Messages in this thread | | | Date | Sun, 25 Jan 2004 17:36:57 -0500 | From | Randy Appleton <> | Subject | Re: Unneeded Code Found?? |
| |
Nick Piggin wrote:
> > > Randy Appleton wrote: > >> Nick Piggin wrote: > > > >> >>> Yes it gets used. >>> >>> I think its a lot more common with direct io and when you have lots of >>> processes. >> >> >> >> I'm not arguing, but how do you know this? I'm trying to convince >> myself that the code is used, and at least on my system >> a few days of general use, followed by heavy parallel compiles, >> doesn't use the code even once. >> >> I have not tested direct I/O. Otherwise it looks unused. >> > > Because I have seen it - I have instrumented it. > > Your usage patterns are pretty tame actually. I remember having 100 > processes > randomly reading from the same part of the disk was one of my test cases. > You need direct IO otherwise everything ends up in pagecache. > > I haven't seen workloads where it gets used a lot, but that doesn't > mean they > don't exist, and I've never seen the code cause any problems, so there > is no > need to make any trade offs by removing it. > O.K. That's convincing. Thanks for the time.
- 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/
| |