aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/net/ethernet
AgeCommit message (Collapse)AuthorFilesLines
2022-03-23Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/netJakub Kicinski6-27/+77
Merge in overtime fixes, no conflicts. Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-23drivers: ethernet: cpsw: fix panic when interrupt coaleceing is set via ethtoolSondhauß, Jan1-4/+2
cpsw_ethtool_begin directly returns the result of pm_runtime_get_sync when successful. pm_runtime_get_sync returns -error code on failure and 0 on successful resume but also 1 when the device is already active. So the common case for cpsw_ethtool_begin is to return 1. That leads to inconsistent calls to pm_runtime_put in the call-chain so that pm_runtime_put is called one too many times and as result leaving the cpsw dev behind suspended. The suspended cpsw dev leads to an access violation later on by different parts of the cpsw driver. Fix this by calling the return-friendly pm_runtime_resume_and_get function. Fixes: d43c65b05b84 ("ethtool: runtime-resume netdev parent in ethnl_ops_begin") Signed-off-by: Jan Sondhauss <jan.sondhauss@wago.com> Reviewed-by: Vignesh Raghavendra <vigneshr@ti.com> Link: https://lore.kernel.org/r/20220323084725.65864-1-jan.sondhauss@wago.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-23ice: don't allow to run ice_send_event_to_aux() in atomic ctxAlexander Lobakin1-0/+3
ice_send_event_to_aux() eventually descends to mutex_lock() (-> might_sched()), so it must not be called under non-task context. However, at least two fixes have happened already for the bug splats occurred due to this function being called from atomic context. To make the emergency landings softer, bail out early when executed in non-task context emitting a warn splat only once. This way we trade some events being potentially lost for system stability and avoid any related hangs and crashes. Fixes: 348048e724a0e ("ice: Implement iidc operations") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Tested-by: Michal Kubiak <michal.kubiak@intel.com> Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com> Acked-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-23ice: fix 'scheduling while atomic' on aux critical err interruptAlexander Lobakin2-10/+17
There's a kernel BUG splat on processing aux critical error interrupts in ice_misc_intr(): [ 2100.917085] BUG: scheduling while atomic: swapper/15/0/0x00010000 ... [ 2101.060770] Call Trace: [ 2101.063229] <IRQ> [ 2101.065252] dump_stack+0x41/0x60 [ 2101.068587] __schedule_bug.cold.100+0x4c/0x58 [ 2101.073060] __schedule+0x6a4/0x830 [ 2101.076570] schedule+0x35/0xa0 [ 2101.079727] schedule_preempt_disabled+0xa/0x10 [ 2101.084284] __mutex_lock.isra.7+0x310/0x420 [ 2101.088580] ? ice_misc_intr+0x201/0x2e0 [ice] [ 2101.093078] ice_send_event_to_aux+0x25/0x70 [ice] [ 2101.097921] ice_misc_intr+0x220/0x2e0 [ice] [ 2101.102232] __handle_irq_event_percpu+0x40/0x180 [ 2101.106965] handle_irq_event_percpu+0x30/0x80 [ 2101.111434] handle_irq_event+0x36/0x53 [ 2101.115292] handle_edge_irq+0x82/0x190 [ 2101.119148] handle_irq+0x1c/0x30 [ 2101.122480] do_IRQ+0x49/0xd0 [ 2101.125465] common_interrupt+0xf/0xf [ 2101.129146] </IRQ> ... As Andrew correctly mentioned previously[0], the following call ladder happens: ice_misc_intr() <- hardirq ice_send_event_to_aux() device_lock() mutex_lock() might_sleep() might_resched() <- oops Add a new PF state bit which indicates that an aux critical error occurred and serve it in ice_service_task() in process context. The new ice_pf::oicr_err_reg is read-write in both hardirq and process contexts, but only 3 bits of non-critical data probably aren't worth explicit synchronizing (and they're even in the same byte [31:24]). [0] https://lore.kernel.org/all/YeSRUVmrdmlUXHDn@lunn.ch Fixes: 348048e724a0e ("ice: Implement iidc operations") Signed-off-by: Alexander Lobakin <alexandr.lobakin@intel.com> Tested-by: Michal Kubiak <michal.kubiak@intel.com> Acked-by: Tony Nguyen <anthony.l.nguyen@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-23net: marvell: prestera: add missing destroy_workqueue() in prestera_module_init()Yang Yingliang1-1/+3
Add the missing destroy_workqueue() before return from prestera_module_init() in the error handling case. Fixes: 4394fbcb78cf ("net: marvell: prestera: handle fib notifications") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Link: https://lore.kernel.org/r/20220322090236.1439649-1-yangyingliang@huawei.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-23drivers: net: xgene: Fix regression in CRC strippingStephane Graber1-5/+7
All packets on ingress (except for jumbo) are terminated with a 4-bytes CRC checksum. It's the responsability of the driver to strip those 4 bytes. Unfortunately a change dating back to March 2017 re-shuffled some code and made the CRC stripping code effectively dead. This change re-orders that part a bit such that the datalen is immediately altered if needed. Fixes: 4902a92270fb ("drivers: net: xgene: Add workaround for errata 10GE_8/ENET_11") Cc: stable@vger.kernel.org Signed-off-by: Stephane Graber <stgraber@ubuntu.com> Tested-by: Stephane Graber <stgraber@ubuntu.com> Link: https://lore.kernel.org/r/20220322224205.752795-1-stgraber@ubuntu.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-22net/mlx5e: Fix build warning, detected write beyond size of fieldSaeed Mahameed1-2/+1
When merged with Linus tree, the cited patch below will cause the following build warning: In function 'fortify_memset_chk', inlined from 'mlx5e_xmit_xdp_frame' at drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c:438:3: include/linux/fortify-string.h:242:25: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning] 242 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Fix that by grouping the fields to memeset in struct_group() to avoid the false alarm. Fixes: 9ded70fa1d81 ("net/mlx5e: Don't prefill WQEs in XDP SQ in the multi buffer mode") Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Suggested-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Link: https://lore.kernel.org/r/20220322172224.31849-1-saeed@kernel.org Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-22bnx2x: truncate value to original sizingBill Wendling1-2/+4
The original behavior was to print out unsigned short or unsigned char values. The change in commit d65aea8e8298 ("bnx2x: use correct format characters") prints out the whole value if not truncated. So truncate the value to an unsigned {short|char} to retain the original behavior. Fixes: d65aea8e8298 ("bnx2x: use correct format characters") Link: https://github.com/ClangBuiltLinux/linux/issues/378 Signed-off-by: Bill Wendling <morbo@google.com> Link: https://lore.kernel.org/r/20220321023155.106066-1-morbo@google.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-03-21Merge tag 'spi-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spiLinus Torvalds7-21/+7
Pull spi updates from Mark Brown: "The overwhelming bulk of this pull request is a change from Uwe Kleine-König which changes the return type of the remove() function to void as part of some wider work he's doing to do this for all bus types, causing updates to most SPI device drivers. The branch with that on has been cross merged with a couple of other trees which added new SPI drivers this cycle, I'm not expecting any build issues resulting from the change. Otherwise it's been a relatively quiet release with some new device support, a few minor features and the welcome completion of the conversion of the subsystem to use GPIO descriptors rather than numbers: - Change return type of remove() to void. - Completion of the conversion of SPI controller drivers to use GPIO descriptors rather than numbers. - Quite a few DT schema conversions. - Support for multiple SPI devices on a bus in ACPI systems. - Big overhaul of the PXA2xx SPI driver. - Support for AMD AMDI0062, Intel Raptor Lake, Mediatek MT7986 and MT8186, nVidia Tegra210 and Tegra234, Renesas RZ/V2L, Tesla FSD and Sunplus SP7021" [ And this is obviously where that spi change that snuck into the regulator tree _should_ have been :^] * tag 'spi-v5.18' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (124 commits) spi: fsi: Implement a timeout for polling status spi: Fix erroneous sgs value with min_t() spi: tegra20: Use of_device_get_match_data() spi: mediatek: add ipm design support for MT7986 spi: Add compatible for MT7986 spi: sun4i: fix typos in comments spi: mediatek: support tick_delay without enhance_timing spi: Update clock-names property for arm pl022 spi: rockchip-sfc: fix platform_get_irq.cocci warning spi: s3c64xx: Add spi port configuration for Tesla FSD SoC spi: dt-bindings: samsung: Add fsd spi compatible spi: topcliff-pch: Prevent usage of potentially stale DMA device spi: tegra210-quad: combined sequence mode spi: tegra210-quad: add acpi support spi: npcm-fiu: Fix typo ("npxm") spi: Fix Tegra QSPI example spi: qup: replace spin_lock_irqsave by spin_lock in hard IRQ spi: cadence: fix platform_get_irq.cocci warning spi: Update NXP Flexspi maintainer details dt-bindings: mfd: maxim,max77802: Convert to dtschema ...
2022-03-21Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6Linus Torvalds1-0/+1
Pull crypto updates from Herbert Xu: "API: - hwrng core now credits for low-quality RNG devices. Algorithms: - Optimisations for neon aes on arm/arm64. - Add accelerated crc32_be on arm64. - Add ffdheXYZ(dh) templates. - Disallow hmac keys < 112 bits in FIPS mode. - Add AVX assembly implementation for sm3 on x86. Drivers: - Add missing local_bh_disable calls for crypto_engine callback. - Ensure BH is disabled in crypto_engine callback path. - Fix zero length DMA mappings in ccree. - Add synchronization between mailbox accesses in octeontx2. - Add Xilinx SHA3 driver. - Add support for the TDES IP available on sama7g5 SoC in atmel" * 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (137 commits) crypto: xilinx - Turn SHA into a tristate and allow COMPILE_TEST MAINTAINERS: update HPRE/SEC2/TRNG driver maintainers list crypto: dh - Remove the unused function dh_safe_prime_dh_alg() hwrng: nomadik - Change clk_disable to clk_disable_unprepare crypto: arm64 - cleanup comments crypto: qat - fix initialization of pfvf rts_map_msg structures crypto: qat - fix initialization of pfvf cap_msg structures crypto: qat - remove unneeded assignment crypto: qat - disable registration of algorithms crypto: hisilicon/qm - fix memset during queues clearing crypto: xilinx: prevent probing on non-xilinx hardware crypto: marvell/octeontx - Use swap() instead of open coding it crypto: ccree - Fix use after free in cc_cipher_exit() crypto: ccp - ccp_dmaengine_unregister release dma channels crypto: octeontx2 - fix missing unlock hwrng: cavium - fix NULL but dereferenced coccicheck error crypto: cavium/nitrox - don't cast parameter in bit operations crypto: vmx - add missing dependencies MAINTAINERS: Add maintainer for Xilinx ZynqMP SHA3 driver crypto: xilinx - Add Xilinx SHA3 driver ...
2022-03-21net: bcmgenet: Use stronger register read/writes to assure orderingJeremy Linton1-2/+2
GCC12 appears to be much smarter about its dependency tracking and is aware that the relaxed variants are just normal loads and stores and this is causing problems like: [ 210.074549] ------------[ cut here ]------------ [ 210.079223] NETDEV WATCHDOG: enabcm6e4ei0 (bcmgenet): transmit queue 1 timed out [ 210.086717] WARNING: CPU: 1 PID: 0 at net/sched/sch_generic.c:529 dev_watchdog+0x234/0x240 [ 210.095044] Modules linked in: genet(E) nft_fib_inet nft_fib_ipv4 nft_fib_ipv6 nft_fib nft_reject_inet nf_reject_ipv4 nf_reject_ipv6 nft_reject nft_ct nft_chain_nat] [ 210.146561] ACPI CPPC: PCC check channel failed for ss: 0. ret=-110 [ 210.146927] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G E 5.17.0-rc7G12+ #58 [ 210.153226] CPPC Cpufreq:cppc_scale_freq_workfn: failed to read perf counters [ 210.161349] Hardware name: Raspberry Pi Foundation Raspberry Pi 4 Model B/Raspberry Pi 4 Model B, BIOS EDK2-DEV 02/08/2022 [ 210.161353] pstate: 80400005 (Nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 210.161358] pc : dev_watchdog+0x234/0x240 [ 210.161364] lr : dev_watchdog+0x234/0x240 [ 210.161368] sp : ffff8000080a3a40 [ 210.161370] x29: ffff8000080a3a40 x28: ffffcd425af87000 x27: ffff8000080a3b20 [ 210.205150] x26: ffffcd425aa00000 x25: 0000000000000001 x24: ffffcd425af8ec08 [ 210.212321] x23: 0000000000000100 x22: ffffcd425af87000 x21: ffff55b142688000 [ 210.219491] x20: 0000000000000001 x19: ffff55b1426884c8 x18: ffffffffffffffff [ 210.226661] x17: 64656d6974203120 x16: 0000000000000001 x15: 6d736e617274203a [ 210.233831] x14: 2974656e65676d63 x13: ffffcd4259c300d8 x12: ffffcd425b07d5f0 [ 210.241001] x11: 00000000ffffffff x10: ffffcd425b07d5f0 x9 : ffffcd4258bdad9c [ 210.248171] x8 : 00000000ffffdfff x7 : 000000000000003f x6 : 0000000000000000 [ 210.255341] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 0000000000001000 [ 210.262511] x2 : 0000000000001000 x1 : 0000000000000005 x0 : 0000000000000044 [ 210.269682] Call trace: [ 210.272133] dev_watchdog+0x234/0x240 [ 210.275811] call_timer_fn+0x3c/0x15c [ 210.279489] __run_timers.part.0+0x288/0x310 [ 210.283777] run_timer_softirq+0x48/0x80 [ 210.287716] __do_softirq+0x128/0x360 [ 210.291392] __irq_exit_rcu+0x138/0x140 [ 210.295243] irq_exit_rcu+0x1c/0x30 [ 210.298745] el1_interrupt+0x38/0x54 [ 210.302334] el1h_64_irq_handler+0x18/0x24 [ 210.306445] el1h_64_irq+0x7c/0x80 [ 210.309857] arch_cpu_idle+0x18/0x2c [ 210.313445] default_idle_call+0x4c/0x140 [ 210.317470] cpuidle_idle_call+0x14c/0x1a0 [ 210.321584] do_idle+0xb0/0x100 [ 210.324737] cpu_startup_entry+0x30/0x8c [ 210.328675] secondary_start_kernel+0xe4/0x110 [ 210.333138] __secondary_switched+0x94/0x98 The assumption when these were relaxed seems to be that device memory would be mapped non reordering, and that other constructs (spinlocks/etc) would provide the barriers to assure that packet data and in memory rings/queues were ordered with respect to device register reads/writes. This itself seems a bit sketchy, but the real problem with GCC12 is that it is moving the actual reads/writes around at will as though they were independent operations when in truth they are not, but the compiler can't know that. When looking at the assembly dumps for many of these routines its possible to see very clean, but not strictly in program order operations occurring as the compiler would be free to do if these weren't actually register reads/write operations. Its possible to suppress the timeout with a liberal bit of dma_mb()'s sprinkled around but the device still seems unable to reliably send/receive data. A better plan is to use the safer readl/writel everywhere. Since this partially reverts an older commit, which notes the use of the relaxed variants for performance reasons. I would suggest that any performance problems with this commit are targeted at relaxing only the performance critical code paths after assuring proper barriers. Fixes: 69d2ea9c79898 ("net: bcmgenet: Use correct I/O accessors") Reported-by: Peter Robinson <pbrobinson@gmail.com> Signed-off-by: Jeremy Linton <jeremy.linton@arm.com> Acked-by: Peter Robinson <pbrobinson@gmail.com> Tested-by: Peter Robinson <pbrobinson@gmail.com> Acked-by: Florian Fainelli <f.fainelli@gmail.com> Link: https://lore.kernel.org/r/20220310045358.224350-1-jeremy.linton@arm.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-21devlink: hold the instance lock during eswitch_mode callbacksJakub Kicinski3-34/+49
Make the devlink core hold the instance lock during eswitch_mode callbacks. Cheat in case of mlx5 (see the cover letter). Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Reviewed-by: Jiri Pirko <jiri@nvidia.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21bnxt: use the devlink instance lock to protect sriovJakub Kicinski4-11/+4
In prep for .eswitch_mode_set being called with the devlink instance lock held use that lock explicitly instead of creating a local mutex just for the sriov reconfig. Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21i825xx: fix typo "Frame to short" -> "Frame too short"Tong Zhang1-1/+1
"Frame to short" -> "Frame too short" Signed-off-by: Tong Zhang <ztong0001@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21net: sparx5: Add mdb handlersCasper Andersson3-10/+136
Adds mdb handlers. Uses the PGID arbiter to find a free entry in the PGID table for the multicast group port mask. Signed-off-by: Casper Andersson <casper.casan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21net: sparx5: Add arbiter for managing PGID tableCasper Andersson4-1/+85
The PGID (Port Group ID) table holds port masks for different purposes. The first 72 are reserved for port destination masks, flood masks, and CPU forwarding. The rest are shared between multicast, link aggregation, and virtualization profiles. The GLAG area is reserved to not be used by anything else, since it is a subset of the MCAST area. The arbiter keeps track of which entries are in use. You can ask for a free ID or give back one you are done using. Signed-off-by: Casper Andersson <casper.casan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: nfdk: implement xdp tx path for NFDKYinjun Zhang2-5/+208
Due to the different definition of txbuf in NFDK comparing to NFD3, there're no pre-allocated txbufs for xdp use in NFDK's implementation, we just use the existed rxbuf and recycle it when xdp tx is completed. For each packet to transmit in xdp path, we cannot use more than `NFDK_TX_DESC_PER_SIMPLE_PKT` txbufs, one is to stash virtual address, and another is for dma address, so currently the amount of transmitted bytes is not accumulated. Also we borrow the last bit of virtual addr to indicate a new transmitted packet due to address's alignment attribution. Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: add support for NFDK data pathJakub Kicinski9-6/+1715
Add new data path. The TX is completely different, each packet has multiple descriptor entries (between 2 and 32). TX ring is divided into blocks 32 descriptor, and descritors of one packet can't cross block bounds. The RX side is the same for now. ABI version 5 or later is required. There is no support for VLAN insertion on TX. XDP_TX action and AF_XDP zero-copy is not implemented in NFDK path. Changes to Jakub's work: * Move statistics of hw_csum_tx after jumbo packet's segmentation. * Set L3_CSUM flag to enable recaculating of L3 header checksum in ipv4 case. * Mark the case of TSO a packet with metadata prepended as unsupported. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Xingfeng Hu <xingfeng.hu@corigine.com> Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by: Dianchao Wang <dianchao.wang@corigine.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: choose data path based on versionJakub Kicinski6-19/+41
Prepare for choosing data path based on the firmware version field. Exploit one bit from the reserved byte in the firmware version field as the data path type. We need the firmware version right after vNIC is allocated, so it has to be read inside nfp_net_alloc(), callers don't have to set it afterwards. Following patches will bring the implementation of the second data path. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: add per-data path feature maskJakub Kicinski3-0/+20
Make sure that features supported only by some of the data paths are not enabled for all. Add a mask of supported features into the data path op structure. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: use TX ring pointer write backJakub Kicinski6-7/+56
Newer versions of the PCIe microcode support writing back the position of the TX pointer back into host memory. This speeds up TX completions, because we avoid a read from device memory (replacing PCIe read with DMA coherent read). Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: move tx_ring->qcidx into cold dataJakub Kicinski1-2/+5
QCidx is not used on fast path, move it to the lower cacheline. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: prepare for multi-part descriptorsJakub Kicinski4-5/+10
New datapaths may use multiple descriptor units to describe a single packet. Prepare for that by adding a descriptors per simple frame constant into ring size calculations. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: use callbacks for slow path ring related functionsJakub Kicinski8-114/+148
To reduce the coupling of slow path ring implementations and their callers, use callbacks instead. Changes to Jakub's work: * Also use callbacks for xmit functions Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Yinjun Zhang <yinjun.zhang@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: move the fast path code to separate filesJakub Kicinski12-2552/+2758
In preparation for support for a new datapath format move all ring and fast path logic into separate files. It is basically a verbatim move with some wrapping functions, no new structures and functions added. The current data path is called NFD3 from the initial version of the driver ABI it used. The non-fast path, but ring related functions are moved to nfp_net_dp.c file. Changes to Jakub's work: * Rebase on xsk related code. * Split the patch, move the callback changes to next commit. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21nfp: calculate ring masks without conditionalsJakub Kicinski1-4/+4
Ring enable masks are 64bit long. Replace mask calculation from: block_cnt == 64 ? 0xffffffffffffffffULL : (1 << block_cnt) - 1 with: (U64_MAX >> (64 - block_cnt)) to simplify the code. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21net: sparx5: Use vid 1 when bridge default vid 0 to avoid collisionCasper Andersson1-3/+23
Standalone ports use vid 0. Let the bridge use vid 1 when "vlan_default_pvid 0" is set to avoid collisions. Since no VLAN is created when default pvid is 0 this is set at "PORT_ATTR_SET" and handled in the Switchdev fdb handler. Signed-off-by: Casper Andersson <casper.casan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-21qed: remove unnecessary memset in qed_init_fw_funcsWan Jiabing1-2/+0
allocated_mem is allocated by kcalloc(). The memory is set to zero. It is unnecessary to call memset again. Signed-off-by: Wan Jiabing <wanjiabing@vivo.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-19Merge tag 'mlx5-updates-2022-03-18' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linuxDavid S. Miller13-113/+329
Saeed Mahameed says: ==================== mlx5-updates-2022-03-18 1) XDP multi buffer support This series enables XDP on non-linear legacy RQ in multi buffer mode. When XDP is enabled, fragmentation scheme on non-linear legacy RQ is adjusted to comply to limitations of XDP multi buffer (fragments of the same size). DMA addresses of fragments are stored in struct page for the completion handler to be able to unmap them. XDP_TX is supported. XDP_REDIRECT is not yet supported, the XDP core blocks it for multi buffer packets at the moment. 2) Trivial cleanups ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-18Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queueJakub Kicinski2-6/+5
Tony Nguyen says: ==================== 40GbE Intel Wired LAN Driver Updates 2022-03-17 This series contains updates to i40e and igb drivers. Tom Rix moves a conversion to little endian to occur only when the value is used for i40e. He also zeros out a structure to resolve possible use of garbage value for igb as reported by clang. * '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue: igb: zero hwtstamp by default i40e: little endian only valid checksums ==================== Link: https://lore.kernel.org/r/20220317160236.3534321-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-18qlcnic: remove redundant assignment to variable indexColin Ian King1-1/+1
Variable index is being assigned a value that is never read, it is being re-assigned later in a following for-loop. The assignment is redundant and can be removed. Cleans up clang scan build warning: drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c:1358:17: warning: Although the value stored to 'index' is used in the enclosing expression, the value is never actually read from 'index' [deadcode.DeadStores] Signed-off-by: Colin Ian King <colin.i.king@gmail.com> Link: https://lore.kernel.org/r/20220318012035.89482-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-18atl1c: remove redundant assignment to variable sizeColin Ian King1-1/+1
Variable sie is being assigned a value that is never read. The The assignment is redundant and can be removed. Cleans up clang scan build warning: drivers/net/ethernet/atheros/atl1c/atl1c_main.c:1054:22: warning: Although the value stored to 'size' is used in the enclosing expression, the value is never actually read from 'size' [deadcode.DeadStores] Signed-off-by: Colin Ian King <colin.i.king@gmail.com> Link: https://lore.kernel.org/r/20220318005021.82073-1-colin.i.king@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-03-18net/mlx5e: HTB, remove unused function declarationSaeed Mahameed1-1/+0
There is no function mlx5e_get_sq(), remove the declaration. Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Moshe Tal <moshet@nvidia.com> Reviewed-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
2022-03-18net/mlx5e: Statify function mlx5_cmd_trigger_completionsTariq Toukan2-2/+1
Starting from commit 4cab346bcf74 ("net/mlx5: No command allowed when command interface is not ready"), no calls to mlx5_cmd_trigger_completions() are external to cmd.c anymore. Make it a static function. Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Gal Pressman <gal@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Remove MLX5E_XDP_TX_DS_COUNTMaxim Mikityanskiy2-2/+1
After introducing multi-buffer XDP_TX, the MLX5E_XDP_TX_DS_COUNT define became misleading. It's no longer the DS count of an XDP_TX WQE, this WQE can be longer because of fragments. As this define is only used at one place in mlx5e_open_xdpsq(), it's also not very useful anymore. This commit removes the define and puts the calculation of ds_count for prefilled single-fragment WQEs inline. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Permit XDP with non-linear legacy RQMaxim Mikityanskiy1-12/+27
Now that legacy RQ implements XDP in the non-linear mode, stop blocking this configuration. Allow non-linear mode only for programs aware of multi buffer. XDP performance with linear mode RQ hasn't changed. Baseline (MTU 1500, TX MPWQE, legacy RQ, single core): 60-byte packets, XDP_DROP: 11.25 Mpps 60-byte packets, XDP_TX: 9.0 Mpps 60-byte packets, XDP_PASS: 668 kpps Multi buffer (MTU 9000, TX MPWQE, legacy RQ, single core): 60-byte packets, XDP_DROP: 10.1 Mpps 60-byte packets, XDP_TX: 6.6 Mpps 60-byte packets, XDP_PASS: 658 kpps 8900-byte packets, XDP_DROP: 769 kpps (100% of sent packets) 8900-byte packets, XDP_TX: 674 kpps (100% of sent packets) 8900-byte packets, XDP_PASS: 637 kpps Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Support multi buffer XDP_TXMaxim Mikityanskiy1-8/+31
This commit enables passing multi buffer XDP frames to the TX handlers on XDP_TX. Fragments are DMA synchronized to the device and queued to the xdpi_fifo for a subsequent unmapping. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Unindent the else-block in mlx5e_xmit_xdp_buffMaxim Mikityanskiy1-14/+20
The next commit will add more indentation levels to mlx5e_xmit_xdp_buff. To keep indentation minimal, unindent the else-block of the if-statement by doing an early return. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Implement sending multi buffer XDP framesMaxim Mikityanskiy4-22/+80
xmit_xdp_frame is extended to support sending fragmented XDP frames. The next commit will start using this functionality. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Don't prefill WQEs in XDP SQ in the multi buffer modeMaxim Mikityanskiy6-5/+46
When MPWQE is disabled, mlx5e_open_xdpsq() prefills the common fields of WQEs in the XDP SQ to save time when sending packets. mlx5e_xmit_xdp_frame() runs on the prefilled fields, however, sending multi buffer XDP frames would require changing some of these fields on a per-packet basis. Besides that, mlx5e_xmit_xdp_frame() will be used as a fallback to send multi buffer XDP frames when MPWQE is enabled (MPWQE can only handle linear packets). In order to prepare for XDP multi buffer support, this commit introduces a mode for mlx5e_xmit_xdp_frame() that fills all the fields itself. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Remove assignment of inline_hdr.sz on XDP TXMaxim Mikityanskiy1-1/+0
When MPWQE is disabled, mlx5e_open_xdpsq prefills the common fields of WQEs in the XDP SQ to save time when sending packets. One of such fields is eseg->inline_hdr.sz, which can be either 0 or MLX5E_XDP_MIN_INLINE, depending on the inline mode of the SQ. The inline mode can't change during the lifetime of the SQ, so setting this field again in mlx5e_xmit_xdp_frame is redundant. Moreover, the xmit function only sets it to MLX5E_XDP_MIN_INLINE, but not to 0 in the other case. This commit removes the redundant assignment in mlx5e_xmit_xdp_frame. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Move mlx5e_xdpi_fifo_push out of xmit_xdp_frameMaxim Mikityanskiy4-11/+13
The implementations of xmit_xdp_frame get the xdpi parameter of type struct mlx5e_xdp_info for the sole purpose of calling mlx5e_xdpi_fifo_push() on success. This commit moves this call outside of xmit_xdp_frame, shifting this responsibility to the caller. It will allow more fine-grained handling of XDP info for cases when an xdp_frame is fragmented. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Store DMA address inside struct pageMaxim Mikityanskiy6-33/+33
Use page_pool_set_dma_addr() to store the DMA address of a page inside struct page, in order to avoid passing struct mlx5e_dma_info to XDP handlers. Previously, struct mlx5e_dma_info was used to pass both the DMA address and the page, and it worked well for the single-fragment case. When XDP multi buffer is in use, and a fragmented xdp_frame has to be transmitted, the driver needs to know the DMA addresses of fragments, however, the array of fragments in struct skb_shared_info doesn't contain them. In order to pass the DMA addresses, the driver puts them into struct page itself, which is accessible from the array of fragments in struct skb_shared_info. The existing XDP handlers are modified to remove the dependency on struct mlx5e_dma_info. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Add XDP multi buffer support to the non-linear legacy RQMaxim Mikityanskiy2-0/+18
This commit adds XDP multi buffer support to the RX path in the non-linear legacy RQ mode. mlx5e_xdp_handle is called from mlx5e_skb_from_cqe_nonlinear. XDP_TX action for fragmented XDP frames is not yet supported and blocked. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Use page-sized fragments with XDP multi bufferMaxim Mikityanskiy1-1/+1
The implementation of XDP in mlx5e assumes that the frame size is equal to the page size. Force this limitation in the non-linear mode for XDP multi buffer. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Use fragments of the same size in non-linear legacy RQ with XDPMaxim Mikityanskiy1-9/+19
XDP multi buffer implementation in the kernel assumes that all fragments have the same size. bpf_xdp_frags_increase_tail uses this assumption to get the size of the last fragment, and __xdp_build_skb_from_frame uses it to calculate truesize as nr_frags * xdpf->frame_sz. The current implementation of mlx5e uses fragments of different size in non-linear legacy RQ. Specifically, the last fragment can be larger than the others. It's an optimization for packets smaller than MTU. This commit adapts mlx5e to the kernel limitations and makes it use fragments of the same size, in order to add support for XDP multi buffer. The change is applied only if XDP is active, otherwise the old optimization still applies. Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18net/mlx5e: Prepare non-linear legacy RQ for XDP multi buffer supportMaxim Mikityanskiy1-14/+61
mlx5e_skb_from_cqe_nonlinear creates an xdp_buff first, putting the first fragment as the linear part, and the rest of fragments as fragments to struct skb_shared_info in the tailroom. Then it creates an SKB in place, based on the xdp_buff. The XDP program is not called in this commit yet. This commit contains no functional change, except the SKB is built over the whole frag_stride of the first fragment, instead of the minimal size required (headroom, data and skb_shared_info). Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com> Reviewed-by: Tariq Toukan <tariqt@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2022-03-18ibmvnic: fix race between xmit and resetSukadev Bhattiprolu2-15/+55
There is a race between reset and the transmit paths that can lead to ibmvnic_xmit() accessing an scrq after it has been freed in the reset path. It can result in a crash like: Kernel attempted to read user page (0) - exploit attempt? (uid: 0) BUG: Kernel NULL pointer dereference on read at 0x00000000 Faulting instruction address: 0xc0080000016189f8 Oops: Kernel access of bad area, sig: 11 [#1] ... NIP [c0080000016189f8] ibmvnic_xmit+0x60/0xb60 [ibmvnic] LR [c000000000c0046c] dev_hard_start_xmit+0x11c/0x280 Call Trace: [c008000001618f08] ibmvnic_xmit+0x570/0xb60 [ibmvnic] (unreliable) [c000000000c0046c] dev_hard_start_xmit+0x11c/0x280 [c000000000c9cfcc] sch_direct_xmit+0xec/0x330 [c000000000bfe640] __dev_xmit_skb+0x3a0/0x9d0 [c000000000c00ad4] __dev_queue_xmit+0x394/0x730 [c008000002db813c] __bond_start_xmit+0x254/0x450 [bonding] [c008000002db8378] bond_start_xmit+0x40/0xc0 [bonding] [c000000000c0046c] dev_hard_start_xmit+0x11c/0x280 [c000000000c00ca4] __dev_queue_xmit+0x564/0x730 [c000000000cf97e0] neigh_hh_output+0xd0/0x180 [c000000000cfa69c] ip_finish_output2+0x31c/0x5c0 [c000000000cfd244] __ip_queue_xmit+0x194/0x4f0 [c000000000d2a3c4] __tcp_transmit_skb+0x434/0x9b0 [c000000000d2d1e0] __tcp_retransmit_skb+0x1d0/0x6a0 [c000000000d2d984] tcp_retransmit_skb+0x34/0x130 [c000000000d310e8] tcp_retransmit_timer+0x388/0x6d0 [c000000000d315ec] tcp_write_timer_handler+0x1bc/0x330 [c000000000d317bc] tcp_write_timer+0x5c/0x200 [c000000000243270] call_timer_fn+0x50/0x1c0 [c000000000243704] __run_timers.part.0+0x324/0x460 [c000000000243894] run_timer_softirq+0x54/0xa0 [c000000000ea713c] __do_softirq+0x15c/0x3e0 [c000000000166258] __irq_exit_rcu+0x158/0x190 [c000000000166420] irq_exit+0x20/0x40 [c00000000002853c] timer_interrupt+0x14c/0x2b0 [c000000000009a00] decrementer_common_virt+0x210/0x220 --- interrupt: 900 at plpar_hcall_norets_notrace+0x18/0x2c The immediate cause of the crash is the access of tx_scrq in the following snippet during a reset, where the tx_scrq can be either NULL or an address that will soon be invalid: ibmvnic_xmit() { ... tx_scrq = adapter->tx_scrq[queue_num]; txq = netdev_get_tx_queue(netdev, queue_num); ind_bufp = &tx_scrq->ind_buf; if (test_bit(0, &adapter->resetting)) { ... } But beyond that, the call to ibmvnic_xmit() itself is not safe during a reset and the reset path attempts to avoid this by stopping the queue in ibmvnic_cleanup(). However just after the queue was stopped, an in-flight ibmvnic_complete_tx() could have restarted the queue even as the reset is progressing. Since the queue was restarted we could get a call to ibmvnic_xmit() which can then access the bad tx_scrq (or other fields). We cannot however simply have ibmvnic_complete_tx() check the ->resetting bit and skip starting the queue. This can race at the "back-end" of a good reset which just restarted the queue but has not cleared the ->resetting bit yet. If we skip restarting the queue due to ->resetting being true, the queue would remain stopped indefinitely potentially leading to transmit timeouts. IOW ->resetting is too broad for this purpose. Instead use a new flag that indicates whether or not the queues are active. Only the open/ reset paths control when the queues are active. ibmvnic_complete_tx() and others wake up the queue only if the queue is marked active. So we will have: A. reset/open thread in ibmvnic_cleanup() and __ibmvnic_open() ->resetting = true ->tx_queues_active = false disable tx queues ... ->tx_queues_active = true start tx queues B. Tx interrupt in ibmvnic_complete_tx(): if (->tx_queues_active) netif_wake_subqueue(); To ensure that ->tx_queues_active and state of the queues are consistent, we need a lock which: - must also be taken in the interrupt path (ibmvnic_complete_tx()) - shared across the multiple queues in the adapter (so they don't become serialized) Use rcu_read_lock() and have the reset thread synchronize_rcu() after updating the ->tx_queues_active state. While here, consolidate a few boolean fields in ibmvnic_adapter for better alignment. Based on discussions with Brian King and Dany Madden. Fixes: 7ed5b31f4a66 ("net/ibmvnic: prevent more than one thread from running in reset") Reported-by: Vaishnavi Bhat <vaish123@in.ibm.com> Signed-off-by: Sukadev Bhattiprolu <sukadev@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-18net: lan743x: Add support for PTP-IO Event Output (Periodic Output)Raju Lakkaraju3-2/+250
Add support for PTP-IO Event Output (Periodic Output - perout) for PCI11010/PCI11414 chips Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2022-03-18net: lan743x: Add support for PTP-IO Event Input External Timestamp (extts)Raju Lakkaraju3-34/+384
PTP-IOs block provides for time stamping PTP-IO input events. PTP-IOs are numbered from 0 to 11. When a PTP-IO is enabled by the corresponding bit in the PTP-IO Capture Configuration Register, a rising or falling edge, respectively, will capture the 1588 Local Time Counter Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microchip.com> Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.