From: Matthew Auld <matthew.auld@intel.com>
To: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>,
intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Cc: matthew.brost@intel.com, paulo.r.zanoni@intel.com,
tvrtko.ursulin@intel.com, jani.nikula@intel.com,
lionel.g.landwerlin@intel.com, thomas.hellstrom@intel.com,
jason@jlekstrand.net, andi.shyti@linux.intel.com,
daniel.vetter@intel.com, christian.koenig@amd.com
Subject: Re: [PATCH 14/16] drm/i915/vm_bind: Handle persistent vmas in execbuf3
Date: Fri, 30 Sep 2022 10:47:48 +0100 [thread overview]
Message-ID: <f9bdd880-d14e-cca7-ab9f-53e6535b4522@intel.com> (raw)
In-Reply-To: <20220928061918.6340-15-niranjana.vishwanathapura@intel.com>
On 28/09/2022 07:19, Niranjana Vishwanathapura wrote:
> Handle persistent (VM_BIND) mappings during the request submission
> in the execbuf3 path.
>
> Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@intel.com>
> Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
> ---
> .../gpu/drm/i915/gem/i915_gem_execbuffer3.c | 188 +++++++++++++++++-
> 1 file changed, 187 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 92af88bc8deb..1aeeff5e8540 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer3.c
> @@ -19,6 +19,7 @@
> #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)
>
> @@ -42,7 +43,9 @@
> * 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
> @@ -58,6 +61,13 @@
> * 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.
> */
>
> /**
> @@ -127,6 +137,23 @@ eb_find_vma(struct i915_address_space *vm, u64 addr)
> return i915_gem_vm_bind_lookup_vma(vm, va);
> }
>
> +static void eb_scoop_unbound_vma_all(struct i915_address_space *vm)
> +{
> + struct i915_vma *vma, *vn;
> +
> + /**
> + * Move all unbound vmas back into vm_bind_list so that they are
> + * revalidated.
> + */
> + 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_vma_all(struct i915_execbuffer *eb)
> {
> unsigned int i, current_batch = 0;
> @@ -141,14 +168,121 @@ static int eb_lookup_vma_all(struct i915_execbuffer *eb)
> ++current_batch;
> }
>
> + eb_scoop_unbound_vma_all(eb->context->vm);
> +
> + return 0;
> +}
> +
> +static int eb_lock_vma_all(struct i915_execbuffer *eb)
> +{
> + struct i915_address_space *vm = eb->context->vm;
> + struct i915_vma *vma;
> + int err;
> +
> + err = i915_gem_object_lock(eb->context->vm->root_obj, &eb->ww);
> + if (err)
> + return err;
> +
> + 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_vma_all(struct i915_execbuffer *eb,
> + bool final)
> +{
> + struct i915_address_space *vm = eb->context->vm;
> + struct i915_vma *vma, *vn;
> +
> + lockdep_assert_held(&vm->vm_bind_lock);
> +
> + if (!(eb->args->flags & __EXEC3_HAS_PIN))
> + return;
> +
> + assert_object_held(vm->root_obj);
> +
> + 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_verify_bind_complete(vma))
> + list_move_tail(&vma->vm_bind_link, &vm->vm_bound_list);
> +}
> +
> static void eb_release_vma_all(struct i915_execbuffer *eb, bool final)
> {
> + eb_release_persistent_vma_all(eb, final);
> eb_unpin_engine(eb);
> }
>
> +static int eb_reserve_fence_for_persistent_vma_all(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_vma_all(struct i915_execbuffer *eb)
> +{
> + struct i915_address_space *vm = eb->context->vm;
> + struct i915_vma *vma, *last_pinned_vma = NULL;
> + int ret = 0;
> +
> + lockdep_assert_held(&vm->vm_bind_lock);
> + assert_object_held(vm->root_obj);
> +
> + ret = eb_reserve_fence_for_persistent_vma_all(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;
> +}
> +
> /*
> * Using two helper loops for the order of which requests / batches are created
> * and added the to backend. Requests are created in order from the parent to
> @@ -161,8 +295,43 @@ static void eb_release_vma_all(struct i915_execbuffer *eb, bool final)
> #define for_each_batch_create_order(_eb) \
> for (unsigned int i = 0; i < (_eb)->num_batches; ++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);
See: 842d9346b2fd ("drm/i915: Individualize fences before adding to
dma_resv obj"). Do we not need something similar?
> +}
> +
> +static void eb_move_all_persistent_vma_to_active(struct i915_execbuffer *eb)
> +{
> + /* Add fence to BOs dma-resv fence list */
> + eb_persistent_add_shared_fence(eb);
> +}
> +
> static int eb_move_to_gpu(struct i915_execbuffer *eb)
> {
> + lockdep_assert_held(&eb->context->vm->vm_bind_lock);
> + assert_object_held(eb->context->vm->root_obj);
> +
> + eb_move_all_persistent_vma_to_active(eb);
> +
> /* Unconditionally flush any chipset caches (for streaming writes). */
> intel_gt_chipset_flush(eb->gt);
>
> @@ -478,6 +647,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,
>
> mutex_lock(&eb.context->vm->vm_bind_lock);
>
> +lookup_vmas:
> err = eb_lookup_vma_all(&eb);
> if (err) {
> eb_release_vma_all(&eb, true);
> @@ -494,6 +664,22 @@ i915_gem_do_execbuffer(struct drm_device *dev,
> /* only throttle once, even if we didn't need to throttle */
> throttle = false;
>
> + err = eb_lock_vma_all(&eb);
> + if (err)
> + goto err_validate;
> +
> + /**
> + * No object unbinds possible once the objects are locked. So,
> + * check for any unbinds here, which needs to be scooped up.
> + */
> + if (!list_empty(&eb.context->vm->vm_rebind_list)) {
> + eb_release_vma_all(&eb, true);
> + i915_gem_ww_ctx_fini(&eb.ww);
> + goto lookup_vmas;
> + }
> +
> + err = eb_validate_persistent_vma_all(&eb);
> +
> err_validate:
> if (err == -EDEADLK) {
> eb_release_vma_all(&eb, false);
next prev parent reply other threads:[~2022-09-30 9:48 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-28 6:19 [PATCH 00/16] drm/i915/vm_bind: Add VM_BIND functionality Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 01/16] drm/i915/vm_bind: Expose vm lookup function Niranjana Vishwanathapura
2022-09-28 17:28 ` Matthew Auld
2022-09-28 6:19 ` [PATCH 02/16] drm/i915/vm_bind: Add __i915_sw_fence_await_reservation() Niranjana Vishwanathapura
2022-09-28 17:39 ` Matthew Auld
2022-09-29 5:20 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 03/16] drm/i915/vm_bind: Expose i915_gem_object_max_page_size() Niranjana Vishwanathapura
2022-09-28 17:40 ` Matthew Auld
2022-09-29 5:20 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 04/16] drm/i915/vm_bind: Add support to create persistent vma Niranjana Vishwanathapura
2022-09-28 7:38 ` [Intel-gfx] " Tvrtko Ursulin
2022-09-28 17:05 ` Niranjana Vishwanathapura
2022-09-28 14:44 ` Andi Shyti
2022-09-28 17:07 ` Niranjana Vishwanathapura
2022-09-29 17:04 ` Matthew Auld
2022-09-28 6:19 ` [PATCH 05/16] drm/i915/vm_bind: Implement bind and unbind of object Niranjana Vishwanathapura
2022-09-28 9:43 ` [Intel-gfx] " kernel test robot
2022-09-28 17:52 ` Matthew Auld
2022-09-29 5:24 ` Niranjana Vishwanathapura
2022-09-29 9:03 ` Matthew Auld
2022-09-29 10:51 ` Matthew Auld
2022-09-29 14:24 ` Niranjana Vishwanathapura
2022-09-28 20:06 ` [Intel-gfx] " Welty, Brian
2022-09-29 5:25 ` Niranjana Vishwanathapura
2022-09-29 10:49 ` Matthew Auld
2022-09-29 16:38 ` Niranjana Vishwanathapura
2022-09-29 17:28 ` Matthew Auld
2022-09-29 17:49 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 06/16] drm/i915/vm_bind: Support for VM private BOs Niranjana Vishwanathapura
2022-09-28 17:54 ` Matthew Auld
2022-09-29 14:28 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 07/16] drm/i915/vm_bind: Add support to handle object evictions Niranjana Vishwanathapura
2022-09-29 17:13 ` Matthew Auld
2022-09-28 6:19 ` [PATCH 08/16] drm/i915/vm_bind: Support persistent vma activeness tracking Niranjana Vishwanathapura
2022-09-30 12:00 ` Andi Shyti
2022-09-28 6:19 ` [PATCH 09/16] drm/i915/vm_bind: Add out fence support Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 10/16] drm/i915/vm_bind: Abstract out common execbuf functions Niranjana Vishwanathapura
2022-09-30 10:45 ` Matthew Auld
2022-09-30 16:26 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 11/16] drm/i915/vm_bind: Use common execbuf functions in execbuf path Niranjana Vishwanathapura
2022-09-30 10:47 ` Matthew Auld
2022-09-28 6:19 ` [PATCH 12/16] drm/i915/vm_bind: Implement I915_GEM_EXECBUFFER3 ioctl Niranjana Vishwanathapura
2022-09-29 15:00 ` Matthew Auld
2022-09-29 16:02 ` Niranjana Vishwanathapura
2022-10-03 21:12 ` [Intel-gfx] " Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 13/16] drm/i915/vm_bind: Update i915_vma_verify_bind_complete() Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 14/16] drm/i915/vm_bind: Handle persistent vmas in execbuf3 Niranjana Vishwanathapura
2022-09-30 9:47 ` Matthew Auld [this message]
2022-10-02 6:28 ` Niranjana Vishwanathapura
2022-10-03 8:36 ` Matthew Auld
2022-10-05 5:38 ` Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 15/16] drm/i915/vm_bind: userptr dma-resv changes Niranjana Vishwanathapura
2022-09-28 6:19 ` [PATCH 16/16] drm/i915/vm_bind: Add uapi for user to enable vm_bind_mode Niranjana Vishwanathapura
2022-09-30 10:01 ` Matthew Auld
2022-09-30 16:13 ` Niranjana Vishwanathapura
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=f9bdd880-d14e-cca7-ab9f-53e6535b4522@intel.com \
--to=matthew.auld@intel.com \
--cc=andi.shyti@linux.intel.com \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jani.nikula@intel.com \
--cc=jason@jlekstrand.net \
--cc=lionel.g.landwerlin@intel.com \
--cc=matthew.brost@intel.com \
--cc=niranjana.vishwanathapura@intel.com \
--cc=paulo.r.zanoni@intel.com \
--cc=thomas.hellstrom@intel.com \
--cc=tvrtko.ursulin@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).