If a vma is already bound to a ppgtt, we incorrectly call
allocate_va_range again when doing a PIN_UPDATE, which will result in
over accounting within our paging structures, such that when we do
unbind something we don't actually destroy the structures and end up
inadvertently recycling them. In reality this probably isn't too bad,
but once we start touching PDEs and PDPEs for 64K/2M/1G pages this
apparent recycling will manifest into lots of really, really subtle
bugs.
v2: Fix the testing of vma->flags for aliasing_ppgtt_bind_vma
Fixes:
ff685975d97f ("drm/i915: Move allocate_va_range to GTT")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20170512091423.26085-1-chris@chris-wilson.co.uk
(cherry picked from commit
1f23475c893a85c934143cd64865ebb9b6af383f)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
u32 pte_flags;
int ret;
- ret = vma->vm->allocate_va_range(vma->vm, vma->node.start, vma->size);
- if (ret)
- return ret;
+ if (!(vma->flags & I915_VMA_LOCAL_BIND)) {
+ ret = vma->vm->allocate_va_range(vma->vm, vma->node.start,
+ vma->size);
+ if (ret)
+ return ret;
+ }
vma->pages = vma->obj->mm.pages;
if (flags & I915_VMA_LOCAL_BIND) {
struct i915_hw_ppgtt *appgtt = i915->mm.aliasing_ppgtt;
- if (appgtt->base.allocate_va_range) {
+ if (!(vma->flags & I915_VMA_LOCAL_BIND) &&
+ appgtt->base.allocate_va_range) {
ret = appgtt->base.allocate_va_range(&appgtt->base,
vma->node.start,
vma->node.size);