Eric Lowe writes:
> On Wed, Jan 25, 2006 at 03:04:55AM +1100, Brendan Gregg wrote:
> |
> | The way a freshly written file (as opposed to a freshly mounted file)
> | changed the number of faults really threw me. Still haven't put my finger
> | on why, but I suspect it's a cache placement policy (maybe MPSS - I need
>
> This is probably free-behind kicking in in UFS, which by putting the pages
> at the head of the cachelists is causing them to be reused before they can
> be reactivated. I'm not a UFS expert (thank goodness?) but there is a way to
> turn off UFS' freebehind code to see if it is your culprit.
echo freebehind/W0t0 | mdb -kw
Back on with :
echo freebehind/W0t1 | mdb -kw
>
> (As an aside, the resulting behavior isn't UFS' fault but the VM's. It
> takes the "free" hint a little TOO seriously and essentially discards the
> pages, instead of allowing them to age. It's on the list of things to fix
> in the new VM system.)
There are 2 freebehind thresholds these days; one for the
head of cachelist, one for the tail and it's dynamically
adjusted based on freemem.
-r
>
> | to go test this on SPARC where I have trapstat. :)
>
> The only MPSS support for files is for text and data segments.
>
> - Eric
>
> | #!/usr/sbin/dtrace -s
> | /* todo: hat_memload_array */
> |
> | #pragma D option quiet
> | #pragma D option defaultargs
> |
> | inline int SCREEN = 21;
> |
> | dtrace:::BEGIN
> | {
> | lines = SCREEN + 1;
> | secs = $1 ? $1 : 1;
> | counts = $2 ? $2 : -1;
> | first = 1;
> | @hits = sum(0);
> | @miss = sum(0);
> | }
> |
> | profile:::tick-1sec
> | {
> | secs--;
> | }
> |
> | profile:::tick-1sec
> | /first || (secs == 0 && lines > SCREEN)/
> | {
> | printf("%10s %10s\n", "HITS", "MISSES");
> | first = 0;
> | lines = 0;
> | }
> |
> | fbt::segvn_fault:entry,
> | fbt::segvn_faulta:entry
> | {
> | self->segvn = 1;
> | }
> |
> | fbt::hat_memload:entry
> | /self->segvn && args[2]->p_vnode->v_path != NULL/
> | {
> | /*
> | * hat_memload is interesting (thanks Eric L.) as we use it
> | * to track the total number of page creates in segvn, so
> | * long as our context is segvn_fault. This value minus io
> | * events is used as the hit rate.
> | */
> | @path[execname, stringof(args[2]->p_vnode->v_path)] = count();
> | @hits = sum(1);
> | }
> |
> | io:::start
> | /self->segvn/
> | {
> | /* a segvn miss is an io event within a segvn_fault context */
> | @iobytes[execname, args[2]->fi_pathname,
> | args[0]->b_flags & B_READ ? "R" : "W"] =
> sum(args[0]->b_bcount);
> | @miss = sum(args[0]->b_bcount / `_pagesize);
> | @hits = sum(- (args[0]->b_bcount / `_pagesize));
> | }
> |
> | fbt::segvn_fault:return,
> | fbt::segvn_faulta:return
> | /self->segvn/
> | {
> | self->segvn = 0;
> | }
> |
> | profile:::tick-1sec
> | /secs == 0/
> | {
> | printa("[EMAIL PROTECTED] ", @hits);
> | printa("[EMAIL PROTECTED]", @miss);
> | trunc(@hits);
> | trunc(@miss);
> | @hits = sum(0);
> | @miss = sum(0);
> | secs = $1 ? $1 : 1;
> | lines++;
> | counts--;
> | }
> |
> | dtrace:::END
> | {
> | printf("hat_memload\n-----------\n");
> | printf("%-16s %-50s %8s\n", "CMD", "PATH", "COUNT");
> | printa("%-16s %-50s [EMAIL PROTECTED]", @path);
> |
> | printf("\nio:::start\n----------\n");
> | printf("%-16s %32s %3s %10s\n", "CMD", "FILE", "DIR", "BYTES");
> | printa("%-16s %32s %3s [EMAIL PROTECTED]", @iobytes);
> | }
>
>
> --
> Eric Lowe Solaris Kernel Development Austin, Texas
> Sun Microsystems. We make the net work.
> x64155/+1(512)401-1155
> _______________________________________________
> perf-discuss mailing list
> [email protected]
_______________________________________________
perf-discuss mailing list
[email protected]
