cvs commit: src/sys/vm vm_page.c
silby at silby.com
Tue Nov 8 06:43:26 GMT 2005
On Tue, 8 Nov 2005, Alan Cox wrote:
> If a physical page is mapped by two or more virtual addresses, transmitted
> by the zero-copy sockets method, and written to before the transmission
> completes, we need to destroy all of the existing mappings to the page,
> not just the one that we fault on. Otherwise, the mappings will no longer
> be to the same page and changes made through one of the mappings will not
> be visible through the others.
These kinds of changes could REALLY use regression tests, if possible.
Mike "Silby" Silbersack
More information about the cvs-src