| /linux/drivers/infiniband/ulp/rtrs/ |
| H A D | README | 11 possibility to either write data from an sg list to the remote side
12 or to request ("read") data transfer from the remote side into a given
15 RTRS provides I/O fail-over and load-balancing capabilities by using
17 Documentation/ABI/testing/sysfs-class-rtrs-client).
26 --------
29 server side for a given client for rdma transfer. A session
36 chunks reserved for him on the server side. Their number, size and addresses
49 discussed in LPC RDMA MC 2019. When always_invalidate=Y, on the server side we
55 registration on each IO causes performance drop of up to 20%. A user of the
64 ------------------------
[all …]
|
| /linux/Documentation/admin-guide/hw-vuln/ |
| H A D | spectre.rst | 1 .. SPDX-License-Identifier: GPL-2.0
3 Spectre Side Channels
6 Spectre is a class of side channel attacks that exploit branch prediction
8 bypassing access controls. Speculative execution side channel exploits
14 -------------------
16 Speculative execution side channel methods affect a wide range of modern
22 - Intel Core, Atom, Pentium, and Xeon processors
24 - AMD Phenom, EPYC, and Zen processors
26 - IBM POWER and zSeries processors
28 - Higher end ARM processors
[all …]
|
| /linux/Documentation/misc-devices/ |
| H A D | amd-sbi.rst | 1 .. SPDX-License-Identifier: GPL-2.0
4 AMD SIDE BAND interface
8 functionality via side-band interface (SBI) called
10 based 2-wire processor target interface. APML is used to
12 (SB Remote Management Interface (SB-RMI)
13 and SB Temperature Sensor Interface (SB-TSI)).
18 .. [1] https://docs.amd.com/v/u/en-US/55898_B1_pub_0_50
24 apml_sbrmi driver under the drivers/misc/amd-sbi creates miscdevice
25 /dev/sbrmi-* to let user space programs run APML mailbox, CPUID,
31 .. code-block:: bash
[all …]
|
| /linux/Documentation/kbuild/ |
| H A D | kconfig-macro-language.rst | 6 -------
21 $(CC) -o $(APP) $(SRC)
27 gcc -o foo foo.c
32 The idea is quite similar in Kconfig - it is possible to describe a Kconfig
38 def_bool $(shell, $(srctree)/scripts/gcc-check-foo.sh $(CC))
46 Then, Kconfig moves onto the evaluation stage to resolve inter-symbol
47 dependency as explained in kconfig-language.rst.
51 ---------
56 $( ). The parentheses are required even for single-letter variable names; $X is
63 righthand side is expanded immediately upon reading the line from the Kconfig
[all …]
|
| /linux/include/linux/ |
| H A D | drbd_genl_api.h | 1 /* SPDX-License-Identifier: GPL-2.0 */
6 * struct drbd_genlmsghdr - DRBD specific header used in NETLINK_GENERIC requests
8 * For admin requests (user -> kernel): which minor device to operate on.
10 * (kernel -> user): which minor device the information is about.
14 * @flags: possible operation modifiers (relevant only for user->kernel):
19 * The volume number corresponds to the same volume number on the remote side,
20 * whereas the minor number on the remote side may be different
22 * @ret_code: kernel->userland unicast cfg reply return code (union with flags);
|
| /linux/Documentation/RCU/ |
| H A D | rcu.rst | 6 The basic idea behind RCU (read-copy update) is to split destructive
11 since dropped their references. For example, an RCU-protected deletion
17 --------------------------
19 - Why would anyone want to use RCU?
21 The advantage of RCU's two-part approach is that RCU readers need
26 in read-mostly situations. The fact that RCU readers need not
27 acquire locks can also greatly simplify deadlock-avoidance code.
29 - How can the updater tell when a grace period has completed
33 block, switch to user-mode execution, or enter the idle loop.
36 read-side critical sections. So, if we remove an item from a
[all …]
|
| H A D | checklist.rst | 1 .. SPDX-License-Identifier: GPL-2.0
14 0. Is RCU being applied to a read-mostly situation? If the data
18 tool for the job. Yes, RCU does reduce read-side overhead by
19 increasing write-side overhead, which is exactly why normal uses
27 Yet another exception is where the low real-time latency of RCU's
28 read-side primitives is critically important.
33 counter-intuitive situation where rcu_read_lock() and
49 them -- even x86 allows later loads to be reordered to precede
59 2. Do the RCU read-side critical sections make proper use of
63 under your read-side code, which can greatly increase the
[all …]
|
| /linux/Documentation/gpu/rfc/ |
| H A D | gpusvm.rst | 1 .. SPDX-License-Identifier: (GPL-2.0+ OR MIT)
16 * An example of a driver-specific lock causing issues occurred before
17 fixing do_swap_page to lock the faulting page. A driver-exclusive lock
34 * GPU fault side
41 should not be handled on the fault side by trying to hold locks;
43 exception is holding a BO's dma-resv lock during the initial migration
44 to VRAM, as this is a well-defined lock that can be taken underneath
49 While no current user (Xe) of GPU SVM has such a policy, it is likely
51 core-MM side rather than through a driver-side lock.
57 * The physical memory backpointer (page->zone_device_data) should remain
[all …]
|
| /linux/drivers/usb/gadget/legacy/ |
| H A D | Kconfig | 1 # SPDX-License-Identifier: GPL-2.0
7 # NOTE: Gadget support ** DOES NOT ** depend on host-side CONFIG_USB !!
9 # - Host systems (like PCs) need CONFIG_USB (with "A" jacks).
10 # - Peripherals (like PDAs) need CONFIG_USB_GADGET (with "B" jacks).
11 # - Some systems have both kinds of controllers.
13 # With help from a special transceiver and a "Mini-AB" jack, systems with
14 # both kinds of controller can also support "USB On-the-Go" (CONFIG_USB_OTG).
23 # Gadget drivers are hardware-neutral, or "platform independent",
44 Gadget Zero is a two-configuration device. It either sinks and
47 conformance. The driver needs only two bulk-capable endpoints, so
[all …]
|
| /linux/Documentation/PCI/endpoint/ |
| H A D | pci-ntb-howto.rst | 1 .. SPDX-License-Identifier: GPL-2.0
4 PCI Non-Transparent Bridge (NTB) Endpoint Function (EPF) User Guide
9 This document is a guide to help users use pci-epf-ntb function driver
11 be followed in the host side and EP side is given below. For the hardware
13 Documentation/PCI/endpoint/pci-ntb-function.rst
19 ---------------------------
27 2900000.pcie-ep 2910000.pcie-ep
32 2900000.pcie-ep 2910000.pcie-ep
36 -------------------------
40 # ls /sys/bus/pci-epf/drivers
[all …]
|
| /linux/drivers/misc/sgi-xp/ |
| H A D | xp.h | 7 * Copyright (C) 2004-2008 Silicon Graphics, Inc. All rights reserved.
45 * other partition that is currently up. Over these channels, kernel-level
56 * interface-shared info to fit in one 128-byte cacheline.
68 * Define macro, XPC_MSG_SIZE(), is provided for the user
74 #define XPC_MSG_PAYLOAD_MAX_SIZE (XPC_MSG_MAX_SIZE - XPC_MSG_HDR_MAX_SIZE)
82 * Define the return values and values passed to user's callout functions.
113 xpUnloading, /* 19: this side is unloading XPC module */
119 xpUnregistering, /* 22: this side is unregistering channel */
120 xpOtherUnregistering, /* 23: other side is unregistering channel */
147 xpOtherGoingDown, /* 44: other side going down, reason unknown */
[all …]
|
| /linux/Documentation/arch/x86/ |
| H A D | mds.rst | 7 --------
9 Microarchitectural Data Sampling (MDS) is a family of side channel attacks
12 - Microarchitectural Store Buffer Data Sampling (MSBDS) (CVE-2018-12126)
13 - Microarchitectural Fill Buffer Data Sampling (MFBDS) (CVE-2018-12130)
14 - Microarchitectural Load Port Data Sampling (MLPDS) (CVE-2018-12127)
15 - Microarchitectural Data Sampling Uncacheable Memory (MDSUM) (CVE-2019-11091)
18 dependent load (store-to-load forwarding) as an optimization. The forward
21 buffers are partitioned between Hyper-Threads so cross thread forwarding is
32 Hyper-Threads so cross thread leakage is possible.
39 exploited eventually. Load ports are shared between Hyper-Threads so cross
[all …]
|
| /linux/kernel/futex/ |
| H A D | requeue.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
25 * On the waiter side:
26 * Q_REQUEUE_PI_NONE -> Q_REQUEUE_PI_IGNORE
27 * Q_REQUEUE_PI_IN_PROGRESS -> Q_REQUEUE_PI_WAIT
29 * On the requeue side:
30 * Q_REQUEUE_PI_NONE -> Q_REQUEUE_PI_INPROGRESS
31 * Q_REQUEUE_PI_IN_PROGRESS -> Q_REQUEUE_PI_DONE/LOCKED
32 * Q_REQUEUE_PI_IN_PROGRESS -> Q_REQUEUE_PI_NONE (requeue failed)
33 * Q_REQUEUE_PI_WAIT -> Q_REQUEUE_PI_DONE/LOCKED
34 * Q_REQUEUE_PI_WAIT -> Q_REQUEUE_PI_IGNORE (requeue failed)
[all …]
|
| /linux/include/rdma/ |
| H A D | rdma_cm.h | 1 /* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
65 * 0 - No primary nor alternate path is available
66 * 1 - Only primary path is available
67 * 2 - Both primary and alternate path are available
111 * rdma_cm_event_handler - Callback used to report user events.
115 * non-zero value from the callback will destroy the passed in id.
142 * rdma_create_id - Create an RDMA identifier.
145 * @event_handler: User callback invoked to report events associated with the
147 * @context: User specified context associated with the id.
162 * rdma_destroy_id - Destroys an RDMA identifier.
[all …]
|
| /linux/Documentation/networking/ |
| H A D | j1939.rst | 1 .. SPDX-License-Identifier: (GPL-2.0 OR MIT)
14 ISO-11783 (ISOBUS). This last one specifies the so-called ETP (Extended
16 results in a maximum packet size of ((2 ^ 24) - 1) * 7 bytes == 111 MiB.
19 -------------------
21 * SAE J1939-21 : data link layer
22 * SAE J1939-81 : network management
23 * ISO 11783-6 : Virtual Terminal (Extended Transport Protocol)
25 .. _j1939-motivation:
45 requirement for _every_ user space process that communicates via J1939. This
53 for _every_ user space process that communicates via J1939.
[all …]
|
| /linux/drivers/usb/gadget/ |
| H A D | Kconfig | 1 # SPDX-License-Identifier: GPL-2.0
7 # NOTE: Gadget support ** DOES NOT ** depend on host-side CONFIG_USB !!
9 # - Host systems (like PCs) need CONFIG_USB (with "A" jacks).
10 # - Peripherals (like PDAs) need CONFIG_USB_GADGET (with "B" jacks).
11 # - Some systems have both kinds of controllers.
13 # With help from a special transceiver and a "Mini-AB" jack, systems with
14 # both kinds of controller can also support "USB On-the-Go" (CONFIG_USB_OTG).
25 you can't connect a "to-the-host" connector to a peripheral.
31 familiar host side controllers have names like "EHCI", "OHCI",
37 peripheral/device side bus controller, and a "gadget driver" for
[all …]
|
| /linux/Documentation/userspace-api/media/v4l/ |
| H A D | ext-ctrls-camera.rst | 1 .. SPDX-License-Identifier: GFDL-1.1-no-invariants-or-later
4 .. _camera-controls:
14 .. _camera-control-id:
24 .. _v4l2-exposure-auto-type:
29 enum v4l2_exposure_auto_type -
38 .. flat-table::
39 :header-rows: 0
40 :stub-columns: 0
42 * - ``V4L2_EXPOSURE_AUTO``
43 - Automatic exposure time, automatic iris aperture.
[all …]
|
| /linux/include/linux/usb/ |
| H A D | ch9.h | 1 /* SPDX-License-Identifier: GPL-2.0 */
9 * - the host side Linux-USB kernel driver API;
10 * - the "usbfs" user space API; and
11 * - the Linux "gadget" device/peripheral side driver API.
15 * device side APIs benefit from working well together.
22 * [b] so that accessing bigger-than-a-bytes fields will never
27 * someone that the two other points are non-issues for that
|
| /linux/Documentation/core-api/ |
| H A D | cachetlb.rst | 9 describes its intended purpose, and what side effect is expected
12 The side effects described below are stated for a uniprocessor
15 definition such that the side effect for a particular interface occurs
19 if it can be proven that a user address space has never executed
25 virtual-->physical address translations obtained from the software
43 This interface flushes an entire user address space from
56 Here we are flushing a specific range of (user) virtual
59 modifications for the address space 'vma->vm_mm' in the range
60 'start' to 'end-1' will be visible to the cpu. That is, after
62 virtual addresses in the range 'start' to 'end-1'.
[all …]
|
| /linux/Documentation/driver-api/ |
| H A D | xillybus.rst | 10 - Introduction
11 -- Background
12 -- Xillybus Overview
14 - Usage
15 -- User interface
16 -- Synchronization
17 -- Seekable pipes
19 - Internals
20 -- Source code organization
21 -- Pipe attributes
[all …]
|
| /linux/tools/memory-model/Documentation/ |
| H A D | recipes.txt | 20 ------------------------------------
31 2. Compilers are permitted to use the "as-if" rule. That is, a
33 as long as the results of a single-threaded execution appear
41 your full-ordering warranty, as do undersized accesses that load
51 holding the update-side lock, reads from that variable
62 -------
66 locklessly accessing lock-protected shared variables.
68 Locking is well-known and straightforward, at least if you don't think
157 Counter-intuitive though it might be, it is quite possible to have
192 sufficiently to rule out the counter-intuitive outcome.
[all …]
|
| /linux/Documentation/sound/designs/ |
| H A D | channel-mapping-api.rst | 2 ALSA PCM channel-mapping API
10 The channel mapping API allows user to query the possible channel maps
38 the kernel/user-space ABI perspective. It uses only the existing
54 allow user to change the channel map dynamically.
57 ---
61 ``type data-bytes ch0 ch1 ch2...``
71 type is where pair-wise channels are swappable. For example, when you
88 /* this follows the alsa-lib mixer channel value + 3 */
95 SNDRV_CHMAP_SL, /* side left */
96 SNDRV_CHMAP_SR, /* side right */
[all …]
|
| /linux/Documentation/infiniband/ |
| H A D | tag_matching.rst | 5 The MPI standard defines a set of rules, known as tag-matching, for matching
10 * User tag - wild card may be specified by the receiver
15 message envelopes may match, the pair that includes the earliest posted-send
16 and the earliest posted-receive is the pair that must be used to satisfy the
31 1. The Eager protocol- the complete message is sent when the send is
35 2. The Rendezvous Protocol - the sender sends the tag-matching header,
50 When send is initiated and arrives at the receive side, if there is no
51 pre-posted receive for this arriving message, it is passed to the software and
54 specified receive buffer. This allows overlapping receive-side MPI tag
57 When a receive-message is posted, the communication library will first check
[all …]
|
| /linux/Documentation/admin-guide/cgroup-v1/ |
| H A D | cpuacct.rst | 8 The CPU accounting controller supports multi-hierarchy groups. An accounting
14 # mount -t cgroup -ocpuacct none /sys/fs/cgroup
35 CPU time obtained by the cgroup into user and system times. Currently
38 user: Time spent by tasks of the cgroup in user mode.
41 user and system are in USER_HZ unit.
43 cpuacct controller uses percpu_counter interface to collect user and
44 system times. This has two side effects:
46 - It is theoretically possible to see wrong values for user and system times.
49 - It is possible to see slightly outdated values for user and system times
|
| /linux/drivers/comedi/drivers/ni_routing/ |
| H A D | README | 13 1) The register values are _NOT_ in user documentation, but rather in
15 increasingly hard to find and the NI-MHDDK (comments in in example code).
21 name enums such that a user might know which enums should be used for
22 varying purposes, but the end-user had to gain a knowledge of register
26 programming manuals and vendor-provided documentation are _not_ even
27 close to the same names that are in the end-user documentation.
32 NIDAQmx(-base) c-libraries, nor with register level programming, _nor_
34 information is through the proprietary NI-MAX software, which currently only
36 cannot be exported from NI-MAX, except by screenshot.
41 of signal routing capabilities of National Instruments data-acquisition and
[all …]
|