• Immutable Page
  • Info
  • Attachments

Linux 3.18

Linux 3.18 has been released on Sun, 7 Dec 2014

Summary: This release adds support for overlayfs, which allows to combine two filesystem in a single mount point; support for mapping user space memory into the GPU on Radeon devices, a bpf() syscall that allows to upload BPF-like programs that can be attached to events; a TCP congestion algorithm optimized for data centers; the Geneve virtualization encapsulation, support for embedding IP protocols over UDP, improved networking performance thanks to batching the processing of socket buffers, and optional multi-queue SCSI support. There are also new drivers and many other small improvements.

  1. Prominent features
    1. Overlayfs
    2. Radeon: mapping of user pages into video memory
    3. bpf() syscall for eBFP virtual machine programs
    4. TCP: Data Center TCP congestion algorithm
    5. Networking: Geneve Virtualization Encapsulation
    6. Networking performance optimization: transmission queue batching
    7. Foo-over-UDP support
    8. Optional multiqueue SCSI support
  2. Drivers and architectures
  3. File systems
  4. Security
  5. Block
  6. Memory management
  7. Cryptography
  8. Virtualization
  9. Tracing & perf
  10. Networking
  11. Core (various)

1. Prominent features

1.1. Overlayfs

An overlay filesystem combines two filesystems, an 'upper' filesystem and a 'lower' filesystem, into a single file system namespace, modifications will be done to the upper filesystem. It has many uses, but it is most often used for live CDs, where a read-only OS image is used as lower filesystem and a writeable RAM-backed filesystem is used as the upper one. Any modifications will be done in the upper filesystem, thus allowing users to run the OS image provided normally. Overlayfs differs from other "union filesystem" implementations in that after a file is opened all operations go directly to the underlying, lower or upper, filesystems. This simplifies the implementation and allows native performance in these cases.

It is possible for both directory trees to be in the same filesystem and there is no requirement that the root of a filesystem be given for either upper or lower. The lower filesystem can be any filesystem supported by Linux and does not need to be writable. The lower filesystem can even be another overlayfs. The upper filesystem will normally be writable and if it is it must support the creation of trusted.* extended attributes, and must provide valid d_type in readdir() responses, so NFS is not suitable.

Documentation: commit Code: commit

1.2. Radeon: mapping of user pages into video memory

Linux 3.16 added the ability to map users addresses into the video memory for Intel hardware. In this release, AMD Radeon has also gained support for this feature. Normal application data can be used as a texture source or even as a render target (depending upon the capabilities of the chipset). This has a number of uses, with zero-copy downloads to the GPU and efficient readback making the intermixed streaming of CPU and GPU operations fairly efficient. This ability has many widespread implications from faster rendering of client-side software rasterisers (chromium), mitigation of stalls due to read back (Mozilla Firefox) and to faster pipelining of texture data (such as pixel buffer objects in OpenGL or data blobs in OpenCL).

Code: commit

1.3. bpf() syscall for eBFP virtual machine programs

bpf() syscall is a multiplexor for a range of different operations on eBPF which can be characterized as "universal in-kernel virtual machine". eBPF is similar to original Berkeley Packet Filter used to filter network packets. eBPF "extends" classic BPF in multiple ways including ability to call in-kernel helper functions and access shared data structures like eBPF maps. The programs can be written in a restricted C that is compiled into eBPF bytecode and executed on the eBPF virtual machine or JITed into native instruction set.

eBPF programs are similar to kernel modules. They are loaded by the user process and automatically unloaded when process exits. Each eBPF program is a safe run-to-completion set of instructions. eBPF verifier statically determines that the program terminates and is safe to execute. The programs are attached to different events. These events can be packets, tracepoint events and other types in the future. Beyond storing data the programs may call into in-kernel helper functions which may, for example, dump stack, do trace_printk or other forms of live kernel debugging.

Recommended LWN article: The BPF system call API, version 14

ebfp() man page and design documentation can be read on the merge commit: commit

Code: commit 1, 2, 3, 4, 5, 6, 7, 8, 9, 10

1.4. TCP: Data Center TCP congestion algorithm

This release adds the Data Center TCP (DCTCP) congestion control algorithm. DCTCP is an enhancement to the TCP congestion control algorithm for data center networks. DCTCP has been designed for workloads typical of data center environments to provide/achieve: high burst tolerance, low latency and high throughput.

For more details about DCTCP, see the DCTCP web page

Code: commit

1.5. Networking: Geneve Virtualization Encapsulation

Advent of network virtualization has caused a surge of renewed interest and a corresponding increase in the introduction of new protocols, ranging all the way from VLANs and MPLS through the more recent VXLAN, NVGRE, and STT. Existing tunnel protocols have each attempted to solve different aspects of the new requirements, only to be quickly rendered out of date. This release adds Geneve, a protocol which seeks to avoid these problems by providing a framework for tunneling that provide Layer 2 Networks over Layer 3 Networks.

For more information see http://tools.ietf.org/html/draft-gross-geneve-01

Related VMware blog post: Geneve, VXLAN, and Network Virtualization Encapsulations

Code: commit

1.6. Networking performance optimization: transmission queue batching

This release adds support for deferred flushing of transmission SKBs (socket buffers) to the networking driver. Processing the transmission queue is expensive, so some batching shares that cost with other SKBs. This change allows to achieve 10 Gbit/s full TX wirespeed smallest packet size on a single CPU core (14.8 Mpps). Several drivers already have support for this feature: i40e, igb, ixgbe, mlx4, virtio_net, more will follow in next releases.

Recommended LWN article: Bulk network packet transmission

Recommended blog post Unlocked 10Gbps TX wirespeed smallest packet single core

Code: see the blog post link

1.7. Foo-over-UDP support

This release adds the ability to encapsulate any IP protocol over UDP including tunnels (IPIP, GRE, SIT).

The rationale for this funcionality is that network mechanisms, hardware and optimizations for UDP (such as ECMP and RSS) can be leveraged to provide better service. GRE, IPIP, and SIT have been modified with netlink commands to configure use of FOU on transmit. A new "ip fou" has been added in newer releases of ip to make use of this feature. Details on configuration can be found in the merge link.

Recommended LWN link: Foo over UDP

Merge link: merge

1.8. Optional multiqueue SCSI support

Linux 3.13 added a new design for the block layer that allowed to process multiple IO queues in parallel. This feature, however, wasn't transparent, and required modifications in drivers to support it. In this release, support for the multi-queue layer has been added to the SCSI layer (used by ATA and SATA drivers) as an optional configuration option.

Code: commit

2. Drivers and architectures

All the driver and architecture-specific changes can be found in the Linux_3.18-DriversArch page

3. File systems

  • F2FS

    • Introduce FITRIM discard support commit

    • Support atomic writes commit

    • Support large sector size commit

    • Support volatile operations for transient data commit

  • ext4

    • Support freezing ext2 (nojournal) file systems commit

    • Add sysfs entry showing whether the fs contains errors commit

  • NFS

    • Implement NFS v4.2 SEEK operation, allowing to use lseek() call with either the SEEK_HOLE or SEEK_DATA flags set commit, commit

    • Allow turning off nfsv3 readdir_plus commit

    • nfsd: add a v4_end_grace file to /proc/fs/nfsd that allows a privileged userland process to end the v4 grace period early commit

    • lockd: add a /proc/fs/lockd/nlm_end_grace file that will allow a (privileged) userland process to end the NLM grace period early commit

  • SMB

    • Add mfsymlinks support for SMB2.1/SMB3 commit, commit

    • Allow mknod and mkfifo on SMB2/SMB3 mounts commit

    • Allow conversion of characters in Mac remap range commit, commit

    • Remap reserved POSIX characters by default commit

  • Btrfs

    • Make defragment work with nodatacow option commit

    • Improve free space cache management and space allocation commit

    • Implement repair function when direct read fails commit

  • XFS

  • NTFS

    • Add FIBMAP ioctl support commit

4. Security

  • Smack: Add a new access (aka permissive) mode commit

  • ima: provide 'ima_appraise=log' kernel option that allows logging only mode without fixing it, in order to properly analyze the system commit

5. Block

  • Implement readpages() to optimize sequential reads of /dev block devices commit

  • SCSI target userspace backend that uses UIO + shared memory ring to dispatch I/O and control commands into user-space commit

  • dm raid: add discard support for RAID levels 1 and 10 commit and 4, 5 and 6 commit

  • zram memory size limitation commit

  • zram: report maximum used memory commit

6. Memory management

  • memory-hotplug: add /sys/devices/system/memory/memoryX/valid_zones, shows which zone this memory block can be onlined to commit

  • balloon_compaction: adds three counters into /proc/vmstat: "balloon_inflate", "balloon_deflate" and "balloon_migrate". They accumulate balloon activity. Under option CONFIG_MEMORY_BALLOON commit

  • slab merge: If new creating slab has similar size and property with existent slab, this feature reuse it rather than creating new one commit

  • Support compiling out madvise() and fadvise() commit

  • Support fadvise without CONFIG_MMU commit

  • Introduce dump_vma commit

  • NUMA: Memory zones are allocated by the page allocator in either node or zone order. This release defaults node-ordering on 64-bit NUMA, and to zone-ordering on 32-bit commit

  • Stress test for memory compaction commit

7. Cryptography

  • Software asynchronous crypto daemon that provides the kernel thread to assist multi-buffer crypto algorithms for submitting jobs and flushing jobs in multi-buffer crypto algorithms commit, commit, commit, commit, commit

  • KEYS: Implement binary asymmetric key ID handling commit

8. Virtualization

  • Xen

    • Xen PV SCSI backend driver commit

    • Add Xen PV SCSI frontend driver commit

    • Remap Xen Identity Mapped RAM commit

9. Tracing & perf

  • perf tools

    • Add +field argument support for --field option commit

    • Add +field argument support for --sort option commit

    • Add perf-with-kcore script commit

    • Add report.queue-size config file option commit

    • Add support to new style format of kernel PMU event commit

    • Let a user specify a PMU event without any config terms commit

    • Let default config be defined for a PMU commit

  • top

    • Add -w option for setting column width commit

    • Add a visual cue for toggle zeroing of samples commit

    • Handle 'z' key for toggle zeroing samples in TUI commit

  • perf bench futex: Support operations for shared futexes commit

  • perf kvm stat report: Enable the target.system_wide flag commit

10. Networking

  • ipv4: implement igmp_qrv sysctl to tune IGMP robustness variable commit

  • ipv6: add sysctl_mld_qrv to configure how many retransmit of unsolicited Multicast Listener Discovery retransmit should happen. Admins might want to tune this on lossy links commit

  • ipv6: notify userspace when we added or changed an ipv6 token commit

  • ipvs: Allow heterogeneous pools commit, commit

  • TCP: Restore RFC5961-compliant behavior for SYN packets commit

  • TCP: TCP Small Queues and strange attractors. The purpose is to try to keep number of packets in qdisc as small as possible commit

  • TCP: improve undo on timeout (increases the undo events by 50% on Google servers) commit

  • TIPC: add name distributor resiliency queue, helps to avoid race conditions in TIPC distributed name table updates commit

  • UDP: GRO for UDPv6 commit

  • ip_tunnel: Add GUE support commit

  • ethtool: adds new ethtool cmd, ETHTOOL_GTUNABLE & ETHTOOL_STUNABLE for getting and setting tunable values commit

  • pktgen: add flag NO_TIMESTAMP to disable timestamping commit

  • xfrm: Enable to specify local and remote prefix length thresholds for the policy hash table via a netlink XFRM_MSG_NEWSPDINFO message commit

  • openvswitch: Add recirc and hash action. commit

  • openvswitch: Add support for Geneve tunneling. commit

  • Bluetooth: 6lowpan: Enable multicast support commit

  • bridge: Add a default_pvid sysfs attribute commit

  • bridge: Add filtering support for default_pvid commit

  • Wireless

    • Support DTPC IE (from Cisco Client eXtensions) commit

    • Add Radio Resource Management support to assoc request commit

    • Add WMM traffic stream API commit

    • Re-enable CSA for drivers that support it commit

  • nftables

    • IPv4 masquerading support commit

    • IPv6 masquerading support commit

    • allow to filter from pre-routing and post-routing commit

    • extend NFT_MSG_DELTABLE to support flushing the ruleset commit

    • nfnetlink_acct: add filter support to nfacct counter list/reset commit

    • export rule-set generation ID commit

    • nft_meta: Add CPU attribute support, allows to match packets with CPU number commit

    • nft_meta: add pkttype support for ip, ipv6 and inet families of tables. Allows to match packets based on the link layer information commit

    • nft_meta: add devgroup support in meta expression, lets match device group of a packets incoming or outgoing interface commit

  • netfilter

    • ipset: hash:mac type added to ipset, by which one can store MAC (Ethernet address) elements in a set. commit

    • conntrack: disable generic tracking for known protocols commit

11. Core (various)

  • Add support for gcc 5 commit

  • locktorture: Add infrastructure for torturing read locks commit, mutexes commit, rwlocks commit, rwsems commit

  • kbuild: handle module compression while running 'make modules_install'. commit

  • Increase the buffer-head per-CPU LRU size commit

  • sched: Add default-disabled option to BUG() when stack end location is overwritten commit

  • sched: Improve sysbench performance by fixing spurious active migration commit

  • prctl: introduce PR_SET_MM_MAP operation to allow checkpoint/restore utilities to restore a process memory properties commit

Tell others about this page:

last edited 2015-02-01 14:29:09 by Morot