| /dpdk/usertools/ |
| H A D | dpdk-hugepages.py | 86 pages = get_hugepages(path + '/' + hdir)
87 if pages > 0:
99 pages = get_hugepages(path + '/' + hdir)
100 if pages > 0:
135 def set_numa_pages(pages, hugepgsz, node=None): argument
144 set_hugepages(huge_path, pages)
147 def set_non_numa_pages(pages, hugepgsz): argument
150 set_hugepages(path, pages)
153 def reserve_pages(pages, hugepgsz, node=None): argument
156 set_numa_pages(pages, hugepgsz, node=node)
[all …]
|
| /dpdk/drivers/common/dpaax/ |
| H A D | dpaax_iova_table.c | 231 entry[i].pages = entry[i-1].pages + in dpaax_iova_table_populate()
234 entry[i].pages = (uint64_t *)((unsigned char *)entry + in dpaax_iova_table_populate()
239 i, entry[i].start, entry[i].len, entry[i].pages); in dpaax_iova_table_populate()
309 entry[i].pages[e_offset] = align_vaddr; in dpaax_iova_table_update()
377 i, entry[i].start, entry[i].len, entry[i].pages); in dpaax_iova_table_dump()
380 if (entry[i].pages[j] == 0) in dpaax_iova_table_dump()
384 entry[i].pages[j]); in dpaax_iova_table_dump()
|
| H A D | dpaax_iova_table.h | 30 uint64_t *pages; /**< VA for each physical page in this block */ member
107 if ((void *)(uintptr_t)entry[i].pages[index] == NULL) in dpaax_iova_table_get_va()
110 vaddr = (void *)((uintptr_t)entry[i].pages[index] + offset); in dpaax_iova_table_get_va()
|
| /dpdk/doc/guides/tools/ |
| H A D | hugepages.rst | 47 Set NUMA node to reserve pages on
56 Reserve huge pages.
66 status of huge pages, modifying the setup requires root privileges.
76 To a complete setup of with 2 Gigabyte of 1G huge pages::
|
| /dpdk/lib/eal/common/ |
| H A D | eal_common_dynmem.c | 280 struct rte_memseg **pages; in eal_dynmem_hugepage_init() local
306 pages = malloc(sizeof(*pages) * needed); in eal_dynmem_hugepage_init()
307 if (pages == NULL) { in eal_dynmem_hugepage_init()
313 cur_pages = eal_memalloc_alloc_seg_bulk(pages, in eal_dynmem_hugepage_init()
317 free(pages); in eal_dynmem_hugepage_init()
323 struct rte_memseg *ms = pages[i]; in eal_dynmem_hugepage_init()
327 free(pages); in eal_dynmem_hugepage_init()
|
| /dpdk/doc/guides/faq/ |
| H A D | faq.rst | 12 If I want to change the number of hugepages allocated, how do I remove the original pages allocated?
15 The number of pages allocated can be seen by executing the following command::
19 Once all the pages are mmapped by an application, they stay that way.
20 If you start a test application with less than the maximum, then you have free pages.
21 When you stop and restart the test application, it looks to see if the pages are available in the `…
22 If you look in the directory, you will see ``n`` number of 2M pages files. If you specified 1024, y…
25 If you need to change the number of pages, it is easier to first remove the pages.
31 I have set up a total of 1024 Hugepages (that is, allocated 512 2M pages to each NUMA node).
33 …parameter does not guarantee that huge pages will be reserved on specific sockets. Therefore, allo…
127 This is especially true when using a large number (>512) of 2 MB huge pages. Please increase the op…
|
| /dpdk/doc/guides/nics/ |
| H A D | build_and_test.rst | 50 #. Mount huge pages:
57 #. Request huge pages:
60 hugepage size configured in the system and calculate the number of pages
63 To reserve 1024 pages of 2MB:
|
| /dpdk/doc/guides/linux_gsg/ |
| H A D | sys_reqs.rst | 143 By using hugepage allocations, performance is increased since fewer pages are needed,
156 (assuming that 1024 of 2MB pages are required)::
162 However, pages can also be reserved explicitly on individual NUMA nodes
185 For 2 MB pages, just pass the hugepages option to the kernel. For example, to reserve 1024 pages of…
189 For other hugepage sizes, for example 1G pages, the size must be specified explicitly and
191 For example, to reserve 4G of hugepage memory in the form of four 1G pages, the following options s…
226 to manually create mount points for those hugepage sizes (e.g. 1GB pages).
|
| H A D | nic_perf_intel_platform.rst | 114 #. Reserve 1G huge pages via grub configurations. For example to reserve 8 huge pages of 1G size::
130 1. Reserve huge pages.
|
| H A D | build_sample_apps.rst | 166 Suppose the system has 1024 reserved 2 MB pages in socket 0 and 1024 in socket 1.
167 If the user requests 128 MB of memory, the 64 pages may not match the constraints:
173 * These pages can be located anywhere in physical memory, and, although the DPDK EAL will attempt…
174 …it is possible that the pages will not be contiguous. In this case, the application is not able to…
|
| H A D | enable_func.rst | 32 * RLIMIT_MEMLOCK (amount of pinned pages the process is allowed to have)
|
| /dpdk/doc/guides/windows_gsg/ |
| H A D | run_apps.rst | 7 Grant *Lock pages in memory* Privilege
10 Use of hugepages ("large pages" in Windows terminology) requires
18 2. Open *Local Policies / User Rights Assignment / Lock pages in memory.*
|
| /dpdk/doc/guides/cryptodevs/ |
| H A D | virtio.rst | 85 reserve enough huge pages
93 reserve enough huge pages
|
| H A D | cnxk.rst | 128 * Ensure that sufficient huge pages are available for your application::
|
| /dpdk/drivers/net/qede/base/ |
| H A D | ecore_cxt.c | 2194 u16 i, pages = 0; in ecore_get_cdut_num_pf_init_pages() local
2199 pages += ecore_blk_calculate_pages(p_blk); in ecore_get_cdut_num_pf_init_pages()
2202 return pages; in ecore_get_cdut_num_pf_init_pages()
2209 u16 i, pages = 0; in ecore_get_cdut_num_vf_init_pages() local
2214 pages += ecore_blk_calculate_pages(p_blk); in ecore_get_cdut_num_vf_init_pages()
2217 return pages; in ecore_get_cdut_num_vf_init_pages()
2224 u16 i, pages = 0; in ecore_get_cdut_num_pf_work_pages() local
2229 pages += ecore_blk_calculate_pages(p_blk); in ecore_get_cdut_num_pf_work_pages()
2232 return pages; in ecore_get_cdut_num_pf_work_pages()
2239 u16 pages = 0, i; in ecore_get_cdut_num_vf_work_pages() local
[all …]
|
| /dpdk/doc/guides/howto/ |
| H A D | lm_bond_virtio_sriov.rst | 265 duplicate: 301620 pages
266 skipped: 0 pages
267 normal: 96433 pages
286 duplicate: 303985 pages
287 skipped: 0 pages
288 normal: 94073 pages
|
| H A D | lm_virtio_vhost_user.rst | 211 duplicate: 303985 pages
212 skipped: 0 pages
213 normal: 94073 pages
|
| /dpdk/doc/guides/compressdevs/ |
| H A D | octeontx.rst | 83 reserve enough huge pages
|
| /dpdk/doc/guides/prog_guide/ |
| H A D | env_abstraction_layer.rst | 128 hugepages, and deallocate those, but any preallocated pages will not be freed.
133 ``--single-file-segments`` command-line option. This option will put pages in
269 when all pages mapped from it are freed back to the system.
295 when all pages mapped from it are freed,
328 Another option is to use bigger page sizes. Since fewer pages are required to
927 Also, since a malloc element may span multiple pages, its contents may not
995 in which case newly allocated pages are also added to the heap, merging with any
1050 Any allocation of new pages has to go through primary process. If the
1075 If deallocating pages at runtime is supported, and the free element encloses
1076 one or more pages, those pages can be deallocated and be removed from the heap.
[all …]
|
| H A D | writing_efficient_code.rst | 81 Locking memory pages
84 The underlying operating system is allowed to load/unload memory pages at its own discretion.
|
| /dpdk/doc/guides/vdpadevs/ |
| H A D | ifc.rst | 32 thread when LM happens, this thread will help device to log dirty pages. Thus
|
| /dpdk/doc/guides/platform/ |
| H A D | cnxk.rst | 184 # 64K pages enabled for better performance
187 # huge pages support enabled
|
| /dpdk/doc/guides/rel_notes/ |
| H A D | release_19_11.rst | 15 for mapping pages and resources into its address space.
31 pages by default. Note, this may consume more memory when using small memory
32 pages.
|
| /dpdk/doc/guides/sample_app_ug/ |
| H A D | multi_process.rst | 61 EAL: Requesting 2 pages of size 1073741824
62 EAL: Requesting 768 pages of size 2097152
|
| /dpdk/lib/eal/linux/ |
| H A D | eal_memory.c | 1277 int pages = pages_old - pages_new; local
1279 nr_hugepages -= pages;
|