Hi Ted,
I really like the idea of gathering data about time spent waiting for a disk,
but where do I find this data?
Thank you,
Navan
On Apr 4, 2013, at 7:50 PM, Ted Unangst <t...@tedunangst.com> wrote:
> This is a stupid simple diff to see just how much time you spend
> waiting for your disk. If you're into measuring that kind of thing.
>
> Index: vfs_bio.c
> ===================================================================
> RCS file: /cvs/src/sys/kern/vfs_bio.c,v
> retrieving revision 1.146
> diff -u -p -r1.146 vfs_bio.c
> --- vfs_bio.c 17 Feb 2013 17:39:29 -0000 1.146
> +++ vfs_bio.c 5 Apr 2013 01:34:03 -0000
> @@ -1153,6 +1153,8 @@ buf_daemon(struct proc *p)
> }
> }
>
> +uint64_t biowaittime;
> +
> /*
> * Wait for operations on the buffer to complete.
> * When they do, extract and return the I/O's error value.
> @@ -1161,12 +1163,18 @@ int
> biowait(struct buf *bp)
> {
> int s;
> + struct timespec before, after, diff;
>
> KASSERT(!(bp->b_flags & B_ASYNC));
>
> s = splbio();
> + nanotime(&before);
> while (!ISSET(bp->b_flags, B_DONE))
> tsleep(bp, PRIBIO + 1, "biowait", 0);
> + nanotime(&after);
> + timespecsub(&after, &before, &diff);
> + biowaittime += diff.tv_nsec;
> + biowaittime += diff.tv_sec * 1000000000L;
> splx(s);
>
> /* check for interruption of I/O (e.g. via NFS), then errors. */
>
