* [PATCH i-g-t] i915_hangman: Force error capture
@ 2019-10-03 15:23 Chris Wilson
2019-10-15 13:31 ` [igt-dev] " Andi Shyti
0 siblings, 1 reply; 2+ messages in thread
From: Chris Wilson @ 2019-10-03 15:23 UTC (permalink / raw)
To: intel-gfx; +Cc: igt-dev
For fast preempt-resets, error capture is skipped, so disable
preempt-resets before checking the error state. While thinking ahead, be
prepared for when the modparams are not accessible.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
lib/igt_gt.c | 7 ++++---
tests/i915/i915_hangman.c | 7 ++++---
2 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/lib/igt_gt.c b/lib/igt_gt.c
index 78e3cd089..5ca77471c 100644
--- a/lib/igt_gt.c
+++ b/lib/igt_gt.c
@@ -162,6 +162,7 @@ igt_hang_t igt_allow_hang(int fd, unsigned ctx, unsigned flags)
struct drm_i915_gem_context_param param = {
.ctx_id = ctx,
};
+ int allow_reset;
unsigned ban;
/*
@@ -177,9 +178,7 @@ igt_hang_t igt_allow_hang(int fd, unsigned ctx, unsigned flags)
if (!igt_check_boolean_env_var("IGT_HANG_WITHOUT_RESET", false))
igt_require(has_gpu_reset(fd));
- igt_require(igt_sysfs_set_parameter
- (fd, "reset", "%d", INT_MAX /* any reset method */));
-
+ allow_reset = 1;
if ((flags & HANG_ALLOW_CAPTURE) == 0) {
param.param = I915_CONTEXT_PARAM_NO_ERROR_CAPTURE;
param.value = 1;
@@ -188,7 +187,9 @@ igt_hang_t igt_allow_hang(int fd, unsigned ctx, unsigned flags)
* the right one).
*/
__gem_context_set_param(fd, ¶m);
+ allow_reset = INT_MAX; /* any reset method */
}
+ igt_require(igt_sysfs_set_parameter(fd, "reset", "%d", allow_reset));
ban = context_get_ban(fd, ctx);
if ((flags & HANG_ALLOW_BAN) == 0)
diff --git a/tests/i915/i915_hangman.c b/tests/i915/i915_hangman.c
index 58141fc92..7a158d8db 100644
--- a/tests/i915/i915_hangman.c
+++ b/tests/i915/i915_hangman.c
@@ -140,12 +140,13 @@ static void check_error_state(const char *expected_ring_name,
size_t line_size = 0;
bool found = false;
- igt_debug("%s(expected ring name=%s, expected offset=%"PRIx64")\n",
- __func__, expected_ring_name, expected_offset);
igt_debugfs_dump(device, "i915_error_state");
igt_assert(getline(&line, &line_size, file) != -1);
- igt_assert(strcasecmp(line, "No error state collected"));
+ igt_require(strcasecmp(line, "No error state collected"));
+
+ igt_debug("%s(expected ring name=%s, expected offset=%"PRIx64")\n",
+ __func__, expected_ring_name, expected_offset);
while (getline(&line, &line_size, file) > 0) {
char *dashes;
--
2.23.0
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [igt-dev] [PATCH i-g-t] i915_hangman: Force error capture
2019-10-03 15:23 [PATCH i-g-t] i915_hangman: Force error capture Chris Wilson
@ 2019-10-15 13:31 ` Andi Shyti
0 siblings, 0 replies; 2+ messages in thread
From: Andi Shyti @ 2019-10-15 13:31 UTC (permalink / raw)
To: Chris Wilson; +Cc: igt-dev, intel-gfx
Hi Chris,
On Thu, Oct 03, 2019 at 04:23:02PM +0100, Chris Wilson wrote:
> For fast preempt-resets, error capture is skipped, so disable
> preempt-resets before checking the error state. While thinking ahead, be
> prepared for when the modparams are not accessible.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
and finally...
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Andi
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2019-10-15 13:31 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-10-03 15:23 [PATCH i-g-t] i915_hangman: Force error capture Chris Wilson
2019-10-15 13:31 ` [igt-dev] " Andi Shyti
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox