- 07 Oct, 2016 19 commits
-
-
Jon Mason authored
During the conversion to the feature flags, a check against ci->id != BCMA_CHIP_ID_BCM47162 became bgmac->feature_flags & BGMAC_FEAT_CLKCTLS instead of !(bgmac->feature_flags & BGMAC_FEAT_CLKCTLS) Reported-by:
Rafał Miłecki <zajec5@gmail.com> Signed-off-by:
Jon Mason <jon.mason@broadcom.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Eric Dumazet authored
Since linux-3.15, netlink_dump() can use up to 16384 bytes skb allocations. Due to struct skb_shared_info ~320 bytes overhead, we end up using order-3 (on x86) page allocations, that might trigger direct reclaim and add stress. The intent was really to attempt a large allocation but immediately fallback to a smaller one (order-1 on x86) in case of memory stress. On recent kernels (linux-4.4), we can remove __GFP_DIRECT_RECLAIM to meet the goal. Old kernels would need to remove __GFP_WAIT While we are at it, since we do an order-3 allocation, allow to use all the allocated bytes instead of 16384 to reduce syscalls during large dumps. iproute2 already uses 32KB recvmsg() buffer sizes. Alexei provided an initial patch downsizing to SKB_WITH_OVERHEAD(16384) Fixes: 9063e21f ("netlink: autosize skb lengthes") Signed-off-by:
Eric Dumazet <edumazet@google.com> Reported-by:
Alexei Starovoitov <ast@kernel.org> Cc: Greg Thelen <gthelen@google.com> Reviewed-by:
Greg Rose <grose@lightfleet.com> Acked-by:
-
Anoob Soman authored
If a socket has FANOUT sockopt set, a new proto_hook is registered as part of fanout_add(). When processing a NETDEV_UNREGISTER event in af_packet, __fanout_unlink is called for all sockets, but prot_hook which was registered as part of fanout_add is not removed. Call fanout_release, on a NETDEV_UNREGISTER, which removes prot_hook and removes fanout from the fanout_list. This fixes BUG_ON(!list_empty(&dev->ptype_specific)) in netdev_run_todo() Signed-off-by:
Anoob Soman <anoob.soman@citrix.com> Signed-off-by:
-
Mike Looijmans authored
The KSZ9031 skew registers contain an offset, the chip's default value is "neutral" which does not add any skew. Programming a 0 into a skew property will actually set it the maximal negative adjustment and not to a neutral position as one would expect. Explain this situation in the devicetree binding documentation and list the settings that the chip considers neutral. Changing the implementation to accept negative values would have been a better solution, but would break existing configurations. Signed-off-by:
Mike Looijmans <mike.looijmans@topic.nl> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Raju Lakkaraju authored
Wake-on-LAN (WoL) is an Ethernet networking standard that allows a computer/device to be turned on or awakened by a network message. VSC8531 PHY can support this feature configure by driver set function. WoL status get by driver get function. Tested on Beaglebone Black with VSC 8531 PHY. Signed-off-by:
Raju Lakkaraju <Raju.Lakkaraju@microsemi.com> Signed-off-by:
-
Laurent Pinchart authored
Add a new compatible string for the R8A7796 (M3-W) RAVB. Signed-off-by:
Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Reviewed-by:
Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by:
-
Mugunthan V N authored
Add support to enable CPSW RGMII internal delay (id mode) bits when rgmii internal delay is configured in phy. Signed-off-by:
Mugunthan V N <mugunthanvnm@ti.com> Signed-off-by:
-
Colin Ian King authored
Trival fix, dev_err messages are missing a \n, so add it. Also fix grammer, spelling mistake and add white spaces to various error messages. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Colin Ian King authored
Trival fix, dev_dbg message is missing a \n, so add it. Signed-off-by:
-
Colin Ian King authored
Trival fix, dev_err messages are missing a \n, so add it. Signed-off-by:
-
David S. Miller authored
Paul Durrant says: ==================== xen-netback: guest rx side refactor This series refactors the guest rx side of xen-netback: - The code is moved into its own source module. - The prefix variant of GSO handling is retired (since it is no longer in common use, and alternatives exist). - The code is then simplified and modifications made to improve performance. v2: - Rebased onto refreshed net-next ==================== Signed-off-by: -
Ross Lagerwall authored
This allows full 64K skbuffs (with 1500 mtu ethernet, composed of 45 fragments) to be handled by netback for to-guest rx. Signed-off-by:
Ross Lagerwall <ross.lagerwall@citrix.com> [re-based] Signed-off-by:
Paul Durrant <paul.durrant@citrix.com> Reviewed-by:
David Vrabel <david.vrabel@citrix.com> Signed-off-by:
-
David Vrabel authored
Instead of flushing the copy ops when an packet is complete, complete packets when their copy ops are done. This improves performance by reducing the number of grant copy hypercalls. Latency is still limited by the relatively small size of the copy batch. Signed-off-by:
-
David Vrabel authored
Instead of only placing one skb on the guest rx ring at a time, process a batch of up-to 64. This improves performance by ~10% in some tests. Signed-off-by:
-
David Vrabel authored
When an skb is removed from the guest rx queue, immediately wake the tx queue, instead of after processing them. Signed-off-by:
-
David Vrabel authored
Refactor the to-guest (rx) path to: 1. Push responses for completed skbs earlier, reducing latency. 2. Reduce the per-queue memory overhead by greatly reducing the maximum number of grant copy ops in each hypercall (from 4352 to 64). Each struct xenvif_queue is now only 44 kB instead of 220 kB. 3. Make the code more maintainable. Signed-off-by:
-
Paul Durrant authored
As far as I am aware only very old Windows network frontends make use of this style of passing GSO packets from backend to frontend. These frontends can easily be replaced by the freely available Xen Project Windows PV network frontend, which uses the 'default' mechanism for passing GSO packets, which is also used by all Linux frontends. NOTE: Removal of this feature will not cause breakage in old Windows frontends. They simply will no longer receive GSO packets - the packets instead being fragmented in the backend. Signed-off-by: -
Paul Durrant authored
The netback source module has become very large and somewhat confusing. This patch simply moves all code related to the backend to frontend (i.e guest side rx) data-path into a separate rx source module. This patch contains no functional change, it is code movement and minimal changes to avoid patch style-check issues. Signed-off-by:
-
git://git.freescale.com/ppc/upstream/linuxDavid S. Miller authored
Madalin Bucur says: ==================== fsl/fman: cleanup and small fixes This series contains fixes for the DPAA FMan driver. Adding myself as maintainer of the driver. ==================== Signed-off-by:
-
- 06 Oct, 2016 21 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fsLinus Torvalds authored
Pull f2fs updates from Jaegeuk Kim: "In this round, we've investigated how f2fs deals with errors given by our fault injection facility. With this, we could fix several corner cases. And, in order to improve the performance, we set inline_dentry by default and enhance the exisiting discard issue flow. In addition, we added f2fs_migrate_page for better memory management. Enhancements: - set inline_dentry by default - improve discard issue flow - add more fault injection cases in f2fs - allow block preallocation for encrypted files - introduce migrate_page callback function - avoid truncating the next direct node block at every checkpoint Bug fixes: - set page flag correctly between write_begin and write_end - missing error handling cases detected by fault injection - preallocate blocks regarding to 4KB alignement correctly - dentry and filename handling of encryption - lost xattrs of directories" * tag 'for-f2fs-4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (69 commits) f2fs: introduce update_ckpt_flags to clean up f2fs: don't submit irrelevant page f2fs: fix to commit bio cache after flushing node pages f2fs: introduce get_checkpoint_version for cleanup f2fs: remove dead variable f2fs: remove redundant io plug f2fs: support checkpoint error injection f2fs: fix to recover old fault injection config in ->remount_fs f2fs: do fault injection initialization in default_options f2fs: remove redundant value definition f2fs: support configuring fault injection per superblock f2fs: adjust display format of segment bit f2fs: remove dirty inode pages in error path f2fs: do not unnecessarily null-terminate encrypted symlink data f2fs: handle errors during recover_orphan_inodes f2fs: avoid gc in cp_error case f2fs: should put_page for summary page f2fs: assign return value in f2fs_gc f2fs: add customized migrate_page callback f2fs: introduce cp_lock to protect updating of ckpt_flags ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/kees/linuxLinus Torvalds authored
Pull pstore updates from Kees Cook: - Fix bug in module unloading - Switch to always using spinlock over cmpxchg - Explicitly define pstore backend's supported modes - Remove bounce buffer from pmsg - Switch to using memcpy_to/fromio() - Error checking improvements * tag 'pstore-v4.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux: ramoops: move spin_lock_init after kmalloc error checking pstore/ram: Use memcpy_fromio() to save old buffer pstore/ram: Use memcpy_toio instead of memcpy pstore/pmsg: drop bounce buffer pstore/ram: Set pstore flags dynamically pstore: Split pstore fragile flags pstore/core: drop cmpxchg based updates pstore/ramoops: fixup driver removal
-
git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linuxLinus Torvalds authored
Pull orangefs updates from Mike Marshall: "Miscellaneous improvements: - clean up debugfs globals - remove dead code in sysfs - reorganize duplicated sysfs attribute structs - consolidate sysfs show and store functions - remove duplicated sysfs_ops structures - describe organization of sysfs - make devreq_mutex static - g_orangefs_stats -> orangefs_stats for consistency - rename most remaining global variables Feature negotiation: - enable Orangefs userspace and kernel module to negotiate mutually supported features" * tag 'for-linus-4.9-ofs1' of git://git.kernel.org/pub/scm/linux/kernel/git/hubcap/linux: Revert "orangefs: bump minimum userspace version" orangefs: bump minimum userspace version orangefs: rename most remaining global variables orangefs: g_orangefs_stats -> orangefs_stats for consistency orangefs: make devreq_mutex static orangefs: describe organization of sysfs orangefs: remove duplicated sysfs_ops structures orangefs: consolidate sysfs show and store functions orangefs: reorganize duplicated sysfs attribute structs orangefs: remove dead code in sysfs orangefs: clean up debugfs globals orangefs: do not allow client readahead cache without feature bit orangefs: add features op orangefs: record userspace version for feature compatbility orangefs: add readahead count and size to sysfs orangefs: re-add flush_racache from out-of-tree orangefs: turn param response value into union orangefs: add missing param request ops orangefs: rename remaining bits of mmap readahead cache -
git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-traceLinus Torvalds authored
Pull tracing updates from Steven Rostedt: "This release cycle is rather small. Just a few fixes to tracing. The big change is the addition of the hwlat tracer. It not only detects SMIs, but also other latency that's caused by the hardware. I have detected some latency from large boxes having bus contention" * tag 'trace-v4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: tracing: Call traceoff trigger after event is recorded ftrace/scripts: Add helper script to bisect function tracing problem functions tracing: Have max_latency be defined for HWLAT_TRACER as well tracing: Add NMI tracing in hwlat detector tracing: Have hwlat trace migrate across tracing_cpumask CPUs tracing: Add documentation for hwlat_detector tracer tracing: Added hardware latency tracer ftrace: Access ret_stack->subtime only in the function profiler function_graph: Handle TRACE_BPUTS in print_graph_comment tracing/uprobe: Drop isdigit() check in create_trace_uprobe
-
git://git.kernel.org/pub/scm/linux/kernel/git/xen/tipLinus Torvalds authored
Pull xen updates from David Vrabel: "xen features and fixes for 4.9: - switch to new CPU hotplug mechanism - support driver_override in pciback - require vector callback for HVM guests (the alternate mechanism via the platform device has been broken for ages)" * tag 'for-linus-4.9-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip: xen/x86: Update topology map for PV VCPUs xen/x86: Initialize per_cpu(xen_vcpu, 0) a little earlier xen/pciback: support driver_override xen/pciback: avoid multiple entries in slot list xen/pciback: simplify pcistub device handling xen: Remove event channel notification through Xen PCI platform device xen/events: Convert to hotplug state machine xen/x86: Convert to hotplug state machine x86/xen: add missing \n at end of printk warning message xen/grant-table: Use kmalloc_array() in arch_gnttab_valloc() xen: Make VPMU init message look less scary xen: rename xen_pmu_init() in sys-hypervisor.c hotplug: Prevent alloc/free of irq descriptors during cpu up/down (again) xen/x86: Move irq allocation from Xen smp_op.cpu_up() -
git://git.kernel.org/pub/scm/virt/kvm/kvmLinus Torvalds authored
Pull KVM updates from Radim Krčmář: "All architectures: - move `make kvmconfig` stubs from x86 - use 64 bits for debugfs stats ARM: - Important fixes for not using an in-kernel irqchip - handle SError exceptions and present them to guests if appropriate - proxying of GICV access at EL2 if guest mappings are unsafe - GICv3 on AArch32 on ARMv8 - preparations for GICv3 save/restore, including ABI docs - cleanups and a bit of optimizations MIPS: - A couple of fixes in preparation for supporting MIPS EVA host kernels - MIPS SMP host & TLB invalidation fixes PPC: - Fix the bug which caused guests to falsely report lockups - other minor fixes - a small optimization s390: - Lazy enablement of runtime instrumentation - up to 255 CPUs for nested guests - rework of machine check deliver - cleanups and fixes x86: - IOMMU part of AMD's AVIC for vmexit-less interrupt delivery - Hyper-V TSC page - per-vcpu tsc_offset in debugfs - accelerated INS/OUTS in nVMX - cleanups and fixes" * tag 'kvm-4.9-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (140 commits) KVM: MIPS: Drop dubious EntryHi optimisation KVM: MIPS: Invalidate TLB by regenerating ASIDs KVM: MIPS: Split kernel/user ASID regeneration KVM: MIPS: Drop other CPU ASIDs on guest MMU changes KVM: arm/arm64: vgic: Don't flush/sync without a working vgic KVM: arm64: Require in-kernel irqchip for PMU support KVM: PPC: Book3s PR: Allow access to unprivileged MMCR2 register KVM: PPC: Book3S PR: Support 64kB page size on POWER8E and POWER8NVL KVM: PPC: Book3S: Remove duplicate setting of the B field in tlbie KVM: PPC: BookE: Fix a sanity check KVM: PPC: Book3S HV: Take out virtual core piggybacking code KVM: PPC: Book3S: Treat VTB as a per-subcore register, not per-thread ARM: gic-v3: Work around definition of gic_write_bpr1 KVM: nVMX: Fix the NMI IDT-vectoring handling KVM: VMX: Enable MSR-BASED TPR shadow even if APICv is inactive KVM: nVMX: Fix reload apic access page warning kvmconfig: add virtio-gpu to config fragment config: move x86 kvm_guest.config to a common location arm64: KVM: Remove duplicating init code for setting VMID ARM: KVM: Support vgic-v3 ... -
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespaceLinus Torvalds authored
Pull namespace updates from Eric Biederman: "This set of changes is a number of smaller things that have been overlooked in other development cycles focused on more fundamental change. The devpts changes are small things that were a distraction until we managed to kill off DEVPTS_MULTPLE_INSTANCES. There is an trivial regression fix to autofs for the unprivileged mount changes that went in last cycle. A pair of ioctls has been added by Andrey Vagin making it is possible to discover the relationships between namespaces when referring to them through file descriptors. The big user visible change is starting to add simple resource limits to catch programs that misbehave. With namespaces in general and user namespaces in particular allowing users to use more kinds of resources, it has become important to have something to limit errant programs. Because the purpose of these limits is to catch errant programs the code needs to be inexpensive to use as it always on, and the default limits need to be high enough that well behaved programs on well behaved systems don't encounter them. To this end, after some review I have implemented per user per user namespace limits, and use them to limit the number of namespaces. The limits being per user mean that one user can not exhause the limits of another user. The limits being per user namespace allow contexts where the limit is 0 and security conscious folks can remove from their threat anlysis the code used to manage namespaces (as they have historically done as it root only). At the same time the limits being per user namespace allow other parts of the system to use namespaces. Namespaces are increasingly being used in application sand boxing scenarios so an all or nothing disable for the entire system for the security conscious folks makes increasing use of these sandboxes impossible. There is also added a limit on the maximum number of mounts present in a single mount namespace. It is nontrivial to guess what a reasonable system wide limit on the number of mount structure in the kernel would be, especially as it various based on how a system is using containers. A limit on the number of mounts in a mount namespace however is much easier to understand and set. In most cases in practice only about 1000 mounts are used. Given that some autofs scenarious have the potential to be 30,000 to 50,000 mounts I have set the default limit for the number of mounts at 100,000 which is well above every known set of users but low enough that the mount hash tables don't degrade unreaonsably. These limits are a start. I expect this estabilishes a pattern that other limits for resources that namespaces use will follow. There has been interest in making inotify event limits per user per user namespace as well as interest expressed in making details about what is going on in the kernel more visible" * 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (28 commits) autofs: Fix automounts by using current_real_cred()->uid mnt: Add a per mount namespace limit on the number of mounts netns: move {inc,dec}_net_namespaces into #ifdef nsfs: Simplify __ns_get_path tools/testing: add a test to check nsfs ioctl-s nsfs: add ioctl to get a parent namespace nsfs: add ioctl to get an owning user namespace for ns file descriptor kernel: add a helper to get an owning user namespace for a namespace devpts: Change the owner of /dev/pts/ptmx to the mounter of /dev/pts devpts: Remove sync_filesystems devpts: Make devpts_kill_sb safe if fsi is NULL devpts: Simplify devpts_mount by using mount_nodev devpts: Move the creation of /dev/pts/ptmx into fill_super devpts: Move parse_mount_options into fill_super userns: When the per user per user namespace limit is reached return ENOSPC userns; Document per user per user namespace limits. mntns: Add a limit on the number of mount namespaces. netns: Add a limit on the number of net namespaces cgroupns: Add a limit on the number of cgroup namespaces ipcns: Add a limit on the number of ipc namespaces ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfsLinus Torvalds authored
Pull xfs and iomap updates from Dave Chinner: "The main things in this update are the iomap-based DAX infrastructure, an XFS delalloc rework, and a chunk of fixes to how log recovery schedules writeback to prevent spurious corruption detections when recovery of certain items was not required. The other main chunk of code is some preparation for the upcoming reflink functionality. Most of it is generic and cleanups that stand alone, but they were ready and reviewed so are in this pull request. Speaking of reflink, I'm currently planning to send you another pull request next week containing all the new reflink functionality. I'm working through a similar process to the last cycle, where I sent the reverse mapping code in a separate request because of how large it was. The reflink code merge is even bigger than reverse mapping, so I'll be doing the same thing again.... Summary for this update: - change of XFS mailing list to linux-xfs@vger.kernel.org - iomap-based DAX infrastructure w/ XFS and ext2 support - small iomap fixes and additions - more efficient XFS delayed allocation infrastructure based on iomap - a rework of log recovery writeback scheduling to ensure we don't fail recovery when trying to replay items that are already on disk - some preparation patches for upcoming reflink support - configurable error handling fixes and documentation - aio access time update race fixes for XFS and generic_file_read_iter" * tag 'xfs-for-linus-4.9-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs: (40 commits) fs: update atime before I/O in generic_file_read_iter xfs: update atime before I/O in xfs_file_dio_aio_read ext2: fix possible integer truncation in ext2_iomap_begin xfs: log recovery tracepoints to track current lsn and buffer submission xfs: update metadata LSN in buffers during log recovery xfs: don't warn on buffers not being recovered due to LSN xfs: pass current lsn to log recovery buffer validation xfs: rework log recovery to submit buffers on LSN boundaries xfs: quiesce the filesystem after recovery on readonly mount xfs: remote attribute blocks aren't really userdata ext2: use iomap to implement DAX ext2: stop passing buffer_head to ext2_get_blocks xfs: use iomap to implement DAX xfs: refactor xfs_setfilesize xfs: take the ilock shared if possible in xfs_file_iomap_begin xfs: fix locking for DAX writes dax: provide an iomap based fault handler dax: provide an iomap based dax read/write path dax: don't pass buffer_head to copy_user_dax dax: don't pass buffer_head to dax_insert_mapping ... -
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds authored
Pull networking fixups from David Miller: "Here are the build and merge fixups for the networking stuff" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: phy: micrel.c: Enable ksz9031 energy-detect power-down mode netfilter: merge fixup for "nf_tables_netdev: remove redundant ip_hdr assignment" netfilter: nft_limit: fix divided by zero panic netfilter: fix namespace handling in nf_log_proc_dostring netfilter: xt_hashlimit: Fix link error in 32bit arch because of 64bit division netfilter: accommodate different kconfig in nf_set_hooks_head netfilter: Fix potential null pointer dereference
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparcLinus Torvalds authored
Pull sparc updates from David Miller: "Besides some cleanups the major thing here is supporting relaxed ordering PCIe transactions on newer sparc64 machines, from Chris Hyser" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc: sparc: fixing ident and beautifying code sparc64: Enable setting "relaxed ordering" in IOMMU mappings sparc64: Enable PCI IOMMU version 2 API sparc: migrate exception table users off module.h and onto extable.h
-
git://git.armlinux.org.uk/~rmk/linux-armLinus Torvalds authored
Pull ARM updates from Russell King: - Correct ARMs dma-mapping to use the correct printk format strings. - Avoid defining OBJCOPYFLAGS globally which upsets lkdtm rodata testing. - Cleanups to ARMs asm/memory.h include. - L2 cache cleanups. - Allow flat nommu binaries to be executed on ARM MMU systems. - Kernel hardening - add more read-only after init annotations, including making some kernel vdso variables const. - Ensure AMBA primecell clocks are appropriately defaulted. - ARM breakpoint cleanup. - Various StrongARM 11x0 and companion chip (SA1111) updates to bring this legacy platform to use more modern APIs for (eg) GPIOs and interrupts, which will allow us in the future to reduce some of the board-level driver clutter and elimate function callbacks into board code via platform data. There still appears to be interest in these platforms! - Remove the now redundant secure_flush_area() API. - Module PLT relocation optimisations. Ard says: This series of 4 patches optimizes the ARM PLT generation code that is invoked at module load time, to get rid of the O(n^2) algorithm that results in pathological load times of 10 seconds or more for large modules on certain STB platforms. - ARMv7M cache maintanence support. - L2 cache PMU support * 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm: (35 commits) ARM: sa1111: provide to_sa1111_device() macro ARM: sa1111: add sa1111_get_irq() ARM: sa1111: clean up duplication in IRQ chip implementation ARM: sa1111: implement a gpio_chip for SA1111 GPIOs ARM: sa1111: move irq cleanup to separate function ARM: sa1111: use devm_clk_get() ARM: sa1111: use devm_kzalloc() ARM: sa1111: ensure we only touch RAB bus type devices when removing ARM: 8611/1: l2x0: add PMU support ARM: 8610/1: V7M: Add dsb before jumping in handler mode ARM: 8609/1: V7M: Add support for the Cortex-M7 processor ARM: 8608/1: V7M: Indirect proc_info construction for V7M CPUs ARM: 8607/1: V7M: Wire up caches for V7M processors with cache support. ARM: 8606/1: V7M: introduce cache operations ARM: 8605/1: V7M: fix notrace variant of save_and_disable_irqs ARM: 8604/1: V7M: Add support for reading the CTR with read_cpuid_cachetype() ARM: 8603/1: V7M: Add addresses for mem-mapped V7M cache operations ARM: 8602/1: factor out CSSELR/CCSIDR operations that use cp15 directly ARM: kernel: avoid brute force search on PLT generation ARM: kernel: sort relocation sections before allocating PLTs ...
-
git://git.armlinux.org.uk/~rmk/linux-armLinus Torvalds authored
Pull ARM fix from Russell King: "A 5% error in delay calculation was introduced during the last merge window, which had gone un-noticed until yesterday" * 'fixes' of git://git.armlinux.org.uk/~rmk/linux-arm: ARM: fix delays
-
Boris Ostrovsky authored
Early during boot topology_update_package_map() computes logical_pkg_ids for all present processors. Later, when processors are brought up, identify_cpu() updates these values based on phys_pkg_id which is a function of initial_apicid. On PV guests the latter may point to a non-existing node, causing logical_pkg_ids to be set to -1. Intel's RAPL uses logical_pkg_id (as topology_logical_package_id()) to index its arrays and therefore in this case will point to index 65535 (since logical_pkg_id is a u16). This could lead to either a crash or may actually access random memory location. As a workaround, we recompute topology during CPU bringup to reset logical_pkg_id to a valid value. (The reason for initial_apicid being bogus is because it is initial_apicid of the processor from which the guest is launched. This value is CPUID(1).EBX[31:24]) Signed-off-by:
Boris Ostrovsky <boris.ostrovsky@oracle.com> Cc: stable@vger.kernel.org Signed-off-by:
-
Russell King authored
-
Russell King authored
-
Russell King authored
Commit 215e362d ("ARM: 8306/1: loop_udelay: remove bogomips value limitation") tried to increase the bogomips limitation, but in doing so messed up udelay such that it always gives about a 5% error in the delay, even if we use a timer. The calculation is: loops = UDELAY_MULT * us_delay * ticks_per_jiffy >> UDELAY_SHIFT Originally, UDELAY_MULT was ((UL(2199023) * HZ) >> 11) and UDELAY_SHIFT 30. Assuming HZ=100, us_delay of 1000 and ticks_per_jiffy of 1660000 (eg, 166MHz timer, 1ms delay) this would calculate: ((UL(2199023) * HZ) >> 11) * 1000 * 1660000 >> 30 => 165999 With the new values of 2047 * HZ + 483648 * HZ / 1000000 and 31, we get: (2047 * HZ + 483648 * HZ / 1000000) * 1000 * 1660000 >> 31 => 158269 which is incorrect. This is due to a typo - correcting it gives: (2147 * HZ + 483648 * HZ / 1000000) * 1000 * 1660000 >> 31 => 165999 i.o.w, the original value. Fixes: 215e362d ("ARM: 8306/1: loop_udelay: remove bogomips value limitation") Cc: <stable@vger.kernel.org> Reviewed-by:
Nicolas Pitre <nico@linaro.org> Signed-off-by:
Russell King <rmk+kernel@armlinux.org.uk>
-
netmonk@netmonk.org authored
Good evening, Following LinuxCodingStyle documentation and with the help of Sam, fixed severals identation issues in the code, and few others cosmetic changes And last and i hope least fixing my name :) Signed-off-by : Dominique Carrel <netmonk@netmonk.org> Acked-by:
Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
-
David S. Miller authored
Chris Hyser says: ==================== sparc64: PCIe performance enhancements Ver 2 is redone for 4.8 where commit 00085f1e changed DMA attributes from struct pointer to unsigned long. This set of patches initiates a series of PCIe performance enhancement patch submittals. Patch 1/2 enables version 2 of the SPARC sun4v IOMMU I/O address translation services need for subsequent enhancements. Patch 2/2 allows drivers to specify DMA_ATTR_WEAK_ORDERING via DMA attributes to the SPARC DMA mapping routines enabling "relaxed ordering" for the buffer being mapped. [Still relevant write-up] PCI-Express Relaxed Ordering and the Sun SPARC Enterprise M-class Servers https://blogs.oracle.com/olympus/entry/relaxed_ordering ==================== Signed-off-by:
-
chris hyser authored
Enable relaxed ordering for memory writes in IOMMU TSB entry from dma_4v_alloc_coherent(), dma_4v_map_page() and dma_4v_map_sg() when dma_attrs DMA_ATTR_WEAK_ORDERING is set. This requires PCI IOMMU I/O Translation Services version 2.0 API. Many PCIe devices allow enabling relaxed-ordering (memory writes bypassing other memory writes) for various DMA buffers. A notable exception is the Mellanox mlx4 IB adapter. Due to the nature of x86 HW this appears to have little performance impact there. On SPARC HW however, this results in major performance degradation getting only about 3Gbps. Enabling RO in the IOMMU entries corresponding to mlx4 data buffers increases the throughput to about 13 Gbps. Orabug: 19245907 Signed-off-by:
Chris Hyser <chris.hyser@oracle.com> Signed-off-by:
-
chris hyser authored
Enable Version 2 of the PCI IOMMU API needed for advanced features such as PCI Relaxed Ordering and greater than 2 GB DMA address space per root complex. Signed-off-by:
-
Paul Gortmaker authored
These files were only including module.h for exception table related functions. We've now separated that content out into its own file "extable.h" so now move over to that and avoid all the extra header content in module.h that we don't really need to compile these files. Cc: "David S. Miller" <davem@davemloft.net> Cc: sparclinux@vger.kernel.org Signed-off-by:
Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by:
-
