Thanks for the clarification. If the buffer and page cache are unified, what is lru ?. Is it part of the cpu cache, or part of the buffer ?.
Shameem ----- Original Message ---- > From: Mulyadi Santosa <[email protected]> > To: Shameem Ahamed <[email protected]> > Cc: [email protected] > Sent: Monday, September 28, 2009 9:00:28 PM > Subject: Re: buffer and page retrievel > > Hi... > > this is also an area that confused me from long time ago, but > hopefully now I got better. Here's my wording (let's assume all page > mentioned here are file backed ones): > > if page fault happens, page cache will be checked to determined which > page hold the corresponding disk block. If one is found, this page > will be mapped into the related faulting process' VMA > > otherwise, if the data is not found in page cache, disk read will be > engaged. Most likely it will be done in async style. Once it is loaded > into page cache, this page will also be mapped into faulting process' > VMA. > > Notice that I use the term "page cache" here, since in 2.6 buffer and > page cache are unified and called as page cache. > > > -- > regards, > > Mulyadi Santosa > Freelance Linux trainer > blog: the-hydra.blogspot.com -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to [email protected] Please read the FAQ at http://kernelnewbies.org/FAQ
