Re: [Mesa-dev] [RFC 2/2] drm/doc/rfc: i915 new parallel submission uAPI plan

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

 



śr., 19 maj 2021 o 01:41 Matthew Brost <matthew.brost@xxxxxxxxx> napisał(a):
>
> Add entry fpr i915 new parallel submission uAPI plan.

s/fpr/for/

>
> v2:
>  (Daniel Vetter):
>   - Expand logical order explaination
>   - Add dummy header
>   - Only allow N BBs in execbuf IOCTL
>   - Configure parallel submission per slot not per gem context
>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> Cc: Tony Ye <tony.ye@xxxxxxxxx>
> CC: Carl Zhang <carl.zhang@xxxxxxxxx>
> Cc: Daniel Vetter <daniel.vetter@xxxxxxxxx>
> Cc: Jason Ekstrand <jason@xxxxxxxxxxxxxx>
> Signed-off-by: Matthew Brost <matthew.brost@xxxxxxxxx>
> ---
>  Documentation/gpu/rfc/i915_parallel_execbuf.h | 144 ++++++++++++++++++
>  Documentation/gpu/rfc/i915_scheduler.rst      |  53 ++++++-
>  2 files changed, 196 insertions(+), 1 deletion(-)
>  create mode 100644 Documentation/gpu/rfc/i915_parallel_execbuf.h
>
> diff --git a/Documentation/gpu/rfc/i915_parallel_execbuf.h b/Documentation/gpu/rfc/i915_parallel_execbuf.h
> new file mode 100644
> index 000000000000..8c64b983ccad
> --- /dev/null
> +++ b/Documentation/gpu/rfc/i915_parallel_execbuf.h
> @@ -0,0 +1,144 @@
> +#define I915_CONTEXT_ENGINES_EXT_PARALLEL_SUBMIT 2 /* see i915_context_engines_parallel_submit */
> +
> +/*
> + * i915_context_engines_parallel_submit:
> + *
> + * Setup a slot to allow multiple BBs to be submitted in a single execbuf IOCTL.
> + * Those BBs will then be scheduled to run on the GPU in parallel. Multiple
> + * hardware contexts are created internally in the i915 run these BBs. Once a
> + * slot is configured for N BBs only N BBs can be submitted in each execbuf
> + * IOCTL and this is implict behavior (e.g. the user doesn't tell the execbuf
> + * IOCTL there are N BBs, the execbuf IOCTL know how many BBs there are based on
> + * the slots configuration).
> + *
> + * Their are two currently defined ways to control the placement of the

s/Their/There/

> + * hardware contexts on physical engines: default behavior (no flags) and
> + * I915_PARALLEL_IMPLICT_BONDS (a flag). More flags may be added the in the

IMPLICIT? This typo is repeated multiple times
s/the in the/in the/

> + * future as new hardware / use cases arise. Details of how to use this
> + * interface below above the flags.

"below above"? :)

> + *
> + * Returns -EINVAL if hardware context placement configuration invalid or if the

is invalid

> + * placement configuration isn't supported on the platform / submission
> + * interface.
> + * Returns -ENODEV if extension isn't supported on the platform / submission
> + * inteface.
> + */
> +struct i915_context_engines_parallel_submit {
> +       struct i915_user_extension base;
> +
> +       __u16 engine_index;     /* slot for parallel engine */
> +       __u16 width;            /* number of contexts per parallel engine */
> +       __u16 num_siblings;     /* number of siblings per context */
> +       __u16 mbz16;
> +/*
> + * Default placement behvavior (currently unsupported):

behavior

> + *
> + * Rather than restricting parallel submission to a single class with a
> + * logically contiguous placement (I915_PARALLEL_IMPLICT_BONDS), add a mode that
> + * enables parallel submission across multiple engine classes. In this case each
> + * context's logical engine mask indicates where that context can placed. It is

can be placed

> + * implied in this mode that all contexts have mutual exclusive placement (e.g.
> + * if one context is running CS0 no other contexts can run on CS0).
> + *
> + * Example 1 pseudo code:
> + * CSX[Y] = engine class X, logical instance Y
> + * INVALID = I915_ENGINE_CLASS_INVALID, I915_ENGINE_CLASS_INVALID_NONE
> + * set_engines(INVALID)
> + * set_parallel(engine_index=0, width=2, num_siblings=2,
> + *             engines=CS0[0],CS0[1],CS1[0],CS1[1])
> + *
> + * Results in the following valid placements:
> + * CS0[0], CS1[0]
> + * CS0[0], CS1[1]
> + * CS0[1], CS1[0]
> + * CS0[1], CS1[1]
> + *
> + * This can also be though of as 2 virtual engines:

thought
(This typo repeats multiple times)

> + * VE[0] = CS0[0], CS0[1]
> + * VE[1] = CS1[0], CS1[1]
> + *
> + * Example 2 pseudo code:
> + * CS[X] = generic engine of same class, logical instance X
> + * INVALID = I915_ENGINE_CLASS_INVALID, I915_ENGINE_CLASS_INVALID_NONE
> + * set_engines(INVALID)
> + * set_parallel(engine_index=0, width=2, num_siblings=3,
> + *             engines=CS[0],CS[1],CS[2],CS[0],CS[1],CS[2])
> + *
> + * Results in the following valid placements:
> + * CS[0], CS[1]
> + * CS[0], CS[2]
> + * CS[1], CS[0]
> + * CS[1], CS[2]
> + * CS[2], CS[0]
> + * CS[2], CS[1]
> + *
> + *
> + * This can also be though of as 2 virtual engines:
> + * VE[0] = CS[0], CS[1], CS[2]
> + * VE[1] = CS[0], CS[1], CS[2]
> +
> + * This enables a use case where all engines are created equally, we don't care
> + * where they are scheduled, we just want a certain number of resources, for
> + * those resources to be scheduled in parallel, and possibly across multiple
> + * engine classes.
> + */
> +
> +/*
> + * I915_PARALLEL_IMPLICT_BONDS - Create implict bonds between each context.
> + * Each context must have the same number sibling and bonds are implictly create

number of siblings

> + * of the siblings.

"bonds are implictly create of the siblings."
-ENOPARSE

(also typo in implicitly)

> + *
> + * All of the below examples are in logical space.
> + *
> + * Example 1 pseudo code:
> + * CS[X] = generic engine of same class, logical instance X
> + * INVALID = I915_ENGINE_CLASS_INVALID, I915_ENGINE_CLASS_INVALID_NONE
> + * set_engines(INVALID)
> + * set_parallel(engine_index=0, width=2, num_siblings=1,
> + *             engines=CS[0],CS[1], flags=I915_PARALLEL_IMPLICT_BONDS)
> + *
> + * Results in the following valid placements:
> + * CS[0], CS[1]
> + *
> + * Example 2 pseudo code:
> + * CS[X] = generic engine of same class, logical instance X
> + * INVALID = I915_ENGINE_CLASS_INVALID, I915_ENGINE_CLASS_INVALID_NONE
> + * set_engines(INVALID)
> + * set_parallel(engine_index=0, width=2, num_siblings=2,
> + *             engines=CS[0],CS[2],CS[1],CS[3], flags=I915_PARALLEL_IMPLICT_BONDS)
> + *
> + * Results in the following valid placements:
> + * CS[0], CS[1]
> + * CS[2], CS[3]
> + *
> + * This can also be though of as 2 virtual engines:
> + * VE[0] = CS[0], CS[2]
> + * VE[1] = CS[1], CS[3]
> + *
> + * This enables a use case where all engines are not equal and certain placement
> + * rules are required (i.e. split-frame requires all contexts to be placed in a
> + * logically contiguous order on the VCS engines on gen11+ platforms). This use
> + * case (logically contiguous placement, within a single engine class) is
> + * supported when using GuC submission. Execlist mode could support all possible
> + * bonding configurations but currently doesn't support this extension.
> + */
> +#define I915_PARALLEL_IMPLICT_BONDS                    (1<<0)
> +/*
> + * Do not allow BBs to be preempted mid BB rather insert coordinated preemption
> + * points on all hardware contexts between each set of BBs. An example use case
> + * of this feature is split-frame on gen11+ hardware. When using this feature a
> + * BB must be submitted on each hardware context in the parallel gem context.
> + * The execbuf2 IOCTL enforces the user adheres to policy.
> + */
> +#define I915_PARALLEL_NO_PREEMPT_MID_BATCH             (1<<1)
> +#define __I915_PARALLEL_UNKNOWN_FLAGS  (-(I915_PARALLEL_NO_PREEMPT_MID_BATCH << 1))
> +       __u64 flags;            /* all undefined flags must be zero */
> +       __u64 mbz64[3];         /* reserved for future use; must be zero */
> +
> +       /*
> +        * width (i) * num_siblings (j) in length
> +        * index = j + i * num_siblings
> +        */
> +       struct i915_engine_class_instance engines[0];
> +} __attribute__ ((packed));
> +
> diff --git a/Documentation/gpu/rfc/i915_scheduler.rst b/Documentation/gpu/rfc/i915_scheduler.rst
> index 7faa46cde088..64c539486ee4 100644
> --- a/Documentation/gpu/rfc/i915_scheduler.rst
> +++ b/Documentation/gpu/rfc/i915_scheduler.rst
> @@ -82,4 +82,55 @@ https://spec.oneapi.com/level-zero/latest/core/api.html#ze-command-queue-priorit
>
>  New parallel submission uAPI
>  ============================
> -Details to come in a following patch.
> +The existing bonding uAPI is completely broken with GuC submission because
> +whether a submission is a single context submit or parallel submit isn't known
> +until execbuf time activated via the I915_SUBMIT_FENCE. To submit multiple
> +contexts in parallel with the GuC the context must be explictly registered with

explicitly

> +N contexts and all N contexts must be submitted in a single command to the GuC.
> +These interfaces doesn't support dynamically changing between N contexts as the

"this interface" or "don't support", I'm not sure what "these" refers to...

"changing" what? maybe "swapping"?

> +bonding uAPI does. Hence the need for a new parallel submission interface. Also
> +the legacy bonding uAPI is quite confusing and not intuitive at all.
> +
> +The new parallel submission uAPI consists of 3 parts:
> +
> +* Export engines logical mapping
> +* A 'set_parallel' extension to configure contexts for parallel
> +  submission
> +* Extend execbuf2 IOCTL to support submitting N BBs in a single IOCTL
> +
> +Export engines logical mapping
> +------------------------------
> +Certain use cases require BBs to be placed on engine instances in logical order
> +(e.g. split-frame on gen11+). The logical mapping of engine instances can change
> +based on fusing. Rather than making UMDs be aware of fusing, simply expose the
> +logical mapping with the existing query engine info IOCTL. Also the GuC
> +submission interface currently only supports submitting multiple contexts to
> +engines in logical order which is a new requirement compared to execlists.
> +Lastly, all current platforms have at most 2 instances and the logical order is

2 instances of what?

> +the same a uABI order. This will change on platforms with more than 2 instances.

probably s/a uABI/as uAPI/ (note: 2 typos)

> +
> +A single bit will be added to drm_i915_engine_info.flags indicating that the
> +logical instance has been returned and a new field,
> +drm_i915_engine_info.logical_instance, returns the logical instance.
> +
> +A 'set_parallel' extension to configure contexts for parallel submission
> +------------------------------------------------------------------------
> +The 'set_parallel' extension configures a slot for parallel submission of N BBs.
> +It is setup step that should be called before using any of the contexts. See
> +I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE or I915_CONTEXT_ENGINES_EXT_BOND for
> +similar existing examples. Once a slot is configured for parallel submission the
> +execbuf2 IOCTL can be called submiting N BBs in a single IOCTL. Initially only

submitting

> +support GuC submission. Execlist support can be added later if needed.
> +
> +Add I915_CONTEXT_ENGINES_EXT_PARALLEL_SUBMIT and
> +i915_context_engines_parallel_submit to the uAPI to implement this extension.
> +
> +Extend execbuf2 IOCTL to support submitting N BBs in a single IOCTL
> +-------------------------------------------------------------------
> +Contexts that have been configured with the 'set_parallel' extension are allowed
> +to submit N BBs in a single execbuf2 IOCTL. The BBs are either the last N
> +objects in the drm_i915_gem_exec_object2 list or the first N if
> +I915_EXEC_BATCH_FIRST is set. The number of BBs is implict based on the slot
> +submitted and how it has been configured by 'set_parallel' or other extensions.
> +No uAPI changes in the execbuf IOCTL but worth mentioning the new behavior of
> +the IOCTL.

worth mentioning where? :)

> --
> 2.28.0




[Index of Archives]     [Linux DRI Users]     [Linux Intel Graphics]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux