Messages in this thread | | | Date | Wed, 22 Feb 2006 14:50:00 -0700 | From | "Jeff V. Merkey" <> | Subject | Re: udevd is killing file write performance. |
| |
Andrew Morton wrote:
>Robin Holt <holt@sgi.com> wrote: > > >>Let me reiterate, I know _VERY_ little about filesystems. Can the >> dentry->d_lock be changed to a read/write lock? >> >> > >Well, it could, but I suspect that won't help - the hold times in there >will be very short so the problem is more likely acquisition frequency. > >However it's a bit strange that this function is the bottleneck. If their >workload is doing large numbers of reads or writes from large numbers of >processes against the same file then they should be hitting heavy >contention on other locks, such as i_sem and/or tree_lock and/or lru_lock >and others. > >Can you tell us more about the kernel-visible behaviour of this app? > >
I have also seen this problem, and it's hard to reproduce. What you will see is udev getting spawned multiple times as reported by top. I have found its related to intermittent failures of the hard drive and the hotpluger for some reason invoking udev multiple times in response to this. I saw it on a Compaq laptop right before my hard drive croaked, and it seems BIOS specific as well, since I have never seen it or been able to reproduce it reliably.
Jeff
>- >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/ > > >
- 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/
| |