On 20/11/2019 09:32, Chris Wilson wrote:
Bonded request submission is designed to allow requests to execute in
parallel as laid out by the user. If the master request is already
finished before its bonded pair is submitted, the pair were not destined
to run in parallel and we lose the information about the master engine
to dictate selection of the secondary. If the second request was
required to be run on a particular engine in a virtual set, that should
have been specified, rather than left to the whims of a random
unconnected requests!
In the selftest, I made the mistake of not ensuring the master would
overlap with its bonded pairs, meaning that it could indeed complete
before we submitted the bonds. Those bonds were then free to select any
available engine in their virtual set, and not the one expected by the
test.
There is a submit await which ensures master is not runnable before
bonded pairs are submitted. Why was that not enough? Are the sporadic
test failures?
Regards,
Tvrtko
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
---
drivers/gpu/drm/i915/gt/selftest_lrc.c | 23 ++++++++++++++++++++---
1 file changed, 20 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c
index 16ebe4d2308e..f3b0610d1f10 100644
--- a/drivers/gpu/drm/i915/gt/selftest_lrc.c
+++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c
@@ -3036,15 +3036,21 @@ static int bond_virtual_engine(struct intel_gt *gt,
struct i915_gem_context *ctx;
struct i915_request *rq[16];
enum intel_engine_id id;
+ struct igt_spinner spin;
unsigned long n;
int err;
GEM_BUG_ON(nsibling >= ARRAY_SIZE(rq) - 1);
- ctx = kernel_context(gt->i915);
- if (!ctx)
+ if (igt_spinner_init(&spin, gt))
return -ENOMEM;
+ ctx = kernel_context(gt->i915);
+ if (!ctx) {
+ err = -ENOMEM;
+ goto err_spin;
+ }
+
err = 0;
rq[0] = ERR_PTR(-ENOMEM);
for_each_engine(master, gt, id) {
@@ -3055,7 +3061,7 @@ static int bond_virtual_engine(struct intel_gt *gt,
memset_p((void *)rq, ERR_PTR(-EINVAL), ARRAY_SIZE(rq));
- rq[0] = igt_request_alloc(ctx, master);
+ rq[0] = spinner_create_request(&spin, ctx, master, MI_NOOP);
if (IS_ERR(rq[0])) {
err = PTR_ERR(rq[0]);
goto out;
@@ -3068,10 +3074,17 @@ static int bond_virtual_engine(struct intel_gt *gt,
&fence,
GFP_KERNEL);
}
+
i915_request_add(rq[0]);
if (err < 0)
goto out;
+ if (!(flags & BOND_SCHEDULE) &&
+ !igt_wait_for_spinner(&spin, rq[0])) {
+ err = -EIO;
+ goto out;
+ }
+
for (n = 0; n < nsibling; n++) {
struct intel_context *ve;
@@ -3119,6 +3132,8 @@ static int bond_virtual_engine(struct intel_gt *gt,
}
}
onstack_fence_fini(&fence);
+ intel_engine_flush_submission(master);
+ igt_spinner_end(&spin);
if (i915_request_wait(rq[0], 0, HZ / 10) < 0) {
pr_err("Master request did not execute (on %s)!\n",
@@ -3156,6 +3171,8 @@ static int bond_virtual_engine(struct intel_gt *gt,
err = -EIO;
kernel_context_close(ctx);
+err_spin:
+ igt_spinner_fini(&spin);
return err;
}
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
