Now that the xr_serial got support for other models, add their USB IDs
as well.
The Maxlinear/Exar USB UARTs can be used in either ACM mode using the
cdc-acm driver or in "custom driver" mode in which further features such
as hardware and software flow control, GPIO control and in-band
line-status reporting are available.
In ACM mode the device always enables RTS/CTS flow control, something
which could prevent transmission in case the CTS input isn't wired up
correctly.
Ensure that cdc_acm will not bind to these devices if the custom
USB-serial driver is enabled.
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/5155887a764cbc11f8da0217fe08a24a77d120b4.1616571453.git.mchehab+huawei@kernel.org
[ johan: rewrite commit message, clean up entries ]
Cc: Oliver Neukum <oneukum@suse.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Reset the transmit and receive FIFOs before enabling the UARTs as part
of open() in order to flush any stale data.
Note that the XR21V141X needs a type-specific implementation due to its
UART Manager registers.
Signed-off-by: Johan Hovold <johan@kernel.org>
The XR22801, XR22802 and XR22804 are compound devices with an embedded
hub and up to seven downstream USB devices including one, two or four
UARTs respectively.
The UART function is similar to XR21B142X but most registers are offset
by 0x40, the register requests are different and are directed at the
device rather than interface, and 5 and 6-bit words are not supported.
Signed-off-by: Johan Hovold <johan@kernel.org>
The single-port XR21B1411 is similar to the XR21B142X type but uses
12-bit registers and 16-bit register addresses, the register requests
are different and are directed at the device rather than interface, and
5 and 6-bit words are not supported.
The register layout is very similar to XR21B142X except that most
registers are offset by 0xc00 (corresponding to a channel index of 12 in
the MSB of wIndex). As the device is single-port so that the derived
channel index is 0, the current register accessors can be reused after
simply changing the address width.
Signed-off-by: Johan Hovold <johan@kernel.org>
The XR21B1421, XR21B1422 and XR21B1424 are the one-, two- and four-port
models of a second XR21B142X type of the Maxlinear/Exar USB UARTs.
The XR21B142X type differs from XR21V141X in several ways, including:
- register layout
- register width (16-bit instead of 8-bit)
- vendor register requests
- UART enable/disable sequence
- custom-driver mode flag
- three additional GPIOs (9 instead of 6)
As for XR21V141X, the XR21B142X vendor requests encode the channel index
in the MSB of wIndex, but it lacks the UART Manager registers which
have been replaced by regular UART registers. The new type also uses the
interface number of the control interface (0, 2, 4, 6) as channel index
instead of the channel number (0, 1, 2, 3).
The XR21B142X lacks the divisor and format registers used by XR21V141X
and instead uses the CDC SET_LINE_CONTROL request to configure the line
settings.
Note that the currently supported XR21V141X type lacks the custom-driver
mode flag that prevents the device from entering CDC-ACM mode when a CDC
requests is received. This specifically means that the SET_LINE_CONTROL
request cannot be used with XR21V141X even though it is otherwise
supported.
The UART enable sequence for XR21B142X does not involve explicitly
enabling the FIFOs, but according to datasheet the UART must be disabled
when writing any register but GPIO_SET, GPIO_CLEAR, TX_BREAK and
ERROR_STATUS.
Signed-off-by: Johan Hovold <johan@kernel.org>
There are at least four types of Maxlinear/Exar USB UARTs which differ
in various ways such as in their register layouts:
XR21V141X
XR21B142X
XR21B1411
XR22804
It is not clear whether the device type can be inferred from the
descriptors so encode it in the device-id table for now.
Add a type structure that can be used to abstract the register layout
and other features, and use it when accessing the XR21V141X UART
registers that are shared by all types.
Note that the currently supported XR21V141X type is the only type that
has a set of UART Manager registers and that these will need to be
handled specifically.
Similarly, XR21V141X is the only type which has the divisor registers
and that needs to use the format register when configuring the line
settings.
Signed-off-by: Johan Hovold <johan@kernel.org>
In preparation for adding support for further types, drop the type
prefix from defines that are not specific to XR21V141X.
Signed-off-by: Johan Hovold <johan@kernel.org>
There's no need to configure the pins on every open and judging from the
vendor driver and datasheet it can be done before enabling the UART.
Move pin configuration from open() to port probe and make sure to
deassert DTR and RTS after configuring all pins as GPIO.
Signed-off-by: Johan Hovold <johan@kernel.org>
Add support for the two- and four-port variants of XR21V1410.
Use the interface number of each control interface (e.g. 0, 2, 4, 6) to
derive the zero-based channel index:
XR21V1410 0
XR21V1412 0, 1
XR21V1414 0, 1, 2, 3
Note that the UART registers reside in separate blocks per channel,
while the UART Manager functionality is implemented using per-channel
registers.
Signed-off-by: Johan Hovold <johan@kernel.org>
The kernel does not use any keys besides IA so we don't need to
install IB/DA/DB/GA on kernel exit if we arrange to install them
on task switch instead, which we can expect to happen an order of
magnitude less often.
Furthermore we can avoid installing the user IA in the case where the
user task has IA disabled and just leave the kernel IA installed. This
also lets us avoid needing to install IA on kernel entry.
On an Apple M1 under a hypervisor, the overhead of kernel entry/exit
has been measured to be reduced by 15.6ns in the case where IA is
enabled, and 31.9ns in the case where IA is disabled.
Signed-off-by: Peter Collingbourne <pcc@google.com>
Link: https://linux-review.googlesource.com/id/Ieddf6b580d23c9e0bed45a822dabe72d2ffc9a8e
Link: https://lore.kernel.org/r/2d653d055f38f779937f2b92f8ddd5cf9e4af4f4.1616123271.git.pcc@google.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
This change introduces a prctl that allows the user program to control
which PAC keys are enabled in a particular task. The main reason
why this is useful is to enable a userspace ABI that uses PAC to
sign and authenticate function pointers and other pointers exposed
outside of the function, while still allowing binaries conforming
to the ABI to interoperate with legacy binaries that do not sign or
authenticate pointers.
The idea is that a dynamic loader or early startup code would issue
this prctl very early after establishing that a process may load legacy
binaries, but before executing any PAC instructions.
This change adds a small amount of overhead to kernel entry and exit
due to additional required instruction sequences.
On a DragonBoard 845c (Cortex-A75) with the powersave governor, the
overhead of similar instruction sequences was measured as 4.9ns when
simulating the common case where IA is left enabled, or 43.7ns when
simulating the uncommon case where IA is disabled. These numbers can
be seen as the worst case scenario, since in more realistic scenarios
a better performing governor would be used and a newer chip would be
used that would support PAC unlike Cortex-A75 and would be expected
to be faster than Cortex-A75.
On an Apple M1 under a hypervisor, the overhead of the entry/exit
instruction sequences introduced by this patch was measured as 0.3ns
in the case where IA is left enabled, and 33.0ns in the case where
IA is disabled.
Signed-off-by: Peter Collingbourne <pcc@google.com>
Reviewed-by: Dave Martin <Dave.Martin@arm.com>
Link: https://linux-review.googlesource.com/id/Ibc41a5e6a76b275efbaa126b31119dc197b927a5
Link: https://lore.kernel.org/r/d6609065f8f40397a4124654eb68c9f490b4d477.1616123271.git.pcc@google.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Remove the random white space from the CSIZE switch.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Add two port-command helpers to handle the UART module-id parameter
instead of open coding.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Make the vendor-request helpers data parameters be void pointers and
drop the caller casts.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Drop unnecessary packed attributes from structures that don't need it.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Drop unnecessary packed attributes from structures that don't need it
and use the __packed macro consistently.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Use kernel types consistently by replacing the remaining __uXX types.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Add a read-port-command helper analogous to the send-port-command
helper to take care of the UART module id instead of open coding.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Add a send-port-command helper which takes care of determining the UART
module id when sending commands instead of doing so at every call site.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Clean up the vendor-request helpers by using kernel-types consistently
and using void pointers for the data arguments, which allows removing
a cast from one caller.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Unlike the TUSB5052, the TUSB3410 has an LSR TEMT bit to tell if both
the transmitter data and shift registers are empty.
Make sure to check also the shift register on TUSB3410 when waiting for
the transmit buffer to drain during close and drop the time-based
one-char delay which is otherwise needed (e.g. 90 ms at 110 bps).
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
The three-character drain delay was added by commit f1175daa53 ("USB:
ti_usb_3410_5052: kill custom closing_wait") when removing the custom
closing-wait implementation, which used a fixed 20 ms poll period and
drain delay.
This was likely a bit too conservative as a one-character timeout (e.g.
33 ms at 300 bps) should be enough to compensate for the lack of a
transmitter empty bit in the TUSB5052 line-status register.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Document that the device line-status register doesn't tell when the
transmitter shift register has emptied and that this is why the
one-character drain delay is needed.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
The f81232 driver now waits for the transmit FIFO to drain during close
so there is no need to keep the time-based drain delay, which would add
up to two seconds on every close for low line speeds.
Fixes: 98405f8103 ("USB: serial: f81232: add tx_empty function")
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
As a flush operation is implemented inside destroy_workqueue(), there is
no need to do flush operation before.
Fixes: bfcc681bd0 ("IB/hns: Fix the bug when free mr")
Fixes: 0425e3e6e0 ("RDMA/hns: Support flush cqe for hip08 in kernel space")
Link: https://lore.kernel.org/r/1618305087-30799-1-git-send-email-liweihang@huawei.com
Signed-off-by: Yixian Liu <liuyixian@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In case of error, the function devm_kasprintf() returns NULL
pointer not ERR_PTR(). The IS_ERR() test in the return value
check should be replaced with NULL test.
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Link: https://lore.kernel.org/r/20210409094901.1903622-1-weiyongjun1@huawei.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
The only way to get out of io_iopoll_getevents() and continue iterating
is to have empty iopoll_list, otherwise the main loop would just exit.
So, instead of the unlock on 8th time heuristic, do that based on
iopoll_list.
Also, as no one can add new requests to iopoll_list while
io_iopoll_check() hold uring_lock, it's useless to spin with the list
empty, return in that case.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/5b8ebe84f5fff7ffa1f708952dfef7fc78b668e2.1618278933.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
As CQE overflows are now untied from requests and so don't hold any
ref, we don't need to handle exiting/exec'ing cases there anymore.
Moreover, it's much nicer in regards to userspace to save overflowed
CQEs whenever possible, so remove failing on in_idle.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d873b7dab75c7f3039ead9628a745bea01f2cfd2.1618278933.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Don't save return values of hrtimer_try_to_cancel() in a variable, but
use right away. It's in general safer to not have an intermediate
variable, which may be reused and passed out wrongly, but it be
contracted out. Also clean io_timeout_extract().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/d2566ef7ce632e6882dc13e022a26249b3fd30b5.1618278933.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
If io_sqe_files_unregister() faults on io_rsrc_ref_quiesce(), it will
fail to do unregister leaving files referenced. And that may well happen
because of a strayed signal or just because it does allocations inside.
In io_ring_ctx_free() do an unsafe version of unregister, as it's
guaranteed to not have requests by that point and so quiesce is useless.
Cc: stable@vger.kernel.org
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/e696e9eade571b51997d0dc1d01f144c6d685c05.1618278933.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The sparse tool complains as follows:
drivers/hwtracing/coresight/coresight-core.c:26:1: warning:
symbol '__pcpu_scope_csdev_sink' was not declared. Should it be static?
As csdev_sink is not used outside of coresight-core.c after the
introduction of coresight_[set|get]_percpu_sink() helpers, this
change marks it static.
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Link: https://lore.kernel.org/r/20210409094900.1902783-1-weiyongjun1@huawei.com
Signed-off-by: Mathieu Poirier <mathieu.poirier@linaro.org>
Lightnvm was an innovative idea to expose more low-level control over SSDs.
But it failed to get properly standardized and remains a non-standarized
extension to NVMe that requires vendor specific quirks for a few now mostly
obsolete SSD devices. The standardized ZNS command set for NVMe has take
over a lot of the approaches and allows for fully standardized operation.
Remove the Linux code to support open channel SSDs as the few production
deployments of the above mentioned SSDs are using userspace driver stacks
instead of the fairly limited Linux support.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Javier González <javier@javigon.com>
Signed-off-by: Matias Bjørling <mb@lightnvm.io>
Link: https://lore.kernel.org/r/20210413105257.159260-5-matias.bjorling@wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
'linux/blkdev.h' and 'uapi/linux/lightnvm.h' included in 'lightnvm.h'
is duplicated.It is also included in the 5th and 7th line.
Signed-off-by: Zhang Yunkai <zhang.yunkai@zte.com.cn>
Signed-off-by: Matias Bjørling <matias.bjorling@wdc.com>
Link: https://lore.kernel.org/r/20210413105257.159260-4-matias.bjorling@wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
When memdup_user returns an error, memdup_user has two different return
values, use PTR_ERR to get the correct return value.
Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: Matias Bjørling <matias.bjorling@wdc.com>
Link: https://lore.kernel.org/r/20210413105257.159260-3-matias.bjorling@wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Normally, If the smi-larb HW need work, we should enable the smi-common
HW power and clock firstly.
This patch adds device-link between the smi-larb dev and the smi-common
dev. then If pm_runtime_get_sync(smi-larb-dev), the pm_runtime_get_sync
(smi-common-dev) will be called automatically.
Also, Add DL_FLAG_STATELESS to avoid the smi-common clocks be gated when
probe.
CC: Matthias Brugger <matthias.bgg@gmail.com>
Suggested-by: Tomasz Figa <tfiga@chromium.org>
Signed-off-by: Yong Wu <yong.wu@mediatek.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Link: https://lore.kernel.org/r/20210410091128.31823-5-yong.wu@mediatek.com
