radv: don't use SQ_NON_EVENT before GE_PC_ALLOC for better perf on Navi1x

Seems it make the perf worse.
Ported from RadeonSI.

Signed-off-by: Samuel Pitoiset <samuel.pitoiset@gmail.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12466>
This commit is contained in:
Samuel Pitoiset 2021-08-19 08:40:19 +02:00 committed by Marge Bot
parent 77ffdf41b1
commit 2dc90ca8a4

View file

@ -1986,12 +1986,6 @@ radv_get_num_input_vertices(nir_shader **nir)
static void
gfx10_emit_ge_pc_alloc(struct radeon_cmdbuf *cs, enum chip_class chip_class, uint32_t oversub_pc_lines)
{
if (chip_class == GFX10) {
/* SQ_NON_EVENT must be emitted before GE_PC_ALLOC is written. */
radeon_emit(cs, PKT3(PKT3_EVENT_WRITE, 0, 0));
radeon_emit(cs, EVENT_TYPE(V_028A90_SQ_NON_EVENT) | EVENT_INDEX(0));
}
radeon_set_uconfig_reg(
cs, R_030980_GE_PC_ALLOC,
S_030980_OVERSUB_EN(oversub_pc_lines > 0) | S_030980_NUM_PC_LINES(oversub_pc_lines - 1));