[Intel-gfx] [RFC PATCH 106/162] drm/i915: i915 returns -EBUSY on thread contention

[Matthew Auld]

From: CQ Tang <cq.tang at intel.com>

During high threads contention, the same object had been pinned
with a different type. A new pinning will catch -EBUSY if the
FORCE flag is not specified.

This error was observed on DG1 silicon during PO.

Cc: Matthew Auld <matthew.auld at intel.com>
Cc: Lucas De Marchi <lucas.demarchi at intel.com>
Cc: Abdiel Janulgue <abdiel.janulgue at linux.intel.com>
Cc: Balestrieri, Francesco <francesco.balestrieri at intel.com>
Cc: Niranjana Vishwanathapura <niranjana.vishwanathapura at intel.com>
Cc: Venkata S Dhanalakota <venkata.s.dhanalakota at intel.com>
Cc: Neel Desai <neel.desai at intel.com>
Cc: Matthew Brost <matthew.brost at intel.com>
Cc: Sudeep Dutt <sudeep.dutt at intel.com>
Signed-off-by: CQ Tang <cq.tang at intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_object_blt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
index b41b076f6864..1096f27627d4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object_blt.c
@@ -57,7 +57,7 @@ struct i915_vma *intel_emit_vma_fill_blt(struct intel_context *ce,
 	/* we pinned the pool, mark it as such */
 	intel_gt_buffer_pool_mark_used(pool);
 
-	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
+	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_FORCE_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
 		goto out_unpin;
@@ -297,7 +297,7 @@ struct i915_vma *intel_emit_vma_copy_blt(struct intel_context *ce,
 	/* we pinned the pool, mark it as such */
 	intel_gt_buffer_pool_mark_used(pool);
 
-	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_WC);
+	cmd = i915_gem_object_pin_map(pool->obj, I915_MAP_FORCE_WC);
 	if (IS_ERR(cmd)) {
 		err = PTR_ERR(cmd);
 		goto out_unpin;
-- 
2.26.2