Messages in this thread | | | Date | Fri, 3 Mar 2006 12:32:22 -0800 | From | Andrew Morton <> | Subject | Re: [PATCH 0/4] map multiple blocks in get_block() and mpage_readpages() |
| |
Badari Pulavarty <pbadari@us.ibm.com> wrote: > > On Wed, 2006-03-01 at 17:52 -0800, Andrew Morton wrote: > > Badari Pulavarty <pbadari@us.ibm.com> wrote: > > > > > > I noticed decent improvements (reduced sys time) on JFS, XFS and ext3. > > > (on simple "dd" read tests). > > > > > > (rc3.mm1) (rc3.mm1 + patches) > > > real 0m18.814s 0m18.482s > > > user 0m0.000s 0m0.004s > > > sys 0m3.240s 0m2.912s > > > > With which filesystem? XFS and JFS implement a larger-than-b_size > > ->get_block, but ext3 doesn't. I'd expect ext3 system time to increase a > > bit, if anything? > > These numbers are on JFS. With the current ext3 (mainline) - I did > find not-really-noticible increase in sys time (due to code overhead). > > I tested on ext3 with Mingming's ext3 getblocks() support in -mm also, > which showed reduction in sys time. >
OK, no surprises there. Things will improve when someone gets around to doing multi-block get_block for ext3. - 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/
| |