[ Upstream commit 46a334a98f ]
Perform a cache flush on the SEV-ES TMR memory after allocation to prevent
any possibility of the firmware encountering an error should dirty cache
lines be present. Use clflush_cache_range() to flush the SEV-ES TMR memory.
Fixes: 97f9ac3db6 ("crypto: ccp - Add support for SEV-ES to the PSP driver")
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 445110941e ]
led_put() is used to "undo" a successful of_led_get() call,
of_led_get() uses class_find_device_by_of_node() which returns
a reference to the device which must be free-ed with put_device()
when the caller is done with it.
Add a put_device() call to led_put() to free the reference returned
by class_find_device_by_of_node().
And also add a put_device() in the error-exit case of try_module_get()
failing.
Fixes: 699a8c7c4b ("leds: Add of_led_get() and led_put()")
Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Lee Jones <lee@kernel.org>
Link: https://lore.kernel.org/r/20230120114524.408368-2-hdegoede@redhat.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 51c082514c ]
As it is xts only handles the special return value of EINPROGRESS,
which means that in all other cases it will free data related to the
request.
However, as the caller of xts may specify MAY_BACKLOG, we also need
to expect EBUSY and treat it in the same way. Otherwise backlogged
requests will trigger a use-after-free.
Fixes: 8083b1bf81 ("crypto: xts - add support for ciphertext stealing")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit be565ec71d ]
Fix following coccicheck warning:
WARNING: Function "for_each_child_of_node"
should have of_node_put() before return.
Early exits from for_each_child_of_node should decrement the
node reference counter.
Signed-off-by: Wang Qing <wangqing@vivo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8fbc2f9edc ]
Use new dev_err_probe() API to handle deferred probe properly and simplify
the code.
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 97067aaf12 ]
The current implementation uses .ndo_set_features() callback to track
NETIF_F_HW_CSUM feature changes and update generic
CPSW_P0_CONTROL_REG.RX_CHECKSUM_EN option accordingly. It's not going to
work in case of multi-port devices as TX csum offload can be changed per
netdev.
On K3 CPSWxG devices TX csum offload enabled in the following way:
- the CPSW_P0_CONTROL_REG.RX_CHECKSUM_EN option enables TX csum offload in
generic and affects all TX DMA channels and packets;
- corresponding fields in TX DMA descriptor have to be filed properly when
upper layer wants to offload TX csum (skb->ip_summed == CHECKSUM_PARTIAL)
and it's per-packet option.
The Linux Network core is expected to never request TX csum offload if
netdev NETIF_F_HW_CSUM feature is disabled, and, as result, TX DMA
descriptors should not be modified, and per-packet TX csum offload will be
disabled (or enabled) on per-netdev basis. Which, in turn, makes it safe to
enable the CPSW_P0_CONTROL_REG.RX_CHECKSUM_EN option unconditionally.
Hence, fix TX csum offload for multi-port devices by:
- enabling the CPSW_P0_CONTROL_REG.RX_CHECKSUM_EN option in
am65_cpsw_nuss_common_open() unconditionally
- and removing .ndo_set_features() callback implementation, which was used
only NETIF_F_HW_CSUM feature update purposes
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit c0dd9245aa ]
The kernel caches each CPU's feature bits at boot in an x86_capability[]
structure. However, the capabilities in the BSP's copy can be turned off
as a result of certain command line parameters or configuration
restrictions, for example the SGX bit. This can cause a mismatch when
comparing the values before and after the microcode update.
Another example is X86_FEATURE_SRBDS_CTRL which gets added only after
microcode update:
# --- cpuid.before 2023-01-21 14:54:15.652000747 +0100
# +++ cpuid.after 2023-01-21 14:54:26.632001024 +0100
# @@ -10,7 +10,7 @@ CPU:
# 0x00000004 0x04: eax=0x00000000 ebx=0x00000000 ecx=0x00000000 edx=0x00000000
# 0x00000005 0x00: eax=0x00000040 ebx=0x00000040 ecx=0x00000003 edx=0x11142120
# 0x00000006 0x00: eax=0x000027f7 ebx=0x00000002 ecx=0x00000001 edx=0x00000000
# - 0x00000007 0x00: eax=0x00000000 ebx=0x029c6fbf ecx=0x40000000 edx=0xbc002400
# + 0x00000007 0x00: eax=0x00000000 ebx=0x029c6fbf ecx=0x40000000 edx=0xbc002e00
^^^
and which proves for a gazillionth time that late loading is a bad bad
idea.
microcode_check() is called after an update to report any previously
cached CPUID bits which might have changed due to the update.
Therefore, store the cached CPU caps before the update and compare them
with the CPU caps after the microcode update has succeeded.
Thus, the comparison is done between the CPUID *hardware* bits before
and after the upgrade instead of using the cached, possibly runtime
modified values in BSP's boot_cpu_data copy.
As a result, false warnings about CPUID bits changes are avoided.
[ bp:
- Massage.
- Add SRBDS_CTRL example.
- Add kernel-doc.
- Incorporate forgotten review feedback from dhansen.
]
Fixes: 1008c52c09 ("x86/CPU: Add a microcode loader callback")
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230109153555.4986-3-ashok.raj@intel.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 2089f34f8c ]
The functions get_online_cpus() and put_online_cpus() have been
deprecated during the CPU hotplug rework. They map directly to
cpus_read_lock() and cpus_read_unlock().
Replace deprecated CPU-hotplug functions with the official version.
The behavior remains unchanged.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210803141621.780504-9-bigeasy@linutronix.de
Stable-dep-of: c0dd9245aa ("x86/microcode: Check CPU capabilities after late microcode update correctly")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b1efd0ff4b ]
SEV-ES guests require properly setup task register with which the TSS
descriptor in the GDT can be located so that the IST-type #VC exception
handler which they need to function properly, can be executed.
This setup needs to happen before attempting to load microcode in
ucode_cpu_init() on secondary CPUs which can cause such #VC exceptions.
Simplify the machinery by running that exception setup from a new function
cpu_init_secondary() and explicitly call cpu_init_exception_handling() for
the boot CPU before cpu_init(). The latter prepares for fixing and
simplifying the exception/IST setup on the boot CPU.
There should be no functional changes resulting from this patch.
[ tglx: Reworked it so cpu_init_exception_handling() stays seperate ]
Signed-off-by: Borislav Petkov <bp@suse.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Lai Jiangshan <laijs@linux.alibaba.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/87k0o6gtvu.ffs@nanos.tec.linutronix.de
Stable-dep-of: c0dd9245aa ("x86/microcode: Check CPU capabilities after late microcode update correctly")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1b6599f741 ]
In the error path after calling dev_set_name(), the device
name is leaked. To fix this, calling dev_set_name() before
device_register(), and call put_device() if it returns error.
All the resources is released in powercap_release(), so it
can return from powercap_register_zone() directly.
Fixes: 75d2364ea0 ("PowerCap: Add class driver")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 32e62025e5 ]
As it is seqiv only handles the special return value of EINPROGERSS,
which means that in all other cases it will free data related to the
request.
However, as the caller of seqiv may specify MAY_BACKLOG, we also need
to expect EBUSY and treat it in the same way. Otherwise backlogged
requests will trigger a use-after-free.
Fixes: 0a270321db ("[CRYPTO] seqiv: Add Sequence Number IV Generator")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b5a772adf4 ]
As it is essiv only handles the special return value of EINPROGERSS,
which means that in all other cases it will free data related to the
request.
However, as the caller of essiv may specify MAY_BACKLOG, we also need
to expect EBUSY and treat it in the same way. Otherwise backlogged
requests will trigger a use-after-free.
Fixes: be1eb7f78a ("crypto: essiv - create wrapper template...")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit f2ac14b5f1 ]
When encountering a string bigger than the destination buffer (32 bytes),
the string is not properly NUL-terminated, causing buffer overreads later.
This for example happens on the Inspiron 3505, where the battery
model name is larger than 32 bytes, which leads to sysfs showing
the model name together with the serial number string (which is
NUL-terminated and thus prevents worse).
Fix this by using strscpy() which ensures that the result is
always NUL-terminated.
Fixes: 106449e870 ("ACPI: Battery: Allow extract string from integer")
Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 8a2f35b983 ]
Fix a stack-out-of-bounds write that occurs in a WMI response callback
function that is called after a timeout occurs in ath9k_wmi_cmd().
The callback writes to wmi->cmd_rsp_buf, a stack-allocated buffer that
could no longer be valid when a timeout occurs. Set wmi->last_seq_id to
0 when a timeout occurred.
Found by a modified version of syzkaller.
BUG: KASAN: stack-out-of-bounds in ath9k_wmi_ctrl_rx
Write of size 4
Call Trace:
memcpy
ath9k_wmi_ctrl_rx
ath9k_htc_rx_msg
ath9k_hif_usb_reg_in_cb
__usb_hcd_giveback_urb
usb_hcd_giveback_urb
dummy_timer
call_timer_fn
run_timer_softirq
__do_softirq
irq_exit_rcu
sysvec_apic_timer_interrupt
Fixes: fb9987d0f7 ("ath9k_htc: Support for AR9271 chipset.")
Signed-off-by: Minsuk Kang <linuxlovemin@yonsei.ac.kr>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230104124130.10996-1-linuxlovemin@yonsei.ac.kr
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 0af54343a7 ]
Syzkaller detected a memory leak of skbs in ath9k_hif_usb_rx_stream().
While processing skbs in ath9k_hif_usb_rx_stream(), the already allocated
skbs in skb_pool are not freed if ath9k_hif_usb_rx_stream() fails. If we
have an incorrect pkt_len or pkt_tag, the input skb is considered invalid
and dropped. All the associated packets already in skb_pool should be
dropped and freed. Added a comment describing this issue.
The patch also makes remain_skb NULL after being processed so that it
cannot be referenced after potential free. The initialization of hif_dev
fields which are associated with remain_skb (rx_remain_len,
rx_transfer_len and rx_pad_len) is moved after a new remain_skb is
allocated.
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
Fixes: 6ce708f54c ("ath9k: Fix out-of-bound memcpy in ath9k_hif_usb_rx_stream")
Fixes: 44b23b488d ("ath9k: hif_usb: Reduce indent 1 column")
Reported-by: syzbot+e9632e3eb038d93d6bc6@syzkaller.appspotmail.com
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/20230104123615.51511-1-pchelkin@ispras.ru
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit d7fc76039b ]
I've changed *STAT_* macros a bit in previous patch and I seems like
they become really unreadable. Align these macros definitions to make
code cleaner and fix folllowing checkpatch warning
ERROR: Macros with complex values should be enclosed in parentheses
Also, statistics macros now accept an hif_dev as argument, since
macros that depend on having a local variable with a magic name
don't abide by the coding style.
No functional change
Suggested-by: Jeff Johnson <quic_jjohnson@quicinc.com>
Signed-off-by: Pavel Skripkin <paskripkin@gmail.com>
Acked-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Kalle Valo <quic_kvalo@quicinc.com>
Link: https://lore.kernel.org/r/ebb2306d06a496cd1b032155ae52fdc5fa8cc2c5.1655145743.git.paskripkin@gmail.com
Stable-dep-of: 0af54343a7 ("wifi: ath9k: hif_usb: clean up skbs if ath9k_hif_usb_rx_stream() fails")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 1e346cbb09 ]
There is currently no return check for writing an authentication
type (HERMES_AUTH_SHARED_KEY or HERMES_AUTH_OPEN). It looks like
it was accidentally skipped.
This patch adds a return check similar to the other checks in
__orinoco_hw_setup_enc() for hermes_write_wordrec().
Detected using the static analysis tool - Svace.
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Alexey Kodanev <aleksei.kodanev@bell-sw.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/20221227133306.201356-1-aleksei.kodanev@bell-sw.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit b39f662ce1 ]
The wifi + bluetooth combo chip RTL8723BU can leak memory (especially?)
when it's connected to a bluetooth audio device. The busy bluetooth
traffic generates lots of C2H (card to host) messages, which are not
freed correctly.
To fix this, move the dev_kfree_skb() call in rtl8xxxu_c2hcmd_callback()
inside the loop where skb_dequeue() is called.
The RTL8192EU leaks memory because the C2H messages are added to the
queue and left there forever. (This was fine in the past because it
probably wasn't sending any C2H messages until commit e542e66b7c
("wifi: rtl8xxxu: gen2: Turn on the rate control"). Since that commit
it sends a C2H message when the TX rate changes.)
To fix this, delete the check for rf_paths > 1 and the goto. Let the
function process the C2H messages from RTL8192EU like the ones from
the other chips.
Theoretically the RTL8188FU could also leak like RTL8723BU, but it
most likely doesn't send C2H messages frequently enough.
This change was tested with RTL8723BU by Erhard F. I tested it with
RTL8188FU and RTL8192EU.
Reported-by: Erhard F. <erhard_f@mailbox.org>
Tested-by: Erhard F. <erhard_f@mailbox.org>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215197
Fixes: e542e66b7c ("rtl8xxxu: add bluetooth co-existence support for single antenna")
Signed-off-by: Bitterblue Smith <rtl8821cerfe2@gmail.com>
Reviewed-by: Ping-Ke Shih <pkshih@realtek.com>
Signed-off-by: Kalle Valo <kvalo@kernel.org>
Link: https://lore.kernel.org/r/03b099c1-c671-d252-36f4-57b70d721f9d@gmail.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit ca843a4c79 ]
Previously acpi_ns_simple_repair() would crash if expected_btypes
contained any combination of ACPI_RTYPE_NONE with a different type,
e.g | ACPI_RTYPE_INTEGER because of slightly incorrect logic in the
!return_object branch, which wouldn't return AE_AML_NO_RETURN_VALUE
for such cases.
Found by Linux Verification Center (linuxtesting.org) with the SVACE
static analysis tool.
Link: https://github.com/acpica/acpica/pull/811
Fixes: 61db45ca21 ("ACPICA: Restore code that repairs NULL package elements in return values.")
Signed-off-by: Daniil Tatianin <d-tatianin@yandex-team.ru>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 91dfd98216 ]
For SEV_GET_ID2, the user provided length does not have a specified
limitation because the length of the ID may change in the future. The
kernel memory allocation, however, is implicitly limited to 4MB on x86 by
the page allocator, otherwise the kzalloc() will fail.
When this happens, it is best not to spam the kernel log with the warning.
Simply fail the allocation and return ENOMEM to the user.
Fixes: d6112ea0cb ("crypto: ccp - introduce SEV_GET_ID2 command")
Reported-by: Andy Nguyen <theflow@google.com>
Reported-by: Peter Gonda <pgonda@google.com>
Suggested-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 13dc15a3f5 ]
For some sev ioctl interfaces, input may be passed that is less than or
equal to SEV_FW_BLOB_MAX_SIZE, but larger than the data that PSP
firmware returns. In this case, kmalloc will allocate memory that is the
size of the input rather than the size of the data. Since PSP firmware
doesn't fully overwrite the buffer, the sev ioctl interfaces with the
issue may return uninitialized slab memory.
Currently, all of the ioctl interfaces in the ccp driver are safe, but
to prevent future problems, change all ioctl interfaces that allocate
memory with kmalloc to use kzalloc and memset the data buffer to zero
in sev_ioctl_do_platform_status.
Fixes: 38103671aa ("crypto: ccp: Use the stack and common buffer for status commands")
Fixes: e799035609 ("crypto: ccp: Implement SEV_PEK_CSR ioctl command")
Fixes: 76a2b524a4 ("crypto: ccp: Implement SEV_PDH_CERT_EXPORT ioctl command")
Fixes: d6112ea0cb ("crypto: ccp - introduce SEV_GET_ID2 command")
Cc: stable@vger.kernel.org
Reported-by: Andy Nguyen <theflow@google.com>
Suggested-by: David Rientjes <rientjes@google.com>
Suggested-by: Peter Gonda <pgonda@google.com>
Signed-off-by: John Allen <john.allen@amd.com>
Reviewed-by: Peter Gonda <pgonda@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Stable-dep-of: 91dfd98216 ("crypto: ccp - Avoid page allocation failure warning for SEV_GET_ID2")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 38103671aa ]
Drop the dedicated status_cmd_buf and instead use a local variable for
PLATFORM_STATUS. Now that the low level helper uses an internal buffer
for all commands, using the stack for the upper layers is safe even when
running with CONFIG_VMAP_STACK=y.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210406224952.4177376-7-seanjc@google.com>
Reviewed-by: Brijesh Singh <brijesh.singh@amd.com>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: 91dfd98216 ("crypto: ccp - Avoid page allocation failure warning for SEV_GET_ID2")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e4a9af799e ]
For commands with small input/output buffers, use the local stack to
"allocate" the structures used to communicate with the PSP. Now that
__sev_do_cmd_locked() gracefully handles vmalloc'd buffers, there's no
reason to avoid using the stack, e.g. CONFIG_VMAP_STACK=y will just work.
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20210406224952.4177376-6-seanjc@google.com>
Reviewed-by: Brijesh Singh <brijesh.singh@amd.com>
Acked-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Stable-dep-of: 91dfd98216 ("crypto: ccp - Avoid page allocation failure warning for SEV_GET_ID2")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 7361d1bc30 ]
The helper mpi_read_raw_from_sgl sets the number of entries in
the SG list according to nbytes. However, if the last entry
in the SG list contains more data than nbytes, then it may overrun
the buffer because it only allocates enough memory for nbytes.
Fixes: 2d4d1eea54 ("lib/mpi: Add mpi sgl helpers")
Reported-by: Roberto Sassu <roberto.sassu@huaweicloud.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 28319d6dc5 ]
RCU Tasks and PID-namespace unshare can interact in do_exit() in a
complicated circular dependency:
1) TASK A calls unshare(CLONE_NEWPID), this creates a new PID namespace
that every subsequent child of TASK A will belong to. But TASK A
doesn't itself belong to that new PID namespace.
2) TASK A forks() and creates TASK B. TASK A stays attached to its PID
namespace (let's say PID_NS1) and TASK B is the first task belonging
to the new PID namespace created by unshare() (let's call it PID_NS2).
3) Since TASK B is the first task attached to PID_NS2, it becomes the
PID_NS2 child reaper.
4) TASK A forks() again and creates TASK C which get attached to PID_NS2.
Note how TASK C has TASK A as a parent (belonging to PID_NS1) but has
TASK B (belonging to PID_NS2) as a pid_namespace child_reaper.
5) TASK B exits and since it is the child reaper for PID_NS2, it has to
kill all other tasks attached to PID_NS2, and wait for all of them to
die before getting reaped itself (zap_pid_ns_process()).
6) TASK A calls synchronize_rcu_tasks() which leads to
synchronize_srcu(&tasks_rcu_exit_srcu).
7) TASK B is waiting for TASK C to get reaped. But TASK B is under a
tasks_rcu_exit_srcu SRCU critical section (exit_notify() is between
exit_tasks_rcu_start() and exit_tasks_rcu_finish()), blocking TASK A.
8) TASK C exits and since TASK A is its parent, it waits for it to reap
TASK C, but it can't because TASK A waits for TASK B that waits for
TASK C.
Pid_namespace semantics can hardly be changed at this point. But the
coverage of tasks_rcu_exit_srcu can be reduced instead.
The current task is assumed not to be concurrently reapable at this
stage of exit_notify() and therefore tasks_rcu_exit_srcu can be
temporarily relaxed without breaking its constraints, providing a way
out of the deadlock scenario.
[ paulmck: Fix build failure by adding additional declaration. ]
Fixes: 3f95aa81d2 ("rcu: Make TASKS_RCU handle tasks that are almost done exiting")
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Suggested-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Cc: Eric W . Biederman <ebiederm@xmission.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 4475709295 ]
Ever since the following commit:
5a41344a3d ("srcu: Simplify __srcu_read_unlock() via this_cpu_dec()")
SRCU doesn't rely anymore on preemption to be disabled in order to
modify the per-CPU counter. And even then it used to be done from the API
itself.
Therefore and after checking further, it appears to be safe to remove
the preemption disablement around __srcu_read_[un]lock() in
exit_tasks_rcu_start() and exit_tasks_rcu_finish()
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Suggested-by: Paul E. McKenney <paulmck@kernel.org>
Suggested-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Stable-dep-of: 28319d6dc5 ("rcu-tasks: Fix synchronize_rcu_tasks() VS zap_pid_ns_processes()")
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit e1d9148582 ]
Microsoft introduced support in Windows XP for blocking port I/O
to various regions. For Windows compatibility ACPICA has adopted
the same protections and will disallow writes to those
(presumably) the same regions.
On some systems the AML included with the firmware will issue 4 byte
long writes to 0x80. These writes aren't making it over because of this
blockage. The first 4 byte write attempt is rejected, and then
subsequently 1 byte at a time each offset is tried. The first at 0x80
works, but then the next 3 bytes are rejected.
This manifests in bizarre failures for devices that expected the AML to
write all 4 bytes. Trying the same AML on Windows 10 or 11 doesn't hit
this failure and all 4 bytes are written.
Either some of these regions were wrong or some point after Windows XP
some of these regions blocks have been lifted.
In the last 15 years there doesn't seem to be any reports popping up of
this error in the Windows event viewer anymore. There is no documentation
at Microsoft's developer site indicating that Windows ACPI interpreter
blocks these regions. Between the lack of documentation and the fact that
the writes actually do work in Windows 10 and 11, it's quite likely
Windows doesn't actually enforce this anymore.
So to help the issue, only enforce Windows XP specific entries if the
latest _OSI supported is Windows XP. Continue to enforce the
ALWAYS_ILLEGAL entries.
Link: https://github.com/acpica/acpica/pull/817
Fixes: 7f07190390 ("ACPICA: New: I/O port protection")
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
[ Upstream commit 116db2704c ]
The key can be unaligned, so use the unaligned memory access helpers.
Fixes: 8ceee72808 ("crypto: ghash-clmulni-intel - use C implementation for setkey()")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <sashal@kernel.org>
