[Mesa-dev] [PATCH v2 08/14] gallium: add PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK

[Nicolai Hähnle]

On 11.09.2016 20:45, Samuel Pitoiset wrote:
> Signed-off-by: Samuel Pitoiset <samuel.pitoiset at gmail.com>
> ---
>  src/gallium/docs/source/screen.rst             | 4 ++++
>  src/gallium/drivers/ilo/ilo_screen.c           | 2 ++
>  src/gallium/drivers/nouveau/nv50/nv50_screen.c | 2 ++
>  src/gallium/drivers/nouveau/nvc0/nvc0_screen.c | 2 ++
>  src/gallium/drivers/radeon/r600_pipe_common.c  | 1 +
>  src/gallium/drivers/softpipe/sp_screen.c       | 1 +
>  src/gallium/include/pipe/p_defines.h           | 3 ++-
>  7 files changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/src/gallium/docs/source/screen.rst b/src/gallium/docs/source/screen.rst
> index 5dff650..cfc0a1b 100644
> --- a/src/gallium/docs/source/screen.rst
> +++ b/src/gallium/docs/source/screen.rst
> @@ -498,6 +498,10 @@ pipe_screen::get_compute_param.
>    threads. Also known as wavefront size, warp size or SIMD width.
>  * ``PIPE_COMPUTE_CAP_ADDRESS_BITS``: The default compute device address space
>    size specified as an unsigned integer value in bits.
> +* ``PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK``: Maximum variable number
> +  of threads that a single block can contain. This is similar to
> +  PIPE_COMPUTE_CAP_MAX_THREADS_PER_BLOCK, except that the variable size is not
> +  known a compile-time but at dispatch-time.
>
>  .. _pipe_bind:
>
> diff --git a/src/gallium/drivers/ilo/ilo_screen.c b/src/gallium/drivers/ilo/ilo_screen.c
> index b9e5ad6..85357fa 100644
> --- a/src/gallium/drivers/ilo/ilo_screen.c
> +++ b/src/gallium/drivers/ilo/ilo_screen.c
> @@ -303,6 +303,8 @@ ilo_get_compute_param(struct pipe_screen *screen,
>        ptr = &val.subgroup_size;
>        size = sizeof(val.subgroup_size);
>        break;
> +   case PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK:
> +      /* fallthrough */
>     default:
>        ptr = NULL;
>        size = 0;
> diff --git a/src/gallium/drivers/nouveau/nv50/nv50_screen.c b/src/gallium/drivers/nouveau/nv50/nv50_screen.c
> index 1ec791d..6eb18ea 100644
> --- a/src/gallium/drivers/nouveau/nv50/nv50_screen.c
> +++ b/src/gallium/drivers/nouveau/nv50/nv50_screen.c
> @@ -418,6 +418,8 @@ nv50_screen_get_compute_param(struct pipe_screen *pscreen,
>        RET((uint32_t []) { 512 }); /* FIXME: arbitrary limit */
>     case PIPE_COMPUTE_CAP_ADDRESS_BITS:
>        RET((uint32_t []) { 32 });
> +   case PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK:
> +      RET((uint64_t []) { 0 });
>     default:
>        return 0;
>     }
> diff --git a/src/gallium/drivers/nouveau/nvc0/nvc0_screen.c b/src/gallium/drivers/nouveau/nvc0/nvc0_screen.c
> index 1757cbb..df6c6af 100644
> --- a/src/gallium/drivers/nouveau/nvc0/nvc0_screen.c
> +++ b/src/gallium/drivers/nouveau/nvc0/nvc0_screen.c
> @@ -478,6 +478,8 @@ nvc0_screen_get_compute_param(struct pipe_screen *pscreen,
>        RET((uint32_t []) { 512 }); /* FIXME: arbitrary limit */
>     case PIPE_COMPUTE_CAP_ADDRESS_BITS:
>        RET((uint32_t []) { 64 });
> +   case PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK:
> +      RET((uint64_t []) { 0 });
>     default:
>        return 0;
>     }
> diff --git a/src/gallium/drivers/radeon/r600_pipe_common.c b/src/gallium/drivers/radeon/r600_pipe_common.c
> index 6d7cc1b..a7c6729 100644
> --- a/src/gallium/drivers/radeon/r600_pipe_common.c
> +++ b/src/gallium/drivers/radeon/r600_pipe_common.c
> @@ -955,6 +955,7 @@ static int r600_get_compute_param(struct pipe_screen *screen,
>  		}
>  		return sizeof(uint32_t);
>  	case PIPE_COMPUTE_CAP_MAX_PRIVATE_SIZE:
> +	case PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK:
>  		break; /* unused */
>  	case PIPE_COMPUTE_CAP_SUBGROUP_SIZE:
>  		if (ret) {
> diff --git a/src/gallium/drivers/softpipe/sp_screen.c b/src/gallium/drivers/softpipe/sp_screen.c
> index cd4269f..29e898b 100644
> --- a/src/gallium/drivers/softpipe/sp_screen.c
> +++ b/src/gallium/drivers/softpipe/sp_screen.c
> @@ -522,6 +522,7 @@ softpipe_get_compute_param(struct pipe_screen *_screen,
>     case PIPE_COMPUTE_CAP_IMAGES_SUPPORTED:
>     case PIPE_COMPUTE_CAP_SUBGROUP_SIZE:
>     case PIPE_COMPUTE_CAP_ADDRESS_BITS:
> +   case PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK;
>        break;

Now that I'm actually trying to compile this: there's a semi-colon here ;p

Nicolai

>     }
>     return 0;
> diff --git a/src/gallium/include/pipe/p_defines.h b/src/gallium/include/pipe/p_defines.h
> index 88aa050..655995e 100644
> --- a/src/gallium/include/pipe/p_defines.h
> +++ b/src/gallium/include/pipe/p_defines.h
> @@ -847,7 +847,8 @@ enum pipe_compute_cap
>     PIPE_COMPUTE_CAP_MAX_CLOCK_FREQUENCY,
>     PIPE_COMPUTE_CAP_MAX_COMPUTE_UNITS,
>     PIPE_COMPUTE_CAP_IMAGES_SUPPORTED,
> -   PIPE_COMPUTE_CAP_SUBGROUP_SIZE
> +   PIPE_COMPUTE_CAP_SUBGROUP_SIZE,
> +   PIPE_COMPUTE_CAP_MAX_VARIABLE_THREADS_PER_BLOCK,
>  };
>
>  /**
>