[igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Chris Wilson]

If we inject a reset into the target context, there is a risk that the
register state is never saved back to memory. The exact interaction
between reset, the context image and the precise timing of our execution
are not well defined. Since we cannot ensure that the context image
remains valid, force a context switch prior to the reset.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
---
 tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
index d8109aa0..4968e367 100644
--- a/tests/gem_ctx_isolation.c
+++ b/tests/gem_ctx_isolation.c
@@ -522,6 +522,32 @@ static void isolation(int fd,
 #define S4 (4 << 8)
 #define SLEEP_MASK (0xf << 8)
 
+static void inject_reset_context(int fd, unsigned int engine)
+{
+	igt_spin_t *spin;
+	uint32_t ctx;
+
+	/*
+	 * Force a context switch before triggering the reset, or else
+	 * we risk corrupting the target context and we can't blame the
+	 * HW for screwing up if the context was already broken.
+	 */
+
+	ctx = gem_context_create(fd);
+	if (gem_can_store_dword(fd, engine)) {
+		spin = __igt_spin_batch_new_poll(fd, ctx, engine);
+		igt_spin_busywait_until_running(spin);
+	} else {
+		spin = __igt_spin_batch_new(fd, ctx, engine, 0);
+		usleep(1000); /* better than nothing */
+	}
+
+	igt_force_gpu_reset(fd);
+
+	igt_spin_batch_free(fd, spin);
+	gem_context_destroy(fd, ctx);
+}
+
 static void preservation(int fd,
 			 const struct intel_execution_engine2 *e,
 			 unsigned int flags)
@@ -558,7 +584,7 @@ static void preservation(int fd,
 	igt_spin_batch_free(fd, spin);
 
 	if (flags & RESET)
-		igt_force_gpu_reset(fd);
+		inject_reset_context(fd, engine);
 
 	switch (flags & SLEEP_MASK) {
 	case NOSLEEP:
-- 
2.16.3

_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Chris Wilson]

Quoting Chris Wilson (2018-03-27 14:48:43)
> If we inject a reset into the target context, there is a risk that the
> register state is never saved back to memory. The exact interaction
> between reset, the context image and the precise timing of our execution
> are not well defined. Since we cannot ensure that the context image
> remains valid, force a context switch prior to the reset.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
>  tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>  1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
> index d8109aa0..4968e367 100644
> --- a/tests/gem_ctx_isolation.c
> +++ b/tests/gem_ctx_isolation.c
> @@ -522,6 +522,32 @@ static void isolation(int fd,
>  #define S4 (4 << 8)
>  #define SLEEP_MASK (0xf << 8)
>  
> +static void inject_reset_context(int fd, unsigned int engine)
> +{
> +       igt_spin_t *spin;
> +       uint32_t ctx;
> +
> +       /*
> +        * Force a context switch before triggering the reset, or else
> +        * we risk corrupting the target context and we can't blame the
> +        * HW for screwing up if the context was already broken.
> +        */
> +
> +       ctx = gem_context_create(fd);
> +       if (gem_can_store_dword(fd, engine)) {
> +               spin = __igt_spin_batch_new_poll(fd, ctx, engine);
> +               igt_spin_busywait_until_running(spin);
> +       } else {
> +               spin = __igt_spin_batch_new(fd, ctx, engine, 0);
> +               usleep(1000); /* better than nothing */
> +       }

Tvrtko, maybe we want igt_spin_batch_run()? Not sure though, so far we
have an example where you need precise control and a couple of examples
where we just want a running spinner.
-Chris
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Chris Wilson]

Quoting Chris Wilson (2018-03-27 14:52:20)
> Quoting Chris Wilson (2018-03-27 14:48:43)
> > If we inject a reset into the target context, there is a risk that the
> > register state is never saved back to memory. The exact interaction
> > between reset, the context image and the precise timing of our execution
> > are not well defined. Since we cannot ensure that the context image
> > remains valid, force a context switch prior to the reset.
> > 
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> > ---
> >  tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
> >  1 file changed, 27 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
> > index d8109aa0..4968e367 100644
> > --- a/tests/gem_ctx_isolation.c
> > +++ b/tests/gem_ctx_isolation.c
> > @@ -522,6 +522,32 @@ static void isolation(int fd,
> >  #define S4 (4 << 8)
> >  #define SLEEP_MASK (0xf << 8)
> >  
> > +static void inject_reset_context(int fd, unsigned int engine)
> > +{
> > +       igt_spin_t *spin;
> > +       uint32_t ctx;
> > +
> > +       /*
> > +        * Force a context switch before triggering the reset, or else
> > +        * we risk corrupting the target context and we can't blame the
> > +        * HW for screwing up if the context was already broken.
> > +        */
> > +
> > +       ctx = gem_context_create(fd);
> > +       if (gem_can_store_dword(fd, engine)) {
> > +               spin = __igt_spin_batch_new_poll(fd, ctx, engine);
> > +               igt_spin_busywait_until_running(spin);
> > +       } else {
> > +               spin = __igt_spin_batch_new(fd, ctx, engine, 0);
> > +               usleep(1000); /* better than nothing */
> > +       }
> 
> Tvrtko, maybe we want igt_spin_batch_run()? Not sure though, so far we
> have an example where you need precise control and a couple of examples
> where we just want a running spinner.

Also this whole function might want to make its way into lib/i915 as the
basis of all hang/reset injection. Some improvement required to set
context parameters (e.g. disabling error state capturing) and skipping
if no reset is available or disabled (although that's the job for the
fixture, so may not be required for the library function). I'm sure
someone will catch me reusing this chunk over and over again...
-Chris
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

On 27/03/2018 14:48, Chris Wilson wrote:
> If we inject a reset into the target context, there is a risk that the
> register state is never saved back to memory. The exact interaction
> between reset, the context image and the precise timing of our execution
> are not well defined. Since we cannot ensure that the context image
> remains valid, force a context switch prior to the reset.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>   1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
> index d8109aa0..4968e367 100644
> --- a/tests/gem_ctx_isolation.c
> +++ b/tests/gem_ctx_isolation.c
> @@ -522,6 +522,32 @@ static void isolation(int fd,
>   #define S4 (4 << 8)
>   #define SLEEP_MASK (0xf << 8)
>   
> +static void inject_reset_context(int fd, unsigned int engine)
> +{
> +	igt_spin_t *spin;
> +	uint32_t ctx;
> +
> +	/*
> +	 * Force a context switch before triggering the reset, or else
> +	 * we risk corrupting the target context and we can't blame the
> +	 * HW for screwing up if the context was already broken.
> +	 */
> +
> +	ctx = gem_context_create(fd);
> +	if (gem_can_store_dword(fd, engine)) {
> +		spin = __igt_spin_batch_new_poll(fd, ctx, engine);
> +		igt_spin_busywait_until_running(spin);
> +	} else {
> +		spin = __igt_spin_batch_new(fd, ctx, engine, 0);
> +		usleep(1000); /* better than nothing */
> +	}
> +
> +	igt_force_gpu_reset(fd);
> +
> +	igt_spin_batch_free(fd, spin);
> +	gem_context_destroy(fd, ctx);
> +}
> +
>   static void preservation(int fd,
>   			 const struct intel_execution_engine2 *e,
>   			 unsigned int flags)
> @@ -558,7 +584,7 @@ static void preservation(int fd,
>   	igt_spin_batch_free(fd, spin);
>   
>   	if (flags & RESET)
> -		igt_force_gpu_reset(fd);
> +		inject_reset_context(fd, engine);
>   
>   	switch (flags & SLEEP_MASK) {
>   	case NOSLEEP:
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

On 27/03/2018 14:59, Chris Wilson wrote:
> Quoting Chris Wilson (2018-03-27 14:52:20)
>> Quoting Chris Wilson (2018-03-27 14:48:43)
>>> If we inject a reset into the target context, there is a risk that the
>>> register state is never saved back to memory. The exact interaction
>>> between reset, the context image and the precise timing of our execution
>>> are not well defined. Since we cannot ensure that the context image
>>> remains valid, force a context switch prior to the reset.
>>>
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>>> ---
>>>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>>>   1 file changed, 27 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
>>> index d8109aa0..4968e367 100644
>>> --- a/tests/gem_ctx_isolation.c
>>> +++ b/tests/gem_ctx_isolation.c
>>> @@ -522,6 +522,32 @@ static void isolation(int fd,
>>>   #define S4 (4 << 8)
>>>   #define SLEEP_MASK (0xf << 8)
>>>   
>>> +static void inject_reset_context(int fd, unsigned int engine)
>>> +{
>>> +       igt_spin_t *spin;
>>> +       uint32_t ctx;
>>> +
>>> +       /*
>>> +        * Force a context switch before triggering the reset, or else
>>> +        * we risk corrupting the target context and we can't blame the
>>> +        * HW for screwing up if the context was already broken.
>>> +        */
>>> +
>>> +       ctx = gem_context_create(fd);
>>> +       if (gem_can_store_dword(fd, engine)) {
>>> +               spin = __igt_spin_batch_new_poll(fd, ctx, engine);
>>> +               igt_spin_busywait_until_running(spin);
>>> +       } else {
>>> +               spin = __igt_spin_batch_new(fd, ctx, engine, 0);
>>> +               usleep(1000); /* better than nothing */
>>> +       }
>>
>> Tvrtko, maybe we want igt_spin_batch_run()? Not sure though, so far we
>> have an example where you need precise control and a couple of examples
>> where we just want a running spinner.

I wasn't sure it is in good taste to put a thing with that usleep in 
lib/, since it incurs a delay to unsuspecting callers. And I couldn't 
decide how to handle it better - skip from lib/? Not so great. Return 
value and make callers handle it - even more cumbersome.

It only affects old gens, but do we want tests there suddenly take much 
longer because someone spotted a cool new API and went to use it?

But it is a third user now so not great to copy paste it all around. I 
don't know. Make the lib functions skip where unsupported and add double 
underscore prefix flavour which sleeps where not supported?

> Also this whole function might want to make its way into lib/i915 as the
> basis of all hang/reset injection. Some improvement required to set
> context parameters (e.g. disabling error state capturing) and skipping
> if no reset is available or disabled (although that's the job for the
> fixture, so may not be required for the library function). I'm sure
> someone will catch me reusing this chunk over and over again...

Sounds OK to me.

Regards,

Tvrtko


_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

[resend for typo in cc]

On 27/03/2018 14:48, Chris Wilson wrote:
> If we inject a reset into the target context, there is a risk that the
> register state is never saved back to memory. The exact interaction
> between reset, the context image and the precise timing of our execution
> are not well defined. Since we cannot ensure that the context image
> remains valid, force a context switch prior to the reset.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>   1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
> index d8109aa0..4968e367 100644
> --- a/tests/gem_ctx_isolation.c
> +++ b/tests/gem_ctx_isolation.c
> @@ -522,6 +522,32 @@ static void isolation(int fd,
>   #define S4 (4 << 8)
>   #define SLEEP_MASK (0xf << 8)
>   
> +static void inject_reset_context(int fd, unsigned int engine)
> +{
> +	igt_spin_t *spin;
> +	uint32_t ctx;
> +
> +	/*
> +	 * Force a context switch before triggering the reset, or else
> +	 * we risk corrupting the target context and we can't blame the
> +	 * HW for screwing up if the context was already broken.
> +	 */
> +
> +	ctx = gem_context_create(fd);
> +	if (gem_can_store_dword(fd, engine)) {
> +		spin = __igt_spin_batch_new_poll(fd, ctx, engine);
> +		igt_spin_busywait_until_running(spin);
> +	} else {
> +		spin = __igt_spin_batch_new(fd, ctx, engine, 0);
> +		usleep(1000); /* better than nothing */
> +	}
> +
> +	igt_force_gpu_reset(fd);
> +
> +	igt_spin_batch_free(fd, spin);
> +	gem_context_destroy(fd, ctx);
> +}
> +
>   static void preservation(int fd,
>   			 const struct intel_execution_engine2 *e,
>   			 unsigned int flags)
> @@ -558,7 +584,7 @@ static void preservation(int fd,
>   	igt_spin_batch_free(fd, spin);
>   
>   	if (flags & RESET)
> -		igt_force_gpu_reset(fd);
> +		inject_reset_context(fd, engine);
>   
>   	switch (flags & SLEEP_MASK) {
>   	case NOSLEEP:
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

[resend for typo in cc]

On 27/03/2018 14:48, Chris Wilson wrote:
> If we inject a reset into the target context, there is a risk that the
> register state is never saved back to memory. The exact interaction
> between reset, the context image and the precise timing of our execution
> are not well defined. Since we cannot ensure that the context image
> remains valid, force a context switch prior to the reset.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>   1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
> index d8109aa0..4968e367 100644
> --- a/tests/gem_ctx_isolation.c
> +++ b/tests/gem_ctx_isolation.c
> @@ -522,6 +522,32 @@ static void isolation(int fd,
>   #define S4 (4 << 8)
>   #define SLEEP_MASK (0xf << 8)
>   
> +static void inject_reset_context(int fd, unsigned int engine)
> +{
> +	igt_spin_t *spin;
> +	uint32_t ctx;
> +
> +	/*
> +	 * Force a context switch before triggering the reset, or else
> +	 * we risk corrupting the target context and we can't blame the
> +	 * HW for screwing up if the context was already broken.
> +	 */
> +
> +	ctx = gem_context_create(fd);
> +	if (gem_can_store_dword(fd, engine)) {
> +		spin = __igt_spin_batch_new_poll(fd, ctx, engine);
> +		igt_spin_busywait_until_running(spin);
> +	} else {
> +		spin = __igt_spin_batch_new(fd, ctx, engine, 0);
> +		usleep(1000); /* better than nothing */
> +	}
> +
> +	igt_force_gpu_reset(fd);
> +
> +	igt_spin_batch_free(fd, spin);
> +	gem_context_destroy(fd, ctx);
> +}
> +
>   static void preservation(int fd,
>   			 const struct intel_execution_engine2 *e,
>   			 unsigned int flags)
> @@ -558,7 +584,7 @@ static void preservation(int fd,
>   	igt_spin_batch_free(fd, spin);
>   
>   	if (flags & RESET)
> -		igt_force_gpu_reset(fd);
> +		inject_reset_context(fd, engine);
>   
>   	switch (flags & SLEEP_MASK) {
>   	case NOSLEEP:
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

[resend for typo in cc]

On 27/03/2018 14:59, Chris Wilson wrote:
> Quoting Chris Wilson (2018-03-27 14:52:20)
>> Quoting Chris Wilson (2018-03-27 14:48:43)
>>> If we inject a reset into the target context, there is a risk that the
>>> register state is never saved back to memory. The exact interaction
>>> between reset, the context image and the precise timing of our execution
>>> are not well defined. Since we cannot ensure that the context image
>>> remains valid, force a context switch prior to the reset.
>>>
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>>> ---
>>>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>>>   1 file changed, 27 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
>>> index d8109aa0..4968e367 100644
>>> --- a/tests/gem_ctx_isolation.c
>>> +++ b/tests/gem_ctx_isolation.c
>>> @@ -522,6 +522,32 @@ static void isolation(int fd,
>>>   #define S4 (4 << 8)
>>>   #define SLEEP_MASK (0xf << 8)
>>>   
>>> +static void inject_reset_context(int fd, unsigned int engine)
>>> +{
>>> +       igt_spin_t *spin;
>>> +       uint32_t ctx;
>>> +
>>> +       /*
>>> +        * Force a context switch before triggering the reset, or else
>>> +        * we risk corrupting the target context and we can't blame the
>>> +        * HW for screwing up if the context was already broken.
>>> +        */
>>> +
>>> +       ctx = gem_context_create(fd);
>>> +       if (gem_can_store_dword(fd, engine)) {
>>> +               spin = __igt_spin_batch_new_poll(fd, ctx, engine);
>>> +               igt_spin_busywait_until_running(spin);
>>> +       } else {
>>> +               spin = __igt_spin_batch_new(fd, ctx, engine, 0);
>>> +               usleep(1000); /* better than nothing */
>>> +       }
>>
>> Tvrtko, maybe we want igt_spin_batch_run()? Not sure though, so far we
>> have an example where you need precise control and a couple of examples
>> where we just want a running spinner.

I wasn't sure it is in good taste to put a thing with that usleep in 
lib/, since it incurs a delay to unsuspecting callers. And I couldn't 
decide how to handle it better - skip from lib/? Not so great. Return 
value and make callers handle it - even more cumbersome.

It only affects old gens, but do we want tests there suddenly take much 
longer because someone spotted a cool new API and went to use it?

But it is a third user now so not great to copy paste it all around. I 
don't know. Make the lib functions skip where unsupported and add double 
underscore prefix flavour which sleeps where not supported?

> Also this whole function might want to make its way into lib/i915 as the
> basis of all hang/reset injection. Some improvement required to set
> context parameters (e.g. disabling error state capturing) and skipping
> if no reset is available or disabled (although that's the job for the
> fixture, so may not be required for the library function). I'm sure
> someone will catch me reusing this chunk over and over again...

Sounds OK to me.

Regards,

Tvrtko


_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [Intel-gfx] [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Tvrtko Ursulin]

[resend for typo in cc]

On 27/03/2018 14:59, Chris Wilson wrote:
> Quoting Chris Wilson (2018-03-27 14:52:20)
>> Quoting Chris Wilson (2018-03-27 14:48:43)
>>> If we inject a reset into the target context, there is a risk that the
>>> register state is never saved back to memory. The exact interaction
>>> between reset, the context image and the precise timing of our execution
>>> are not well defined. Since we cannot ensure that the context image
>>> remains valid, force a context switch prior to the reset.
>>>
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105270
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105457
>>> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105545
>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>>> ---
>>>   tests/gem_ctx_isolation.c | 28 +++++++++++++++++++++++++++-
>>>   1 file changed, 27 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/tests/gem_ctx_isolation.c b/tests/gem_ctx_isolation.c
>>> index d8109aa0..4968e367 100644
>>> --- a/tests/gem_ctx_isolation.c
>>> +++ b/tests/gem_ctx_isolation.c
>>> @@ -522,6 +522,32 @@ static void isolation(int fd,
>>>   #define S4 (4 << 8)
>>>   #define SLEEP_MASK (0xf << 8)
>>>   
>>> +static void inject_reset_context(int fd, unsigned int engine)
>>> +{
>>> +       igt_spin_t *spin;
>>> +       uint32_t ctx;
>>> +
>>> +       /*
>>> +        * Force a context switch before triggering the reset, or else
>>> +        * we risk corrupting the target context and we can't blame the
>>> +        * HW for screwing up if the context was already broken.
>>> +        */
>>> +
>>> +       ctx = gem_context_create(fd);
>>> +       if (gem_can_store_dword(fd, engine)) {
>>> +               spin = __igt_spin_batch_new_poll(fd, ctx, engine);
>>> +               igt_spin_busywait_until_running(spin);
>>> +       } else {
>>> +               spin = __igt_spin_batch_new(fd, ctx, engine, 0);
>>> +               usleep(1000); /* better than nothing */
>>> +       }
>>
>> Tvrtko, maybe we want igt_spin_batch_run()? Not sure though, so far we
>> have an example where you need precise control and a couple of examples
>> where we just want a running spinner.

I wasn't sure it is in good taste to put a thing with that usleep in 
lib/, since it incurs a delay to unsuspecting callers. And I couldn't 
decide how to handle it better - skip from lib/? Not so great. Return 
value and make callers handle it - even more cumbersome.

It only affects old gens, but do we want tests there suddenly take much 
longer because someone spotted a cool new API and went to use it?

But it is a third user now so not great to copy paste it all around. I 
don't know. Make the lib functions skip where unsupported and add double 
underscore prefix flavour which sleeps where not supported?

> Also this whole function might want to make its way into lib/i915 as the
> basis of all hang/reset injection. Some improvement required to set
> context parameters (e.g. disabling error state capturing) and skipping
> if no reset is available or disabled (although that's the job for the
> fixture, so may not be required for the library function). I'm sure
> someone will catch me reusing this chunk over and over again...

Sounds OK to me.

Regards,

Tvrtko


_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Chris Wilson]

Quoting Tvrtko Ursulin (2018-03-27 16:41:14)
> 
> [resend for typo in cc]

The small advantage in redundancy, I guess.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [igt-dev] [PATCH igt] igt/gem_ctx_isolation: Reset a scratch context

[Chris Wilson]

Quoting Tvrtko Ursulin (2018-03-27 16:41:14)
> 
> [resend for typo in cc]

The small advantage in redundancy, I guess.
-Chris
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] ✓ Fi.CI.BAT: success for igt/gem_ctx_isolation: Reset a scratch context

[Patchwork]

== Series Details ==

Series: igt/gem_ctx_isolation: Reset a scratch context
URL   : https://patchwork.freedesktop.org/series/40740/
State : success

== Summary ==

IGT patchset tested on top of latest successful build
a9741da52ad1963f7632ef1e852cbe1c3bcc601e tests/perf_pmu: Improve accuracy by waiting on spinner to start

with latest DRM-Tip kernel build CI_DRM_3989
ff7820832182 drm-tip: 2018y-03m-27d-14h-31m-00s UTC integration manifest

No testlist changes.

---- Known issues:

Test kms_flip:
        Subgroup basic-flip-vs-wf_vblank:
                pass       -> FAIL       (fi-cfl-s3) fdo#100368
Test kms_pipe_crc_basic:
        Subgroup suspend-read-crc-pipe-b:
                dmesg-warn -> PASS       (fi-cnl-y3) fdo#104951

fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
fdo#104951 https://bugs.freedesktop.org/show_bug.cgi?id=104951

fi-bdw-5557u     total:285  pass:264  dwarn:0   dfail:0   fail:0   skip:21  time:435s
fi-bdw-gvtdvm    total:285  pass:261  dwarn:0   dfail:0   fail:0   skip:24  time:441s
fi-blb-e6850     total:285  pass:220  dwarn:1   dfail:0   fail:0   skip:64  time:382s
fi-bsw-n3050     total:285  pass:239  dwarn:0   dfail:0   fail:0   skip:46  time:546s
fi-bwr-2160      total:285  pass:180  dwarn:0   dfail:0   fail:0   skip:105 time:300s
fi-bxt-dsi       total:285  pass:255  dwarn:0   dfail:0   fail:0   skip:30  time:516s
fi-bxt-j4205     total:285  pass:256  dwarn:0   dfail:0   fail:0   skip:29  time:515s
fi-byt-j1900     total:285  pass:250  dwarn:0   dfail:0   fail:0   skip:35  time:528s
fi-byt-n2820     total:285  pass:246  dwarn:0   dfail:0   fail:0   skip:39  time:515s
fi-cfl-8700k     total:285  pass:257  dwarn:0   dfail:0   fail:0   skip:28  time:410s
fi-cfl-s3        total:285  pass:258  dwarn:0   dfail:0   fail:1   skip:26  time:552s
fi-cfl-u         total:285  pass:259  dwarn:0   dfail:0   fail:0   skip:26  time:512s
fi-cnl-y3        total:285  pass:259  dwarn:0   dfail:0   fail:0   skip:26  time:586s
fi-elk-e7500     total:285  pass:225  dwarn:1   dfail:0   fail:0   skip:59  time:429s
fi-gdg-551       total:285  pass:177  dwarn:0   dfail:0   fail:0   skip:108 time:321s
fi-glk-1         total:285  pass:257  dwarn:0   dfail:0   fail:0   skip:28  time:538s
fi-hsw-4770      total:285  pass:258  dwarn:0   dfail:0   fail:0   skip:27  time:405s
fi-ilk-650       total:285  pass:225  dwarn:0   dfail:0   fail:0   skip:60  time:424s
fi-ivb-3520m     total:285  pass:256  dwarn:0   dfail:0   fail:0   skip:29  time:475s
fi-ivb-3770      total:285  pass:252  dwarn:0   dfail:0   fail:0   skip:33  time:429s
fi-kbl-7500u     total:285  pass:260  dwarn:1   dfail:0   fail:0   skip:24  time:477s
fi-kbl-7567u     total:285  pass:265  dwarn:0   dfail:0   fail:0   skip:20  time:471s
fi-kbl-r         total:285  pass:258  dwarn:0   dfail:0   fail:0   skip:27  time:519s
fi-pnv-d510      total:285  pass:219  dwarn:1   dfail:0   fail:0   skip:65  time:659s
fi-skl-6260u     total:285  pass:265  dwarn:0   dfail:0   fail:0   skip:20  time:450s
fi-skl-6600u     total:285  pass:258  dwarn:0   dfail:0   fail:0   skip:27  time:534s
fi-skl-6700k2    total:285  pass:261  dwarn:0   dfail:0   fail:0   skip:24  time:506s
fi-skl-6770hq    total:285  pass:265  dwarn:0   dfail:0   fail:0   skip:20  time:514s
fi-skl-guc       total:285  pass:257  dwarn:0   dfail:0   fail:0   skip:28  time:430s
fi-skl-gvtdvm    total:285  pass:262  dwarn:0   dfail:0   fail:0   skip:23  time:449s
fi-snb-2520m     total:285  pass:245  dwarn:0   dfail:0   fail:0   skip:40  time:586s
fi-snb-2600      total:285  pass:245  dwarn:0   dfail:0   fail:0   skip:40  time:405s
Blacklisted hosts:
fi-cnl-psr       total:285  pass:256  dwarn:3   dfail:0   fail:0   skip:26  time:535s
fi-glk-j4005     total:285  pass:256  dwarn:0   dfail:0   fail:0   skip:29  time:489s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_1196/issues.html
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev

[igt-dev] ✓ Fi.CI.IGT: success for igt/gem_ctx_isolation: Reset a scratch context

[Patchwork]

== Series Details ==

Series: igt/gem_ctx_isolation: Reset a scratch context
URL   : https://patchwork.freedesktop.org/series/40740/
State : success

== Summary ==

---- Known issues:

Test kms_cursor_legacy:
        Subgroup flip-vs-cursor-atomic:
                fail       -> PASS       (shard-hsw) fdo#102670
Test kms_flip:
        Subgroup 2x-flip-vs-wf_vblank:
                fail       -> PASS       (shard-hsw) fdo#100368 +2
Test kms_rotation_crc:
        Subgroup sprite-rotation-180:
                pass       -> FAIL       (shard-snb) fdo#103925
Test kms_sysfs_edid_timing:
                pass       -> WARN       (shard-apl) fdo#100047

fdo#102670 https://bugs.freedesktop.org/show_bug.cgi?id=102670
fdo#100368 https://bugs.freedesktop.org/show_bug.cgi?id=100368
fdo#103925 https://bugs.freedesktop.org/show_bug.cgi?id=103925
fdo#100047 https://bugs.freedesktop.org/show_bug.cgi?id=100047

shard-apl        total:3495 pass:1831 dwarn:1   dfail:0   fail:7   skip:1655 time:12938s
shard-hsw        total:3495 pass:1781 dwarn:1   dfail:0   fail:3   skip:1709 time:11630s
shard-snb        total:3495 pass:1373 dwarn:1   dfail:0   fail:4   skip:2117 time:6987s
Blacklisted hosts:
shard-kbl        total:3495 pass:1955 dwarn:1   dfail:1   fail:10  skip:1528 time:9700s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_1196/shards.html
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev