diff options
| author | Samuel Pitoiset <[email protected]> | 2019-05-02 14:56:46 +0200 |
|---|---|---|
| committer | Samuel Pitoiset <[email protected]> | 2019-05-02 15:55:46 +0200 |
| commit | 08be23bfdec9fb447c58ae48bf9cc1b91ecba128 (patch) | |
| tree | 0b0b5a9f46db642220d7a4483d4c7009cc045859 /src/amd/vulkan/si_cmd_buffer.c | |
| parent | 48107b5a2b0c911fb60a367458806e3db84283f2 (diff) | |
radv: set WD_SWITCH_ON_EOP=1 when drawing primitives from a stream output buffer
According to RadeonSI, this seems to be required by the hardware
to avoid GPU hangs. I think I just forgot to set that bit when I
implemented VK_EXT_transform_feedback.
This fixes a GPU hang with Space Engineers and DXVK.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=110291
Fixes: b4eb029062a ("radv: implement VK_EXT_transform_feedback")
Signed-off-by: Samuel Pitoiset <[email protected]>
Reviewed-by: Bas Nieuwenhuizen <[email protected]>
Diffstat (limited to 'src/amd/vulkan/si_cmd_buffer.c')
| -rw-r--r-- | src/amd/vulkan/si_cmd_buffer.c | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/src/amd/vulkan/si_cmd_buffer.c b/src/amd/vulkan/si_cmd_buffer.c index e75c6d127d6..e73c13762e5 100644 --- a/src/amd/vulkan/si_cmd_buffer.c +++ b/src/amd/vulkan/si_cmd_buffer.c @@ -561,6 +561,7 @@ radv_prims_for_vertices(struct radv_prim_vertex_count *info, unsigned num) uint32_t si_get_ia_multi_vgt_param(struct radv_cmd_buffer *cmd_buffer, bool instanced_draw, bool indirect_draw, + bool count_from_stream_output, uint32_t draw_vertex_count) { enum chip_class chip_class = cmd_buffer->device->physical_device->rad_info.chip_class; @@ -622,6 +623,12 @@ si_get_ia_multi_vgt_param(struct radv_cmd_buffer *cmd_buffer, (instanced_draw || indirect_draw)) partial_vs_wave = true; + /* Hardware requirement when drawing primitives from a stream + * output buffer. + */ + if (count_from_stream_output) + wd_switch_on_eop = true; + /* If the WD switch is false, the IA switch must be false too. */ assert(wd_switch_on_eop || !ia_switch_on_eop); } |