On fr., aug. 31, 2018 at 7:35 PM, Gurchetan Singh <gurchetansi...@chromium.org> wrote:
On Thu, Aug 30, 2018 at 6:41 AM Erik Faye-Lund <erik.faye-l...@collabora.com> wrote:From: Tomeu Vizoso <tomeu.viz...@collabora.com>Emulating atomics on top of ssbos can lead to too small max SSBO count, so let's use the hw-atomics mechanism to expose atomic buffers instead.Signed-off-by: Erik Faye-Lund <erik.faye-l...@collabora.com> ---src/gallium/drivers/virgl/virgl_context.c | 37 ++++++++++++++++++++++src/gallium/drivers/virgl/virgl_context.h | 2 ++ src/gallium/drivers/virgl/virgl_encode.c | 23 ++++++++++++++ src/gallium/drivers/virgl/virgl_encode.h | 3 ++ src/gallium/drivers/virgl/virgl_hw.h | 5 +++ src/gallium/drivers/virgl/virgl_protocol.h | 9 ++++++ src/gallium/drivers/virgl/virgl_screen.c | 14 +++++--- 7 files changed, 88 insertions(+), 5 deletions(-)diff --git a/src/gallium/drivers/virgl/virgl_context.c b/src/gallium/drivers/virgl/virgl_context.cindex edc03f5dcf..30cd0e4331 100644 --- a/src/gallium/drivers/virgl/virgl_context.c +++ b/src/gallium/drivers/virgl/virgl_context.c@@ -196,6 +196,19 @@ static void virgl_attach_res_shader_images(struct virgl_context *vctx,} }+static void virgl_attach_res_atomic_buffers(struct virgl_context *vctx)+{ + struct virgl_winsys *vws = virgl_screen(vctx->base.screen)->vws; + struct virgl_resource *res; + unsigned i; + for (i = 0; i < PIPE_MAX_SHADER_BUFFERS; i++) {Why not PIPE_MAX_HW_ATOMIC_BUFFERS?
Yeah, that seems better. Fixed locally.
+ res = virgl_resource(vctx->atomic_buffers[i]); + if (res) { + vws->emit_res(vws, vctx->cbuf, res->hw_res, FALSE); + } + } +} + /* * after flushing, the hw context still has a bunch of * resources bound, so we need to rebind those here.@@ -214,6 +227,7 @@ static void virgl_reemit_res(struct virgl_context *vctx)virgl_attach_res_shader_buffers(vctx, shader_type); virgl_attach_res_shader_images(vctx, shader_type); } + virgl_attach_res_atomic_buffers(vctx); virgl_attach_res_vertex_buffers(vctx); virgl_attach_res_so_targets(vctx); }@@ -952,6 +966,28 @@ static void virgl_blit(struct pipe_context *ctx,blit); } +static void virgl_set_hw_atomic_buffers(struct pipe_context *ctx, + unsigned start_slot, + unsigned count,+ const struct pipe_shader_buffer *buffers)nit: mixing tabs and spaces
Thanks, fixed locally.
+{ + struct virgl_context *vctx = virgl_context(ctx); + + for (unsigned i = 0; i < count; i++) { + unsigned idx = start_slot + i; + + if (buffers) { + if (buffers[i].buffer) { + pipe_resource_reference(&vctx->atomic_buffers[idx], + buffers[i].buffer); + continue; + } + } + pipe_resource_reference(&vctx->atomic_buffers[idx], NULL); + }+ virgl_encode_set_hw_atomic_buffers(vctx, start_slot, count, buffers);+} + static void virgl_set_shader_buffers(struct pipe_context *ctx, enum pipe_shader_type shader,unsigned start_slot, unsigned count, @@ -1209,6 +1245,7 @@ struct pipe_context *virgl_context_create(struct pipe_screen *pscreen,vctx->base.blit = virgl_blit; vctx->base.set_shader_buffers = virgl_set_shader_buffers; + vctx->base.set_hw_atomic_buffers = virgl_set_hw_atomic_buffers; vctx->base.set_shader_images = virgl_set_shader_images; vctx->base.memory_barrier = virgl_memory_barrier;diff --git a/src/gallium/drivers/virgl/virgl_context.h b/src/gallium/drivers/virgl/virgl_context.hindex 38d1f450e1..20988baa3c 100644 --- a/src/gallium/drivers/virgl/virgl_context.h +++ b/src/gallium/drivers/virgl/virgl_context.h @@ -75,6 +75,8 @@ struct virgl_context { int num_draws; struct list_head to_flush_bufs;+ struct pipe_resource *atomic_buffers[PIPE_MAX_HW_ATOMIC_BUFFERS];+ struct primconvert_context *primconvert; uint32_t hw_sub_ctx_id; };diff --git a/src/gallium/drivers/virgl/virgl_encode.c b/src/gallium/drivers/virgl/virgl_encode.cindex bcb14d8939..c9cc099061 100644 --- a/src/gallium/drivers/virgl/virgl_encode.c +++ b/src/gallium/drivers/virgl/virgl_encode.c@@ -958,6 +958,29 @@ int virgl_encode_set_shader_buffers(struct virgl_context *ctx,return 0; } +int virgl_encode_set_hw_atomic_buffers(struct virgl_context *ctx,+ unsigned start_slot, unsigned count, + const struct pipe_shader_buffer *buffers)+{ + int i;+ virgl_encoder_write_cmd_dword(ctx, VIRGL_CMD0(VIRGL_CCMD_SET_ATOMIC_BUFFERS, 0, VIRGL_SET_ATOMIC_BUFFER_SIZE(count)));+ + virgl_encoder_write_dword(ctx->cbuf, start_slot); + for (i = 0; i < count; i++) { + if (buffers) {+ struct virgl_resource *res = virgl_resource(buffers[i].buffer); + virgl_encoder_write_dword(ctx->cbuf, buffers[i].buffer_offset); + virgl_encoder_write_dword(ctx->cbuf, buffers[i].buffer_size);+ virgl_encoder_write_res(ctx, res);nit: mixing tabs and spaces
Thanks, fixed locally.
+ } else { + virgl_encoder_write_dword(ctx->cbuf, 0); + virgl_encoder_write_dword(ctx->cbuf, 0); + virgl_encoder_write_dword(ctx->cbuf, 0); + } + } + return 0; +} + int virgl_encode_set_shader_images(struct virgl_context *ctx, enum pipe_shader_type shader,unsigned start_slot, unsigned count, diff --git a/src/gallium/drivers/virgl/virgl_encode.h b/src/gallium/drivers/virgl/virgl_encode.hindex 999123f426..40e62d453b 100644 --- a/src/gallium/drivers/virgl/virgl_encode.h +++ b/src/gallium/drivers/virgl/virgl_encode.h@@ -267,6 +267,9 @@ int virgl_encode_set_shader_images(struct virgl_context *ctx,enum pipe_shader_type shader,unsigned start_slot, unsigned count, const struct pipe_image_view *images);+int virgl_encode_set_hw_atomic_buffers(struct virgl_context *ctx,+ unsigned start_slot, unsigned count, + const struct pipe_shader_buffer *buffers);int virgl_encode_memory_barrier(struct virgl_context *ctx, unsigned flags); int virgl_encode_launch_grid(struct virgl_context *ctx,diff --git a/src/gallium/drivers/virgl/virgl_hw.h b/src/gallium/drivers/virgl/virgl_hw.hindex 9a358dd844..7736ceb935 100644 --- a/src/gallium/drivers/virgl/virgl_hw.h +++ b/src/gallium/drivers/virgl/virgl_hw.h @@ -351,6 +351,11 @@ struct virgl_caps_v2 { uint32_t max_texture_2d_size; uint32_t max_texture_3d_size; uint32_t max_texture_cube_size; + uint32_t max_combined_shader_buffers; + uint32_t max_atomic_counters[6]; + uint32_t max_atomic_counter_buffers[6]; + uint32_t max_combined_atomic_counters; + uint32_t max_combined_atomic_counter_buffers; }; union virgl_caps {diff --git a/src/gallium/drivers/virgl/virgl_protocol.h b/src/gallium/drivers/virgl/virgl_protocol.hindex dd1a4ee54d..8d99c5ed47 100644 --- a/src/gallium/drivers/virgl/virgl_protocol.h +++ b/src/gallium/drivers/virgl/virgl_protocol.h @@ -91,6 +91,7 @@ enum virgl_context_cmd { VIRGL_CCMD_LAUNCH_GRID, VIRGL_CCMD_SET_FRAMEBUFFER_STATE_NO_ATTACH, VIRGL_CCMD_TEXTURE_BARRIER, + VIRGL_CCMD_SET_ATOMIC_BUFFERS, }; /* @@ -544,4 +545,12 @@ enum virgl_context_cmd { #define VIRGL_TEXTURE_BARRIER_SIZE 1 #define VIRGL_TEXTURE_BARRIER_FLAGS 1 +/* hw atomics */ +#define VIRGL_SET_ATOMIC_BUFFER_ELEMENT_SIZE 3+#define VIRGL_SET_ATOMIC_BUFFER_SIZE(x) (VIRGL_SET_ATOMIC_BUFFER_ELEMENT_SIZE * (x)) + 1+#define VIRGL_SET_ATOMIC_BUFFER_START_SLOT 1+#define VIRGL_SET_ATOMIC_BUFFER_OFFSET(x) ((x) * VIRGL_SET_ATOMIC_BUFFER_ELEMENT_SIZE + 2) +#define VIRGL_SET_ATOMIC_BUFFER_LENGTH(x) ((x) * VIRGL_SET_ATOMIC_BUFFER_ELEMENT_SIZE + 3) +#define VIRGL_SET_ATOMIC_BUFFER_RES_HANDLE(x) ((x) * VIRGL_SET_ATOMIC_BUFFER_ELEMENT_SIZE + 4)+ #endifdiff --git a/src/gallium/drivers/virgl/virgl_screen.c b/src/gallium/drivers/virgl/virgl_screen.cindex 1a9e4bc383..4bef15045a 100644 --- a/src/gallium/drivers/virgl/virgl_screen.c +++ b/src/gallium/drivers/virgl/virgl_screen.c@@ -247,6 +247,10 @@ virgl_get_param(struct pipe_screen *screen, enum pipe_cap param) return vscreen->caps.caps.v2.capability_bits & VIRGL_CAP_SHADER_CLOCK;case PIPE_CAP_TGSI_ARRAY_COMPONENTS:return vscreen->caps.caps.v2.capability_bits & VIRGL_CAP_TGSI_COMPONENTS;+ case PIPE_CAP_MAX_COMBINED_HW_ATOMIC_COUNTERS: + return vscreen->caps.caps.v2.max_combined_atomic_counters; + case PIPE_CAP_MAX_COMBINED_HW_ATOMIC_COUNTER_BUFFERS:+ return MIN2(vscreen->caps.caps.v2.max_combined_atomic_counter_buffers, 15 * 6);case PIPE_CAP_TEXTURE_GATHER_SM5: case PIPE_CAP_BUFFER_MAP_PERSISTENT_COHERENT: case PIPE_CAP_FAKE_SW_MSAA:@@ -316,15 +320,13 @@ virgl_get_param(struct pipe_screen *screen, enum pipe_cap param)case PIPE_CAP_CONSERVATIVE_RASTER_POST_DEPTH_COVERAGE: case PIPE_CAP_MAX_CONSERVATIVE_RASTER_SUBPIXEL_PRECISION_BIAS: case PIPE_CAP_PROGRAMMABLE_SAMPLE_LOCATIONS: - case PIPE_CAP_MAX_COMBINED_HW_ATOMIC_COUNTERS: - case PIPE_CAP_MAX_COMBINED_HW_ATOMIC_COUNTER_BUFFERS: return 0; case PIPE_CAP_MAX_GS_INVOCATIONS: return 32; case PIPE_CAP_MAX_SHADER_BUFFER_SIZE: return 1 << 27; case PIPE_CAP_MAX_COMBINED_SHADER_BUFFERS: - return 0;+ return MIN2(vscreen->caps.caps.v2.max_combined_shader_buffers, 16 * 6);nit: lose the MIN2, maybe we want to hit the asserts in st_init_limits if there's incorrect reporting on the host.
This isn't about correct/incorrect reporting, it's about the state-tracker having limitations here.
But thinking about it a bit more, it's probably better to clamp when the cap is queried instead, as that keeps these limitations on the same abstraction level.
So I think I will move this to st_extensions.c instead. That way, only the state tracker needs to worry about reading / writing outside the fixed-size array...
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
