Mesa (master): vc4: Stop doing qpu_inst(add, NOP) or qpu_inst(NOP, mul).
Eric Anholt
anholt at kemper.freedesktop.org
Mon Aug 25 05:51:53 UTC 2014
Module: Mesa
Branch: master
Commit: 7317f1185932e2188206069a938c598f9cd18c60
URL: http://cgit.freedesktop.org/mesa/mesa/commit/?id=7317f1185932e2188206069a938c598f9cd18c60
Author: Eric Anholt <eric at anholt.net>
Date: Sun Aug 24 17:58:24 2014 -0700
vc4: Stop doing qpu_inst(add, NOP) or qpu_inst(NOP, mul).
Now that the extra WADDR is set, we can knock this off. Saves a lot of
typing, and makes this code much more legible.
---
src/gallium/drivers/vc4/vc4_qpu_emit.c | 131 ++++++++++++--------------------
1 file changed, 47 insertions(+), 84 deletions(-)
diff --git a/src/gallium/drivers/vc4/vc4_qpu_emit.c b/src/gallium/drivers/vc4/vc4_qpu_emit.c
index cf9cea6..9f5ccdb 100644
--- a/src/gallium/drivers/vc4/vc4_qpu_emit.c
+++ b/src/gallium/drivers/vc4/vc4_qpu_emit.c
@@ -76,8 +76,7 @@ fixup_raddr_conflict(struct qcompile *c,
if ((src0.mux == QPU_MUX_A || src0.mux == QPU_MUX_B) &&
(src1->mux == QPU_MUX_A || src1->mux == QPU_MUX_B) &&
src0.addr != src1->addr) {
- queue(c, qpu_inst(qpu_a_MOV(qpu_r3(), *src1),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_r3(), *src1));
*src1 = qpu_r3();
}
}
@@ -362,15 +361,12 @@ vc4_generate_code(struct qcompile *c)
/* Skip emitting the MOV if it's a no-op. */
if (dst.mux == QPU_MUX_A || dst.mux == QPU_MUX_B ||
dst.mux != src[0].mux || dst.addr != src[0].addr) {
- queue(c, qpu_inst(qpu_a_MOV(dst, src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, src[0]));
}
break;
case QOP_CMP:
- queue(c, qpu_inst(qpu_a_MOV(qpu_ra(QPU_W_NOP),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_ra(QPU_W_NOP), src[0]));
*last_inst(c) |= QPU_SF;
if (dst.mux <= QPU_MUX_R3) {
@@ -384,19 +380,15 @@ vc4_generate_code(struct qcompile *c)
} else {
if (dst.mux == src[1].mux &&
dst.addr == src[1].addr) {
- queue(c, qpu_inst(qpu_a_MOV(dst, src[1]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, src[1]));
- queue(c, qpu_inst(qpu_a_MOV(dst, src[2]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, src[2]));
*last_inst(c) = qpu_set_cond_add(*last_inst(c),
QPU_COND_NC);
} else {
- queue(c, qpu_inst(qpu_a_MOV(dst, src[2]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, src[2]));
- queue(c, qpu_inst(qpu_a_MOV(dst, src[1]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, src[1]));
*last_inst(c) = qpu_set_cond_add(*last_inst(c),
QPU_COND_NS);
}
@@ -408,9 +400,7 @@ vc4_generate_code(struct qcompile *c)
case QOP_SGE:
case QOP_SLT:
fixup_raddr_conflict(c, src[0], &src[1]);
- queue(c, qpu_inst(qpu_a_FSUB(qpu_ra(QPU_W_NOP),
- src[0], src[1]),
- qpu_NOP()));
+ queue(c, qpu_a_FSUB(qpu_ra(QPU_W_NOP), src[0], src[1]));
*last_inst(c) |= QPU_SF;
queue(c, qpu_load_imm_f(dst, 0.0));
@@ -422,13 +412,11 @@ vc4_generate_code(struct qcompile *c)
break;
case QOP_VPM_WRITE:
- queue(c, qpu_inst(qpu_a_MOV(qpu_ra(QPU_W_VPM), src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_ra(QPU_W_VPM), src[0]));
break;
case QOP_VPM_READ:
- queue(c, qpu_inst(qpu_a_MOV(dst, qpu_ra(QPU_R_VPM)),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, qpu_ra(QPU_R_VPM)));
break;
case QOP_RCP:
@@ -437,86 +425,72 @@ vc4_generate_code(struct qcompile *c)
case QOP_LOG2:
switch (qinst->op) {
case QOP_RCP:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIP),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIP),
+ src[0]));
break;
case QOP_RSQ:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIPSQRT),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIPSQRT),
+ src[0]));
break;
case QOP_EXP2:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_SFU_EXP),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_SFU_EXP),
+ src[0]));
break;
case QOP_LOG2:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_SFU_LOG),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_SFU_LOG),
+ src[0]));
break;
default:
abort();
}
- queue(c, qpu_inst(qpu_a_MOV(dst, qpu_r4()),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, qpu_r4()));
break;
case QOP_PACK_COLORS:
for (int i = 0; i < 4; i++) {
- queue(c, qpu_inst(qpu_NOP(),
- qpu_m_MOV(qpu_r3(), src[i])));
+ queue(c, qpu_m_MOV(qpu_r3(), src[i]));
*last_inst(c) |= QPU_PM;
*last_inst(c) |= QPU_SET_FIELD(QPU_PACK_MUL_8A + i,
QPU_PACK);
}
- queue(c, qpu_inst(qpu_a_MOV(dst, qpu_r3()),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, qpu_r3()));
break;
case QOP_FRAG_X:
- queue(c, qpu_inst(qpu_a_ITOF(dst,
- qpu_ra(QPU_R_XY_PIXEL_COORD)),
- qpu_NOP()));
+ queue(c, qpu_a_ITOF(dst,
+ qpu_ra(QPU_R_XY_PIXEL_COORD)));
break;
case QOP_FRAG_Y:
- queue(c, qpu_inst(qpu_a_ITOF(dst,
- qpu_rb(QPU_R_XY_PIXEL_COORD)),
- qpu_NOP()));
+ queue(c, qpu_a_ITOF(dst,
+ qpu_rb(QPU_R_XY_PIXEL_COORD)));
break;
case QOP_FRAG_Z:
- queue(c, qpu_inst(qpu_a_ITOF(dst,
- qpu_rb(QPU_R_FRAG_PAYLOAD_ZW)),
- qpu_NOP()));
+ queue(c, qpu_a_ITOF(dst,
+ qpu_rb(QPU_R_FRAG_PAYLOAD_ZW)));
break;
case QOP_FRAG_RCP_W:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIP),
- qpu_ra(QPU_R_FRAG_PAYLOAD_ZW)),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_SFU_RECIP),
+ qpu_ra(QPU_R_FRAG_PAYLOAD_ZW)));
- queue(c, qpu_inst(qpu_a_MOV(dst, qpu_r4()),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, qpu_r4()));
break;
case QOP_TLB_DISCARD_SETUP:
discard = true;
- queue(c, qpu_inst(qpu_a_MOV(src[0], src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(src[0], src[0]));
*last_inst(c) |= QPU_SF;
break;
case QOP_TLB_PASSTHROUGH_Z_WRITE:
- queue(c, qpu_inst(qpu_a_MOV(qpu_ra(QPU_W_TLB_Z),
- qpu_rb(QPU_R_FRAG_PAYLOAD_ZW)),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_ra(QPU_W_TLB_Z),
+ qpu_rb(QPU_R_FRAG_PAYLOAD_ZW)));
if (discard) {
*last_inst(c) = qpu_set_cond_add(*last_inst(c),
QPU_COND_ZS);
@@ -531,9 +505,7 @@ vc4_generate_code(struct qcompile *c)
break;
case QOP_TLB_COLOR_WRITE:
- queue(c, qpu_inst(qpu_a_MOV(qpu_tlbc(),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_tlbc(), src[0]));
if (discard) {
*last_inst(c) = qpu_set_cond_add(*last_inst(c),
QPU_COND_ZS);
@@ -541,18 +513,14 @@ vc4_generate_code(struct qcompile *c)
break;
case QOP_VARY_ADD_C:
- queue(c, qpu_inst(qpu_a_FADD(dst,
- src[0], qpu_r5()),
- qpu_NOP()));
+ queue(c, qpu_a_FADD(dst, src[0], qpu_r5()));
break;
case QOP_PACK_SCALED: {
- uint64_t a = (qpu_inst(qpu_a_MOV(dst, src[0]),
- qpu_NOP()) |
+ uint64_t a = (qpu_a_MOV(dst, src[0]) |
QPU_SET_FIELD(QPU_PACK_A_16A,
QPU_PACK));
- uint64_t b = (qpu_inst(qpu_a_MOV(dst, src[1]),
- qpu_NOP()) |
+ uint64_t b = (qpu_a_MOV(dst, src[1]) |
QPU_SET_FIELD(QPU_PACK_A_16B,
QPU_PACK));
@@ -570,11 +538,9 @@ vc4_generate_code(struct qcompile *c)
case QOP_TEX_T:
case QOP_TEX_R:
case QOP_TEX_B:
- queue(c, qpu_inst(qpu_a_MOV(qpu_rb(QPU_W_TMU0_S +
- (qinst->op -
- QOP_TEX_S)),
- src[0]),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(qpu_rb(QPU_W_TMU0_S +
+ (qinst->op - QOP_TEX_S)),
+ src[0]));
break;
case QOP_TEX_RESULT:
@@ -588,8 +554,7 @@ vc4_generate_code(struct qcompile *c)
case QOP_R4_UNPACK_B:
case QOP_R4_UNPACK_C:
case QOP_R4_UNPACK_D:
- queue(c, qpu_inst(qpu_a_MOV(dst, qpu_r4()),
- qpu_NOP()));
+ queue(c, qpu_a_MOV(dst, qpu_r4()));
*last_inst(c) |= QPU_PM;
*last_inst(c) |= QPU_SET_FIELD(QPU_UNPACK_R4_8A +
(qinst->op -
@@ -612,15 +577,13 @@ vc4_generate_code(struct qcompile *c)
fixup_raddr_conflict(c, src[0], &src[1]);
if (translate[qinst->op].is_mul) {
- queue(c, qpu_inst(qpu_NOP(),
- qpu_m_alu2(translate[qinst->op].op,
- dst,
- src[0], src[1])));
+ queue(c, qpu_m_alu2(translate[qinst->op].op,
+ dst,
+ src[0], src[1]));
} else {
- queue(c, qpu_inst(qpu_a_alu2(translate[qinst->op].op,
- dst,
- src[0], src[1]),
- qpu_NOP()));
+ queue(c, qpu_a_alu2(translate[qinst->op].op,
+ dst,
+ src[0], src[1]));
}
break;
}
More information about the mesa-commit
mailing list