[PATCH i-g-t v6 3/3] tests/intel/xe_fault_injection: Do not assert for probe_guc_fail_* functions.

[K V P, Satyanarayana]

Hi.
> -----Original Message-----
> From: Bernatowicz, Marcin <marcin.bernatowicz at linux.intel.com>
> Sent: Thursday, April 17, 2025 5:25 PM
> To: K V P, Satyanarayana <satyanarayana.k.v.p at intel.com>; igt-
> dev at lists.freedesktop.org
> Cc: Wajdeczko, Michal <Michal.Wajdeczko at intel.com>; Dugast, Francois
> <francois.dugast at intel.com>
> Subject: Re: [PATCH i-g-t v6 3/3] tests/intel/xe_fault_injection: Do not assert
> for probe_guc_fail_* functions.
> 
> Hi,
> 
> On 4/15/2025 10:27 AM, Satyanarayana K V P wrote:
> > In the current implementation, test asserts if the captured error is not same
> as
> > injected error. It is possible that the error received is translated to other
> > error which can be returned to application and in some scenarios driver
> retries
> > in case of failure and so, no error might be captured.
> >
> > Considering above cases, added flags to control the assertion after injecting
> > error. Test does not assert for probe_guc_fail_* functions as driver some
> times
> > retries in case of failure. The main idea of injecting errors for these guc
> > functions is to check the robustness of the driver.
> >
> > Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p at intel.com>
> > Cc: Michał Wajdeczko <michal.wajdeczko at intel.com>
> > Cc: Francois Dugast <francois.dugast at intel.com>
> > Reviewed-by: Francois Dugast <francois.dugast at intel.com>
> > ---
> >   tests/intel/xe_fault_injection.c | 14 ++++++++++----
> >   1 file changed, 10 insertions(+), 4 deletions(-)
> >
> > diff --git a/tests/intel/xe_fault_injection.c b/tests/intel/xe_fault_injection.c
> > index 252209308..3c389a268 100644
> > --- a/tests/intel/xe_fault_injection.c
> > +++ b/tests/intel/xe_fault_injection.c
> > @@ -28,6 +28,11 @@
> >   #define BO_SIZE		(1024*1024)
> >   #define INJECT_ITERATIONS	100
> >
> > +enum assert_flags {
> > +	INJECT_ASSERT_ON_FAIL,
> > +	INJECT_NOASSERT_ON_FAIL,
> > +};
> > +
> >   int32_t inject_iters_raw;
> >   struct fault_injection_params {
> >   	/* @probability: Likelihood of failure injection, in percent. */
> > @@ -223,7 +228,7 @@ static void set_retval(const char function_name[],
> long long retval)
> >    * @xe_wopcm_init:			xe_wopcm_init
> >    */
> >   static void
> > -inject_fault_probe(int fd, char pci_slot[], const char function_name[])
> > +inject_fault_probe(int fd, int flags, char pci_slot[], const char
> function_name[])
> >   {
> >   	igt_info("Injecting error \"%s\" (%d) in function \"%s\"\n",
> >   		 strerror(-INJECT_ERRNO), INJECT_ERRNO, function_name);
> > @@ -232,7 +237,8 @@ inject_fault_probe(int fd, char pci_slot[], const char
> function_name[])
> >   	injection_list_add(function_name);
> >   	set_retval(function_name, INJECT_ERRNO);
> >   	xe_sysfs_driver_do(fd, pci_slot, XE_SYSFS_DRIVER_TRY_BIND);
> > -	igt_assert_eq(-errno, INJECT_ERRNO);
> > +	if (flags & INJECT_ASSERT_ON_FAIL)
> > +		igt_assert_eq(-errno, INJECT_ERRNO);
> >   	injection_list_remove(function_name);
> >   }
> 
> Would it make sense to return the probe result from
> inject_fault_probe(...) and decide to assert/not assert outside?
> 
Fixed and sent new version for review.
-Satya.
> >
> > @@ -263,7 +269,7 @@ static void probe_fail_guc(int fd, char pci_slot[],
> const char function_name[],
> >   	for (int i = iter_start; i < iter_end; i++) {
> >   		fault_params->space = i;
> >   		setup_injection_fault(fault_params);
> > -		inject_fault_probe(fd, pci_slot, function_name);
> > +		inject_fault_probe(fd, INJECT_NOASSERT_ON_FAIL, pci_slot,
> function_name);
> 
> 		inject_fault_probe(fd, pci_slot, function_name);
> 
> >   		igt_kmod_unbind("xe", pci_slot);
> >   	}
> >   }
> > @@ -575,7 +581,7 @@ igt_main_args("I:", NULL, help_str, opt_handler,
> NULL)
> >
> >   	for (const struct section *s = probe_fail_functions; s->name; s++)
> >   		igt_subtest_f("inject-fault-probe-function-%s", s->name)
> > -			inject_fault_probe(fd, pci_slot, s->name);
> > +			inject_fault_probe(fd, INJECT_ASSERT_ON_FAIL,
> pci_slot,	s->name);
> 
> 	igt_assert_eq(INJECT_ERRNO, inject_fault_probe(fd, pci_slot,
> function_name));
> 
> >
> >   	for (const struct section *s = guc_fail_functions; s->name; s++)
> >   		igt_subtest_f("probe-fail-guc-%s", s->name) {