/linux-6.14.4/drivers/misc/ |
D | sram-exec.c | 1 // SPDX-License-Identifier: GPL-2.0-only 3 * SRAM protect-exec region helper functions 5 * Copyright (C) 2017 Texas Instruments Incorporated - https://www.ti.com/ 25 unsigned long base = (unsigned long)part->base; in sram_check_protect_exec() 26 unsigned long end = base + block->size; in sram_check_protect_exec() 29 dev_err(sram->dev, in sram_check_protect_exec() 30 "SRAM pool marked with 'protect-exec' is not page aligned and will not be created.\n"); in sram_check_protect_exec() 31 return -ENOMEM; in sram_check_protect_exec() 40 list_add_tail(&part->list, &exec_pool_list); in sram_add_protect_exec() 47 * sram_exec_copy - copy data to a protected executable region of sram [all …]
|
/linux-6.14.4/drivers/gpu/drm/xe/ |
D | xe_hw_engine_group_types.h | 1 /* SPDX-License-Identifier: MIT */ 14 * enum xe_hw_engine_group_execution_mode - possible execution modes of a hw 17 * @EXEC_MODE_LR: execution in long-running mode 26 * struct xe_hw_engine_group - Hardware engine group 33 * @exec_queue_list: list of exec queues attached to this 37 /** @resume_work: worker to resume faulting LR exec queues */ 39 /** @resume_wq: workqueue to resume faulting LR exec queues */ 42 * @mode_sem: used to protect this group's hardware resources and ensure
|
D | xe_exec_queue.c | 1 // SPDX-License-Identifier: MIT 41 if (q->vm) in __xe_exec_queue_free() 42 xe_vm_put(q->vm); in __xe_exec_queue_free() 44 if (q->xef) in __xe_exec_queue_free() 45 xe_file_put(q->xef); in __xe_exec_queue_free() 57 struct xe_gt *gt = hwe->gt; in __xe_exec_queue_alloc() 65 return ERR_PTR(-ENOMEM); in __xe_exec_queue_alloc() 67 kref_init(&q->refcount); in __xe_exec_queue_alloc() 68 q->flags = flags; in __xe_exec_queue_alloc() 69 q->hwe = hwe; in __xe_exec_queue_alloc() [all …]
|
D | xe_vm.c | 1 // SPDX-License-Identifier: MIT 8 #include <linux/dma-fence-array.h> 44 return vm->gpuvm.r_obj; in xe_vm_obj() 48 * xe_vma_userptr_check_repin() - Advisory check for repin needed 53 * without the vm->userptr.notifier_lock held. There is no guarantee that the 57 * Return: 0 if userptr vma is valid, -EAGAIN otherwise; repin recommended. 61 return mmu_interval_check_retry(&uvma->userptr.notifier, in xe_vma_userptr_check_repin() 62 uvma->userptr.notifier_seq) ? in xe_vma_userptr_check_repin() 63 -EAGAIN : 0; in xe_vma_userptr_check_repin() 68 struct xe_vma *vma = &uvma->vma; in xe_vma_userptr_pin_pages() [all …]
|
D | xe_device_types.h | 1 /* SPDX-License-Identifier: MIT */ 3 * Copyright © 2022-2023 Intel Corporation 40 #define GRAPHICS_VER(xe) ((xe)->info.graphics_verx100 / 100) 41 #define MEDIA_VER(xe) ((xe)->info.media_verx100 / 100) 42 #define GRAPHICS_VERx100(xe) ((xe)->info.graphics_verx100) 43 #define MEDIA_VERx100(xe) ((xe)->info.media_verx100) 44 #define IS_DGFX(xe) ((xe)->info.is_dgfx) 55 ((_xe)->info.platform == (_platform) && \ 56 (_xe)->info.step.graphics >= (min_step) && \ 57 (_xe)->info.step.graphics < (max_step)) [all …]
|
/linux-6.14.4/tools/testing/selftests/exec/ |
D | check-exec-tests.sh | 2 # SPDX-License-Identifier: GPL-2.0 7 # samples/check-exec/inc.c 11 set -u -e -o pipefail 14 exec 2>/dev/null 16 DIR="$(dirname $(readlink -f "$0"))" 29 if [[ ${ret} -ne ${expect} ]]; then 33 if [[ ${ret} -eq 0 && "${out}" != "${EXPECTED_OUTPUT}" ]]; then 49 if [[ ${ret} -ne ${expect} ]]; then 53 if [[ ${ret} -eq 0 && "${out}" != "${EXPECTED_OUTPUT}" ]]; then 67 out="$("$@" ./inc -i < "${script}")" || ret=$? [all …]
|
/linux-6.14.4/Documentation/gpu/ |
D | drm-vm-bind-locking.rst | 1 .. SPDX-License-Identifier: (GPL-2.0+ OR MIT) 30 meta-data. Typically one per client (DRM file-private), or one per 33 associated meta-data. The backing storage of a gpu_vma can either be 34 a GEM object or anonymous or page-cache pages mapped also into the CPU 40 is anonymous or page-cache pages as described above. 43 page-table entries point to that backing store. 47 the :doc:`dma-buf doc </driver-api/dma-buf>`. 53 allows deadlock-safe locking of multiple dma_resvs in arbitrary 55 :doc:`dma-buf doc </driver-api/dma-buf>`. 56 * ``exec function``: An exec function is a function that revalidates all [all …]
|
/linux-6.14.4/drivers/gpu/drm/ |
D | drm_gpuvm.c | 1 // SPDX-License-Identifier: GPL-2.0-only OR MIT 47 * The GPU VA manager internally uses a rb-tree to manage the 62 * entries from within dma-fence signalling critical sections it is enough to 63 * pre-allocate the &drm_gpuva structures. 88 * lists are maintained in order to accelerate locking of dma-resv locks and 112 * implement Vulkan 'Sparse Memory Bindings' - drivers UAPIs often refer to this 150 * created such that those mappings are split up and re-mapped partically. 161 * dma-fence signalling critical path. 172 * by drm_gpuvm_sm_map() - it doesn't cover any arbitrary combinations of these. 180 * old: |-----------| (bo_offset=n) [all …]
|
/linux-6.14.4/Documentation/userspace-api/ |
D | check_exec.rst | 1 .. SPDX-License-Identifier: GPL-2.0 12 `samples/check-exec/inc.c`_ example. 41 Programs should always perform this check to apply kernel-level checks against 60 To avoid race conditions leading to time-of-check to time-of-use issues, 76 securebits but without relying on any other user-controlled configuration. 103 of any exec securebits: 122 users. That can protect against unintended script executions (e.g. ``sh 123 /tmp/*.sh``). This makes sense for (semi-restricted) user sessions. 131 dm-verity/IPE) but where access rights might not be ready yet. Indeed, 143 .. _samples/check-exec/inc.c: [all …]
|
D | mseal.rst | 1 .. SPDX-License-Identifier: GPL-2.0 17 such an attacker primitive can break control-flow integrity guarantees 18 since read-only memory that is supposed to be trusted can become writable 29 ----------------------- 34 - The start address must be in an allocated VMA. 35 - The start address must be page aligned. 36 - The end address (**addr** + **len**) must be in an allocated VMA. 37 - no gap (unallocated memory) between start and end address. 44 - **0**: Success. 45 - **-EINVAL**: [all …]
|
/linux-6.14.4/Documentation/devicetree/bindings/arm/omap/ |
D | mpu.txt | 1 * TI - MPU (Main Processor Unit) subsystem 8 - compatible : Should be "ti,omap3-mpu" for OMAP3 9 Should be "ti,omap4-mpu" for OMAP4 10 Should be "ti,omap5-mpu" for OMAP5 11 - ti,hwmods: "mpu" 14 - sram: Phandle to the ocmcram node 17 - pm-sram: Phandles to ocmcram nodes to be used for power management. 18 First should be type 'protect-exec' for the driver to use to copy 25 - For an OMAP5 SMP system: 28 compatible = "ti,omap5-mpu"; [all …]
|
/linux-6.14.4/tools/testing/selftests/powerpc/mm/ |
D | pkey_siginfo.c | 1 // SPDX-License-Identifier: GPL-2.0 44 if (sinfo->si_code != SEGV_PKUERR) { in segv_handler() 50 if (sinfo->si_addr != (void *) fault_addr) { in segv_handler() 68 pgstart = (void *) ((unsigned long) fault_addr & ~(pgsize - 1)); in segv_handler() 72 * reassociate the page with the exec-only pkey since execute in segv_handler() 77 * read-write rights, change the AMR permission bits for the in segv_handler() 97 static void *protect(void *p) in protect() function 105 base = ((struct region *) p)->base; in protect() 106 size = ((struct region *) p)->size; in protect() 119 * Repeatedly try to protect the common region with a permissive in protect() [all …]
|
/linux-6.14.4/Documentation/devicetree/bindings/memory-controllers/ti/ |
D | emif.txt | 3 EMIF - External Memory Interface - is an SDRAM controller used in 11 - compatible : Should be of the form "ti,emif-<ip-rev>" where <ip-rev> 14 "ti,emif-am3352" 15 "ti,emif-am4372" 16 "ti,emif-dra7xx" 17 "ti,emif-keystone" 19 - phy-type : <u32> indicating the DDR phy type. Following are the 24 - device-handle : phandle to a "lpddr2" node representing the memory part 26 - ti,hwmods : For TI hwmods processing and omap device creation 29 - interrupts : interrupt used by the controller [all …]
|
/linux-6.14.4/arch/powerpc/mm/ |
D | pgtable.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later 7 * Copyright (C) 1995-1996 Gary Thomas ([email protected]) 29 #include <asm/pte-walk.h> 41 return current->thread.regs && TRAP(current->thread.regs) == 0x400; in is_exec_fault() 76 /* Server-style MMU handles coherency when hashing if HW exec permission 77 * is supposed per page (currently 64-bit only). If not, then, we always 78 * flush the cache for valid PTEs in set_pte. Embedded CPU without HW exec 90 if (!test_bit(PG_dcache_clean, &folio->flags)) { in set_pte_filter_hash() 92 set_bit(PG_dcache_clean, &folio->flags); in set_pte_filter_hash() 104 /* Embedded type MMU with HW exec support. This is a bit more complicated [all …]
|
/linux-6.14.4/Documentation/arch/x86/ |
D | shstk.rst | 1 .. SPDX-License-Identifier: GPL-2.0 4 Control-flow Enforcement Technology (CET) Shadow Stack 10 Control-flow Enforcement Technology (CET) covers several related x86 processor 12 can protect both applications and the kernel. 20 control-protection fault. IBT verifies indirect CALL/JMP targets are intended 22 Stack and Indirect Branch Tracking. Today in the 64-bit kernel, only userspace 45 from readelf/llvm-readelf output:: 47 readelf -n <application> | grep -a SHSTK 60 on a per-thread basis. The enablement status is inherited on clone, so if the 90 -EPERM if any of the passed feature are locked. [all …]
|
/linux-6.14.4/include/drm/ |
D | drm_gpuvm.h | 1 /* SPDX-License-Identifier: GPL-2.0-only OR MIT */ 28 #include <linux/dma-resv.h> 42 * enum drm_gpuva_flags - flags for struct drm_gpuva 66 * struct drm_gpuva - structure to track a GPU VA mapping 126 * @rb: structure containing data to store &drm_gpuvas in a rb-tree 130 * @rb.node: the rb-tree node 138 * through the rb-tree while doing modifications on the rb-tree 144 * @rb.__subtree_last: needed by the interval tree, holding last-in-subtree 166 va->va.addr = addr; in drm_gpuva_init() 167 va->va.range = range; in drm_gpuva_init() [all …]
|
/linux-6.14.4/drivers/gpu/drm/amd/amdgpu/ |
D | amdgpu_gem.c | 32 #include <linux/dma-buf.h> 49 struct ttm_buffer_object *bo = vmf->vma->vm_private_data; in amdgpu_gem_fault() 50 struct drm_device *ddev = bo->base.dev; in amdgpu_gem_fault() 65 ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot, in amdgpu_gem_fault() 70 ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot); in amdgpu_gem_fault() 72 if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT)) in amdgpu_gem_fault() 76 dma_resv_unlock(bo->base.resv); in amdgpu_gem_fault() 92 ttm_bo_put(&aobj->tbo); in amdgpu_gem_object_free() 124 bo = &ubo->bo; in amdgpu_gem_object_create() 125 *obj = &bo->tbo.base; in amdgpu_gem_object_create() [all …]
|
D | amdgpu_amdkfd_gpuvm.c | 1 // SPDX-License-Identifier: MIT 3 * Copyright 2014-2018 Advanced Micro Devices, Inc. 23 #include <linux/dma-buf.h> 73 #define domain_string(domain) domain_bit_to_string[ffs(domain)-1] 82 list_for_each_entry(entry, &mem->attachments, list) in kfd_mem_is_attached() 83 if (entry->bo_va->base.vm == avm) in kfd_mem_is_attached() 90 * reuse_dmamap() - Check whether adev can share the original 104 return (adev->ram_is_direct_mapped && bo_adev->ram_is_direct_mapped) || in reuse_dmamap() 105 (adev->dev->iommu_group == bo_adev->dev->iommu_group); in reuse_dmamap() 109 * System (TTM + userptr) memory - 15/16th System RAM [all …]
|
/linux-6.14.4/fs/ |
D | exec.c | 1 // SPDX-License-Identifier: GPL-2.0-only 3 * linux/fs/exec.c 9 * #!-checking implemented by tytso. 12 * Demand-loading implemented 01.12.91 - no need to read anything but 14 * "current->executable", and page faults do the actual loading. Clean. 17 * was less than 2 hours work to get demand-loading completely implemented. 20 * current->executable is only used by the procfs. This allows a dispatch 91 insert ? list_add(&fmt->lh, &formats) : in __register_binfmt() 92 list_add_tail(&fmt->lh, &formats); in __register_binfmt() 101 list_del(&fmt->lh); in unregister_binfmt() [all …]
|
/linux-6.14.4/Documentation/devicetree/bindings/sram/ |
D | sram.yaml | 1 # SPDX-License-Identifier: GPL-2.0 3 --- 5 $schema: http://devicetree.org/meta-schemas/core.yaml# 7 title: Generic on-chip SRAM 10 - Rob Herring <[email protected]> 19 Following the generic-names recommended practice, node names should 30 - mmio-sram 31 - amlogic,meson-gxbb-sram 32 - arm,juno-sram-ns 33 - atmel,sama5d2-securam [all …]
|
/linux-6.14.4/include/uapi/linux/ |
D | taskstats.h | 1 /* SPDX-License-Identifier: LGPL-2.1 WITH Linux-syscall-note */ 2 /* taskstats.h - exporting per-task statistics 22 /* Format for per-task data returned to userland when 23 * - a task exits 24 * - listener requests stats for a task 33 * c) add new fields after version comment; maintain 64-bit alignment 78 /* Following four fields atomically updated using task->delays->lock */ 94 /* cpu "wall-clock" running time 133 /* Accumulated RSS usage in duration of a task, in MBytes-usecs. 139 __u64 coremem; /* accumulated RSS usage in MB-usec */ [all …]
|
/linux-6.14.4/drivers/gpu/drm/panthor/ |
D | panthor_mmu.c | 1 // SPDX-License-Identifier: GPL-2.0 or MIT 16 #include <linux/dma-mapping.h> 20 #include <linux/io-pgtable.h> 42 * struct panthor_as_slot - Address space slot 50 * struct panthor_mmu - MMU related data 59 * us to re-assign them to re-assign slots on-demand. 105 * struct panthor_vm_pool - VM pool object 113 * struct panthor_vma - GPU mapping object 133 * struct panthor_vm_op_ctx - VM operation context 135 * With VM operations potentially taking place in a dma-signaling path, we [all …]
|
/linux-6.14.4/Documentation/driver-api/ |
D | dma-buf.rst | 1 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-6.14.4/tools/testing/selftests/drivers/net/ |
D | stats.py | 2 # SPDX-License-Identifier: GPL-2.0 20 def check_pause(cfg) -> None: 24 ethnl.pause_get({"header": {"dev-index": cfg.ifindex}}) 30 data = ethnl.pause_get({"header": {"dev-index": cfg.ifindex, 35 def check_fec(cfg) -> None: 39 ethnl.fec_get({"header": {"dev-index": cfg.ifindex}}) 45 data = ethnl.fec_get({"header": {"dev-index": cfg.ifindex, 50 def pkt_byte_sum(cfg) -> None: 65 for key in ['tx-packets', 'tx-bytes', 'rx-packets', 'rx-bytes']: 72 for key in ['tx-packets', 'tx-bytes', 'rx-packets', 'rx-bytes']: [all …]
|
/linux-6.14.4/Documentation/filesystems/ |
D | affs.rst | 1 .. SPDX-License-Identifier: GPL-2.0 19 in file names are case-insensitive, as they ought to be. 44 protect 107 Amiga -> Linux: 111 - R maps to r for user, group and others. On directories, R implies x. 113 - W maps to w. 115 - E maps to x. 117 - D is ignored. 119 - H, S and P are always retained and ignored under Linux. 121 - A is cleared when a file is written to. [all …]
|