![]() | |||||||||||||
Messages in this thread |
On Tue, 22 Jan 2002, James Washer wrote: > > In __free_pages_ok(), if current->flags & PF_FREE_PAGES, the code adds the > pages being freed to current->local_pages using the following block of code: > ...... > Unless I am misreading this, that first if statement in going to limit > us to at most one block of pages no matter how many times we call > __free_pages_ok() > > However, in balance_classzone(), the code seems to expect that multiple > blocks of pages may be 'cached' in local_pages. > ...... > So, am I misreading this.. or is one of these wrong? If I were to hazard a > guess, __free_pages_ok() should be 'fixed'. I think you read it right. I believe it comes from Linus integrating Andrea's VM, but making changes to it on the fly. So one end has a capability that the other end now prohibits. In Andrea's -aa tree it isn't quite the same; and I don't think it's worth tidying up the mainline here, while there's the possibility that more of Andrea's version is to be merged in. Hugh - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2005-03-22 12:15 [W:0.129 / U:0.010 seconds] ©2003-2008 Jasper Spaans | |||||||||||||