Okay, here is the background.. writepages() interfaces gets used to flush the entire file. ext3 doesn't have writepages() interface. Without the interface, the generic writepage interface calls writepage() in a loop. Since we are operating on one page at a time, we end up submiting one block of IO at a time. Hopefully, IO schedulers do all the merging needed to submit a large IO.
My patch provides, writepages() interface for ext3 writeback mode (ordered mode is little complicated). writepages() could batch up larger IO chunks (since it operates on multiple pages/blocks). In this case, IO scheduler does less work. Thanks, Badari On Mon, 2005-02-14 at 08:50, Thiago Rondon wrote: > Can you explain more what do you wrote ? > > Thanks in advanced! > -Thiago Rondon > > > On Thu, 10 Feb 2005 17:48:55 -0800 (PST), Badari Pulavarty > <[EMAIL PROTECTED]> wrote: > > Hi, > > > > Here is my first cut at adding writepages() support for > > ext3 writeback mode. > > > > I have not done any performance analysis on the patch, > > so try it at your own risk. > > > > Please let me know, if I am completely off or its a > > stupid idea. > > > > Thanks, > > Badari > > > > > > > - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
