summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2019-05-07net_sched: sch_fq: do not assume EDT packets are orderedEric Dumazet
TCP stack makes sure packets for a given flow are monotically increasing, but we want to allow UDP packets to use EDT as well, so that QUIC servers can use in-kernel pacing. This patch adds a per-flow rb-tree on which packets might be stored. We still try to use the linear list for the typical cases where packets are queued with monotically increasing skb->tstamp, since queue/dequeue packets on a standard list is O(1). Note that the ability to store packets in arbitrary EDT order will allow us to implement later a per TCP socket mechanism adding delays (with jitter eventually) and reorders, to implement convenient network emulators. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07IB/core, ipoib: Do not overreact to SM LID change eventDennis Dalessandro
When IPoIB receives an SM LID change event, it reacts by flushing its path record cache and rejoining multicast groups. This is the same behavior it performs when it receives a reregistration event. This behavior is unnecessary as an SM may have database backup or synchronization mechanisms which permit the SM location or LID to change without loss of multicast membership and without impact to path records. Both opensm and the OPA FM issue reregistration events if a new SM is started (or restarted with a new config) or an SM event occurs which results in loss of multicast membership records by the SM (such as opensm failover) or the SM encounters new nodes with Active ports (such as after joining 2 fabrics by connecting switches via ISLs). Hence this event can be depended on as the trigger for IPoIB cache and multicast flushing. It appears that some drivers, such as qib, and hfi1 issue the IB_EVENT_SM_CHANGE but other drivers such as mlx4 and mlx5 do not. Empirical testing on Mellanox EDR using ibv_asyncwatch has confirmed that Mellanox EDR HCAs do not generate SM change events and that opensm does generate reregistration. An SM LID change event is generated by the mentioned drivers to reflect that sm_lid and/or sm_sl in the local port info has changed. The intent of this event is to permit applications and ULPs which have a local copy of this information (or an address handle using it) to update their information. The intent is that the reregistration event (caused by the SM via a bit in Set(PortInfo)) be used to inform nodes that they need to rejoin multicast groups, resubscribe for notices and potentially update path records. When an SM migrates or fails over, a SM LID change event can occur. In response IPoIB discards path records and multicast membership and loses connectivity until these records are restored via SA requests. In very large fabrics, it may take minutes for the SM to be ready and for the SA responses to be supplied. This can result in undesirable and unnecessary IPoIB connectivity impacts. It also can result in an unnecessary storm of SA queries from all nodes in a cluster potentially followed by yet another storm if the SM issues the reregistration request. The fact the Mellanox HCAs do not even generate this event, is further evidence that on modern IB fabrics there will be no ill side effects from the proposed changes below to reduce the reaction by 3 kernel components to this event. So these changes should be benign for Mellanox IB fabrics and will benefit OPA fabrics while also making ib_core and ULP behavor "correct" as intended by the IBTA spec and kernel RDMA event APIs. Address these issues by removing IB_EVENT_SM_CHANGE handling from ipoib. IPoIB does not locally store sm_lid nor sm_sl, so it does not need to do anything on SM LID change. IPoIB makes use of other ib_core components to issue SA requests for it and those components correctly track SM LID and SM LID changes. Also in ib_core multicast handling, remove the test for IB_EVENT_SM_CHANGE. This code is moving all multicast groups to the error state, which will trigger rejoins. This code is used by IPoIB as well as the connection manager and other clients of multicast groups. This kernel module centralizes group membership status and joins since a node can only join a given group once but multiple ULPs or applications may want to join the same group. It makes use of the sa_query.c component in ib_core, which correctly trackes SM LID and SL. This component does not track SM LID nor SL itself and hence need not react to their changes. Similarly in the ib_core cache code remove the handling for the IB_EVENT_SM_CHANGE. In this function. The ib_cache_update function which is ultimately called is updating local copies of the pkey table, gid table and lmc. It does not update nor retain sm_lid nor sm_sl. As such it does not need to be called on an SM LID change. It technically also does not need to be called on a reregistration. The LID_CHANGE, PKEY_CHANGE, GID_CHANGE and port state change events (PORT_ERR, PORT_ACTICE) should be sufficient triggers. It is worth noting that the alternative of simply having the hfi1 and qib drivers not generate the SM LID change event was explored. While this would duplicate what Mellanox drivers do now, it is not the correct behavior and removes the ability for an SM to migrate without requiring reregistration. Since both opensm and OPA SM have mechanisms to backup or synchronize registration information, it is desirable to let them perform SM migrations (with LID or SL changes) without requiring reregistration when they deem it appropriate. Suggested-by: Todd Rimmer <todd.rimmer@intel.com> Tested-by: Michael Brooks <michael.brooks@intel.com> Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com> Reviewed-by: Todd Rimmer <todd.rimmer@intel.com> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2019-05-07Merge tag 'xfs-5.2-merge-4' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linuxLinus Torvalds
Pull xfs updates from Darrick Wong: "Here's a big pile of new stuff for XFS for 5.2. XFS has grown the ability to report metadata health status to userspace after online fsck checks the filesystem. The online metadata checking code is (I really hope) feature complete with the addition of checks for the global fs counters, though it'll remain EXPERIMENTAL for now. There are also fixes for thundering herds of writeback completions and some other deadlocks, fixes for theoretical integer overflow attacks on space accounting, and removal of the long-defunct 'mntpt' option which was deprecated in the mid-2000s and (it turns out) totally broken since 2011 (and nobody complained...). Summary: - Fix some more buffer deadlocks when performing an unmount after a hard shutdown. - Fix some minor space accounting issues. - Fix some use after free problems. - Make the (undocumented) FITRIM behavior consistent with other filesystems. - Embiggen the xfs geometry ioctl's data structure. - Introduce a new AG geometry ioctl. - Introduce a new online health reporting infrastructure and ioctl for userspace to query a filesystem's health status. - Enhance online scrub and repair to update the health reports. - Reduce thundering herd problems when writeback io completes. - Fix some transaction reservation type errors. - Fix integer overflow problems with delayed alloc reservation counters. - Fix some problems where we would exit to userspace without unlocking. - Fix inconsistent behavior when finishing deferred ops fails. - Strengthen scrub to check incore data against ondisk metadata. - Remove long-broken mntpt mount option. - Add an online scrub function for the filesystem summary counters, which should make online metadata scrub more or less feature complete for now. - Various cleanups" * tag 'xfs-5.2-merge-4' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: (38 commits) xfs: change some error-less functions to void types xfs: add online scrub for superblock counters xfs: don't parse the mtpt mount option xfs: always rejoin held resources during defer roll xfs: add missing error check in xfs_prepare_shift() xfs: scrub should check incore counters against ondisk headers xfs: allow scrubbers to pause background reclaim xfs: rename the speculative block allocation reclaim toggle functions xfs: track delayed allocation reservations across the filesystem xfs: fix broken bhold behavior in xrep_roll_ag_trans xfs: unlock inode when xfs_ioctl_setattr_get_trans can't get transaction xfs: kill the xfs_dqtrx_t typedef xfs: widen inode delalloc block counter to 64-bits xfs: widen quota block counters to 64-bit integers xfs: abort unaligned nowait directio early xfs: assert that we don't enter agfl freeing with a non-permanent transaction xfs: make tr_growdata a permanent transaction xfs: merge adjacent io completions of the same type xfs: remove unused m_data_workqueue xfs: implement per-inode writeback completion queues ...
2019-05-07Merge branch 'clk-parent-rewrite-1' into clk-nextStephen Boyd
- Rewrite how clk parents can be specified to be DT/clkdev based instead of just string based * clk-parent-rewrite-1: clk: Cache core in clk_fetch_parent_index() without names clk: fixed-factor: Initialize clk_init_data on stack clk: fixed-factor: Let clk framework find parent clk: Allow parents to be specified via clkspec index clk: Look for parents with clkdev based clk_lookups clk: Allow parents to be specified without string names clk: Add of_clk_hw_register() API for early clk drivers driver core: Let dev_of_node() accept a NULL dev clk: Prepare for clk registration API that uses DT nodes clkdev: Move clk creation outside of 'clocks_mutex'
2019-05-07Merge branch 'clk-ti' into clk-nextStephen Boyd
* clk-ti: clk: Remove CLK_IS_BASIC clk flag clk: ti: dra7: disable the RNG and TIMER12 clkctrl clocks on HS devices clk: ti: dra7x: prevent non-existing clkctrl clocks from registering ARM: omap2+: hwmod: drop CLK_IS_BASIC flag usage clk: ti: export the omap2_clk_is_hw_omap call
2019-05-07Merge branches 'clk-stm32f4', 'clk-tegra', 'clk-at91', 'clk-sifive-fu540' ↵Stephen Boyd
and 'clk-spdx' into clk-next - Support for STM32F769 - Rework AT91 sckc DT bindings - Fix slow RC oscillator issue on sama5d3 - AT91 sam9x60 PMC support - SiFive FU540 PRCI and PLL support * clk-stm32f4: clk: stm32mp1: Add ddrperfm clock clk: stm32: Introduce clocks of STM32F769 board * clk-tegra: clk: tegra: divider: Mark Memory Controller clock as read-only clk: tegra: emc: Replace BUG() with WARN_ONCE() clk: tegra: emc: Fix EMC max-rate clamping clk: tegra: emc: Support multiple RAM codes clk: tegra: emc: Don't enable EMC clock manually clk: tegra124: Remove lock-enable bit from PLLM clk: tegra: Fix PLLM programming on Tegra124+ when PMC overrides divider clk: tegra: Don't enable already enabled PLLs * clk-at91: clk: at91: Mark struct clk_range as const clk: at91: add sam9x60 pmc driver dt-bindings: clk: at91: add bindings for SAM9X60 pmc clk: at91: add sam9x60 PLL driver clk: at91: master: Add sam9x60 support clk: at91: usb: Add sam9x60 support clk: at91: allow configuring generated PCR layout clk: at91: allow configuring peripheral PCR layout clk: at91: sckc: handle different RC startup time clk: at91: modernize sckc binding dt-bindings: clock: at91: new sckc bindings * clk-sifive-fu540: clk: sifive: add a driver for the SiFive FU540 PRCI IP block clk: analogbits: add Wide-Range PLL library dt-bindings: clk: add documentation for the SiFive PRCI driver * clk-spdx: clk: sunxi-ng: Use the correct style for SPDX License Identifier clk: sprd: Use the correct style for SPDX License Identifier clk: renesas: Use the correct style for SPDX License Identifier clk: qcom: Use the correct style for SPDX License Identifier clk: davinci: Use the correct style for SPDX License Identifier clk: actions: Use the correct style for SPDX License Identifier
2019-05-07Merge branches 'clk-hisi', 'clk-lochnagar', 'clk-allwinner', 'clk-rockchip' ↵Stephen Boyd
and 'clk-qoriq' into clk-next - Mark UFS clk as critical on Hi-Silicon hi3660 SoCs - Support for Cirrus Logic Lochnagar clks * clk-hisi: clk: hi3660: Mark clk_gate_ufs_subsys as critical * clk-lochnagar: clk: lochnagar: Add support for the Cirrus Logic Lochnagar clk: lochnagar: Add initial binding documentation * clk-allwinner: clk: sunxi-ng: sun5i: Export the MBUS clock clk: sunxi-ng: a83t: Add pll-video0 as parent of csi-mclk clk: sunxi-ng: h6: Allow video & vpu clocks to change parent rate clk: sunxi-ng: h6: Preset hdmi-cec clock parent clk: sunxi: Add Kconfig options clk: sunxi-ng: f1c100s: fix USB PHY gate bit offset clk: sunxi-ng: Allow DE clock to set parent rate * clk-rockchip: clk: rockchip: undo several noc and special clocks as critical on rk3288 clk: rockchip: add a COMPOSITE_DIV_OFFSET clock-type clk: rockchip: Turn on "aclk_dmac1" for suspend on rk3288 clk: rockchip: Limit use of USB PHY clock to USB on rk3288 clk: rockchip: Fix video codec clocks on rk3288 clk: rockchip: Make rkpwm a critical clock on rk3288 clk: rockchip: fix wrong clock definitions for rk3328 * clk-qoriq: clk: qoriq: increase array size of cmux_to_group dt-bindings: qoriq-clock: Add ls1028a chip compatible string clk: qoriq: Add ls1028a clock configuration clk: qoriq: add more PLL divider clocks support dt-bindings: qoriq-clock: add more PLL divider clocks support
2019-05-07Merge branches 'clk-sa', 'clk-aspeed', 'clk-samsung', 'clk-ingenic' and ↵Stephen Boyd
'clk-zynq' into clk-next - Various static analysis fixes/finds - Video Engine (ECLK) support on Aspeed SoCs - Xilinx ZynqMP Versal platform support - Convert Xilinx ZynqMP driver to be struct oriented * clk-sa: clk: mvebu: fix spelling mistake "gatable" -> "gateable" clk: ux500: add range to usleep_range clk: tegra: Make tegra_clk_super_mux_ops static clk: davinci: cfgchip: use PTR_ERR_OR_ZERO in da8xx_cfgchip_register_div4p5 * clk-aspeed: clk: Aspeed: Setup video engine clocking * clk-samsung: clk: samsung: exynos5410: Add gate clock for ADC clk: samsung: dt-bindings: Add ADC clock ID to Exynos5410 clk: samsung: dt-bindings: Put CLK_UART3 in order * clk-ingenic: clk: ingenic: jz4725b: Add UDC PHY clock dt-bindings: clock: jz4725b-cgu: Add UDC PHY clock * clk-zynq: clk: zynqmp: use structs for clk query responses clk: zynqmp: fix check for fractional clock clk: zynqmp: do not export zynqmp_clk_register_* functions clk: zynqmp: fix kerneldoc of __zynqmp_clock_get_parents drivers: clk: Update clock driver to handle clock attribute drivers: clk: zynqmp: Allow zero divisor value
2019-05-07Merge branches 'clk-doc', 'clk-more-critical', 'clk-meson' and ↵Stephen Boyd
'clk-basic-be' into clk-next - Remove clk_readl() and introduce BE versions of basic clk types * clk-doc: clk: Drop duplicate clk_register() documentation clk: Document and simplify clk_core_get_rate_nolock() clk: Remove 'flags' member of struct clk_fixed_rate clk: nxp: Drop 'flags' on fixed_rate clk macro clk: Document __clk_mux_determine_rate() clk: Document CLK_MUX_READ_ONLY mux flag clk: Document deprecated things clk: Collapse gpio clk kerneldoc * clk-more-critical: clk: highbank: Convert to CLK_IS_CRITICAL * clk-meson: (21 commits) clk: meson: axg-audio: add g12a support clk: meson: axg-audio: don't register inputs in the onecell data clk: meson: axg_audio: replace prefix axg by aud dt-bindings: clk: axg-audio: add g12a support clk: meson: meson8b: add the video decoder clock trees clk: meson: meson8b: add the VPU clock trees clk: meson: meson8b: add support for the GP_PLL clock on Meson8m2 clk: meson: meson8b: use a separate clock table for Meson8m2 dt-bindings: clock: meson8b: export the video decoder clocks clk: meson-g12a: add video decoder clocks dt-bindings: clock: meson8b: export the VPU clock clk: meson-g12a: add PCIE PLL clocks dt-bindings: clock: g12a-aoclk: expose CLKID_AO_CTS_OSCIN clk: meson-pll: add reduced specific clk_ops for G12A PCIe PLL dt-bindings: clock: meson8b: drop the "ABP" clock definition clk: meson: g12a: add cpu clocks dt-bindings: clk: g12a-clkc: add VDEC clock IDs dt-bindings: clock: axg-audio: unexpose controller inputs dt-bindings: clk: g12a-clkc: add PCIE PLL clock ID clk: g12a-aoclk: re-export CLKID_AO_SAR_ADC_SEL clock id ... * clk-basic-be: clk: core: replace clk_{readl,writel} with {readl,writel} clk: core: remove powerpc special handling powerpc/512x: mark clocks as big endian clk: mux: add explicit big endian support clk: multiplier: add explicit big endian support clk: gate: add explicit big endian support clk: fractional-divider: add explicit big endian support clk: divider: add explicit big endian support
2019-05-07Merge branches 'clk-renesas', 'clk-qcom', 'clk-mtk', 'clk-milbeaut' and ↵Stephen Boyd
'clk-imx' into clk-next - Qualcomm QCS404 CDSP clk support - Qualcomm QCS404 Turing clk support - Mediatek MT8183 clock support - Mediatek MT8516 clock support - Milbeaut M10V clk controller support * clk-renesas: clk: renesas: rcar-gen3: Remove unused variable clk: renesas: rcar-gen3: Fix cpg_sd_clock_round_rate() return value clk: renesas: r8a77980: Fix RPC-IF module clock's parent clk: renesas: rcar-gen3: Rename DRIF clocks clk: renesas: rcar-gen3: Correct parent clock of Audio-DMAC clk: renesas: rcar-gen3: Correct parent clock of SYS-DMAC clk: renesas: rcar-gen3: Correct parent clock of HS-USB clk: renesas: rcar-gen3: Correct parent clock of EHCI/OHCI clk: renesas: r8a774c0: Add Z2 clock clk: renesas: r8a77990: Add Z2 clock clk: renesas: rcar-gen3: Support Z and Z2 clocks with high frequency parents math64: New DIV64_U64_ROUND_CLOSEST helper clk: renesas: rcar-gen3: Remove CLK_TYPE_GEN3_Z2 clk: renesas: rcar-gen3: Parameterise Z and Z2 clock offset clk: renesas: rcar-gen3: Parameterise Z and Z2 clock fixed divisor clk: renesas: r9a06g032: Add missing PCI USB clock clk: renesas: r7s9210: Always use readl() clk: renesas: rcar-gen3: Pass name/offset to cpg_sd_clk_register() * clk-qcom: clk: qcom: Skip halt checks on gcc_pcie_0_pipe_clk for 8998 clk: qcom: Add QCS404 TuringCC clk: qcom: branch: Add AON clock ops dt-bindings: clock: Introduce Qualcomm Turing Clock controller clk: qcom: gcc-qcs404: Add CDSP related clocks and resets * clk-mtk: clk: mediatek: add clock driver for MT8516 dt-bindings: mediatek: apmixedsys: add support for MT8516 dt-bindings: mediatek: infracfg: add support for MT8516 dt-bindings: mediatek: topckgen: add support for MT8516 clk: mediatek: Allow changing PLL rate when it is off clk: mediatek: Add MT8183 clock support clk: mediatek: Add configurable pcw_chg_reg to mtk_pll_data clk: mediatek: Add dt-bindings for MT8183 clocks dt-bindings: ARM: Mediatek: Document bindings for MT8183 clk: mediatek: Add configurable pcwibits and fmin to mtk_pll_data clk: mediatek: Add new clkmux register API clk: mediatek: Disable tuner_en before change PLL rate * clk-milbeaut: clock: milbeaut: Add Milbeaut M10V clock controller dt-bindings: clock: milbeaut: add Milbeaut clock description * clk-imx: clk: imx: correct pfdv2 gate_bit/vld_bit operations clk: imx: clk-pllv3: mark expected switch fall-throughs clk: imx8mq: Add dsi_ipg_div clk: imx: pllv4: add fractional-N pll support clk: imx: keep uart clock on during system boot clk: imx: correct i.MX7D AV PLL num/denom offset clk: imx6sll: Fix mispelling uart4_serial as serail clk: imx: pll14xx: drop unused variable clk: imx: rename clk-imx51-imx53.c to clk-imx5.c clk: imx5: Fix i.MX50 ESDHC clock registers clk: imx5: Fix i.MX50 mainbus clock registers clk: imx: Remove unused imx_get_clk_hw_fixed dt-bindings: clock: imx7ulp: remove SNVS clock clk: imx7ulp: remove snvs clock
2019-05-07Merge tag 'iomap-5.2-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linuxLinus Torvalds
Pull iomap updates from Darrick Wong: "Nothing particularly exciting here, just adding some callouts for gfs2 and cleaning a few things. Summary: - Add some extra hooks to the iomap buffered write path to enable gfs2 journalled writes - SPDX conversion - Various refactoring" * tag 'iomap-5.2-merge-2' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux: iomap: move iomap_read_inline_data around iomap: Add a page_prepare callback iomap: Fix use-after-free error in page_done callback fs: Turn __generic_write_end into a void function iomap: Clean up __generic_write_end calling iomap: convert to SPDX identifier
2019-05-07Merge tag 'jfs-5.2' of git://github.com/kleikamp/linux-shaggyLinus Torvalds
Pull jfs updates from Dave Kleikamp: "Several minor jfs fixes" * tag 'jfs-5.2' of git://github.com/kleikamp/linux-shaggy: jfs: fix bogus variable self-initialization fs/jfs: Switch to use new generic UUID API jfs: compare old and new mode before setting update_mode flag jfs: remove incorrect comment in jfs_superblock jfs: fix spelling mistake, EACCESS -> EACCES
2019-05-07Merge tag 'for-5.2-tag' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux Pull btrfs updates from David Sterba: "This time the majority of changes are cleanups, though there's still a number of changes of user interest. User visible changes: - better read time and write checks to catch errors early and before writing data to disk (to catch potential memory corruption on data that get checksummed) - qgroups + metadata relocation: last speed up patch int the series to address the slowness, there should be no overhead comparing balance with and without qgroups - FIEMAP ioctl does not start a transaction unnecessarily, this can result in a speed up and less blocking due to IO - LOGICAL_INO (v1, v2) does not start transaction unnecessarily, this can speed up the mentioned ioctl and scrub as well - fsync on files with many (but not too many) hardlinks is faster, finer decision if the links should be fsynced individually or completely - send tries harder to find ranges to clone - trim/discard will skip unallocated chunks that haven't been touched since the last mount Fixes: - send flushes delayed allocation before start, otherwise it could miss some changes in case of a very recent rw->ro switch of a subvolume - fix fallocate with qgroups that could lead to space accounting underflow, reported as a warning - trim/discard ioctl honours the requested range - starting send and dedupe on a subvolume at the same time will let only one of them succeed, this is to prevent changes that send could miss due to dedupe; both operations are restartable Core changes: - more tree-checker validations, errors reported by fuzzing tools: - device item - inode item - block group profiles - tracepoints for extent buffer locking - async cow preallocates memory to avoid errors happening too deep in the call chain - metadata reservations for delalloc reworked to better adapt in many-writers/low-space scenarios - improved space flushing logic for intense DIO vs buffered workloads - lots of cleanups - removed unused struct members - redundant argument removal - properties and xattrs - extent buffer locking - selftests - use common file type conversions - many-argument functions reduction" * tag 'for-5.2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux: (227 commits) btrfs: Use kvmalloc for allocating compressed path context btrfs: Factor out common extent locking code in submit_compressed_extents btrfs: Set io_tree only once in submit_compressed_extents btrfs: Replace clear_extent_bit with unlock_extent btrfs: Make compress_file_range take only struct async_chunk btrfs: Remove fs_info from struct async_chunk btrfs: Rename async_cow to async_chunk btrfs: Preallocate chunks in cow_file_range_async btrfs: reserve delalloc metadata differently btrfs: track DIO bytes in flight btrfs: merge calls of btrfs_setxattr and btrfs_setxattr_trans in btrfs_set_prop btrfs: delete unused function btrfs_set_prop_trans btrfs: start transaction in xattr_handler_set_prop btrfs: drop local copy of inode i_mode btrfs: drop old_fsflags in btrfs_ioctl_setflags btrfs: modify local copy of btrfs_inode flags btrfs: drop useless inode i_flags copy and restore btrfs: start transaction in btrfs_ioctl_setflags() btrfs: export btrfs_set_prop btrfs: refactor btrfs_set_props to validate externally ...
2019-05-07Merge branch 'stable-fodder' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs Pull vfs stable fodder fixes from Al Viro: - acct_on() fix for deadlock caught by overlayfs folks - autofs RCU use-after-free SNAFU (->d_manage() can be called locklessly, so we need to RCU-delay freeing the objects it looks at) - (hopefully) the end of "do we need freeing this dentry RCU-delayed" whack-a-mole. * 'stable-fodder' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: autofs: fix use-after-free in lockless ->d_manage() dcache: sort the freeing-without-RCU-delay mess for good. acct_on(): don't mess with freeze protection
2019-05-07Merge branch 'work.icache' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs Pull vfs inode freeing updates from Al Viro: "Introduction of separate method for RCU-delayed part of ->destroy_inode() (if any). Pretty much as posted, except that destroy_inode() stashes ->free_inode into the victim (anon-unioned with ->i_fops) before scheduling i_callback() and the last two patches (sockfs conversion and folding struct socket_wq into struct socket) are excluded - that pair should go through netdev once davem reopens his tree" * 'work.icache' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (58 commits) orangefs: make use of ->free_inode() shmem: make use of ->free_inode() hugetlb: make use of ->free_inode() overlayfs: make use of ->free_inode() jfs: switch to ->free_inode() fuse: switch to ->free_inode() ext4: make use of ->free_inode() ecryptfs: make use of ->free_inode() ceph: use ->free_inode() btrfs: use ->free_inode() afs: switch to use of ->free_inode() dax: make use of ->free_inode() ntfs: switch to ->free_inode() securityfs: switch to ->free_inode() apparmor: switch to ->free_inode() rpcpipe: switch to ->free_inode() bpf: switch to ->free_inode() mqueue: switch to ->free_inode() ufs: switch to ->free_inode() coda: switch to ->free_inode() ...
2019-05-07Merge branch 'hns3-next'David S. Miller
Huazhong Tan says: ==================== cleanup & optimizations & bugfixes for HNS3 driver This patchset contains some cleanup related to hns3_enet_ring struct and tx bd filling process, optimizations related to rx page reusing, barrier using and tso handling process, bugfixes related to tunnel type handling and error handling for desc filling. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: use devm_kcalloc when allocating desc_cbYunsheng Lin
This patch uses devm_kcalloc instead of kcalloc when allocating ring->desc_cb, because devm_kcalloc not only ensure to free the memory when the dev is deallocted, but also allocate the memory from it's device memory node. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: some cleanup for struct hns3_enet_ringYunsheng Lin
This patch removes some unused field in struct hns3_enet_ring, use ring->dev for ring_to_dev macro, and use dev consistently in hns3_fill_desc. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: unify the page reusing for page size 4K and 64KYunsheng Lin
When page size is 64K, RX buffer is currently not reused when the page_offset is moved to last buffer. This patch adds checking to decide whether the buffer page can be reused when last_offset is moved beyond last offset. If the driver is the only user of page when page_offset is moved to beyond last offset, then buffer can be reused and page_offset is set to zero. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: optimize the barrier using when cleaning TX BDYunsheng Lin
Currently, a barrier is used when cleaning each TX BD, which may cause performance degradation. This patch optimizes it to use one barrier when cleaning TX BD each round. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: fix error handling for desc fillingYunsheng Lin
When desc filling fails in hns3_nic_net_xmit, it will call hns3_clear_desc to unmap the dma mapping. But currently the ring->next_to_use points to the desc where the desc filling or dma mapping return error, which means the desc that ring->next_to_use points to has not done the dma mapping, the desc that need unmapping is before the ring->next_to_use. This patch fixes it by calling ring_ptr_move_bw(next_to_use) before doing unmapping operation, and set desc_cb->dma to zero to avoid freeing it again when unloading. Also, when filling skb head or frag fails, both need to unmap all the way back to next_to_use_head, so remove one desc filling error handling. Fixes: 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: combine len and checksum handling for inner and outer header.Yunsheng Lin
When filling len and checksum info to description, there is some similar checking or calculation. So this patch adds hns3_set_l2l3l4 to fill the inner(/normal) header's len and checksum info. If it is a encapsulation skb, it calls hns3_set_outer_l2l3l4 to handle the outer header's len and checksum info, in order to avoid some similar checking or calculation. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: refactor BD filling for l2l3l4 infoYunsheng Lin
This patch separates the inner and outer l2l3l4 len handling in hns3_set_l2l3l4_len, this is a preparation to combine the l2l3l4 len and checksum handling for inner and outer header. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: fix for tunnel type handling in hns3_rx_checksumYunsheng Lin
According to hardware user manual, the tunnel packet type is available in the rx.ol_info field of struct hns3_desc. Currently the tunnel packet type is decided by the rx.l234_info, which may cause RX checksum handling error. This patch fixes it by using the correct field in struct hns3_desc to decide the tunnel packet type. Fixes: 76ad4f0ee747 ("net: hns3: Add support of HNS3 Ethernet Driver for hip08 SoC") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: add linearizing checking for TSO caseYunsheng Lin
HW requires every continuous 8 buffer data to be larger than MSS, we simplify it by ensuring skb_headlen + the first continuous 7 frags to to be larger than GSO header len + mss, and the remaining continuous 7 frags to be larger than MSS except the last 7 frags. This patch adds hns3_skb_need_linearized to handle it for TSO case. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: add counter for times RX pages gets allocatedYunsheng Lin
Currently, using "ethtool --statistics" can show how many time RX page have been reused, but there is no counter for RX page not being reused. This patch adds non_reuse_pg counter to better debug the performance issue, because it is hard to determine when the RX page is reused or not if there is no such counter. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: use napi_schedule_irqoff in hard interrupts handlersYunsheng Lin
napi_schedule_irqoff is introduced to be used from hard interrupts handlers or when irqs are already masked, see: https://lists.openwall.net/netdev/2014/10/29/2 So this patch replaces napi_schedule with napi_schedule_irqoff. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: hns3: unify maybe_stop_tx for TSO and non-TSO caseYunsheng Lin
Currently, maybe_stop_tx ops for TSO and non-TSO case share some BD calculation code, so this patch unifies the maybe_stop_tx by removing the maybe_stop_tx ops. skb_is_gso() can be used to differentiate the case between TSO and non-TSO case if there is need to handle special case for TSO case. This patch also add tx_copy field in "ethtool --statistics" to help better debug the performance issue caused by calling skb_copy. Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07xtensa: implement initialize_cacheattr for MPU coresMax Filippov
Use CONFIG_MEMMAP_CACHEATTR to initialize MPU as described in the Xtensa LSP RM document. Coalesce adjacent regions with the same cacheattr. Update Kconfig help text. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2019-05-07xtensa: add exclusive atomics supportMax Filippov
Implement atomic primitives using exclusive access opcodes available in the recent xtensa cores. Since l32ex/s32ex don't have any memory ordering guarantees don't define __smp_mb__before_atomic/__smp_mb__after_atomic to make them use memw. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2019-05-07Merge branch 'net-dsa-lantiq-Add-bridge-offloading'David S. Miller
Hauke Mehrtens says: ==================== net: dsa: lantiq: Add bridge offloading This adds bridge offloading for the Intel / Lantiq GSWIP 2.1 switch. Changes since: v2: - Added Fixes tag to patch 1 - Fixed typo - added GSWIP_TABLE_MAC_BRIDGE_STATIC and made use of it - used GSWIP_TABLE_MAC_BRIDGE in more places v1: - fix typo signle -> single ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: dsa: lantiq: Add Forwarding Database accessHauke Mehrtens
This adds functions to add and remove static entries to and from the forwarding database and dump the full forwarding database. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: dsa: lantiq: Add fast age functionHauke Mehrtens
Fast aging per port is not supported directly by the hardware, it is only possible to configure a global aging time. Do the fast aging by iterating over the MAC forwarding table and remove all dynamic entries for a given port. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: dsa: lantiq: Add VLAN aware bridge offloadingHauke Mehrtens
The VLAN aware bridge offloading is similar to the VLAN unaware offloading, this makes it possible to offload the VLAN bridge functionalities. The hardware supports up to 64 VLAN bridge entries, we already use one entry for each LAN port to prevent forwarding of packets between the ports when the ports are not in a bridge, so in the end we have 57 possible VLANs. The VLAN filtering is currently only active when the ports are in a bridge, VLAN filtering for ports not in a bridge is not implemented. It is currently not possible to change between VLAN filtering and not filtering while the port is already in a bridge, this would make the driver more complicated. The VLANs are only defined on bridge entries, so we will not add anything into the hardware when the port joins a bridge if it is doing VLAN filtering, but only when an allowed VLAN is added. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: dsa: lantiq: Add VLAN unaware bridge offloadingHauke Mehrtens
This allows to offload bridges with DSA to the switch hardware and do the packet forwarding in hardware. This implements generic functions to access the switch hardware tables, which are used to control many features of the switch. This patch activates the MAC learning by removing the MAC address table lock, to prevent uncontrolled forwarding of packets between all the LAN ports, they are added into individual bridge tables entries with individual flow ids and the switch will do the MAC learning for each port separately before they are added to a real bridge. Each bridge consist of an entry in the active VLAN table and the VLAN mapping table, table entries with the same index are matching. In the VLAN unaware mode we configure everything with VLAN ID 0, but we use different flow IDs, the switch should handle all VLANs as normal payload and ignore them. When the hardware looks for the port of the destination MAC address it only takes the entries which have the same flow ID of the ingress packet. The bridges are configured with 64 possible entries with these information: Table Index, 0...63 VLAN ID, 0...4095: VLAN ID 0 is untagged flow ID, 0..63: Same flow IDs share entries in MAC learning table port map, one bit for each port number tagged port map, one bit for each port number Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07net: dsa: lantiq: Allow special tags only on CPU portHauke Mehrtens
Allow the special tag in ingress only on the CPU port and not on all ports. A packet with a special tag could circumvent the hardware forwarding and should only be allowed on the CPU port where Linux controls the port. Fixes: 14fceff4771e ("net: dsa: Add Lantiq / Intel DSA driver for vrx200)" Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07xtensa: clean up inline assembly in futex.hMax Filippov
Replace numeric inline assembly parameters with named parameters. Drop unused parameters from the futex_atomic_cmpxchg_inatomic. Use new temporary variable to hold target address in the fixup code in futex_atomic_cmpxchg_inatomic. Conditionalize function bodies so that only 'return -ENOSYS' is left in configurations without futex support. Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
2019-05-07Merge branch 'x86-fpu-for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull x86 FPU state handling updates from Borislav Petkov: "This contains work started by Rik van Riel and brought to fruition by Sebastian Andrzej Siewior with the main goal to optimize when to load FPU registers: only when returning to userspace and not on every context switch (while the task remains in the kernel). In addition, this optimization makes kernel_fpu_begin() cheaper by requiring registers saving only on the first invocation and skipping that in following ones. What is more, this series cleans up and streamlines many aspects of the already complex FPU code, hopefully making it more palatable for future improvements and simplifications. Finally, there's a __user annotations fix from Jann Horn" * 'x86-fpu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (29 commits) x86/fpu: Fault-in user stack if copy_fpstate_to_sigframe() fails x86/pkeys: Add PKRU value to init_fpstate x86/fpu: Restore regs in copy_fpstate_to_sigframe() in order to use the fastpath x86/fpu: Add a fastpath to copy_fpstate_to_sigframe() x86/fpu: Add a fastpath to __fpu__restore_sig() x86/fpu: Defer FPU state load until return to userspace x86/fpu: Merge the two code paths in __fpu__restore_sig() x86/fpu: Restore from kernel memory on the 64-bit path too x86/fpu: Inline copy_user_to_fpregs_zeroing() x86/fpu: Update xstate's PKRU value on write_pkru() x86/fpu: Prepare copy_fpstate_to_sigframe() for TIF_NEED_FPU_LOAD x86/fpu: Always store the registers in copy_fpstate_to_sigframe() x86/entry: Add TIF_NEED_FPU_LOAD x86/fpu: Eager switch PKRU state x86/pkeys: Don't check if PKRU is zero before writing it x86/fpu: Only write PKRU if it is different from current x86/pkeys: Provide *pkru() helpers x86/fpu: Use a feature number instead of mask in two more helpers x86/fpu: Make __raw_xsave_addr() use a feature number instead of mask x86/fpu: Add an __fpregs_load_activate() internal helper ...
2019-05-07vfio/mdev: Avoid inline get and put parent helpersParav Pandit
As section 15 of Documentation/process/coding-style.rst clearly describes that compiler will be able to optimize code. Hence drop inline for get and put helpers for parent. Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Fix aborting mdev child device removal if one failsParav Pandit
device_for_each_child() stops executing callback function for remaining child devices, if callback hits an error. Each child mdev device is independent of each other. While unregistering parent device, mdev core must remove all child mdev devices. Therefore, mdev_device_remove_cb() always returns success so that device_for_each_child doesn't abort if one child removal hits error. While at it, improve remove and unregister functions for below simplicity. There isn't need to pass forced flag pointer during mdev parent removal which invokes mdev_device_remove(). So simplify the flow. mdev_device_remove() is called from two paths. 1. mdev_unregister_driver() mdev_device_remove_cb() mdev_device_remove() 2. remove_store() mdev_device_remove() Fixes: 7b96953bc640 ("vfio: Mediated device Core driver") Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Follow correct remove sequenceParav Pandit
mdev_remove_sysfs_files() should follow exact mirror sequence of a create, similar to what is followed in error unwinding path of mdev_create_sysfs_files(). Fixes: 6a62c1dfb5c7 ("vfio/mdev: Re-order sysfs attribute creation") Reviewed-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Avoid masking error code to EBUSYParav Pandit
Instead of masking return error to -EBUSY, return actual error returned by the driver. Reviewed-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Drop redundant extern for exported symbolsParav Pandit
There is no need use 'extern' for exported functions. Acked-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Removed unused krefParav Pandit
Remove unused kref from the mdev_device structure. Fixes: 7b96953bc640 ("vfio: Mediated device Core driver") Reviewed-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Kirti Wankhede <kwankhede@nvidia.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07vfio/mdev: Avoid release parent reference during error pathParav Pandit
During mdev parent registration in mdev_register_device(), if parent device is duplicate, it releases the reference of existing parent device. This is incorrect. Existing parent device should not be touched. Fixes: 7b96953bc640 ("vfio: Mediated device Core driver") Reviewed-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Kirti Wankhede <kwankhede@nvidia.com> Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by: Parav Pandit <parav@mellanox.com> Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
2019-05-07Merge tag 'ktest-v5.1' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest Pull ktest updates from Steven Rostedt: "Minor updates to ktest.pl - Handle meta characters in grub memu - Use configurable reboot return code for handling ssh reboots - Display names and iteration number on error message" * tag 'ktest-v5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-ktest: ktest: introduce REBOOT_RETURN_CODE to confirm the result of REBOOT ktest: Add support for meta characters in GRUB_MENU ktest: Show name and iteration on errors
2019-05-07Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-nextDavid S. Miller
Daniel Borkmann says: ==================== pull-request: bpf-next 2019-05-06 The following pull-request contains BPF updates for your *net-next* tree. The main changes are: 1) Two AF_XDP libbpf fixes for socket teardown; first one an invalid munmap and the other one an invalid skmap cleanup, both from Björn. 2) More graceful CONFIG_DEBUG_INFO_BTF handling when pahole is not present in the system to generate vmlinux btf info, from Andrii. 3) Fix libbpf and thus fix perf build error with uClibc on arc architecture, from Vineet. 4) Fix missing libbpf_util.h header install in libbpf, from William. 5) Exclude bash-completion/bpftool from .gitignore pattern, from Masahiro. 6) Fix up rlimit in test_libbpf_open kselftest test case, from Yonghong. 7) Minor misc cleanups. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpfDavid S. Miller
Daniel Borkmann says: ==================== pull-request: bpf 2019-05-06 The following pull-request contains BPF updates for your *net* tree. The main changes are: 1) Two x32 JIT fixes: one which has buggy signed comparisons in 64 bit conditional jumps and another one for 64 bit negation, both from Wang. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
2019-05-07Merge tag 'printk-for-5.2' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk Pull printk updates from Petr Mladek: - Allow state reset of printk_once() calls. - Prevent crashes when dereferencing invalid pointers in vsprintf(). Only the first byte is checked for simplicity. - Make vsprintf warnings consistent and inlined. - Treewide conversion of obsolete %pf, %pF to %ps, %pF printf modifiers. - Some clean up of vsprintf and test_printf code. * tag 'printk-for-5.2' of git://git.kernel.org/pub/scm/linux/kernel/git/pmladek/printk: lib/vsprintf: Make function pointer_string static vsprintf: Limit the length of inlined error messages vsprintf: Avoid confusion between invalid address and value vsprintf: Prevent crash when dereferencing invalid pointers vsprintf: Consolidate handling of unknown pointer specifiers vsprintf: Factor out %pO handler as kobject_string() vsprintf: Factor out %pV handler as va_format() vsprintf: Factor out %p[iI] handler as ip_addr_string() vsprintf: Do not check address of well-known strings vsprintf: Consistent %pK handling for kptr_restrict == 0 vsprintf: Shuffle restricted_pointer() printk: Tie printk_once / printk_deferred_once into .data.once for reset treewide: Switch printk users from %pf and %pF to %ps and %pS, respectively lib/test_printf: Switch to bitmap_zalloc()
2019-05-07RDMA/device: Don't fire uevent before device is fully initializedLeon Romanovsky
When the refcount is 0 the device is invisible to netlink. However in the patch below the refcount = 1 was moved to after the device_add(). This creates a race where userspace can issue a netlink query after the device_add() event and not see the device as visible. Ensure that no uevent is fired before device is fully registered. Fixes: d79af7242bb2 ("RDMA/device: Expose ib_device_try_get(()") Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
generated by cgit (git 2.34.1) at 2025-09-14 08:38:27 +0000