LSFMM: Page forking

Posted Apr 29, 2013 19:25 UTC (Mon) by luto (guest, #39314)
Parent article: LSFMM: Page forking

Whenever a page is written back, the kernel would have to find every page table entry referring to it and turn that entry read-only so that the copy-on-write semantics could be enforced.

This is, for better or for worse, already the case. (This is a perennial cause of pain for me.) x86 hardware, at least, has real dirty tracking, but the kernel doesn't use it; instead, it write protects every mapping of a page before writeback. The gory details are in a function called page_mkclean.