From: Boris Brezillon <boris.brezillon@collabora.com>
To: Danilo Krummrich <dakr@redhat.com>
Cc: matthew.brost@intel.com, thomas.hellstrom@linux.intel.com,
sarah.walker@imgtec.com, nouveau@lists.freedesktop.org,
linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org,
donald.robson@imgtec.com, christian.koenig@amd.com,
faith.ekstrand@collabora.com
Subject: Re: [PATCH drm-misc-next v4 0/8] [RFC] DRM GPUVA Manager GPU-VM features
Date: Thu, 28 Sep 2023 14:09:13 +0200 [thread overview]
Message-ID: <20230928140913.7cb78ed4@collabora.com> (raw)
In-Reply-To: <20230920144343.64830-1-dakr@redhat.com>
On Wed, 20 Sep 2023 16:42:33 +0200
Danilo Krummrich <dakr@redhat.com> wrote:
> So far the DRM GPUVA manager offers common infrastructure to track GPU VA
> allocations and mappings, generically connect GPU VA mappings to their
> backing buffers and perform more complex mapping operations on the GPU VA
> space.
>
> However, there are more design patterns commonly used by drivers, which
> can potentially be generalized in order to make the DRM GPUVA manager
> represent a basic GPU-VM implementation. In this context, this patch series
> aims at generalizing the following elements.
>
> 1) Provide a common dma-resv for GEM objects not being used outside of
> this GPU-VM.
>
> 2) Provide tracking of external GEM objects (GEM objects which are
> shared with other GPU-VMs).
>
> 3) Provide functions to efficiently lock all GEM objects dma-resv the
> GPU-VM contains mappings of.
>
> 4) Provide tracking of evicted GEM objects the GPU-VM contains mappings
> of, such that validation of evicted GEM objects is accelerated.
>
> 5) Provide some convinience functions for common patterns.
>
> The implementation introduces struct drm_gpuvm_bo, which serves as abstraction
> combining a struct drm_gpuvm and struct drm_gem_object, similar to what
> amdgpu does with struct amdgpu_bo_vm. While this adds a bit of complexity it
> improves the efficiency of tracking external and evicted GEM objects.
>
> This patch series also renames struct drm_gpuva_manager to struct drm_gpuvm
> including corresponding functions. This way the GPUVA manager's structures align
> better with the documentation of VM_BIND [1] and VM_BIND locking [2]. It also
> provides a better foundation for the naming of data structures and functions
> introduced for implementing the features of this patch series.
>
> This patch series is also available at [3].
>
> [1] Documentation/gpu/drm-vm-bind-async.rst
> [2] Documentation/gpu/drm-vm-bind-locking.rst
> [3] https://gitlab.freedesktop.org/nouvelles/kernel/-/commits/gpuvm-next
>
> Changes in V2:
> ==============
> - rename 'drm_gpuva_manager' -> 'drm_gpuvm' which generally leads to more
> consistent naming
> - properly separate commits (introduce common dma-resv, drm_gpuvm_bo
> abstraction, etc.)
> - remove maple tree for tracking external objects, use a list drm_gpuvm_bos
> per drm_gpuvm instead
> - rework dma-resv locking helpers (Thomas)
> - add a locking helper for a given range of the VA space (Christian)
> - make the GPUVA manager buildable as module, rather than drm_exec
> builtin (Christian)
>
> Changes in V3:
> ==============
> - rename missing function and files (Boris)
> - warn if vm_obj->obj != obj in drm_gpuva_link() (Boris)
> - don't expose drm_gpuvm_bo_destroy() (Boris)
> - unlink VM_BO from GEM in drm_gpuvm_bo_destroy() rather than
> drm_gpuva_unlink() and link within drm_gpuvm_bo_obtain() to keep
> drm_gpuvm_bo instances unique
> - add internal locking to external and evicted object lists to support drivers
> updating the VA space from within the fence signalling critical path (Boris)
> - unlink external objects and evicted objects from the GPUVM's list in
> drm_gpuvm_bo_destroy()
> - add more documentation and fix some kernel doc issues
>
> Changes in V4:
> ==============
> - add a drm_gpuvm_resv() helper (Boris)
> - add a drm_gpuvm::<list_name>::local_list field (Boris)
> - remove drm_gpuvm_bo_get_unless_zero() helper (Boris)
> - fix missing NULL assignment in get_next_vm_bo_from_list() (Boris)
> - keep a drm_gem_object reference on potential vm_bo destroy (alternatively we
> could free the vm_bo and drop the vm_bo's drm_gem_object reference through
> async work)
> - introduce DRM_GPUVM_RESV_PROTECTED flag to indicate external locking through
> the corresponding dma-resv locks to optimize for drivers already holding
> them when needed; add the corresponding lock_assert_held() calls (Thomas)
> - make drm_gpuvm_bo_evict() per vm_bo and add a drm_gpuvm_bo_gem_evict()
> helper (Thomas)
> - pass a drm_gpuvm_bo in drm_gpuvm_ops::vm_bo_validate() (Thomas)
> - documentation fixes
>
> Danilo Krummrich (8):
> drm/gpuvm: rename struct drm_gpuva_manager to struct drm_gpuvm
> drm/gpuvm: allow building as module
> drm/nouveau: uvmm: rename 'umgr' to 'base'
> drm/gpuvm: add common dma-resv per struct drm_gpuvm
> drm/gpuvm: add an abstraction for a VM / BO combination
> drm/gpuvm: add drm_gpuvm_flags to drm_gpuvm
> drm/gpuvm: generalize dma_resv/extobj handling and GEM validation
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
> drm/nouveau: GPUVM dma-resv/extobj handling, GEM validation
>
> drivers/gpu/drm/Kconfig | 7 +
> drivers/gpu/drm/Makefile | 2 +-
> drivers/gpu/drm/drm_debugfs.c | 16 +-
> drivers/gpu/drm/drm_gpuva_mgr.c | 1725 --------------
> drivers/gpu/drm/drm_gpuvm.c | 2600 +++++++++++++++++++++
> drivers/gpu/drm/nouveau/Kconfig | 1 +
> drivers/gpu/drm/nouveau/nouveau_bo.c | 4 +-
> drivers/gpu/drm/nouveau/nouveau_debugfs.c | 2 +-
> drivers/gpu/drm/nouveau/nouveau_exec.c | 52 +-
> drivers/gpu/drm/nouveau/nouveau_exec.h | 4 -
> drivers/gpu/drm/nouveau/nouveau_gem.c | 5 +-
> drivers/gpu/drm/nouveau/nouveau_sched.h | 4 +-
> drivers/gpu/drm/nouveau/nouveau_uvmm.c | 207 +-
> drivers/gpu/drm/nouveau/nouveau_uvmm.h | 8 +-
> include/drm/drm_debugfs.h | 6 +-
> include/drm/drm_gem.h | 32 +-
> include/drm/drm_gpuva_mgr.h | 706 ------
> include/drm/drm_gpuvm.h | 1142 +++++++++
> 18 files changed, 3934 insertions(+), 2589 deletions(-)
> delete mode 100644 drivers/gpu/drm/drm_gpuva_mgr.c
> create mode 100644 drivers/gpu/drm/drm_gpuvm.c
> delete mode 100644 include/drm/drm_gpuva_mgr.h
> create mode 100644 include/drm/drm_gpuvm.h
>
>
> base-commit: 1c7a387ffef894b1ab3942f0482dac7a6e0a909c
WARNING: multiple messages have this Message-ID (diff)
From: Boris Brezillon <boris.brezillon@collabora.com>
To: Danilo Krummrich <dakr@redhat.com>
Cc: airlied@gmail.com, daniel@ffwll.ch, matthew.brost@intel.com,
thomas.hellstrom@linux.intel.com, sarah.walker@imgtec.com,
donald.robson@imgtec.com, christian.koenig@amd.com,
faith.ekstrand@collabora.com, dri-devel@lists.freedesktop.org,
nouveau@lists.freedesktop.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH drm-misc-next v4 0/8] [RFC] DRM GPUVA Manager GPU-VM features
Date: Thu, 28 Sep 2023 14:09:13 +0200 [thread overview]
Message-ID: <20230928140913.7cb78ed4@collabora.com> (raw)
In-Reply-To: <20230920144343.64830-1-dakr@redhat.com>
On Wed, 20 Sep 2023 16:42:33 +0200
Danilo Krummrich <dakr@redhat.com> wrote:
> So far the DRM GPUVA manager offers common infrastructure to track GPU VA
> allocations and mappings, generically connect GPU VA mappings to their
> backing buffers and perform more complex mapping operations on the GPU VA
> space.
>
> However, there are more design patterns commonly used by drivers, which
> can potentially be generalized in order to make the DRM GPUVA manager
> represent a basic GPU-VM implementation. In this context, this patch series
> aims at generalizing the following elements.
>
> 1) Provide a common dma-resv for GEM objects not being used outside of
> this GPU-VM.
>
> 2) Provide tracking of external GEM objects (GEM objects which are
> shared with other GPU-VMs).
>
> 3) Provide functions to efficiently lock all GEM objects dma-resv the
> GPU-VM contains mappings of.
>
> 4) Provide tracking of evicted GEM objects the GPU-VM contains mappings
> of, such that validation of evicted GEM objects is accelerated.
>
> 5) Provide some convinience functions for common patterns.
>
> The implementation introduces struct drm_gpuvm_bo, which serves as abstraction
> combining a struct drm_gpuvm and struct drm_gem_object, similar to what
> amdgpu does with struct amdgpu_bo_vm. While this adds a bit of complexity it
> improves the efficiency of tracking external and evicted GEM objects.
>
> This patch series also renames struct drm_gpuva_manager to struct drm_gpuvm
> including corresponding functions. This way the GPUVA manager's structures align
> better with the documentation of VM_BIND [1] and VM_BIND locking [2]. It also
> provides a better foundation for the naming of data structures and functions
> introduced for implementing the features of this patch series.
>
> This patch series is also available at [3].
>
> [1] Documentation/gpu/drm-vm-bind-async.rst
> [2] Documentation/gpu/drm-vm-bind-locking.rst
> [3] https://gitlab.freedesktop.org/nouvelles/kernel/-/commits/gpuvm-next
>
> Changes in V2:
> ==============
> - rename 'drm_gpuva_manager' -> 'drm_gpuvm' which generally leads to more
> consistent naming
> - properly separate commits (introduce common dma-resv, drm_gpuvm_bo
> abstraction, etc.)
> - remove maple tree for tracking external objects, use a list drm_gpuvm_bos
> per drm_gpuvm instead
> - rework dma-resv locking helpers (Thomas)
> - add a locking helper for a given range of the VA space (Christian)
> - make the GPUVA manager buildable as module, rather than drm_exec
> builtin (Christian)
>
> Changes in V3:
> ==============
> - rename missing function and files (Boris)
> - warn if vm_obj->obj != obj in drm_gpuva_link() (Boris)
> - don't expose drm_gpuvm_bo_destroy() (Boris)
> - unlink VM_BO from GEM in drm_gpuvm_bo_destroy() rather than
> drm_gpuva_unlink() and link within drm_gpuvm_bo_obtain() to keep
> drm_gpuvm_bo instances unique
> - add internal locking to external and evicted object lists to support drivers
> updating the VA space from within the fence signalling critical path (Boris)
> - unlink external objects and evicted objects from the GPUVM's list in
> drm_gpuvm_bo_destroy()
> - add more documentation and fix some kernel doc issues
>
> Changes in V4:
> ==============
> - add a drm_gpuvm_resv() helper (Boris)
> - add a drm_gpuvm::<list_name>::local_list field (Boris)
> - remove drm_gpuvm_bo_get_unless_zero() helper (Boris)
> - fix missing NULL assignment in get_next_vm_bo_from_list() (Boris)
> - keep a drm_gem_object reference on potential vm_bo destroy (alternatively we
> could free the vm_bo and drop the vm_bo's drm_gem_object reference through
> async work)
> - introduce DRM_GPUVM_RESV_PROTECTED flag to indicate external locking through
> the corresponding dma-resv locks to optimize for drivers already holding
> them when needed; add the corresponding lock_assert_held() calls (Thomas)
> - make drm_gpuvm_bo_evict() per vm_bo and add a drm_gpuvm_bo_gem_evict()
> helper (Thomas)
> - pass a drm_gpuvm_bo in drm_gpuvm_ops::vm_bo_validate() (Thomas)
> - documentation fixes
>
> Danilo Krummrich (8):
> drm/gpuvm: rename struct drm_gpuva_manager to struct drm_gpuvm
> drm/gpuvm: allow building as module
> drm/nouveau: uvmm: rename 'umgr' to 'base'
> drm/gpuvm: add common dma-resv per struct drm_gpuvm
> drm/gpuvm: add an abstraction for a VM / BO combination
> drm/gpuvm: add drm_gpuvm_flags to drm_gpuvm
> drm/gpuvm: generalize dma_resv/extobj handling and GEM validation
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
> drm/nouveau: GPUVM dma-resv/extobj handling, GEM validation
>
> drivers/gpu/drm/Kconfig | 7 +
> drivers/gpu/drm/Makefile | 2 +-
> drivers/gpu/drm/drm_debugfs.c | 16 +-
> drivers/gpu/drm/drm_gpuva_mgr.c | 1725 --------------
> drivers/gpu/drm/drm_gpuvm.c | 2600 +++++++++++++++++++++
> drivers/gpu/drm/nouveau/Kconfig | 1 +
> drivers/gpu/drm/nouveau/nouveau_bo.c | 4 +-
> drivers/gpu/drm/nouveau/nouveau_debugfs.c | 2 +-
> drivers/gpu/drm/nouveau/nouveau_exec.c | 52 +-
> drivers/gpu/drm/nouveau/nouveau_exec.h | 4 -
> drivers/gpu/drm/nouveau/nouveau_gem.c | 5 +-
> drivers/gpu/drm/nouveau/nouveau_sched.h | 4 +-
> drivers/gpu/drm/nouveau/nouveau_uvmm.c | 207 +-
> drivers/gpu/drm/nouveau/nouveau_uvmm.h | 8 +-
> include/drm/drm_debugfs.h | 6 +-
> include/drm/drm_gem.h | 32 +-
> include/drm/drm_gpuva_mgr.h | 706 ------
> include/drm/drm_gpuvm.h | 1142 +++++++++
> 18 files changed, 3934 insertions(+), 2589 deletions(-)
> delete mode 100644 drivers/gpu/drm/drm_gpuva_mgr.c
> create mode 100644 drivers/gpu/drm/drm_gpuvm.c
> delete mode 100644 include/drm/drm_gpuva_mgr.h
> create mode 100644 include/drm/drm_gpuvm.h
>
>
> base-commit: 1c7a387ffef894b1ab3942f0482dac7a6e0a909c
next prev parent reply other threads:[~2023-09-28 12:09 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-20 14:42 [Nouveau] [PATCH drm-misc-next v4 0/8] [RFC] DRM GPUVA Manager GPU-VM features Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 1/8] drm/gpuvm: rename struct drm_gpuva_manager to struct drm_gpuvm Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-21 6:48 ` [Nouveau] " Christian König
2023-09-21 6:48 ` Christian König
2023-09-21 6:48 ` Christian König
2023-09-25 0:42 ` [Nouveau] " Dave Airlie
2023-09-25 0:42 ` Dave Airlie
2023-09-25 0:42 ` Dave Airlie
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 2/8] drm/gpuvm: allow building as module Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-25 0:42 ` [Nouveau] " Dave Airlie
2023-09-25 0:42 ` Dave Airlie
2023-09-25 0:42 ` Dave Airlie
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 3/8] drm/nouveau: uvmm: rename 'umgr' to 'base' Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-25 0:43 ` [Nouveau] " Dave Airlie
2023-09-25 0:43 ` Dave Airlie
2023-09-25 0:43 ` Dave Airlie
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 4/8] drm/gpuvm: add common dma-resv per struct drm_gpuvm Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-21 7:39 ` [Nouveau] " Christian König
2023-09-21 7:39 ` Christian König
2023-09-21 7:39 ` Christian König
2023-09-21 13:34 ` [Nouveau] " Danilo Krummrich
2023-09-21 13:34 ` Danilo Krummrich
2023-09-21 13:34 ` Danilo Krummrich
2023-09-21 14:21 ` [Nouveau] " Christian König
2023-09-21 14:21 ` Christian König
2023-09-21 14:21 ` Christian König
2023-09-21 14:25 ` Boris Brezillon
2023-09-21 14:25 ` Boris Brezillon
2023-09-21 14:34 ` [Nouveau] " Christian König
2023-09-21 14:34 ` Christian König
2023-09-21 14:34 ` Christian König
2023-09-21 15:27 ` Boris Brezillon
2023-09-21 15:27 ` Boris Brezillon
2023-09-21 15:30 ` [Nouveau] " Danilo Krummrich
2023-09-21 15:30 ` Danilo Krummrich
2023-09-21 15:30 ` Danilo Krummrich
2023-09-21 14:38 ` [Nouveau] " Danilo Krummrich
2023-09-21 14:38 ` Danilo Krummrich
2023-09-21 14:38 ` Danilo Krummrich
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 5/8] drm/gpuvm: add an abstraction for a VM / BO combination Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 6/8] drm/gpuvm: add drm_gpuvm_flags to drm_gpuvm Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 16:40 ` [Nouveau] " kernel test robot
2023-09-20 16:40 ` kernel test robot
2023-09-20 16:40 ` kernel test robot
2023-09-22 11:42 ` Boris Brezillon
2023-09-22 11:42 ` Boris Brezillon
2023-09-22 11:58 ` Boris Brezillon
2023-09-22 11:58 ` Boris Brezillon
2023-09-27 16:52 ` [Nouveau] " Danilo Krummrich
2023-09-27 16:52 ` Danilo Krummrich
2023-09-27 16:52 ` Danilo Krummrich
2023-09-28 12:19 ` Boris Brezillon
2023-09-28 12:19 ` Boris Brezillon
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 7/8] drm/gpuvm: generalize dma_resv/extobj handling and GEM validation Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-22 11:45 ` Boris Brezillon
2023-09-22 11:45 ` Boris Brezillon
2023-09-27 16:59 ` [Nouveau] " Danilo Krummrich
2023-09-27 16:59 ` Danilo Krummrich
2023-09-27 16:59 ` Danilo Krummrich
2023-09-20 14:42 ` [Nouveau] [PATCH drm-misc-next v4 8/8] drm/nouveau: GPUVM dma-resv/extobj handling, " Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-20 14:42 ` Danilo Krummrich
2023-09-28 12:09 ` Boris Brezillon [this message]
2023-09-28 12:09 ` [PATCH drm-misc-next v4 0/8] [RFC] DRM GPUVA Manager GPU-VM features Boris Brezillon
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=20230928140913.7cb78ed4@collabora.com \
--to=boris.brezillon@collabora.com \
--cc=christian.koenig@amd.com \
--cc=dakr@redhat.com \
--cc=donald.robson@imgtec.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=faith.ekstrand@collabora.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matthew.brost@intel.com \
--cc=nouveau@lists.freedesktop.org \
--cc=sarah.walker@imgtec.com \
--cc=thomas.hellstrom@linux.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.