[v5,0/6] drm/i915: Remove short term pins from execbuf by requiring lock to unbind.

Message

Maarten Lankhorst Jan. 13, 2022, 11:44 a.m. UTC

Previously, short term pinning in execbuf was required because i915_vma was
effectively independent from objects, and has its own refcount, locking,
lifetime rules and pinning.

This series removes the separate locking, by requiring vma->obj->resv to be
held when pinning and unbinding. This will also be required for VM_BIND work.
Some patches have already been merged, but this contains the mremainder of
the conversion.

With pinning required for pinning and unbinding, the lock is enough to prevent
unbinding when trying to pin with the lock held, for example in execbuf.

This makes binding/unbinding similar to ttm_bo_validate()'s use, which just
cares that an object is in a certain place, without pinning it in place.

Having the VMA part of gem bo removes a lot of the vma refcounting, and makes
i915_vma more a part of the bo, instead of its own floating object that just
happens to be part of a bo. This is also required to make it more compatible
with TTM, and migration in general.

For future work, it makes things a lot simpler and clear. We want to end up
with i915_vma just being a specific mapping of the BO, just like is the
case in other drivers. i915_vma->active removal is the next step there,
and makes it when object is destroyed, the bindings are destroyed (after idle),
instead of object being destroyed when bindings are idle. 

Changes since previous version:
* As a special case, we allow unbinding pages when the object is dead.
  This will allow us to free objects without complications, which is
  important because many IGT tests free and then immediately put a new
  object in the same address. This is a shortcoming of softpin, as
  there is no way to tell when it's safe to re-use the address.

  As a potential performance optimization, i915 should use the random
  allocator by default, instead of the high to low allocator,
  which causes the unintentional address re-use.

Maarten Lankhorst (6):
  drm/i915: Call i915_gem_evict_vm in vm_fault_gtt to prevent new ENOSPC
    errors, v2.
  drm/i915: Add locking to i915_gem_evict_vm()
  drm/i915: Add object locking to i915_gem_evict_for_node and
    i915_gem_evict_something, v2.
  drm/i915: Add i915_vma_unbind_unlocked, and take obj lock for
    i915_vma_unbind, v2.
  drm/i915: Remove support for unlocked i915_vma unbind
  drm/i915: Remove short-term pins from execbuf, v6.

 drivers/gpu/drm/i915/display/intel_fb_pin.c   |   2 +-
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    | 220 +++++++++---------
 drivers/gpu/drm/i915/gem/i915_gem_mman.c      |  18 +-
 .../gpu/drm/i915/gem/selftests/huge_pages.c   |   2 +-
 .../i915/gem/selftests/i915_gem_client_blt.c  |   2 +-
 .../drm/i915/gem/selftests/i915_gem_mman.c    |   6 +
 drivers/gpu/drm/i915/gt/intel_ggtt.c          |  47 +++-
 drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c  |   1 -
 drivers/gpu/drm/i915/gt/selftest_hangcheck.c  |   2 +-
 drivers/gpu/drm/i915/gvt/aperture_gm.c        |   2 +-
 drivers/gpu/drm/i915/i915_gem.c               |   2 +
 drivers/gpu/drm/i915/i915_gem_evict.c         | 107 ++++++++-
 drivers/gpu/drm/i915/i915_gem_evict.h         |   6 +-
 drivers/gpu/drm/i915/i915_gem_gtt.c           |   8 +-
 drivers/gpu/drm/i915/i915_gem_gtt.h           |   4 +
 drivers/gpu/drm/i915/i915_vgpu.c              |   2 +-
 drivers/gpu/drm/i915/i915_vma.c               | 122 +++++-----
 drivers/gpu/drm/i915/i915_vma.h               |   1 +
 .../gpu/drm/i915/selftests/i915_gem_evict.c   |  27 ++-
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c |  28 +--
 drivers/gpu/drm/i915/selftests/i915_vma.c     |   8 +-
 21 files changed, 393 insertions(+), 224 deletions(-)