd0] [c0000000001dbc34] sysfs_kf_write+0x64/0x78
[ 241.223756] [c0000003f07c3cf0] [c0000000001da858] kernfs_fop_write+0x170/0x1b8
[ 241.223773] [c0000003f07c3d40] [c0000000001621fc] __vfs_write+0x34/0x60
[ 241.223787] [c0000003f07c3d60] [c000000000163c2c] vfs_write+0xa8/0xcc
[ 241.223802] [c0000003f07c3db0] [c000000000163df4] ksys_write+0x70/0xbc
[ 241.223816] [c0000003f07c3e20] [c00000000000b40c] system_call+0x5c/0x68
As a part of the scan process Linux will allocate and configure a
scsi_device for each target to be scanned. If the device is not present,
then the scsi_device is torn down. As a part of scsi_device teardown a
workqueue item will be scheduled and the lockups we see are because there
are 250k workqueue items to be processed. Accoding to the specification of
SIS-64 sas controller, max_channel should be decreased on SIS-64 adapters
to 4.
The patch fixes softlockup issue.
Thanks for Oliver Halloran's help with debugging and explanation!
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Wen Xiong <[email protected]>
Signed-off-by: Martin K. Petersen <[email protected]>
Signed-off-by: Sasha Levin <[email protected]>
