Am Freitag, den 08.12.2017, 16:30 +1000 schrieb Dave Airlie:
> [snip]
>
> So I haven't commited these yet, because I wanted to see if I could
> get sb to work.
Well, it was very much work in progress, I didn't expect it to be
committed as is anyway.
>
> https://cgit.freedesktop.org/~airlied/mesa/log/?h=r600-sb-lds-wip
>
> is my non functional attempt, so far, biut it gpu hangs on the nop
> shader.
I've played aound it a bit and added some hacks to make it not hang,
i.e. sb scheduls calls into any slot, but LDS read/write should go only
into SLOT_X, and not splitting up the fetch seemed to be important
(patch attached).
However, gcm moves around the LSD_OQ* loads changing the order without
changing the order of the according LDS_READ_RET calls. At least for
this the nop shader still fails.
I tried to persuade the optimizer to not reorder these move
instructions by adding a "use" to the dst-value of a node that reads
from a LDS_OQ to the next node that reads from the same queue, but to
no avail. I guess I didn't figure out how to count these extra uses
properly when the instructuions are scheduled.
Best,
Gert
diff --git a/src/gallium/drivers/r600/r600_shader.c b/src/gallium/drivers/r600/r600_shader.c
index 5236e7110c..4ec7900398 100644
diff --git a/src/gallium/drivers/r600/sb/sb_bc.h b/src/gallium/drivers/r600/sb/sb_bc.h
index 3a3bae9d44..b35671bf0f 100644
--- a/src/gallium/drivers/r600/sb/sb_bc.h
+++ b/src/gallium/drivers/r600/sb/sb_bc.h
@@ -711,6 +711,9 @@ public:
mask = 0x0F;
if (!is_cayman() && (slot_flags & AF_S))
mask |= 0x10;
+ /* Force LDS_IDX ops into SLOT_X */
+ if (op_ptr->opcode[0] == -1 && ((op_ptr->opcode[1] & 0xFF) == 0x11))
+ mask = 0x01;
return mask;
}
diff --git a/src/gallium/drivers/r600/sb/sb_bc_parser.cpp b/src/gallium/drivers/r600/sb/sb_bc_parser.cpp
index bde4bde8fe..4f7dc2dc51 100644
--- a/src/gallium/drivers/r600/sb/sb_bc_parser.cpp
+++ b/src/gallium/drivers/r600/sb/sb_bc_parser.cpp
@@ -388,7 +389,7 @@ int bc_parser::prepare_alu_group(cf_node* cf, alu_group_node *g) {
n->flags |= NF_DONT_HOIST | NF_DONT_MOVE |
NF_DONT_KILL | NF_SCHEDULE_EARLY;
} else if (flags & AF_LDS) {
- n->flags |= NF_DONT_MOVE | NF_DONT_HOIST | NF_SCHEDULE_EARLY;
+ n->flags |= NF_DONT_MOVE | NF_DONT_HOIST;
/* all non-read operations have side effects */
if (n->bc.op != LDS_OP2_LDS_READ2_RET &&
n->bc.op != LDS_OP1_LDS_READ_REL_RET &&
@@ -523,6 +524,7 @@ int bc_parser::prepare_alu_group(cf_node* cf, alu_group_node *g) {
n->src[s] = sh->get_special_value(SV_LDS_OQB);
break;
}
+ n->flags |= NF_DONT_MOVE;
} else {
switch (src.sel) {
case ALU_SRC_0:
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
