- 22 Sep, 2020 18 commits
-
-
Joakim Zhang authored
This patch adds CAN FD BitRate Switch (BRS) support to driver. Signed-off-by:
Joakim Zhang <qiangqing.zhang@nxp.com> Link: https://lore.kernel.org/r/20190712075926.7357-5-qiangqing.zhang@nxp.comSigned-off-by:
Marc Kleine-Budde <mkl@pengutronix.de>
-
Joakim Zhang authored
ISO CAN FD is introduced to increase the failture detection capability than non-ISO CAN FD. The non-ISO CAN FD is still supported by FlexCAN so that it can be used mainly during an intermediate phase, for evaluation and development purposes. Therefore, it is strongly recommended to configure FlexCAN to the ISO CAN FD protocol by setting the ISOCANFDEN field in the CTRL2 register. NOTE: If you only set "fd on", driver will use ISO FD mode by default. You should set "fd-non-iso on" after setting "fd on" if you want to use NON ISO FD mode. Signed-off-by:
-
Joakim Zhang authored
This patch adds CAN-FD mode support to the driver, it means that payload size can extend up to 64 bytes. Bit timing always set in the CBT register, not in the CTRL1 register any more. This has an extend range of all CAN bit timing variables (PRESDIV, PROPSEG, PSEG1, PSEG2 and RJW), which will improve the bit timing accuracy. Signed-off-by:
-
Joakim Zhang authored
This patch prepares for CAN FD mode, using struct canfd_frame can both for classic format frame and fd format frame. Signed-off-by:
-
Marc Kleine-Budde authored
This is a patch prepares for the CAN-FD support. In a later patch the setup for canfd bittiming will be added, with this patch the change is easier to read. Signed-off-by:
-
Joakim Zhang authored
On the i.MX7D in LPSR mode, the controller will be powered off and the configuration state is lost after system resume. Upcoming i.MX8QM/QXP will also completely power off the domain, the controller state is lost and needs restore, too. So we need to set the pinctrl state again and re-start chip to re-configuration after resume. For the wakeup case, it should not set pinctrl to sleep state by pinctrl_pm_select_sleep_state. If the interface is down before suspend, we don't need to re-configure it as it will be configured if the interface is brought up later. Signed-off-by:
-
Joakim Zhang authored
Disable clocks while CAN core is in stop mode. Signed-off-by:
Sean Nyekjaer <sean@geanix.com> Link: https://lore.kernel.org/r/20191210085721.9853-2-qiangqing.zhang@nxp.comSigned-off-by:
-
Joakim Zhang authored
This patch implements error handling and propagates the error value of flexcan_chip_stop(). This function will be called from flexcan_suspend() in an upcoming patch in some SoCs which support LPSR mode. Add a new function flexcan_chip_stop_disable_on_error() that tries to disable the chip even in case of errors. Signed-off-by:
-
Joakim Zhang authored
commit cdce8448 ("can: flexcan: add vf610 support for FlexCAN") From above commit by Stefan Agner, the patch just disables non-correctable errors interrupt and freeze mode. It still can correct the correctable errors since ECC enabled by default after reset (MECR[ECCDIS]=0, enable memory error correct) if HW supports ECC. commit 5e269324 ("can: flexcan: disable completely the ECC mechanism") From above commit by Joakim Zhang, the patch disables ECC completely (assert MECR[ECCDIS]) according to the explanation of FLEXCAN_QUIRK_DISABLE_MECR that disable memory error detection. This cause correctable errors cannot be corrected even HW supports ECC. The error correction mechanism ensures that in this 13-bit word, errors in one bit can be corrected (correctable errors) and errors in two bits can be detected but not corrected (non-correctable errors). Errors in more than two bits may not be detected. If HW supports ECC, we can use this to correct the correctable errors detected from FlexCAN memory. Then disable non-correctable errors interrupt and freeze mode to avoid that put FlexCAN in freeze mode. This patch adds correctable errors correction when HW supports ECC, and modify explanation for FLEXCAN_QUIRK_DISABLE_MECR. Signed-off-by:
-
Joakim Zhang authored
CAN FD can transmit up to 8Mbps, but some transceivers only can support 5Mbps, so add check in driver. Signed-off-by:
-
Marc Kleine-Budde authored
As the transcevier regulator is optional, this patch switches from devm_regulator_get() to devm_regulator_get_optional(). This gets rid of "using dummy regulator" warning message from the regulator core, if no regulator is available. Signed-off-by:
-
Joakim Zhang authored
As FLEXCAN_ESR_ALL_INT is for all bus errors and state change IRQ sources, strictly speaking FLEXCAN_ESR_WAK_INT does not belong to these. So add wakeup interrupt ack separately to existing ack of the interrupts. Suggested-by:
-
Marc Kleine-Budde authored
This patch reformats the quirks to get rid of long lines. Signed-off-by:
-
Marc Kleine-Budde authored
This patch documents which registers are not affected by a soft reset of the flexcan IP core. Signed-off-by:
-
Marc Kleine-Budde authored
This patch adds some new register names and tries to ensure with a static_assert that the documented offset is correct. Signed-off-by:
-
Marc Kleine-Budde authored
This patch removes a stray empty line from the flexcan_exit_stop_mode() function. Signed-off-by:
-
Marc Kleine-Budde authored
This patch sorts the include files alphabetically. Signed-off-by:
-
Florian Fainelli authored
BCM72113 features a 28nm integrated EPHY, add an entry to the driver for it. Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- 21 Sep, 2020 22 commits
-
-
David S. Miller authored
Merge tag 'linux-can-next-for-5.10-20200921' of git://git.kernel.org/pub/scm/linux/kernel/git/mkl/linux-can-next Marc Kleine-Budde says: ==================== pull-request: can-next 2020-09-21 this is a pull request of 38 patches for net-next. the first 5 patches are by Colin Ian King, Alexandre Belloni and me and they fix various spelling mistakes. The next patch is by me and fixes the indention in the CAN raw protocol according to the kernel coding style. Diego Elio Pettenò contributes two patches to fix dead links in CAN's Kconfig. Masahiro Yamada's patch removes the "WITH Linux-syscall-note" from SPDX tag of C files. AThe next 4 patches are by me and target the CAN device infrastructure and add error propagation and improve the output of various messages to ease driver development and debugging. YueHaibing's patch for the c_can driver removes an unused inline function. Next follows another patch by Colin Ian King, which removes the unneeded initialization of a variable in the mcba_usb driver. A patch by me annotates a fallthrough in the mscan driver. The ti_hecc driver is converted to use devm_platform_ioremap_resource_byname() in a patch by Dejin Zheng. Liu Shixin's patch converts the pcan_usb_pro driver to make use of le32_add_cpu() instead of open coding it. Wang Hai's patch for the peak_pciefd_main driver removes an unused makro. Vaibhav Gupta's patch converts the pch_can driver to generic power management. Stephane Grosjean improves the pcan_usb usb driver by first documenting the commands sent to the device and by adding support of rxerr/txerr counters. The next patch is by me and cleans up the Kconfig of the CAN SPI drivers. The next 6 patches all target the mcp251x driver, they are by Timo Schlüßler, Andy Shevchenko, Tim Harvey and me. They update the DT bindings documentation, sort the include files alphabetically, add GPIO support, make use of the readx_poll_timeout() helper, and add support for half duplex SPI-controllers. Wolfram Sang contributes a patch to update the contact email address in the mscan driver, while Zhang Changzhong updates the clock handling. The next patch is by and updates the rx-offload infrastructure to support callback less usage. The last 6 patches add support for the mcp25xxfd CAN SPI driver. First the dt-bindings are added by Oleksij Rempel, the regmap infrastructure and the main driver is contributed by me. Kurt Van Dijck adds listen-only support, Manivannan Sadhasivam adds himself as maintainer, and Thomas Kopp himself as a reviewer. ==================== Signed-off-by:
-
David S. Miller authored
Merge tag 'mac80211-next-for-net-next-2020-09-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211-next Johannes Berg says: ==================== This time we have: * some AP-side infrastructure for FILS discovery and unsolicited probe resonses * a major rework of the encapsulation/header conversion offload from Felix, to fit better with e.g. AP_VLAN interfaces * performance fix for VHT A-MPDU size, don't limit to HT * some initial patches for S1G (sub 1 GHz) support, more will come in this area * minor cleanups ==================== Signed-off-by:
-
Zheng Yongjun authored
Fixes gcc '-Wunused-but-set-variable' warning: drivers/net/ethernet/natsemi/ns83820.c: In function ns83820_get_link_ksettings: drivers/net/ethernet/natsemi/ns83820.c:1210:11: warning: variable ‘tanar’ set but not used [-Wunused-but-set-variable] `tanar` is never used, so remove it. Signed-off-by:
Zheng Yongjun <zhengyongjun3@huawei.com> Signed-off-by:
-
Jing Xiangfeng authored
After commit 37ab4fa7 ("net: unix: allow bind to fail on mutex lock"), the assignment to err is redundant. So remove it. Signed-off-by:
Jing Xiangfeng <jingxiangfeng@huawei.com> Signed-off-by:
-
Jisheng Zhang authored
Enable ALDPS(Advanced Link Down Power Saving) to save power when link down. Signed-off-by:
Jisheng Zhang <Jisheng.Zhang@synaptics.com> Reviewed-by:
-
Linus Walleij authored
There is an off-by-one error in rtl8366rb_is_vlan_valid() making VLANs 0..4094 valid while it should be 1..4095. Fix it. Fixes: d8652956 ("net: dsa: realtek-smi: Add Realtek SMI driver") Signed-off-by:
Linus Walleij <linus.walleij@linaro.org> Reviewed-by:
-
Alex Dewar authored
In mt7531_cpu_port_config(), if the variable port is neither 5 nor 6, then variable interface will be used uninitialised. Change the function to return -EINVAL in this case. As the return value of mt7531_cpu_port_config() is never checked (even though it returns an int) add a check in the correct place so that the error can be passed up the call stack. Now that we correctly handle errors thrown in this function, also check the return value of mt7531_mac_config() in case an error occurs here. Also add misisng checks to mt7530_setup() and mt7531_setup(), which are another level further up the call stack. Fixes: c288575f ("net: dsa: mt7530: Add the support of MT7531 switch") Addresses-Coverity: 1496993 ("Uninitialized variables") Signed-off-by:
Alex Dewar <alex.dewar90@gmail.com> Signed-off-by:
-
Paolo Abeni authored
Currently the backlog status in not exposed to user-space. Since long backlogs (or simply not empty ones) can be a source of noticeable latency, -RT deployments need some way to inspect it. Additionally, there isn't a direct match between 'softnet_stat' lines and the related CPU - sd for offline CPUs are not dumped - so this patch also includes the CPU id into such entry. Signed-off-by:
Paolo Abeni <pabeni@redhat.com> Signed-off-by:
-
David S. Miller authored
Shay Agroskin says: ==================== Update license and polish ENA driver code This series adds the following: - Change driver's license into SPDX format - Capitalize all log prints in ENA driver - Fix issues raised by static checkers - Improve code readability by adding functions, fix spelling mistakes etc. - Update driver's documentation Changed from previous version: v1->v2: dropped patch that transforms pr_* log prints into dev_* prints ==================== Signed-off-by: -
Shay Agroskin authored
The PCI vendor IDs in the documentation inaccurately describe the ENA devices. For example, the 1d0f:ec20 can have LLQ support. The driver loads in LLQ mode by default, and a message is printed to the kernel ring if the mode isn't supported by the device, so the device table isn't needed. Also, LLQ can support various entry sizes, so the documentation is updated to reflect that. Interrupt moderation description is also updated to be more accurate. Signed-off-by:
Shay Agroskin <shayagr@amazon.com> Signed-off-by:
-
Shay Agroskin authored
After running Sparse checker on the driver using make C=1 M=drivers/net/ethernet/amazon/ena the only error that is thrown is: sparse: sparse: Using plain integer as NULL pointer about the line struct ena_calc_queue_size_ctx calc_queue_ctx = { 0 }; This patch fixes this warning, thus making our driver free (for now) of Sparse errors/warnings. To make a more complete work, this patch also fixes all static warnings that were found using an internal static checker. Signed-off-by:Ido Segev <idose@amazon.com> Signed-off-by:
-
Shay Agroskin authored
The formal name changes to "ENA_ADMIN_RSS_INDIRECTION_TABLE_CONFIG". Indirection is the ability to reference "something" using "something else" instead of the value itself. Indirection table, as the name implies, is the ability to reference CPU/Queue value using hash-to-CPU table instead of CPU/Queue itself. This patch renames the variable keys_num, which describes the number of words in the RSS hash key, to key_parts which makes its purpose clearer in RSS context. Signed-off-by:
Amit Bernstein <amitbern@amazon.com> Signed-off-by:
-
Shay Agroskin authored
The placement policy is printed in the process of queue creation in ena_up(). No need to print it in ena_probe(). Signed-off-by:
Arthur Kiyanovski <akiyano@amazon.com> Signed-off-by:
-
Shay Agroskin authored
Capitalize all log strings printed by the ena driver to make their format uniform across it. Also fix indentation, spelling mistakes and comments to improve code readability. This also includes adding comments to macros/enums whose purpose might be difficult to understand. Separate some code into functions to make it easier to understand the purpose of these lines. Signed-off-by:
-
Shay Agroskin authored
Make log prints in ena_netdev use the same log functions as the rest of the driver. For the sake of consistency, all prints in ena_netdev file were converted into netif_* format except where netdev struct isn't yet defined. For these places, dev_* log functions are used (similar to the patch for ena_com files). This commit leaves some corner cases which would be changed in a future patch. Signed-off-by:
-
Shay Agroskin authored
All ena files should now use SPDX format in their license string. This doesn't change the license of the files, but rather states the same license in fewer words. Also update the license years in some of the files. Signed-off-by:
-
Qinglang Miao authored
Simplify the return expression. Signed-off-by:
Qinglang Miao <miaoqinglang@huawei.com> Signed-off-by:
-
Qinglang Miao authored
Simplify the return expression. Signed-off-by:
-
Qinglang Miao authored
Simplify the return expression. Signed-off-by:
Claudiu Manoil <claudiu.manoil@nxp.com> Signed-off-by:
-
Qinglang Miao authored
Simplify the return expression. Signed-off-by:
-
Qinglang Miao authored
Simplify the return expression. Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Signed-off-by:
-
Qinglang Miao authored
Signed-off-by:
-
