[DTrace-devel] [PATCH] test: skip tst.alignring.d

Kris Van Hees kris.van.hees at oracle.com
Thu Apr 21 15:19:34 UTC 2022


If the test is described as "meaningless test producing excessive output",
why is this patch not simply removing it?  The main point behind the test
(originally, in Solaris) seems to have been to test that when the ring
buffer causes a wrap-around condition, the data is still being written to
properly aligned locations.

We actually *do* use a ring buffer, so the comment below is actually not
accurate, but we use kernel mechanisms to write to it (because it is
provided by the perf subsystem).  As such, any issues with alignment of
data written to it are generally an issue with that implementation and we
do not need to provide testing for that.

Either way, this test is not relevant to the current implementation and
*if* we want to add some testing of the underlying kernel implementation,
we should add a specific test for that rather than try to get this one to
serve that purpose.

On Mon, Jan 10, 2022 at 09:12:22PM -0500, eugene.loh--- via DTrace-devel wrote:
> From: Eugene Loh <eugene.loh at oracle.com>
> 
> Signed-off-by: Eugene Loh <eugene.loh at oracle.com>
> ---
>  test/unittest/buffering/tst.alignring.d | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/test/unittest/buffering/tst.alignring.d b/test/unittest/buffering/tst.alignring.d
> index a2f9d549..22786fec 100644
> --- a/test/unittest/buffering/tst.alignring.d
> +++ b/test/unittest/buffering/tst.alignring.d
> @@ -18,6 +18,12 @@
>  /* Incurs swapping, so
>   *  @@timeout: 15
>   */
> +/* This test
> + *   - produces excessive output (writing 100Ks of lines to runtest.log)
> + *   - tests nothing (the output is not checked)
> + *   - targets ring buffering, which we do not support
> + */
> +/* @@skip: meaningless test producing excessive output */
>  
>  #pragma D option bufpolicy=ring
>  #pragma D option bufsize=4k
> -- 
> 2.18.4
> 
> 
> _______________________________________________
> DTrace-devel mailing list
> DTrace-devel at oss.oracle.com
> https://oss.oracle.com/mailman/listinfo/dtrace-devel



More information about the DTrace-devel mailing list