Re: [RFC 07/10] drm/i915/vm_bind: Handle persistent vmas in execbuf3

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, 2022-07-08 at 05:44 -0700, Niranjana Vishwanathapura wrote:
> On Thu, Jul 07, 2022 at 07:54:16AM -0700, Hellstrom, Thomas wrote:
> > On Fri, 2022-07-01 at 15:50 -0700, Niranjana Vishwanathapura wrote:
> > > Handle persistent (VM_BIND) mappings during the request
> > > submission
> > > in the execbuf3 path.
> > > 
> > > Signed-off-by: Niranjana Vishwanathapura
> > > <niranjana.vishwanathapura@xxxxxxxxx>
> > > ---
> > >  .../gpu/drm/i915/gem/i915_gem_execbuffer3.c   | 176
> > > +++++++++++++++++-
> > >  1 file changed, 175 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> > > b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> > > index 13121df72e3d..2079f5ca9010 100644
> > > --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> > > +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> > > @@ -22,6 +22,7 @@
> > >  #include "i915_gem_vm_bind.h"
> > >  #include "i915_trace.h"
> > > 
> > > +#define __EXEC3_HAS_PIN                        BIT_ULL(33)
> > >  #define __EXEC3_ENGINE_PINNED          BIT_ULL(32)
> > >  #define __EXEC3_INTERNAL_FLAGS         (~0ull << 32)
> > > 
> > > @@ -45,7 +46,9 @@
> > >   * execlist. Hence, no support for implicit sync.
> > >   *
> > >   * The new execbuf3 ioctl only works in VM_BIND mode and the
> > > VM_BIND
> > > mode only
> > > - * works with execbuf3 ioctl for submission.
> > > + * works with execbuf3 ioctl for submission. All BOs mapped on
> > > that
> > > VM (through
> > > + * VM_BIND call) at the time of execbuf3 call are deemed
> > > required
> > > for that
> > > + * submission.
> > >   *
> > >   * The execbuf3 ioctl directly specifies the batch addresses
> > > instead
> > > of as
> > >   * object handles as in execbuf2 ioctl. The execbuf3 ioctl will
> > > also
> > > not
> > > @@ -61,6 +64,13 @@
> > >   * So, a lot of code supporting execbuf2 ioctl, like
> > > relocations, VA
> > > evictions,
> > >   * vma lookup table, implicit sync, vma active reference
> > > tracking
> > > etc., are not
> > >   * applicable for execbuf3 ioctl.
> > > + *
> > > + * During each execbuf submission, request fence is added to all
> > > VM_BIND mapped
> > > + * objects with DMA_RESV_USAGE_BOOKKEEP. The
> > > DMA_RESV_USAGE_BOOKKEEP
> > > usage will
> > > + * prevent over sync (See enum dma_resv_usage). Note that
> > > DRM_I915_GEM_WAIT and
> > > + * DRM_I915_GEM_BUSY ioctls do not check for
> > > DMA_RESV_USAGE_BOOKKEEP
> > > usage and
> > > + * hence should not be used for end of batch check. Instead, the
> > > execbuf3
> > > + * timeline out fence should be used for end of batch check.
> > >   */
> > > 
> > >  struct eb_fence {
> > > @@ -124,6 +134,19 @@ eb_find_vma(struct i915_address_space *vm,
> > > u64
> > > addr)
> > >         return i915_gem_vm_bind_lookup_vma(vm, va);
> > >  }
> > > 
> > > +static void eb_scoop_unbound_vmas(struct i915_address_space *vm)
> > > +{
> > > +       struct i915_vma *vma, *vn;
> > > +
> > > +       spin_lock(&vm->vm_rebind_lock);
> > > +       list_for_each_entry_safe(vma, vn, &vm->vm_rebind_list,
> > > vm_rebind_link) {
> > > +               list_del_init(&vma->vm_rebind_link);
> > > +               if (!list_empty(&vma->vm_bind_link))
> > > +                       list_move_tail(&vma->vm_bind_link, &vm-
> > > > vm_bind_list);
> > > +       }
> > > +       spin_unlock(&vm->vm_rebind_lock);
> > > +}
> > > +
> > >  static int eb_lookup_vmas(struct i915_execbuffer *eb)
> > >  {
> > >         unsigned int i, current_batch = 0;
> > > @@ -138,11 +161,118 @@ static int eb_lookup_vmas(struct
> > > i915_execbuffer *eb)
> > >                 ++current_batch;
> > >         }
> > > 
> > > +       eb_scoop_unbound_vmas(eb->context->vm);
> > > +
> > > +       return 0;
> > > +}
> > > +
> > > +static int eb_lock_vmas(struct i915_execbuffer *eb)
> > > +{
> > > +       struct i915_address_space *vm = eb->context->vm;
> > > +       struct i915_vma *vma;
> > > +       int err;
> > > +
> > > +       err = i915_gem_vm_priv_lock(eb->context->vm, &eb->ww);
> > > +       if (err)
> > > +               return err;
> > > +
> > 
> > See comment in review for 08/10 about re-checking the rebind list
> > here.
> > 
> > 
> > 
> > > +       list_for_each_entry(vma, &vm->non_priv_vm_bind_list,
> > > +                           non_priv_vm_bind_link) {
> > > +               err = i915_gem_object_lock(vma->obj, &eb->ww);
> > > +               if (err)
> > > +                       return err;
> > > +       }
> > > +
> > >         return 0;
> > >  }
> > > 
> > > +static void eb_release_persistent_vmas(struct i915_execbuffer
> > > *eb,
> > > bool final)
> > > +{
> > > +       struct i915_address_space *vm = eb->context->vm;
> > > +       struct i915_vma *vma, *vn;
> > > +
> > > +       assert_vm_bind_held(vm);
> > > +
> > > +       if (!(eb->args->flags & __EXEC3_HAS_PIN))
> > > +               return;
> > > +
> > > +       assert_vm_priv_held(vm);
> > > +
> > > +       list_for_each_entry(vma, &vm->vm_bind_list, vm_bind_link)
> > > +               __i915_vma_unpin(vma);
> > > +
> > > +       eb->args->flags &= ~__EXEC3_HAS_PIN;
> > > +       if (!final)
> > > +               return;
> > > +
> > > +       list_for_each_entry_safe(vma, vn, &vm->vm_bind_list,
> > > vm_bind_link)
> > > +               if (i915_vma_is_bind_complete(vma))
> > > +                       list_move_tail(&vma->vm_bind_link, &vm-
> > > > vm_bound_list);
> > > +}
> > > +
> > >  static void eb_release_vmas(struct i915_execbuffer *eb, bool
> > > final)
> > >  {
> > > +       eb_release_persistent_vmas(eb, final);
> > > +       eb_unpin_engine(eb);
> > > +}
> > > +
> > > +static int eb_reserve_fence_for_persistent_vmas(struct
> > > i915_execbuffer *eb)
> > > +{
> > > +       struct i915_address_space *vm = eb->context->vm;
> > > +       struct i915_vma *vma;
> > > +       int ret;
> > > +
> > > +       ret = dma_resv_reserve_fences(vm->root_obj->base.resv,
> > > 1);
> > > +       if (ret)
> > > +               return ret;
> > > +
> > > +       list_for_each_entry(vma, &vm->non_priv_vm_bind_list,
> > > +                           non_priv_vm_bind_link) {
> > > +               ret = dma_resv_reserve_fences(vma->obj-
> > > >base.resv,
> > > 1);
> > > +               if (ret)
> > > +                       return ret;
> > > +       }
> > > +
> > > +       return 0;
> > > +}
> > > +
> > > +static int eb_validate_persistent_vmas(struct i915_execbuffer
> > > *eb)
> > > +{
> > > +       struct i915_address_space *vm = eb->context->vm;
> > > +       struct i915_vma *vma, *last_pinned_vma = NULL;
> > > +       int ret = 0;
> > > +
> > > +       assert_vm_bind_held(vm);
> > > +       assert_vm_priv_held(vm);
> > > +
> > > +       ret = eb_reserve_fence_for_persistent_vmas(eb);
> > > +       if (ret)
> > > +               return ret;
> > > +
> > > +       if (list_empty(&vm->vm_bind_list))
> > > +               return 0;
> > > +
> > > +       list_for_each_entry(vma, &vm->vm_bind_list, vm_bind_link)
> > > {
> > > +               u64 pin_flags = vma->start | PIN_OFFSET_FIXED |
> > > PIN_USER;
> > > +
> > > +               ret = i915_vma_pin_ww(vma, &eb->ww, 0, 0,
> > > pin_flags);
> > > +               if (ret)
> > > +                       break;
> > > +
> > > +               last_pinned_vma = vma;
> > > +       }
> > > +
> > > +       if (ret && last_pinned_vma) {
> > > +               list_for_each_entry(vma, &vm->vm_bind_list,
> > > vm_bind_link) {
> > > +                       __i915_vma_unpin(vma);
> > > +                       if (vma == last_pinned_vma)
> > > +                               break;
> > > +               }
> > > +       } else if (last_pinned_vma) {
> > > +               eb->args->flags |= __EXEC3_HAS_PIN;
> > > +       }
> > > +
> > > +       return ret;
> > >  }
> > > 
> > >  static int eb_validate_vmas(struct i915_execbuffer *eb)
> > > @@ -162,8 +292,17 @@ static int eb_validate_vmas(struct
> > > i915_execbuffer *eb)
> > >         /* only throttle once, even if we didn't need to throttle
> > > */
> > >         throttle = false;
> > > 
> > > +       err = eb_lock_vmas(eb);
> > > +       if (err)
> > > +               goto err;
> > > +
> > > +       err = eb_validate_persistent_vmas(eb);
> > > +       if (err)
> > > +               goto err;
> > > +
> > >  err:
> > >         if (err == -EDEADLK) {
> > > +               eb_release_vmas(eb, false);
> > >                 err = i915_gem_ww_ctx_backoff(&eb->ww);
> > >                 if (!err)
> > >                         goto retry;
> > > @@ -187,8 +326,43 @@ static int eb_validate_vmas(struct
> > > i915_execbuffer *eb)
> > >         BUILD_BUG_ON(!typecheck(int, _i)); \
> > >         for ((_i) = (_eb)->num_batches - 1; (_i) >= 0; --(_i))
> > > 
> > > +static void __eb_persistent_add_shared_fence(struct
> > > drm_i915_gem_object *obj,
> > > +                                            struct dma_fence
> > > *fence)
> > > +{
> > > +       dma_resv_add_fence(obj->base.resv, fence,
> > > DMA_RESV_USAGE_BOOKKEEP);
> > > +       obj->write_domain = 0;
> > > +       obj->read_domains |= I915_GEM_GPU_DOMAINS;
> > > +       obj->mm.dirty = true;
> > > +}
> > > +
> > > +static void eb_persistent_add_shared_fence(struct
> > > i915_execbuffer
> > > *eb)
> > > +{
> > > +       struct i915_address_space *vm = eb->context->vm;
> > > +       struct dma_fence *fence;
> > > +       struct i915_vma *vma;
> > > +
> > > +       fence = eb->composite_fence ? eb->composite_fence :
> > > +               &eb->requests[0]->fence;
> > > +
> > > +       __eb_persistent_add_shared_fence(vm->root_obj, fence);
> > > +       list_for_each_entry(vma, &vm->non_priv_vm_bind_list,
> > > +                           non_priv_vm_bind_link)
> > > +               __eb_persistent_add_shared_fence(vma->obj,
> > > fence);
> > > +}
> > > +
> > > +static void eb_persistent_vmas_move_to_active(struct
> > > i915_execbuffer
> > > *eb)
> > > +{
> > > +       /* Add fence to BOs dma-resv fence list */
> > > +       eb_persistent_add_shared_fence(eb);
> > 
> > This means we don't add any fences to the vma active trackers.
> > While
> > this works fine for TTM delayed buffer destruction, unbinding at
> > eviction and shrinking wouldn't wait for gpu activity to idle
> > before
> > unbinding?
> 
> Eviction and shrinker will wait for gpu activity to idle before
> unbinding.
> The i915_vma_is_active() and i915_vma_sync() have been updated to
> handle
> the persistent vmas to differntly (by checking/waiting for dma-resv
> fence
> list).

Ah, yes. Now I see. Still the async unbinding __i915_vma_unbind_async()
needs update? Should probably add a i915_sw_fence_await_reservation()
(modified if needed for bookkeeping fences) of the vm's reservation
object there?

/Thomas

> 
> Niranjana
> 
> > 
> > 
> > /Thomas
> > 





[Index of Archives]     [AMD Graphics]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux