zfs code - Aug 2006 - Bug in arc.c revision 1.15

Hi,

I''ve received a bug report in zfs-fuse that doesn''t seem to be
specific to the
port.
The problem was introduced in arc.c revision 1.15. There''s a new static
variable arc_min_prefetch_lifespan that initially is 1 second but it is 
converted to clock ticks in arc_init():

	/* Convert seconds to clock ticks */
	arc_min_prefetch_lifespan *= hz;

However, arc_init() is called more than once in ztest, so 
arc_min_prefetch_lifespan keeps getting multiplied by hz (119).

This results in an almost infinite loop in arc_flush(), since arc_evict() 
skips prefetch buffers with a lifespan less than (in my particular test-case) 
119*119*119 seconds..

Ricardo,

We have also discovered this bug:

6453172 ztest turns into a sloth due to massive arc_min_prefetch_lifespan

I believe Neil has a fix in the works.

Thanks,
George

Ricardo Correia wrote:
> Hi,
> 
> I''ve received a bug report in zfs-fuse that doesn''t seem
to be specific to the
> port.
> The problem was introduced in arc.c revision 1.15. There''s a new
static
> variable arc_min_prefetch_lifespan that initially is 1 second but it is 
> converted to clock ticks in arc_init():
> 
> 	/* Convert seconds to clock ticks */
> 	arc_min_prefetch_lifespan *= hz;
> 
> However, arc_init() is called more than once in ztest, so 
> arc_min_prefetch_lifespan keeps getting multiplied by hz (119).
> 
> This results in an almost infinite loop in arc_flush(), since arc_evict() 
> skips prefetch buffers with a lifespan less than (in my particular
test-case)
> 119*119*119 seconds..
> _______________________________________________
> zfs-code mailing list
> zfs-code at opensolaris.org
> http://opensolaris.org/mailman/listinfo/zfs-code