From aece34cd576c7625181b0488a8129c1e165355f7 Mon Sep 17 00:00:00 2001 From: Paul Burton Date: Fri, 19 Jan 2018 16:40:48 +0100 Subject: dt-bindings: Document mti,mips-cpc binding Document a binding for the MIPS Cluster Power Controller (CPC) that allows the device tree to specify where the CPC registers are located. Signed-off-by: Paul Burton Signed-off-by: Aleksandar Markovic Reviewed-by: Rob Herring Cc: linux-mips@linux-mips.org Cc: devicetree@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/18512/ Signed-off-by: James Hogan --- Documentation/devicetree/bindings/power/mti,mips-cpc.txt | 8 ++++++++ 1 file changed, 8 insertions(+) create mode 100644 Documentation/devicetree/bindings/power/mti,mips-cpc.txt (limited to 'Documentation') diff --git a/Documentation/devicetree/bindings/power/mti,mips-cpc.txt b/Documentation/devicetree/bindings/power/mti,mips-cpc.txt new file mode 100644 index 000000000000..c6b82511ae8a --- /dev/null +++ b/Documentation/devicetree/bindings/power/mti,mips-cpc.txt @@ -0,0 +1,8 @@ +Binding for MIPS Cluster Power Controller (CPC). + +This binding allows a system to specify where the CPC registers are +located. + +Required properties: +compatible : Should be "mti,mips-cpc". +regs: Should describe the address & size of the CPC register region. -- cgit v1.2.3 From 6a546c7e69ff0b69581377cc70d7e8a601b98fce Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Fri, 9 Feb 2018 15:30:56 -0500 Subject: membarrier-sync-core: Document architecture support Ensure we gather architecture requirements about each architecture supporting the "sync_core" membarrier command in a single file under Documentation/features. Signed-off-by: Mathieu Desnoyers Cc: Andrea Parri Cc: Andrew Hunter Cc: Andy Lutomirski Cc: Avi Kivity Cc: Benjamin Herrenschmidt Cc: Boqun Feng Cc: Dave Watson Cc: David Sehr Cc: Greg Hackmann Cc: Linus Torvalds Cc: Maged Michael Cc: Michael Ellerman Cc: Paul E. McKenney Cc: Paul Mackerras Cc: Peter Zijlstra (Intel) Cc: Peter Zijlstra Cc: Russell King Cc: Thomas Gleixner Cc: Will Deacon Cc: linux-api@vger.kernel.org Cc: linux-arch@vger.kernel.org Link: http://lkml.kernel.org/r/1518208256-22034-1-git-send-email-mathieu.desnoyers@efficios.com Signed-off-by: Ingo Molnar --- .../sched/membarrier-sync-core/arch-support.txt | 62 ++++++++++++++++++++++ 1 file changed, 62 insertions(+) create mode 100644 Documentation/features/sched/membarrier-sync-core/arch-support.txt (limited to 'Documentation') diff --git a/Documentation/features/sched/membarrier-sync-core/arch-support.txt b/Documentation/features/sched/membarrier-sync-core/arch-support.txt new file mode 100644 index 000000000000..2c815a7f1ba7 --- /dev/null +++ b/Documentation/features/sched/membarrier-sync-core/arch-support.txt @@ -0,0 +1,62 @@ +# +# Feature name: membarrier-sync-core +# Kconfig: ARCH_HAS_MEMBARRIER_SYNC_CORE +# description: arch supports core serializing membarrier +# +# Architecture requirements +# +# * arm64 +# +# Rely on eret context synchronization when returning from IPI handler, and +# when returning to user-space. +# +# * x86 +# +# x86-32 uses IRET as return from interrupt, which takes care of the IPI. +# However, it uses both IRET and SYSEXIT to go back to user-space. The IRET +# instruction is core serializing, but not SYSEXIT. +# +# x86-64 uses IRET as return from interrupt, which takes care of the IPI. +# However, it can return to user-space through either SYSRETL (compat code), +# SYSRETQ, or IRET. +# +# Given that neither SYSRET{L,Q}, nor SYSEXIT, are core serializing, we rely +# instead on write_cr3() performed by switch_mm() to provide core serialization +# after changing the current mm, and deal with the special case of kthread -> +# uthread (temporarily keeping current mm into active_mm) by issuing a +# sync_core_before_usermode() in that specific case. +# + ----------------------- + | arch |status| + ----------------------- + | alpha: | TODO | + | arc: | TODO | + | arm: | TODO | + | arm64: | ok | + | blackfin: | TODO | + | c6x: | TODO | + | cris: | TODO | + | frv: | TODO | + | h8300: | TODO | + | hexagon: | TODO | + | ia64: | TODO | + | m32r: | TODO | + | m68k: | TODO | + | metag: | TODO | + | microblaze: | TODO | + | mips: | TODO | + | mn10300: | TODO | + | nios2: | TODO | + | openrisc: | TODO | + | parisc: | TODO | + | powerpc: | TODO | + | s390: | TODO | + | score: | TODO | + | sh: | TODO | + | sparc: | TODO | + | tile: | TODO | + | um: | TODO | + | unicore32: | TODO | + | x86: | ok | + | xtensa: | TODO | + ----------------------- -- cgit v1.2.3 From 79e902382637a2f421b7f295dcf9934d80d84d7d Mon Sep 17 00:00:00 2001 From: Juri Lelli Date: Fri, 9 Feb 2018 17:01:14 +0100 Subject: Documentation/locking/mutex-design: Update to reflect latest changes Commit 3ca0ff571b09 ("locking/mutex: Rework mutex::owner") reworked the basic mutex implementation to deal with several problems. Documentation was however left unchanged and became stale. Update mutex-design.txt to reflect changes introduced by the above commit. Signed-off-by: Juri Lelli Cc: Andrew Morton Cc: Davidlohr Bueso Cc: Jonathan Corbet Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-doc@vger.kernel.org Link: http://lkml.kernel.org/r/20180209160114.19980-1-juri.lelli@redhat.com [ Small readability tweaks to the text. ] Signed-off-by: Ingo Molnar --- Documentation/locking/mutex-design.txt | 49 ++++++++++++---------------------- 1 file changed, 17 insertions(+), 32 deletions(-) (limited to 'Documentation') diff --git a/Documentation/locking/mutex-design.txt b/Documentation/locking/mutex-design.txt index 60c482df1a38..818aca19612f 100644 --- a/Documentation/locking/mutex-design.txt +++ b/Documentation/locking/mutex-design.txt @@ -21,37 +21,23 @@ Implementation -------------- Mutexes are represented by 'struct mutex', defined in include/linux/mutex.h -and implemented in kernel/locking/mutex.c. These locks use a three -state atomic counter (->count) to represent the different possible -transitions that can occur during the lifetime of a lock: - - 1: unlocked - 0: locked, no waiters - negative: locked, with potential waiters - -In its most basic form it also includes a wait-queue and a spinlock -that serializes access to it. CONFIG_SMP systems can also include -a pointer to the lock task owner (->owner) as well as a spinner MCS -lock (->osq), both described below in (ii). +and implemented in kernel/locking/mutex.c. These locks use an atomic variable +(->owner) to keep track of the lock state during its lifetime. Field owner +actually contains 'struct task_struct *' to the current lock owner and it is +therefore NULL if not currently owned. Since task_struct pointers are aligned +at at least L1_CACHE_BYTES, low bits (3) are used to store extra state (e.g., +if waiter list is non-empty). In its most basic form it also includes a +wait-queue and a spinlock that serializes access to it. Furthermore, +CONFIG_MUTEX_SPIN_ON_OWNER=y systems use a spinner MCS lock (->osq), described +below in (ii). When acquiring a mutex, there are three possible paths that can be taken, depending on the state of the lock: -(i) fastpath: tries to atomically acquire the lock by decrementing the - counter. If it was already taken by another task it goes to the next - possible path. This logic is architecture specific. On x86-64, the - locking fastpath is 2 instructions: - - 0000000000000e10 : - e21: f0 ff 0b lock decl (%rbx) - e24: 79 08 jns e2e - - the unlocking fastpath is equally tight: - - 0000000000000bc0 : - bc8: f0 ff 07 lock incl (%rdi) - bcb: 7f 0a jg bd7 - +(i) fastpath: tries to atomically acquire the lock by cmpxchg()ing the owner with + the current task. This only works in the uncontended case (cmpxchg() checks + against 0UL, so all 3 state bits above have to be 0). If the lock is + contended it goes to the next possible path. (ii) midpath: aka optimistic spinning, tries to spin for acquisition while the lock owner is running and there are no other tasks ready @@ -143,11 +129,10 @@ Test if the mutex is taken: Disadvantages ------------- -Unlike its original design and purpose, 'struct mutex' is larger than -most locks in the kernel. E.g: on x86-64 it is 40 bytes, almost twice -as large as 'struct semaphore' (24 bytes) and tied, along with rwsems, -for the largest lock in the kernel. Larger structure sizes mean more -CPU cache and memory footprint. +Unlike its original design and purpose, 'struct mutex' is among the largest +locks in the kernel. E.g: on x86-64 it is 32 bytes, where 'struct semaphore' +is 24 bytes and rw_semaphore is 40 bytes. Larger structure sizes mean more CPU +cache and memory footprint. When to use mutexes ------------------- -- cgit v1.2.3 From b6d8ef86cb7b8b6920b6815ebf1352757d3adb87 Mon Sep 17 00:00:00 2001 From: Aishwarya Pant Date: Wed, 7 Feb 2018 19:04:36 +0530 Subject: Documentation/ABI: update cpuidle sysfs documentation Update cpuidle documentation using git logs and existing documentation in Documentation/cpuidle/sysfs.txt. This might be useful for scripting and tracking changes in the ABI. Signed-off-by: Aishwarya Pant Signed-off-by: Rafael J. Wysocki --- Documentation/ABI/testing/sysfs-devices-system-cpu | 77 +++++++++++++++++++++- 1 file changed, 75 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index bfd29bc8d37a..4ed63b6cfb15 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -108,6 +108,8 @@ Description: CPU topology files that describe a logical CPU's relationship What: /sys/devices/system/cpu/cpuidle/current_driver /sys/devices/system/cpu/cpuidle/current_governer_ro + /sys/devices/system/cpu/cpuidle/available_governors + /sys/devices/system/cpu/cpuidle/current_governor Date: September 2007 Contact: Linux kernel mailing list Description: Discover cpuidle policy and mechanism @@ -119,13 +121,84 @@ Description: Discover cpuidle policy and mechanism Idle policy (governor) is differentiated from idle mechanism (driver) - current_driver: displays current idle mechanism + current_driver: (RO) displays current idle mechanism - current_governor_ro: displays current idle policy + current_governor_ro: (RO) displays current idle policy + + With the cpuidle_sysfs_switch boot option enabled (meant for + developer testing), the following three attributes are visible + instead: + + current_driver: same as described above + + available_governors: (RO) displays a space separated list of + available governors + + current_governor: (RW) displays current idle policy. Users can + switch the governor at runtime by writing to this file. See files in Documentation/cpuidle/ for more information. +What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/name + /sys/devices/system/cpu/cpuX/cpuidle/stateN/latency + /sys/devices/system/cpu/cpuX/cpuidle/stateN/power + /sys/devices/system/cpu/cpuX/cpuidle/stateN/time + /sys/devices/system/cpu/cpuX/cpuidle/stateN/usage +Date: September 2007 +KernelVersion: v2.6.24 +Contact: Linux power management list +Description: + The directory /sys/devices/system/cpu/cpuX/cpuidle contains per + logical CPU specific cpuidle information for each online cpu X. + The processor idle states which are available for use have the + following attributes: + + name: (RO) Name of the idle state (string). + + latency: (RO) The latency to exit out of this idle state (in + microseconds). + + power: (RO) The power consumed while in this idle state (in + milliwatts). + + time: (RO) The total time spent in this idle state (in microseconds). + + usage: (RO) Number of times this state was entered (a count). + + +What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/desc +Date: February 2008 +KernelVersion: v2.6.25 +Contact: Linux power management list +Description: + (RO) A small description about the idle state (string). + + +What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/disable +Date: March 2012 +KernelVersion: v3.10 +Contact: Linux power management list +Description: + (RW) Option to disable this idle state (bool). The behavior and + the effect of the disable variable depends on the implementation + of a particular governor. In the ladder governor, for example, + it is not coherent, i.e. if one is disabling a light state, then + all deeper states are disabled as well, but the disable variable + does not reflect it. Likewise, if one enables a deep state but a + lighter state still is disabled, then this has no effect. + + +What: /sys/devices/system/cpu/cpuX/cpuidle/stateN/residency +Date: March 2014 +KernelVersion: v3.15 +Contact: Linux power management list +Description: + (RO) Display the target residency i.e. the minimum amount of + time (in microseconds) this cpu should spend in this idle state + to make the transition worth the effort. + + What: /sys/devices/system/cpu/cpu#/cpufreq/* Date: pre-git history Contact: linux-pm@vger.kernel.org -- cgit v1.2.3 From 6b4af818c7d7a35a861c94596e05e43596e5fd28 Mon Sep 17 00:00:00 2001 From: Aishwarya Pant Date: Sat, 10 Feb 2018 14:27:19 +0530 Subject: ACPI / DPTF: Document dptf_power sysfs atttributes The descriptions have been collected from git commit logs and reading through code. Signed-off-by: Aishwarya Pant Signed-off-by: Rafael J. Wysocki --- Documentation/ABI/testing/sysfs-platform-dptf | 40 +++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-platform-dptf (limited to 'Documentation') diff --git a/Documentation/ABI/testing/sysfs-platform-dptf b/Documentation/ABI/testing/sysfs-platform-dptf new file mode 100644 index 000000000000..325dc0667dbb --- /dev/null +++ b/Documentation/ABI/testing/sysfs-platform-dptf @@ -0,0 +1,40 @@ +What: /sys/bus/platform/devices/INT3407:00/dptf_power/charger_type +Date: Jul, 2016 +KernelVersion: v4.10 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) The charger type - Traditional, Hybrid or NVDC. + +What: /sys/bus/platform/devices/INT3407:00/dptf_power/adapter_rating_mw +Date: Jul, 2016 +KernelVersion: v4.10 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Adapter rating in milliwatts (the maximum Adapter power). + Must be 0 if no AC Adaptor is plugged in. + +What: /sys/bus/platform/devices/INT3407:00/dptf_power/max_platform_power_mw +Date: Jul, 2016 +KernelVersion: v4.10 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Maximum platform power that can be supported by the battery + in milliwatts. + +What: /sys/bus/platform/devices/INT3407:00/dptf_power/platform_power_source +Date: Jul, 2016 +KernelVersion: v4.10 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Display the platform power source + 0x00 = DC + 0x01 = AC + 0x02 = USB + 0x03 = Wireless Charger + +What: /sys/bus/platform/devices/INT3407:00/dptf_power/battery_steady_power +Date: Jul, 2016 +KernelVersion: v4.10 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) The maximum sustained power for battery in milliwatts. -- cgit v1.2.3 From 22029845ad81033115910cdef35170de6a10a1eb Mon Sep 17 00:00:00 2001 From: Aishwarya Pant Date: Sat, 10 Feb 2018 14:27:38 +0530 Subject: ACPI: dock: document sysfs interface Description has been collected from git commit history and reading through code. Signed-off-by: Aishwarya Pant Signed-off-by: Rafael J. Wysocki --- .../ABI/testing/sysfs-devices-platform-dock | 39 ++++++++++++++++++++++ 1 file changed, 39 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-devices-platform-dock (limited to 'Documentation') diff --git a/Documentation/ABI/testing/sysfs-devices-platform-dock b/Documentation/ABI/testing/sysfs-devices-platform-dock new file mode 100644 index 000000000000..1d8c18f905c7 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-devices-platform-dock @@ -0,0 +1,39 @@ +What: /sys/devices/platform/dock.N/docked +Date: Dec, 2006 +KernelVersion: 2.6.19 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Value 1 or 0 indicates whether the software believes the + laptop is docked in a docking station. + +What: /sys/devices/platform/dock.N/undock +Date: Dec, 2006 +KernelVersion: 2.6.19 +Contact: linux-acpi@vger.kernel.org +Description: + (WO) Writing to this file causes the software to initiate an + undock request to the firmware. + +What: /sys/devices/platform/dock.N/uid +Date: Feb, 2007 +KernelVersion: v2.6.21 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Displays the docking station the laptop is docked to. + +What: /sys/devices/platform/dock.N/flags +Date: May, 2007 +KernelVersion: v2.6.21 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Show dock station flags, useful for checking if undock + request has been made by the user (from the immediate_undock + option). + +What: /sys/devices/platform/dock.N/type +Date: Aug, 2008 +KernelVersion: v2.6.27 +Contact: linux-acpi@vger.kernel.org +Description: + (RO) Display the dock station type- dock_station, ata_bay or + battery_bay. -- cgit v1.2.3 From 61e02392d3c7ecac1f91c0a90a8043d67e081846 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 13 Feb 2018 13:30:19 +0000 Subject: locking/atomic/bitops: Document and clarify ordering semantics for failed test_and_{}_bit() A test_and_{}_bit() operation fails if the value of the bit is such that the modification does not take place. For example, if test_and_set_bit() returns 1. In these cases, follow the behaviour of cmpxchg and allow the operation to be unordered. This also applies to test_and_set_bit_lock() if the lock is found to be be taken already. Signed-off-by: Will Deacon Acked-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Paul E. McKenney Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1518528619-20049-1-git-send-email-will.deacon@arm.com Signed-off-by: Ingo Molnar --- Documentation/atomic_bitops.txt | 7 ++++++- include/asm-generic/bitops/lock.h | 3 ++- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'Documentation') diff --git a/Documentation/atomic_bitops.txt b/Documentation/atomic_bitops.txt index 5550bfdcce5f..be70b32c95d9 100644 --- a/Documentation/atomic_bitops.txt +++ b/Documentation/atomic_bitops.txt @@ -58,7 +58,12 @@ Like with atomic_t, the rule of thumb is: - RMW operations that have a return value are fully ordered. -Except for test_and_set_bit_lock() which has ACQUIRE semantics and + - RMW operations that are conditional are unordered on FAILURE, + otherwise the above rules apply. In the case of test_and_{}_bit() operations, + if the bit in memory is unchanged by the operation then it is deemed to have + failed. + +Except for a successful test_and_set_bit_lock() which has ACQUIRE semantics and clear_bit_unlock() which has RELEASE semantics. Since a platform only has a single means of achieving atomic operations diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h index bc397573c43a..67ab280ad134 100644 --- a/include/asm-generic/bitops/lock.h +++ b/include/asm-generic/bitops/lock.h @@ -7,7 +7,8 @@ * @nr: Bit to set * @addr: Address to count from * - * This operation is atomic and provides acquire barrier semantics. + * This operation is atomic and provides acquire barrier semantics if + * the returned value is 0. * It can be used to implement bit locks. */ #define test_and_set_bit_lock(nr, addr) test_and_set_bit(nr, addr) -- cgit v1.2.3 From a65820e6956782af6c5330749ae37222350d8d3f Mon Sep 17 00:00:00 2001 From: Daniel Axtens Date: Wed, 14 Feb 2018 18:05:31 +1100 Subject: docs: segmentation-offloads.txt: update for UFO depreciation UFO is deprecated except for tuntap and packet per 0c19f846d582, ("net: accept UFO datagrams from tuntap and packet"). Update UFO docs to reflect this. Signed-off-by: Daniel Axtens Signed-off-by: David S. Miller --- Documentation/networking/segmentation-offloads.txt | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'Documentation') diff --git a/Documentation/networking/segmentation-offloads.txt b/Documentation/networking/segmentation-offloads.txt index 2f09455a993a..2cda12ab7075 100644 --- a/Documentation/networking/segmentation-offloads.txt +++ b/Documentation/networking/segmentation-offloads.txt @@ -49,6 +49,10 @@ datagram into multiple IPv4 fragments. Many of the requirements for UDP fragmentation offload are the same as TSO. However the IPv4 ID for fragments should not increment as a single IPv4 datagram is fragmented. +UFO is deprecated: modern kernels will no longer generate UFO skbs, but can +still receive them from tuntap and similar devices. Offload of UDP-based +tunnel protocols is still supported. + IPIP, SIT, GRE, UDP Tunnel, and Remote Checksum Offloads ======================================================== -- cgit v1.2.3 From bc3c2431d4173816240679a02fd4d74685e94bc8 Mon Sep 17 00:00:00 2001 From: Daniel Axtens Date: Wed, 14 Feb 2018 18:05:32 +1100 Subject: docs: segmentation-offloads.txt: Fix ref to SKB_GSO_TUNNEL_REMCSUM The doc originally called it SKB_GSO_REMCSUM. Fix it. Fixes: f7a6272bf3cb ("Documentation: Add documentation for TSO and GSO features") Signed-off-by: Daniel Axtens Signed-off-by: David S. Miller --- Documentation/networking/segmentation-offloads.txt | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'Documentation') diff --git a/Documentation/networking/segmentation-offloads.txt b/Documentation/networking/segmentation-offloads.txt index 2cda12ab7075..b247471a183c 100644 --- a/Documentation/networking/segmentation-offloads.txt +++ b/Documentation/networking/segmentation-offloads.txt @@ -87,10 +87,10 @@ SKB_GSO_UDP_TUNNEL_CSUM. These two additional tunnel types reflect the fact that the outer header also requests to have a non-zero checksum included in the outer header. -Finally there is SKB_GSO_REMCSUM which indicates that a given tunnel header -has requested a remote checksum offload. In this case the inner headers -will be left with a partial checksum and only the outer header checksum -will be computed. +Finally there is SKB_GSO_TUNNEL_REMCSUM which indicates that a given tunnel +header has requested a remote checksum offload. In this case the inner +headers will be left with a partial checksum and only the outer header +checksum will be computed. Generic Segmentation Offload ============================ -- cgit v1.2.3 From a677088922831d94d292ca3891b148a8ba0b5fa1 Mon Sep 17 00:00:00 2001 From: Daniel Axtens Date: Wed, 14 Feb 2018 18:05:33 +1100 Subject: docs: segmentation-offloads.txt: add SCTP info Most of this is extracted from 90017accff61 ("sctp: Add GSO support"), with some extra text about GSO_BY_FRAGS and the need to check for it. Cc: Marcelo Ricardo Leitner Signed-off-by: Daniel Axtens Acked-by: Marcelo Ricardo Leitner Signed-off-by: David S. Miller --- Documentation/networking/segmentation-offloads.txt | 26 ++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'Documentation') diff --git a/Documentation/networking/segmentation-offloads.txt b/Documentation/networking/segmentation-offloads.txt index b247471a183c..d47480b61ac6 100644 --- a/Documentation/networking/segmentation-offloads.txt +++ b/Documentation/networking/segmentation-offloads.txt @@ -13,6 +13,7 @@ The following technologies are described: * Generic Segmentation Offload - GSO * Generic Receive Offload - GRO * Partial Generic Segmentation Offload - GSO_PARTIAL + * SCTP accelleration with GSO - GSO_BY_FRAGS TCP Segmentation Offload ======================== @@ -132,3 +133,28 @@ values for if the header was simply duplicated. The one exception to this is the outer IPv4 ID field. It is up to the device drivers to guarantee that the IPv4 ID field is incremented in the case that a given header does not have the DF bit set. + +SCTP accelleration with GSO +=========================== + +SCTP - despite the lack of hardware support - can still take advantage of +GSO to pass one large packet through the network stack, rather than +multiple small packets. + +This requires a different approach to other offloads, as SCTP packets +cannot be just segmented to (P)MTU. Rather, the chunks must be contained in +IP segments, padding respected. So unlike regular GSO, SCTP can't just +generate a big skb, set gso_size to the fragmentation point and deliver it +to IP layer. + +Instead, the SCTP protocol layer builds an skb with the segments correctly +padded and stored as chained skbs, and skb_segment() splits based on those. +To signal this, gso_size is set to the special value GSO_BY_FRAGS. + +Therefore, any code in the core networking stack must be aware of the +possibility that gso_size will be GSO_BY_FRAGS and handle that case +appropriately. (For size checks, the skb_gso_validate_*_len family of +helpers do this automatically.) + +This also affects drivers with the NETIF_F_FRAGLIST & NETIF_F_GSO_SCTP bits +set. Note also that NETIF_F_GSO_SCTP is included in NETIF_F_GSO_SOFTWARE. -- cgit v1.2.3