[Beignet] [PATCH] intel: Check that we can reserve the zero-offset

[Chris Wilson]

commit ff57cee0519d ("ocl20/runtime: take the first 64KB page table
entries") tries to allocate a bo at 0 offset, but failed to take into
account that something may already be allocated there that it is not
allowed to evict (particularly when not using full-ppgtt separation).
Failure to do so causes all execution to subsequentally fail with
"drm_intel_gem_bo_context_exec() failed: Device or resource busy"

Reported-by: Kenneth Johansson <ken at kenjo.org>
Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=98647
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
 src/intel/intel_driver.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/src/intel/intel_driver.c b/src/intel/intel_driver.c
index 363e018..a6074ae 100644
--- a/src/intel/intel_driver.c
+++ b/src/intel/intel_driver.c
@@ -143,10 +143,14 @@ intel_driver_context_init(intel_driver_t *driver)
 #ifdef HAS_BO_SET_SOFTPIN
   drm_intel_bo *bo = dri_bo_alloc(driver->bufmgr, "null_bo", 64*1024, 4096);
   drm_intel_bo_set_softpin_offset(bo, 0);
-  // don't reuse it, that would make two bo trying to bind to same address,
-  // which is un-reasonable.
-  drm_intel_bo_disable_reuse(bo);
-  driver->null_bo = bo;
+  drm_intel_bo_map(bo, true);
+  *(uint32_t *)bo->virtual = MI_BATCH_BUFFER_END;
+  if (drm_intel_gem_bo_context_exec(bo, driver->ctx, 0, 0) == 0) {
+    drm_intel_bo_map(bo, true);
+    driver->null_bo = memset(bo->virtual, 0, 64*1024);
+  } else {
+    drm_intel_bo_unreference(bo);
+  }
 #endif
 }
 
-- 
2.10.2