Re: [PATCH 2/2] panthor: use drm_gpuva_unlink_defer()

[Alice Ryhl <aliceryhl@google.com>]

On Fri, Sep 5, 2025 at 2:53 PM Boris Brezillon
<boris.brezillon@collabora.com> wrote:
>
> On Fri, 05 Sep 2025 12:11:29 +0000
> Alice Ryhl <aliceryhl@google.com> wrote:
>
> >  static void panthor_vm_cleanup_op_ctx(struct panthor_vm_op_ctx *op_ctx,
> >                                     struct panthor_vm *vm)
> >  {
> > -     struct panthor_vma *vma, *tmp_vma;
> > -
> >       u32 remaining_pt_count = op_ctx->rsvd_page_tables.count -
> >                                op_ctx->rsvd_page_tables.ptr;
> >
> > @@ -1134,16 +1091,12 @@ static void panthor_vm_cleanup_op_ctx(struct panthor_vm_op_ctx *op_ctx,
> >       kfree(op_ctx->rsvd_page_tables.pages);
> >
> >       if (op_ctx->map.vm_bo)
> > -             panthor_vm_bo_put(op_ctx->map.vm_bo);
> > +             drm_gpuvm_bo_put_deferred(op_ctx->map.vm_bo);
> >
> >       for (u32 i = 0; i < ARRAY_SIZE(op_ctx->preallocated_vmas); i++)
> >               kfree(op_ctx->preallocated_vmas[i]);
> >
> > -     list_for_each_entry_safe(vma, tmp_vma, &op_ctx->returned_vmas, node) {
> > -             list_del(&vma->node);
> > -             panthor_vm_bo_put(vma->base.vm_bo);
> > -             kfree(vma);
>
> Maybe I'm missing something, but I don't see the VMAs being freed in
> this new version.

Sorry you are right. We can kfree the vma right away after unlink(),
but I forgot to add that.

Alice