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).
