Lines Matching full:pages
3 /// See [madvise()](https://man7.org/linux/man-pages/man2/madvise.2.html) map page.
20 /// Expect page references in sequential order. (Hence, pages
28 /// good idea to read some pages ahead.)
33 /// Do not make the pages in this range available to the child
51 /// Enable Kernel Samepage Merging (KSM) for the pages in the
54 /// mergeable, looking for pages with identical content.
58 /// anonymous pages (see mmap(2)).
75 /// the specified address range; KSM unmerges whatever pages
83 /// Enable Transparent Huge Pages (THP) for pages in the range
85 /// Pages work only with private anonymous pages (see
87 /// as huge page candidates to replace them with huge pages.
88 /// The kernel will also allocate huge pages directly when the
129 /// Exclude from a core dump those pages in the range
147 /// Poison the pages in the range specified by addr and length
148 /// and handle subsequent references to those pages like a
163 /// pages in the range just as if manually reading from each
171 /// file content from disk; however, pages won't be dirtied and
173 /// evicting the pages from memory.
198 /// pages in the range just as if manually writing to each each
229 /// Indicates that the application would like the wired pages in this address range to be
230 /// zeroed out if the address range is deallocated without first unwiring the pages (i.e.
240 /// and effects on the mapped pages which are conceptually writes,
241 /// i.e. the change the observable contents of these pages which
246 /// does not keep any borrows of the mapping active while the mapped pages
249 /// [man_page]: https://man7.org/linux/man-pages/man2/madvise.2.html
261 /// subsequent accesses of pages in the range will succeed,
266 /// segments) or zero-fill-on-demand pages for anonymous
270 /// might not lead to immediate freeing of the pages in the
271 /// range. The kernel is free to delay freeing the pages
276 /// **MADV_DONTNEED** cannot be applied to locked pages, Huge TLB
277 /// pages, or VM_PFNMAP pages. (Pages marked with the kernel-
279 /// not managed by the virtual memory subsystem. Such pages
280 /// are typically created by device drivers that map the pages
286 /// mapped pages, i.e. borrowing the mapping when the pages
295 /// The application no longer requires the pages in the range
297 /// pages, but the freeing could be delayed until memory
298 /// pressure occurs. For each of the pages that has been
302 /// data (i.e., dirty, unwritten pages) will be lost when the
303 /// kernel frees the pages. However, subsequent writes to
304 /// pages in the range will succeed and then kernel cannot
305 /// free those dirtied pages, so that the caller can always
307 /// the kernel can free the pages at any time. Once pages in
309 /// on-demand pages upon subsequent page references.
312 /// anonymous pages (see mmap(2)). In Linux before version
313 /// 4.12, when freeing pages on a swapless system, the pages
320 /// mapped pages, i.e. borrowing the mapping while the pages
327 /// Free up a given range of pages and its associated backing
334 /// writable. This flag cannot be applied to locked pages,
335 /// Huge TLB pages, or VM_PFNMAP pages.
347 /// mapped pages, i.e. borrowing the mapping when the pages
354 /// Behaves like **MADV_FREE**, but the freed pages are accounted for in the RSS of the process.
359 /// mapped pages, i.e. borrowing the mapping while the pages
367 /// for the pages in the RSS of the process. Pages that have been freed will be replaced by
368 /// zero-filled pages on demand, other pages will be left as is.
373 /// mapped pages, i.e. borrowing the mapping while the pages
391 /// [`madvise(2)`]: https://man7.org/linux/man-pages/man2/madvise.2.html#VERSIONS
402 /// [`madvise(2)`]: https://man7.org/linux/man-pages/man2/madvise.2.html#VERSIONS