Home
last modified time | relevance | path

Searched +full:compute +full:- (Results 1 – 25 of 1123) sorted by relevance

12345678910>>...45

/linux/scripts/coccinelle/misc/
H A Darray_size_dup.cocci1 // SPDX-License-Identifier: GPL-2.0-only
5 /// 1. An opencoded expression is used before array_size() to compute the same size
6 /// 2. An opencoded expression is used after array_size() to compute the same size
13 // Options: --no-includes --include-headers --no-loops
43 msg = "WARNING: array_size is used later (line %s) to compute the same size" % (p2[0].line)
51 msg = "WARNING: array_size is used later (line %s) to compute the same size" % (p2[0].line)
72 msg = "WARNING: array_size is already used (line %s) to compute the same size" % (p1[0].line)
80 msg = "WARNING: array_size is already used (line %s) to compute the same size" % (p1[0].line)
108 msg = "WARNING: array3_size is used later (line %s) to compute the same size" % (p2[0].line)
116 msg = "WARNING: array3_size is used later (line %s) to compute the same size" % (p2[0].line)
[all …]
/linux/Documentation/arch/x86/
H A Dtopology.rst1 .. SPDX-License-Identifier: GPL-2.0
11 The architecture-agnostic topology definitions are in
12 Documentation/admin-guide/cputopology.rst. This file holds x86-specific
17 Needless to say, code should use the generic functions - this file is *only*
35 - packages
36 - cores
37 - threads
48 Package-related topology information in the kernel:
50 - topology_num_threads_per_package()
54 - topology_num_cores_per_package()
[all …]
/linux/tools/perf/pmu-events/arch/s390/cf_z16/
H A Dpai_crypto.json3 "Unit": "PAI-CRYPTO",
10 "Unit": "PAI-CRYPTO",
14 "PublicDescription": "KM-DEA function ending with CC=0"
17 "Unit": "PAI-CRYPTO",
21 "PublicDescription": "KM-TDEA-128 function ending with CC=0"
24 "Unit": "PAI-CRYPTO",
28 "PublicDescription": "KM-TDEA-192 function ending with CC=0"
31 "Unit": "PAI-CRYPTO",
35 "PublicDescription": "KM-Encrypted-DEA function ending with CC=0"
38 "Unit": "PAI-CRYPTO",
[all …]
/linux/drivers/gpu/drm/xe/
H A Dxe_gt_ccs_mode.c1 // SPDX-License-Identifier: MIT
32 * For example, if there are four compute slices available, the in __xe_gt_apply_ccs_mode()
33 * assignment of compute slices to compute engines would be, in __xe_gt_apply_ccs_mode()
48 for (width = num_slices / num_engines; width; width--) { in __xe_gt_apply_ccs_mode()
53 if (hwe->class != XE_ENGINE_CLASS_COMPUTE) in __xe_gt_apply_ccs_mode()
56 if (hwe->logical_instance >= num_engines) in __xe_gt_apply_ccs_mode()
59 config |= BIT(hwe->instance) << XE_HW_ENGINE_CCS0; in __xe_gt_apply_ccs_mode()
66 mode |= CCS_MODE_CSLICE(cslice, hwe->instance); in __xe_gt_apply_ccs_mode()
77 xe_mmio_write32(&gt->mmio, CCS_MODE, mode); in __xe_gt_apply_ccs_mode()
85 if (!gt->ccs_mode || IS_SRIOV_VF(gt_to_xe(gt))) in xe_gt_apply_ccs_mode()
[all …]
/linux/Documentation/gpu/
H A Ddrm-compute.rst2 Long running workloads and compute
5 Long running workloads (compute) are workloads that will not complete in 10
10 Some hardware may schedule compute jobs, and have no way to pre-empt them, or
14 This means that it differs from what is described in driver-api/dma-buf.rst.
16 As with normal compute jobs, dma-fence may not be used at all. In this case,
18 from the long compute job's address space on unbind immediately, not even
26 The first approach you will likely try is to pin all buffers used by compute.
34 older compute jobs to start a new one.
41 driver-allocated CPU memory would be accounted to the correct cgroup, and
/linux/tools/perf/pmu-events/arch/s390/cf_z17/
H A Dpai_crypto.json3 "Unit": "PAI-CRYPTO",
10 "Unit": "PAI-CRYPTO",
14 "PublicDescription": "KM-DEA function ending with CC=0"
17 "Unit": "PAI-CRYPTO",
21 "PublicDescription": "KM-TDEA-128 function ending with CC=0"
24 "Unit": "PAI-CRYPTO",
28 "PublicDescription": "KM-TDEA-192 function ending with CC=0"
31 "Unit": "PAI-CRYPTO",
35 "PublicDescription": "KM-Encrypted-DEA function ending with CC=0"
38 "Unit": "PAI-CRYPTO",
[all …]
/linux/drivers/iio/common/inv_sensors/
H A Dinv_sensors_timestamp.c1 // SPDX-License-Identifier: GPL-2.0-or-later
13 /* compute jitter, min and max following jitter in per mille */
17 (((_val) * (1000 - (_jitter))) / 1000)
27 acc->values[acc->idx++] = val; in inv_update_acc()
28 if (acc->idx >= ARRAY_SIZE(acc->values)) in inv_update_acc()
29 acc->idx = 0; in inv_update_acc()
31 /* compute the mean of all stored values, use 0 as empty slot */ in inv_update_acc()
32 for (i = 0; i < ARRAY_SIZE(acc->values); ++i) { in inv_update_acc()
33 if (acc->values[i] == 0) in inv_update_acc()
35 sum += acc->values[i]; in inv_update_acc()
[all …]
/linux/Documentation/driver-api/
H A Ddma-buf.rst1 Buffer Sharing and Synchronization (dma-buf)
4 The dma-buf subsystem provides the framework for sharing buffers for
14 interact with the three main primitives offered by dma-buf:
16 - dma-buf, representing a sg_table and exposed to userspace as a file
19 - dma-fence, providing a mechanism to signal when an asynchronous
21 - dma-resv, which manages a set of dma-fences for a particular dma-buf
22 allowing implicit (kernel-ordered) synchronization of work to
27 --------------------------------
29 For more details on how to design your subsystem's API for dma-buf use, please
30 see Documentation/userspace-api/dma-buf-alloc-exchange.rst.
[all …]
/linux/Documentation/hwmon/
H A Dfam15h_power.rst16 - BIOS and Kernel Developer's Guide (BKDG) For AMD Family 15h Processors
17 - BIOS and Kernel Developer's Guide (BKDG) For AMD Family 16h Processors
18 - AMD64 Architecture Programmer's Manual Volume 2: System Programming
23 -----------
55 On multi-node processors the calculated value is for the entire
57 attributes only for internal node0 of a multi-node processor.
67 compute unit power accumulator sample period
76 the ratio of compute unit power accumulator sample period to the
80 max compute unit accumulated power which is indicated by
84 compute unit accumulated power which is indicated by
[all …]
/linux/tools/testing/selftests/bpf/
H A Dnetwork_helpers.h1 /* SPDX-License-Identifier: GPL-2.0 */
31 /* +ve: Passed to listen() as-is.
35 * Most tests only have one on-going connection.
36 * -ve: It is changed to 0 before passing to listen().
94 * open_netns() - Switch to specified network namespace by name.
106 * append_tid() - Append thread ID to the given string. in csum_fold()
113 * Returns -1 on errors, 0 otherwise in csum_partial()
142 iph->chec
[all...]
/linux/Documentation/devicetree/bindings/arm/bcm/
H A Dbcm2835.yaml1 # SPDX-License-Identifier: GPL-2.0
3 ---
5 $schema: http://devicetree.org/meta-schemas/core.yaml#
10 - Eric Anholt <eric@anholt.net>
11 - Stefan Wahren <wahrenst@gmx.net>
18 - description: BCM2711 based Boards
20 - enum:
21 - raspberrypi,400
22 - raspberrypi,4-compute-module
23 - raspberrypi,4-model-b
[all …]
/linux/lib/
H A Dsiphash.c1 // SPDX-License-Identifier: (GPL-2.0-only OR BSD-3-Clause)
2 /* Copyright (C) 2016-2022 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
4 * SipHash: a fast short-input PRF
7 * This implementation is specifically for SipHash2-4 for a secure PRF
8 * and HalfSipHash1-3/SipHash1-3 for an insecure PRF only suitable for
17 #include <asm/word-at-a-time.h>
28 v3 ^= key->key[1]; \
29 v2 ^= key->key[0]; \
30 v1 ^= key->key[1]; \
31 v0 ^= key->key[0];
[all …]
H A Dbch.c15 * Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
24 * Bose-Chaudhuri-Hocquenghem (BCH) codes.
30 * Call bch_encode to compute and store ecc parity bytes to a given buffer.
50 * b. Error locator polynomial computation using Berlekamp-Massey algorithm
56 * (BTA) down to a certain degree (4), after which ad hoc low-degree polynomial
63 * - WEWoRC 2009, Graz, Austria, LNCS, Springer, July 2009, to appear.
81 #define GF_N(_p) ((1 << (CONFIG_BCH_CONST_M))-1)
85 #define GF_M(_p) ((_p)->m)
86 #define GF_T(_p) ((_p)->t)
87 #define GF_N(_p) ((_p)->n)
[all …]
/linux/tools/perf/Documentation/
H A Dperf-diff.txt1 perf-diff(1)
5 ----
6 perf-diff - Read perf.data files and display the differential profile
9 --------
14 -----------
29 -------
30 -D::
31 --dump-raw-trace::
34 --kallsyms=<file>::
37 -m::
[all …]
/linux/fs/xfs/libxfs/
H A Dxfs_rtbitmap.c1 // SPDX-License-Identifier: GPL-2.0
3 * Copyright (c) 2000-2005 Silicon Graphics, Inc.
37 struct xfs_mount *mp = bp->b_mount; in xfs_rtbuf_verify()
38 struct xfs_rtbuf_blkinfo *hdr = bp->b_addr; in xfs_rtbuf_verify()
40 if (!xfs_verify_magic(bp, hdr->rt_magic)) in xfs_rtbuf_verify()
46 if (!uuid_equal(&hdr->rt_uuid, &mp->m_sb.sb_meta_uuid)) in xfs_rtbuf_verify()
48 if (hdr->rt_blkno != cpu_to_be64(xfs_buf_daddr(bp))) in xfs_rtbuf_verify()
57 struct xfs_mount *mp = bp->b_mount; in xfs_rtbuf_verify_read()
58 struct xfs_rtbuf_blkinfo *hdr = bp->b_addr; in xfs_rtbuf_verify_read()
64 if (!xfs_log_check_lsn(mp, be64_to_cpu(hdr->rt_lsn))) { in xfs_rtbuf_verify_read()
[all …]
/linux/arch/x86/include/asm/
H A Dchecksum_64.h1 /* SPDX-License-Identifier: GPL-2.0 */
6 * Checksums for x86-64
8 * with some code from asm-x86/checksum.h
15 * csum_fold - Fold and invert a 32bit checksum.
41 * ip_fast_csum - Compute the IPv4 header checksum efficiently.
76 * csum_tcpup_nofold - Compute an IPv4 pseudo header checksum.
102 * csum_tcpup_magic - Compute an IPv4 pseudo header checksum.
120 * csum_partial - Compute an internet checksum.
139 * ip_compute_csum - Compute an 16bit IP checksum.
149 * csum_ipv6_magic - Compute checksum of an IPv6 pseudo header.
/linux/drivers/gpu/drm/amd/amdgpu/
H A Damdgpu_doorbell.h47 * For asic before vega10, doorbell is 32-bit, so the
49 * can be 64-bit, so the index defined is in qword.
121 /* Compute + GFX: 0~255 */
176 /* 8 compute rings per GC. Max to 0x1CE */
188 /* Compute + GFX: 0~255 */
244 …* All compute related doorbells: kiq, hiq, diq, traditional compute queue, user queue, should loca…
246 * Compute related doorbells are allocated from 0x00 to 0x8a
257 /* Compute engines */
277 * default non-graphics QWORD index is 0xe0 - 0xFF inclusive
328 /* Compute: 0x08 ~ 0x20 */
/linux/Documentation/devicetree/bindings/interconnect/
H A Dqcom,rpmh.yaml1 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
3 ---
5 $schema: http://devicetree.org/meta-schemas/core.yaml#
7 title: Qualcomm RPMh Network-On-Chip Interconnect
10 - Georgi Djakov <georgi.djakov@linaro.org>
11 - Odelu Kukatla <quic_okukatla@quicinc.com>
27 - qcom,sc7180-aggre1-noc
28 - qcom,sc7180-aggre2-noc
29 - qcom,sc7180-camnoc-virt
30 - qcom,sc7180-compute-noc
[all …]
/linux/tools/perf/
H A Dbuiltin-diff.c1 // SPDX-License-Identifier: GPL-2.0
3 * builtin-diff.c
23 #include "util/time-utils.h"
27 #include "util/block-info.h"
33 #include <subcmd/parse-options.h>
113 COMPUTE_STREAM, /* After COMPUTE_MAX to avoid use current compute arrays */
118 [COMPUTE_DELTA_ABS] = "delta-abs",
124 static int compute = COMPUTE_DELTA_ABS; variable
186 int ret = -EINVAL; in setup_compute_opt_wdiff()
205 pr_debug("compute wdiff w1(%" PRId64 ") w2(%" PRId64 ")\n", in setup_compute_opt_wdiff()
[all …]
/linux/lib/raid6/
H A Dmktables.c1 // SPDX-License-Identifier: GPL-2.0-or-later
2 /* -*- linux-c -*- ------------------------------------------------------- *
4 * Copyright 2002-2007 H. Peter Anvin - All Rights Reserved
6 * ----------------------------------------------------------------------- */
11 * Make RAID-6 tables. This is a host user space program to be run at
64 /* Compute multiplication table */ in main()
83 /* Compute vector multiplication table */ in main()
108 /* Compute power-of-2 table (exponent) */ in main()
127 /* Compute log-of-2 table */ in main()
147 /* Compute inverse table x^-1 == x^254 */ in main()
[all …]
/linux/arch/xtensa/lib/
H A Dstrncpy_user.S8 * Returns: -EFAULT if exception before terminator, N if the entire
60 bbsi.l a3, 0, .Lsrc1mod2 # if only 8-bit aligned
61 bbsi.l a3, 1, .Lsrc2mod4 # if only 16-bit aligned
62 .Lsrcaligned: # return here when src is word-aligned
74 addi a4, a4, -1 # decrement len
76 bbci.l a3, 1, .Lsrcaligned # if src is now word-aligned
80 /* 1-cycle interlock */
84 addi a4, a4, -1 # decrement len
91 addi a4, a4, -1 # decrement len
94 sub a2, a11, a2 # compute strlen
[all …]
/linux/Documentation/accel/
H A Dintroduction.rst1 .. SPDX-License-Identifier: GPL-2.0
7 The Linux compute accelerators subsystem is designed to expose compute
8 accelerators in a common way to user-space and provide a common set of
11 These devices can be either stand-alone ASICs or IP blocks inside an SoC/GPU.
13 Machine-Learning (ML) and/or Deep-Learning (DL) computations, the accel layer
16 Typically, a compute accelerator will belong to one of the following
19 - Edge AI - doing inference at an edge device. It can be an embedded ASIC/FPGA,
23 - Inference data-center - single/multi user devices in a large server. This
24 type of device can be stand-alone or an IP inside a SoC or a GPU. It will
25 have on-board DRAM (to hold the DL topology), DMA engines and
[all …]
/linux/drivers/clk/ti/
H A Ddpll44xx.c1 // SPDX-License-Identifier: GPL-2.0-only
3 * OMAP4-specific DPLL control functions
20 * can supported when using the DPLL low-power mode. Frequencies are
22 * Status, and Low-Power Operation Mode".
45 mask = clk->flags & CLOCK_CLKOUTX2 ? in omap4_dpllmx_allow_gatectrl()
49 v = ti_clk_ll_ops->clk_readl(&clk->clksel_reg); in omap4_dpllmx_allow_gatectrl()
52 ti_clk_ll_ops->clk_writel(v, &clk->clksel_reg); in omap4_dpllmx_allow_gatectrl()
63 mask = clk->flags & CLOCK_CLKOUTX2 ? in omap4_dpllmx_deny_gatectrl()
67 v = ti_clk_ll_ops->clk_readl(&clk->clksel_reg); in omap4_dpllmx_deny_gatectrl()
70 ti_clk_ll_ops->clk_writel(v, &clk->clksel_reg); in omap4_dpllmx_deny_gatectrl()
[all …]
/linux/Documentation/admin-guide/mm/
H A Dnumaperf.rst8 Some platforms may have multiple types of memory attached to a compute
17 CPUs, they may still be local to one or more compute nodes relative to
18 other nodes. The following diagram shows one such example of two compute
19 nodes with local memory and a memory only node for each of compute node::
21 +------------------+ +------------------+
22 | Compute Node 0 +-----+ Compute Node 1 |
24 +--------+---------+ +--------+---------+
26 +--------+---------+ +--------+---------+
28 +------------------+ +--------+---------+
36 performance when accessing a given memory target. Each initiator-target
[all …]
/linux/Documentation/userspace-api/media/v4l/
H A Dmetafmt-vsp1-hgo.rst1 .. SPDX-License-Identifier: GFDL-1.1-no-invariants-or-later
3 .. _v4l2-meta-fmt-vsp1-hgo:
9 Renesas R-Car VSP1 1-D Histogram Data
15 This format describes histogram data generated by the Renesas R-Car VSP1 1-D
20 computes the minimum, maximum and sum of all pixels as well as per-channel
23 The HGO can compute histograms independently per channel, on the maximum of the
28 - In *64 bins normal mode*, the HGO operates on the three channels independently
29 to compute three 64-bins histograms. RGB, YCbCr and HSV image formats are
31 - In *64 bins maximum mode*, the HGO operates on the maximum of the (R, G, B)
32 channels to compute a single 64-bins histogram. Only the RGB image format is
[all …]

12345678910>>...45