Re: [PATCH v2 6/6] selftests/sgx: Add a bpftrace script for tracking allocation errors

[Jarkko Sakkinen]

On Wed, Aug 31, 2022 at 01:09:21PM -0700, Reinette Chatre wrote:
> Hi Jarkko,
> 
> On 8/31/2022 10:38 AM, Jarkko Sakkinen wrote:
> > Signed-off-by: Jarkko Sakkinen <jarkko@xxxxxxxxxx>
> > ---
> > v2:
> > * Added comments.
> > ---
> >  tools/testing/selftests/sgx/alloc-error.bt | 9 +++++++++
> >  1 file changed, 9 insertions(+)
> >  create mode 100644 tools/testing/selftests/sgx/alloc-error.bt
> > 
> > diff --git a/tools/testing/selftests/sgx/alloc-error.bt b/tools/testing/selftests/sgx/alloc-error.bt
> > new file mode 100644
> > index 000000000000..0cc8b2e41852
> > --- /dev/null
> > +++ b/tools/testing/selftests/sgx/alloc-error.bt
> > @@ -0,0 +1,9 @@
> > +/* EPC allocation */
> > +kr:sgx_alloc_epc_page /(uint64)retval >= (uint64)(-4095)/ {
> > +		 printf("sgx_alloc_epc_page: retval=%d\n", (int64)retval);
> > +}
> > +
> > +/* kzalloc for struct sgx_encl_page */
> > +kr:sgx_encl_page_alloc /(uint64)retval >= (uint64)(-4095)/ {
> > +		 printf("sgx_encl_page_alloc: retval=%d\n", (int64)retval);
> > +}
> 
> I did see you response in [1]. I continue to find this
> addition very cryptic in that it assumes global familiarity
> with BPF scripting which I do not think can be assumed. I
> still think this script can benefit from a comment to describe
> what the script does and how to use it.


I think I follow Dave's advice to use perf here. I'll
make a header which explains what the script does.

BR, Jarkko