summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-01-16net/mlx5: Add copy header action struct layoutHamdan Igbaria
Add definition for copy header action, copy action is used to copy header fields from source to destination. Signed-off-by: Hamdan Igbaria <hamdani@mellanox.com> Signed-off-by: Alex Vesker <valex@mellanox.com> Reviewed-by: Alex Vesker <valex@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-01-16net/mlx5: Expose resource dump register mappingAya Levin
Add new register enumeration for resource dump. Add layout mapping for resource dump: access command and response. Signed-off-by: Aya Levin <ayal@mellanox.com> Reviewed-by: Moshe Shemesh <moshe@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-01-16net/mlx5: Add structures and defines for MIRC registerEran Ben Elisha
Add needed structures, layouts and defines for MIRC (Management Image Re-activation Control) register. This structure will be used for the FSM reactivation flow in the downstream patches. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-01-16net/mlx5: Read MCAM register groups 1 and 2Eran Ben Elisha
On load, Driver caches MCAM (Management Capabilities Mask Register) registers. in addition to the only MCAM register group (0) the driver already reads, here we add support for reading groups 1 and 2. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-01-16net/mlx5: Add structures layout for new MCAM access reg groupsEran Ben Elisha
MCAM has 3 access_reg_groups (0-2). Defines data structures in order to read and parse access_reg_groups #1 and #2. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
2020-01-16Merge tag 'qcom-arm64-defconfig-for-5.6' of ↵Olof Johansson
https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into arm/defconfig Qualcomm ARM64 defconfig updates for v5.6 * Enable NVMEM and OSM CPUfreq drivers * Enable CPR driver * Enable HFPLL driver * Enable ATH10k SNOC driver * Enable PMIC thermal driver * Enable wakeup controller driver * Enable watchdog driver * Enable PRNG driver * Enable SN65DSI86 DSI to DisplayPort bridge driver * Enable QCA Bluetooth driver * Enable Qualcomm SoCinfo driver * Enable SPI and QSPI drivers * Enable drivers providing remoteproc dependencies * tag 'qcom-arm64-defconfig-for-5.6' of https://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux: arm64: defconfig: enable CONFIG_ARM_QCOM_CPUFREQ_NVMEM arm64: defconfig: enable CONFIG_QCOM_CPR arm64: defconfig: Enable HFPLL arm64: defconfig: Enable ATH10K_SNOC arm64: defconfig: Enable QCOM PMIC thermal arm64: defconfig: enable PDC interrupt controller for Qualcomm SDM845 arm64: defconfig: Enable Qualcomm watchdog driver arm64: defconfig: Enable Qualcomm pseudo rng arm64: defconfig: Enable SN65DSI86 display bridge arm64: defconfig: Enable QCA Bluetooth over UART arm64: defconfig: Enable Qualcomm CPUfreq HW driver arm64: defconfig: Enable Qualcomm socinfo driver arm64: defconfig: Enable Qualcomm SPI and QSPI controller arm64: defconfig: Enable Qualcomm remoteproc dependencies Link: https://lore.kernel.org/r/20200113204130.GA3325@yoga Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16Merge tag 'at91-5.6-soc' of ↵Olof Johansson
git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux into arm/soc AT91 SoC for 5.5 - Document new SoC: sam9x60 - rework sam9x60 Kconfig option * tag 'at91-5.6-soc' of git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux: ARM: at91: Documentation: add sam9x60 product and datasheet ARM: at91: pm: use of_device_id array to find the proper shdwc node ARM: at91: pm: use SAM9X60 PMC's compatible ARM: debug-ll: select DEBUG_AT91_RM9200_DBGU for sam9x60 drivers: soc: atmel: select POWER_RESET_AT91_SAMA5D2_SHDWC for sam9x60 power: reset: Kconfig: select POWER_RESET_AT91_RESET for sam9x60 drivers: soc: atmel: move sam9x60 under its own config flag ARM: at91: pm: move SAM9X60's PM under its own SoC config flag ARM: at91: Kconfig: add config flag for SAM9X60 SoC ARM: at91: Kconfig: add sam9x60 pll config flag Link: https://lore.kernel.org/r/20200113161612.GA1358903@piout.net Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16Merge tag 'at91-5.6-defconfig' of ↵Olof Johansson
git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux into arm/defconfig AT91 defconfig for 5.6 - Add sam9x60 to at91_dt_defconfig * tag 'at91-5.6-defconfig' of git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux: ARM: configs: at91: enable config flags for sam9x60 SoC ARM: configs: at91: use savedefconfig Link: https://lore.kernel.org/r/20200113161033.GA1358651@piout.net Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16Merge tag 'at91-5.6-dt-1' of ↵Olof Johansson
git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux into arm/dt AT91 DT for 5.6 - Fix sama5d3 peripheral clock rate range - New boards: Overkiz Smartikz and Kizbox Mini, Microchip SAMA5D27 wlsom1-ek - sama5d2 sdmcc fixes * tag 'at91-5.6-dt-1' of git://git.kernel.org/pub/scm/linux/kernel/git/at91/linux: ARM: dts: at91: sama5d3: define clock rate range for tcb1 ARM: dts: at91: sama5d3: fix maximum peripheral clock rates ARM: dts: at91: nattis 2: remove unnecessary include ARM: dts: at91: add smartkiz support and a common kizboxmini dtsi file dt-bindings: arm: at91: Document Kizboxmini and Smartkiz boards binding ARM: dts: at91: rearrange kizbox dts using aliases nodes ARM: dts: at91: sama5d27_som1_ek: add the microchip,sdcal-inverted on sdmmc0 ARM: dts: at91: Reenable UART TX pull-ups ARM: dts: at91: sama5d2: set the sdmmc gclk frequency ARM: dts: at91: sama5d27_som1_ek: add i2c filters properties ARM: dts: at91: sama5d27_wlsom1: add SAMA5D27 wlsom1 and wlsom1-ek dt-bindings: ARM: at91: Document SAMA5D27 WLSOM1 and Evaluation Kit ARM: dts: at91: sama5d2: mark secumod as a GPIO controller ARM: dts: at91: sama5d2: disable pwm0 by default Link: https://lore.kernel.org/r/20200113155423.GA1357189@piout.net Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16selftests/bpf: Fix test_progs send_signal flakiness with nmi modeYonghong Song
Alexei observed that test_progs send_signal may fail if run with command line "./test_progs" and the tests will pass if just run "./test_progs -n 40". I observed similar issue with nmi subtest failure and added a delay 100 us in Commit ab8b7f0cb358 ("tools/bpf: Add self tests for bpf_send_signal_thread()") and the problem is gone for me. But the issue still exists in Alexei's testing environment. The current code uses sample_freq = 50 (50 events/second), which may not be enough. But if the sample_freq value is larger than sysctl kernel/perf_event_max_sample_rate, the perf_event_open syscall will fail. This patch changed nmi perf testing to use sample_period = 1, which means trying to sampling every event. This seems fixing the issue. Fixes: ab8b7f0cb358 ("tools/bpf: Add self tests for bpf_send_signal_thread()") Signed-off-by: Yonghong Song <yhs@fb.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Andrii Nakryiko <andriin@fb.com> Link: https://lore.kernel.org/bpf/20200116174004.1522812-1-yhs@fb.com
2020-01-16Merge tag 'timers-v5.5-rc6' of ↵Thomas Gleixner
https://git.linaro.org/people/daniel.lezcano/linux into timers/core Pull clocksource/events driver updates from Daniel Lezcano - Add suspend/resume for Hyper-V clocksource (Dexuan Cui) - Fix Kconfig indentation (Krzysztof Kozlowski) - Add SoC specific bindings for RZ/G2N (r8a774b1) (Biju Das) - Add Microchip PIT64B support (Claudiu Beznea) - Convert the cadence ttc driver to a platform driver to initialize later (Rajan Vaja) - Fix a memory leak when the initialization fails on bcm2835 (Colin Ian King) - Use the combo devm_platform_ioremap_resource() function for em-sti, ti-dm and switch to platform_get_irq (Yangtao Li) - Fix Exynos naming in the driver (Krzysztof Kozlowski) - Fix an uninitialized pointer access in ti-dm timer (Tony Lindgren) - Fix a sparse warning in microchip-pit64b (Claudiu Beznea) - Code reorg without functional changes for Hyper-V clocksource (Andrea Parri) - Decrease the Hyper-V clocksource rating in order to let the stable TSC to be selected instead (Andrea Parri)
2020-01-16selinux: do not allocate ancillary buffer on first loadOndrej Mosnacek
In security_load_policy(), we can defer allocating the newpolicydb ancillary array to after checking state->initialized, thereby avoiding the pointless allocation when loading policy the first time. Signed-off-by: Ondrej Mosnacek <omosnace@redhat.com> [PM: merged portions by hand] Reviewed-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Paul Moore <paul@paul-moore.com>
2020-01-16EDAC/mce_amd: Make fam_ops static globalBorislav Petkov
... and do not kmalloc a three-pointer struct. Which simplifies mce_amd_init() a bit. No functional changes. Signed-off-by: Borislav Petkov <bp@suse.de> Link: https://lkml.kernel.org/r/20200116163403.GF27148@zn.tnic
2020-01-16Merge tag 'v5.5-next-dts64' of ↵Olof Johansson
https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux into arm/dt mt8173: - add dynamic power coefficient to the cpu clusters - add jpeg decoder node mt8183: - add node for the Global Command Engine (gce) - add reset cells to the infracfg node * tag 'v5.5-next-dts64' of https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux: arm64: dts: mt8183: add reset-cells in infracfg arm64: dts: mt8173: add Mediatek JPEG Codec arm64: dts: add gce node for mt8183 arm64: dts: mt8173: Add dynamic power node. Link: https://lore.kernel.org/r/46c1a244-3f74-8069-6600-8ced02775677@gmail.com Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16Merge tag 'sunxi-dt-for-5.6-2' of ↵Olof Johansson
https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux into arm/dt This is our usual set of DT patches for the Allwinner SoCs. It's fairly big this time, but the highlights are: - Enable cpufreq and CPU thermal throttling on the A64 - CLK_CPUX macro usage removed (changed from first pull request) - CSI0 support on the R40 - CSI1 support on the A10 and A20 - SPI support on the R40 - PMU support on the H3, H5, H6 and R40 - MIPI-DSI support on the A64 - PWM support on the H6 - Thermal sensor on the A64, A83t, H3, H5, H6 and R40 - More DT schemas fixes and conversions - New boards: LibreComputer ALL-H5-CC H5, LibreComputer ALL-H3-IT H5, Pine64 H64 Model B, Neutis N5H3 * tag 'sunxi-dt-for-5.6-2' of https://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux: (52 commits) arm64: dts: allwinner: a64: enable DVFS arm64: dts: allwinner: a64: add dtsi with CPU operating points arm64: dts: allwinner: a64: add cooling maps and thermal tripping points arm64: dts: allwinner: a64: add CPU clock to CPU0-3 nodes arm64: dts: allwinner: sun50i-a64: Use macros for newly exported clocks ARM: dts: sunxi: Use macros for references to CCU clocks arm64: dts: allwinner: h5: Add Libre Computer ALL-H5-CC H5 board ARM: dts: sun8i: R40: Add SPI controllers nodes and pinmuxes arm64: dts: allwinner: a64: pinebook: Fix lid wakeup ARM: dts: sun8i: r40: Add device node for CSI0 ARM: dts: sun7i: Add CSI1 controller and pinmux options ARM: dts: sun4i: Add CSI1 controller and pinmux options ARM: dts: sunxi: Add missing LVDS resets and clocks ARM: dts: sun8i: r40: Use tcon top clock index macros ARM: dts: sun8i: R40: Add PMU node ARM: dts: sun8i: R40: Upgrade GICC reg size to 8K arm64: dts: allwinner: h6: Add thermal sensor and thermal zones ARM: dts: sunxi: Add Libre Computer ALL-H3-IT H5 board arm64: dts: allwinner: a64: Add MIPI DSI pipeline arm64: dts: allwinner: a64: Add thermal sensors and thermal zones ... Link: https://lore.kernel.org/r/20200113095555.GA29848@wens.csie.org Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16Merge tag 'soc-fsl-next-v5.6' of ↵Olof Johansson
git://git.kernel.org/pub/scm/linux/kernel/git/leo/linux into arm/drivers NXP/FSL SoC driver updates for v5.6 QUICC Engine drivers - Improve the QE drivers to be compatible with ARM/ARM64/PPC64 architectures - Various cleanups to the QE drivers * tag 'soc-fsl-next-v5.6' of git://git.kernel.org/pub/scm/linux/kernel/git/leo/linux: (49 commits) soc: fsl: qe: remove set but not used variable 'mm_gc' soc: fsl: qe: remove PPC32 dependency from CONFIG_QUICC_ENGINE soc: fsl: qe: remove unused #include of asm/irq.h from ucc.c net: ethernet: freescale: make UCC_GETH explicitly depend on PPC32 net/wan/fsl_ucc_hdlc: reject muram offsets above 64K net/wan/fsl_ucc_hdlc: fix reading of __be16 registers net/wan/fsl_ucc_hdlc: avoid use of IS_ERR_VALUE() soc: fsl: qe: avoid IS_ERR_VALUE in ucc_fast.c soc: fsl: qe: drop pointless check in qe_sdma_init() soc: fsl: qe: drop use of IS_ERR_VALUE in qe_sdma_init() soc: fsl: qe: avoid IS_ERR_VALUE in ucc_slow.c soc: fsl: qe: refactor cpm_muram_alloc_common to prevent BUG on error path soc: fsl: qe: drop broken lazy call of cpm_muram_init() soc: fsl: qe: make cpm_muram_free() ignore a negative offset soc: fsl: qe: make cpm_muram_free() return void soc: fsl: qe: change return type of cpm_muram_alloc() to s32 serial: ucc_uart: access __be32 field using be32_to_cpu serial: ucc_uart: limit brg-frequency workaround to PPC32 serial: ucc_uart: use of_property_read_u32() in ucc_uart_probe() serial: ucc_uart: stub out soft_uart_init for !CONFIG_PPC32 ... Link: https://lore.kernel.org/r/1578608351-23289-1-git-send-email-leoyang.li@nxp.com Signed-off-by: Olof Johansson <olof@lixom.net>
2020-01-16PM: suspend: Add sysfs attribute to control the "sync on suspend" behaviorJonas Meurer
The sysfs attribute `/sys/power/sync_on_suspend` controls, whether or not filesystems are synced by the kernel before system suspend. Congruously, the behaviour of build-time switch CONFIG_SUSPEND_SKIP_SYNC is slightly changed: It now defines the run-tim default for the new sysfs attribute `/sys/power/sync_on_suspend`. The run-time attribute is added because the existing corresponding build-time Kconfig flag for (`CONFIG_SUSPEND_SKIP_SYNC`) is not flexible enough. E.g. Linux distributions that provide pre-compiled kernels usually want to stick with the default (sync filesystems before suspend) but under special conditions this needs to be changed. One example for such a special condition is user-space handling of suspending block devices (e.g. using `cryptsetup luksSuspend` or `dmsetup suspend`) before system suspend. The Kernel trying to sync filesystems after the underlying block device already got suspended obviously leads to dead-locks. Be aware that you have to take care of the filesystem sync yourself before suspending the system in those scenarios. Signed-off-by: Jonas Meurer <jonas@freesources.org> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2020-01-16lib/vdso: Only read hrtimer_res when needed in __cvdso_clock_getres()Christophe Leroy
Only perform READ_ONCE(vd[CS_HRES_COARSE].hrtimer_res) for HRES and RAW clocks. Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/7ac2f0d21652f95e2bbdfa6bd514ae6c7caf53ab.1579196675.git.christophe.leroy@c-s.fr
2020-01-16MIPS: vdso: Define BUILD_VDSO32 when building a 32bit kernelThomas Gleixner
The confinement of the 32bit specific VDSO functions missed to define BUILD_VDSO32 when building a 32bit MIPS kernel: arch/mips/vdso/vgettimeofday.c: In function __vdso_clock_gettime: arch/mips/vdso/vgettimeofday.c:17:9: error: implicit declaration of function __cvdso_clock_gettime32 arch/mips/vdso/vgettimeofday.c: In function __vdso_clock_getres: arch/mips/vdso/vgettimeofday.c:39:9: error: implicit declaration of function __cvdso_clock_getres_time32 Force the define for 32bit builds in the VDSO Makefile. Fixes: bf279849ad59 ("lib/vdso: Build 32 bit specific functions in the right context") Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Paul Burton <paulburton@kernel.org> Link: https://lore.kernel.org/r/87d0bjfaqa.fsf@nanos.tec.linutronix.de
2020-01-16IB/mlx4: Fix memory leak in add_gid error flowJack Morgenstein
In procedure mlx4_ib_add_gid(), if the driver is unable to update the FW gid table, there is a memory leak in the driver's copy of the gid table: the gid entry's context buffer is not freed. If such an error occurs, free the entry's context buffer, and mark the entry as available (by setting its context pointer to NULL). Fixes: e26be1bfef81 ("IB/mlx4: Implement ib_device callbacks") Link: https://lore.kernel.org/r/20200115085050.73746-1-leon@kernel.org Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2020-01-16IB/mlx5: Expose RoCE accelerator countersAvihai Horon
Introduce the following RoCE accelerator counters: * roce_adp_retrans - number of adaptive retransmission for RoCE traffic. * roce_adp_retrans_to - number of times RoCE traffic reached time out due to adaptive retransmission. * roce_slow_restart - number of times RoCE slow restart was used. * roce_slow_restart_cnps - number of times RoCE slow restart generate CNP packets. * roce_slow_restart_trans - number of times RoCE slow restart change state to slow restart. Link: https://lore.kernel.org/r/20200115145459.83280-3-leon@kernel.org Signed-off-by: Avihai Horon <avihaih@mellanox.com> Reviewed-by: Maor Gottlieb <maorg@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2020-01-16kselftest: Minimise dependency of get_size on C library interfacesSiddhesh Poyarekar
It was observed[1] on arm64 that __builtin_strlen led to an infinite loop in the get_size selftest. This is because __builtin_strlen (and other builtins) may sometimes result in a call to the C library function. The C library implementation of strlen uses an IFUNC resolver to load the most efficient strlen implementation for the underlying machine and hence has a PLT indirection even for static binaries. Because this binary avoids the C library startup routines, the PLT initialization never happens and hence the program gets stuck in an infinite loop. On x86_64 the __builtin_strlen just happens to expand inline and avoid the call but that is not always guaranteed. Further, while testing on x86_64 (Fedora 31), it was observed that the test also failed with a segfault inside write() because the generated code for the write function in glibc seems to access TLS before the syscall (probably due to the cancellation point check) and fails because TLS is not initialised. To mitigate these problems, this patch reduces the interface with the C library to just the syscall function. The syscall function still sets errno on failure, which is undesirable but for now it only affects cases where syscalls fail. [1] https://bugs.linaro.org/show_bug.cgi?id=5479 Signed-off-by: Siddhesh Poyarekar <siddhesh@gotplt.org> Reported-by: Masami Hiramatsu <masami.hiramatsu@linaro.org> Tested-by: Masami Hiramatsu <masami.hiramatsu@linaro.org> Reviewed-by: Tim Bird <tim.bird@sony.com> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
2020-01-16selftests/livepatch: Remove unused local variable in set_ftrace_enabled()Miroslav Benes
set_ftrace_enabled() contains unused local variable "sysctl". Remove it. Signed-off-by: Miroslav Benes <mbenes@suse.cz> Reviewed-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
2020-01-16selftests/livepatch: Replace set_dynamic_debug() with setup_config() in READMEMiroslav Benes
Commit 35c9e74cff4c ("selftests/livepatch: Make dynamic debug setup and restore generic") introduced setup_config() to set up the environment for each test. It superseded set_dynamic_debug(). README still mentions set_dynamic_debug(), so update it to setup_config() which should be used now in every test. Signed-off-by: Miroslav Benes <mbenes@suse.cz> Reviewed-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
2020-01-16staging: wfx: update TODOJérôme Pouiller
Some work has been done :) Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-66-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify hif_multi_tx_confirm()Jérôme Pouiller
Usage of the "buf_loc" variable does not simplify the function. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-65-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify wfx_tx_queue_get_num_queued()Jérôme Pouiller
wfx_tx_queue_get_num_queued() can take advantage of BIT() instead of maintaining one variable for a counter and another for a mask. In add, wfx_tx_queue_get_num_queued() has no real reason to return a size_t instead of an int. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-64-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify hif_handle_tx_data()Jérôme Pouiller
Since enum action has now only two cases, it can be dropped. Then hif_handle_tx_data() can be simplified. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-63-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: remove check for interface stateJérôme Pouiller
Obviously, the value of wvif->state as no reason to be wrong. At least, if it the case, dropping the frame is probably not the bast thing to do. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-62-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: remove unused do_probeJérôme Pouiller
The identifier do_probe is unused since "staging: wfx: remove workaround to send probe requests" Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-61-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: ensure that packet_id is uniqueJérôme Pouiller
In current code, packet_id is derivated from mac802.11 packet sequence number, but this number is only unique for a station. It is not sufficient. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-60-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify wfx_ps_notify_sta()Jérôme Pouiller
wfx_ps_notify_sta() is too complex compared to the task it do. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-59-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify handling of IEEE80211_TX_CTL_SEND_AFTER_DTIMJérôme Pouiller
When mac80211 ask for a frame to be sent after a DTIM, driver should: 1. Update TIM with multicast bit set (using update_ie). This function can be called whenever. 2. Keep buffered all frames marked "after dtim" 3. When it receive a suspend_resume indication (see wfx_suspend_resume_mc()), send all the buffered frames. This indication is sent by the firmware 4ms before the dtim. 4. If one of the frames returns status "REQUEUE", it means that the DTIM period was ended before to be able to send the frame. 5. When all the buffered frames were sent or if DTIM period was ended, driver should update the TIM with multicast bit reset. All the mess with the asynchronous works can be dropped. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-58-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify wfx_suspend_resume_mc()Jérôme Pouiller
Indeed, it is not necessary to pass whole hif_ind_suspend_resume_tx to wfx_suspend_resume_mc(). In add, the structure hif_ind_suspend_resume_tx come from hardware API. It is not intended to be manipulated in upper layers of the driver. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-57-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: firmware never return PS status for stationsJérôme Pouiller
At the beginning, firmware could send suspend_resume indication to notify that a station wake up or sleep down. However, mac80211 already handles power save status of stations and this behavior has been removed from the firmware. So now, when suspend_resume indication is received, it is always to notify that a DTIM is about to be sent. So, it is possible to simply wfx_suspend_resume(). Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-56-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: sta and dtimJérôme Pouiller
wfx_ps_notify() is called for any changes in the TIM. However, association ID 0 is a very special case that should be handled independently. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-55-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: pspoll_mask make no senseJérôme Pouiller
pspoll_mask is here to send data buffered in driver. But since station is marked buffered, TIM for this station is 1 and mac80211 will call sta_notify when a ps-poll is received. So pspoll_mask is useless and sta_alseep_mask is sufficient. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-54-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: replace wfx_tx_get_tid() with ieee80211_get_tid()Jérôme Pouiller
wfx_tx_get_tid() was used as a wrapper around ieee80211_get_tid(). It did sometime return WFX_MAX_TID to ask to upper layers to not include the frame in "buffered" counter. The objective of this behavior is not clear, but tests has shown that wfx_tx_get_tid() can be replaced by ieee80211_get_tid() without any regressions. BTW, it is not necessary to save the tid in tx_rpiv since it can be retrieved from the 802.11 header. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-53-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: check that no tx is pending before release staJérôme Pouiller
Just for sanity. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-52-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify the link-id allocationJérôme Pouiller
The "link-id" is a slot number provided to the chip. A link-id is allocated to every station associated with the chip (mainly when the chip is in AP mode). It is more or less the same thing than the association ID, but it is limited to 14 values. Firmware uses the link-id to track the power save status of the stations. The current code try to associate a link-id as soon as data are exchanged with station. It is far easier to rely on sta_add() and sta_remove(). Until now the value WFX_LINK_ID_NO_ASSOC, was only used when no more link-id was available. Now, we also use this value for not-yet-associated stations (that was its primary behavior). Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-51-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: simplify wfx_set_tim_impl()Jérôme Pouiller
Argument provided to wfx_set_tim_impl() is always wvif->aid0_bit_set and there is no reason to provide another argument. Also rename wfx_set_tim_impl() into wfx_update_tim() to reflect the new behavior. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-50-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: fix RCU usageJérôme Pouiller
Indeed, sta was used after call to rcu_unlock() Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-49-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: fix bss_lossJérôme Pouiller
wfx_tx_confirm_cb() retrieves the station associated with a frame using the MAC address from the 802.11 header. In the other side wfx_tx() retrieves the station using sta field from the ieee80211_tx_control argument. In wfx_cqm_bssloss_sm(), wfx_tx() was called directly without valid sta field, but with a valid MAC address in 802.11 header. So there the processing of this packet was unbalanced and may produce weird bugs. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-48-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: relocate "buffered" information to sta_privJérôme Pouiller
It simplify the code if field buffered is hosted in the struct sta_priv instead of in the struct wfx_link_entry. More globally, struct wfx_link_entry has no real reasons to exist and should be dropped soon. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-47-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: remove handling of "early_data"Jérôme Pouiller
It seems that purpose of "early_data" was to prevent sending data to mac80211 before station was completely associated. It is a useless precaution. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-46-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: fix possible overflow on jiffies comparaisonJérôme Pouiller
MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It is recommended to use function time_*() to compare jiffies. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-45-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: fix case where RTS threshold is 0Jérôme Pouiller
If RTS threshold is 0, it currently disables RTS. It should mean "enabled for every frames". Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-44-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: do not update uapsd if not necessaryJérôme Pouiller
wfx_conf_tx() is called for each queue. On every call, the function updates UAPSD mask and PM mode for all queues. It is a pity since the UAPSD configuration very rarely changes and it makes exchanges between the host and the chip more difficult to track. This patch avoid to update UAPSD and Power Mode in most usual cases. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-43-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: with multiple vifs, force PS only if channels differsJérôme Pouiller
When multiple vif are in use (the only supported configuration is one station and one AP), the driver force power save flag on station. This behavior allows the station to leave the station channel and make its business on AP channel. However, this has a big impact on station performances (especially since only legacy PS is supported). When both vifs use the same channel, it is not necessary to keep this restriction. This greatly improve station performances. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-42-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2020-01-16staging: wfx: update power-save per interfaceJérôme Pouiller
mac80211 and the device are both able to control power-save per vif. But, the current code retrieve power-save from wfx_config(). So, it does not allow to setup power-save independently for each vif. Driver just has to rely on wfx_bss_info_changed() instead of wfx_config(). wfx_config() has nothing to do anymore, but we keep it since it is mandatory for mac80211. Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com> Link: https://lore.kernel.org/r/20200115135338.14374-41-Jerome.Pouiller@silabs.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
generated by cgit (git 2.34.1) at 2025-10-01 23:12:23 +0000