iris: fix some SO overflow query bugs and tidy the code a bit

This commit is contained in:
Kenneth Graunke 2018-11-26 14:59:17 -08:00
parent 527e5bcdc7
commit 39d1056d10

View file

@ -109,8 +109,8 @@ struct iris_query_snapshots {
struct iris_query_so_overflow {
uint64_t snapshots_landed;
struct {
uint64_t num_prims[2];
uint64_t prim_storage_needed[2];
uint64_t num_prims[2];
} stream[4];
};
@ -247,15 +247,14 @@ write_overflow_values(struct iris_context *ice, struct iris_query *q, bool end)
PIPE_CONTROL_CS_STALL |
PIPE_CONTROL_STALL_AT_SCOREBOARD);
for (uint32_t i = 0; i < count; i++) {
int s = q->index + i;
int g_idx = offsetof(struct iris_query_so_overflow,
stream[i].num_prims[end]);
stream[s].num_prims[end]);
int w_idx = offsetof(struct iris_query_so_overflow,
stream[i].prim_storage_needed[end]);
ice->vtbl.store_register_mem64(batch,
SO_NUM_PRIMS_WRITTEN(q->index + i),
stream[s].prim_storage_needed[end]);
ice->vtbl.store_register_mem64(batch, SO_NUM_PRIMS_WRITTEN(s),
q->bo, g_idx, false);
ice->vtbl.store_register_mem64(batch,
SO_PRIM_STORAGE_NEEDED(q->index + i),
ice->vtbl.store_register_mem64(batch, SO_PRIM_STORAGE_NEEDED(s),
q->bo, w_idx, false);
}
}
@ -277,6 +276,14 @@ iris_raw_timestamp_delta(uint64_t time0, uint64_t time1)
}
}
static bool
stream_overflowed(struct iris_query_so_overflow *so, int s)
{
return (so->stream[s].prim_storage_needed[1] -
so->stream[s].prim_storage_needed[0]) !=
(so->stream[s].num_prims[1] - so->stream[s].num_prims[0]);
}
static void
calculate_result_on_cpu(const struct gen_device_info *devinfo,
struct iris_query *q)
@ -298,18 +305,13 @@ calculate_result_on_cpu(const struct gen_device_info *devinfo,
q->result &= (1ull << TIMESTAMP_BITS) - 1;
break;
case PIPE_QUERY_SO_OVERFLOW_PREDICATE:
case PIPE_QUERY_SO_OVERFLOW_ANY_PREDICATE: {
struct iris_query_so_overflow *xfb =
(struct iris_query_so_overflow *)q->map;
uint32_t count = q->type == PIPE_QUERY_SO_OVERFLOW_PREDICATE ? 1 : 4;
for (int i = 0; i < count; i++) {
if ((xfb->stream[i].prim_storage_needed[1] - xfb->stream[i].prim_storage_needed[0]) !=
(xfb->stream[i].num_prims[1] - xfb->stream[i].num_prims[0]))
q->result = true;
}
q->result = stream_overflowed((void *) q->map, q->index);
break;
case PIPE_QUERY_SO_OVERFLOW_ANY_PREDICATE:
q->result = false;
for (int i = 0; i < MAX_VERTEX_STREAMS; i++)
q->result |= stream_overflowed((void *) q->map, i);
break;
}
case PIPE_QUERY_OCCLUSION_COUNTER:
case PIPE_QUERY_PRIMITIVES_GENERATED:
case PIPE_QUERY_PRIMITIVES_EMITTED:
@ -402,15 +404,23 @@ calc_overflow_for_stream(struct iris_context *ice)
}
static void
calc_overflow_to_gpr0(struct iris_context *ice, struct iris_query *q,
int count)
overflow_result_to_gpr0(struct iris_context *ice, struct iris_query *q)
{
struct iris_batch *batch = &ice->batches[IRIS_BATCH_RENDER];
ice->vtbl.load_register_imm64(batch, CS_GPR(0), 0ull);
for (int i = 0; i < count; i++) {
load_overflow_data_to_cs_gprs(ice, q, i);
if (q->type == PIPE_QUERY_SO_OVERFLOW_PREDICATE) {
load_overflow_data_to_cs_gprs(ice, q, q->index);
calc_overflow_for_stream(ice);
} else {
for (int i = 0; i < MAX_VERTEX_STREAMS; i++) {
load_overflow_data_to_cs_gprs(ice, q, i);
calc_overflow_for_stream(ice);
}
}
gpr0_to_bool(ice);
}
/**
@ -423,9 +433,7 @@ calculate_result_on_gpu(struct iris_context *ice, struct iris_query *q)
if (q->type == PIPE_QUERY_SO_OVERFLOW_PREDICATE ||
q->type == PIPE_QUERY_SO_OVERFLOW_ANY_PREDICATE) {
uint32_t count = q->type == PIPE_QUERY_SO_OVERFLOW_PREDICATE ? 1 : 4;
calc_overflow_to_gpr0(ice, q, count);
gpr0_to_bool(ice);
overflow_result_to_gpr0(ice, q);
return;
}