When the erratum workaround is triggered, the newly created xdp_frame
structure is stored at the start of the newly allocated buffer. Avoid
the structure from being overwritten by explicitly reserving enough
space in the buffer for storing it.
Account for the fact that the structure's size might increase in time by
aligning the headroom to DPAA_FD_DATA_ALIGNMENT bytes, thus guaranteeing
the data's alignment.
Fixes: ae680bcbd0 ("dpaa_eth: implement the A050385 erratum workaround for XDP")
Signed-off-by: Camelia Groza <camelia.groza@nxp.com>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Acked-by: Madalin Bucur <madalin.bucur@oss.nxp.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit c80794323e ("net: Fix packet reordering caused by GRO and
listified RX cooperation") had the unfortunate effect of adding
latencies in common workloads.
Before the patch, GRO packets were immediately passed to
upper stacks.
After the patch, we can accumulate quite a lot of GRO
packets (depdending on NAPI budget).
My fix is counting in napi->rx_count number of segments
instead of number of logical packets.
Fixes: c80794323e ("net: Fix packet reordering caused by GRO and listified RX cooperation")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Bisected-by: John Sperbeck <jsperbeck@google.com>
Tested-by: Jian Yang <jianyang@google.com>
Cc: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Alexander Lobakin <alobakin@pm.me>
Link: https://lore.kernel.org/r/20210204213146.4192368-1-eric.dumazet@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The variable err is being assigned a value that is never read,
the same error number is being returned at the error return
path via label err1. Clean up the code by removing the assignment.
Addresses-Coverity: ("Unused value")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
There are certain scenarios, where a disconnect event might
occur on a Type-C port with no port partners. This is required
to enable communication to Burnside Bridge USB4 retimers.
Signed-off-by: Rajmohan Mani <rajmohan.mani@intel.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
Link: https://lore.kernel.org/r/20210205195113.20277-3-rajmohan.mani@intel.com
Signed-off-by: Benson Leung <bleung@chromium.org>
For certain needs like updating the USB4 retimer firmware when no
device are connected, the Type-C ports require mux configuration,
to be able to communicate with the retimer. So removed the above
check to allow for mux configuration of Type-C ports, to enable
retimer communication.
Signed-off-by: Rajmohan Mani <rajmohan.mani@intel.com>
Reviewed-by: Prashant Malani <pmalani@chromium.org>
Link: https://lore.kernel.org/r/20210205195113.20277-2-rajmohan.mani@intel.com
Signed-off-by: Benson Leung <bleung@chromium.org>
It doesn't make sense to set the RTC to a default value at probe time. Let
the core handle invalid date and time.
Also, this is basically dead code since commit 22652ba724 ("rtc: stop
validating rtc_time in .read_time")
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/20210202112934.3612726-1-alexandre.belloni@bootlin.com
The driver will compile with warning without OF because armada38x_data and
armada8k_data will be defined but not used. It would be possible to move
then in the #ifdef CONFIG_OF section but then their members will be defined
but not used. Instead of moving most of the driver in the #ifdef, simply
depend on OF.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210202112219.3610853-3-alexandre.belloni@bootlin.com
Disabling all event calls is already handled by the core right before
starting the DAA process. Do not do it again when the DAA process
completes, it is redundant.
Fixes: 1dd728f5d4 ("i3c: master: Add driver for Synopsys DesignWare IP")
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20201228105501.6104-1-miquel.raynal@bootlin.com
Add support for Silvaco I3C dual-role IP. The master role is supported
in SDR mode only. I2C transfers have not been tested but are shared
because they are very close to the I3C transfers in terms of register
configuration.
The IBI processing follows this logic:
- When a slave advertizes an interrupt (SDA pulled low) an interrupt
gets generated by the master. This time is unbounded and may be
deferred.
- The IRQ handler itself does not process anything: it only queues a
work that will be run in non-atomic context. This is needed because
short wait periods must be experienced.
- The IBI job is divided in two parts: the first one is "critical" in
the sense that it may not support getting interrupted. If this
happens, after this first section the driver checks the master error
register and depending on its content either flushes everything and
errors out, or ends the processing (this second section may be
interrupted).
- If the critical section got interrupted, the slave will
automatically respin it's IBI request when it will be allowed to.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210121101808.14654-6-miquel.raynal@bootlin.com
Silvaco provide a dual-role I3C master.
Description is rather simple: it needs a register mapping, three
clocks and an interrupt.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210121101808.14654-5-miquel.raynal@bootlin.com
Update a little bit the content to match the bus binding, including:
- the node title should have been named after the description done
in the historical i3c.txt file, ie: i3c-master@<address>
- child nodes should be accepted even though the drivers do not currently
support it
- #address-cells and #size-cells are also mandatory and have specific
values
Cc: Nicolas Pitre <nico@fluxnic.net>
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210121101808.14654-3-miquel.raynal@bootlin.com
Attempting a conversion of the i3c.txt file to yaml schema with
minimal content changes.
Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20210121101808.14654-2-miquel.raynal@bootlin.com
Michael Kerrisk suggested that, from an API perspective, it is a bad
idea to share the PR_SYS_DISPATCH_ defines between the prctl operation
and the selector variable.
Therefore, define two new constants to be used by SUD's selector variable
and update the corresponding documentation and test cases.
While this changes the API syscall user dispatch has never been part of a
Linux release, it will show up for the first time in 5.11.
Suggested-by: Michael Kerrisk (man-pages) <mtk.manpages@gmail.com>
Signed-off-by: Gabriel Krisman Bertazi <krisman@collabora.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210205184321.2062251-1-krisman@collabora.com
