KonstantinSeurer
/
mesa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

nv50/ir: when merging immediates/consts, load directly 

When a MERGE operation gets its constraint moves added, it
susbstantially extends live ranges to be reusing an immediate from
earlier in the program (not to mention the silliness of loading an
immediate into a register, and then moving into another register).

We detect these scenarios and insert moves that take the immediate or
constbuf load directly into the register. If it's the last use, then we
can just move that operation to the closer location.

With SM35 (255 regs) we get these results:

total instructions in shared programs : 6583670 -> 6580681 (-0.05%)
total gprs used in shared programs    : 950818 -> 944261 (-0.69%)
total shared used in shared programs  : 0 -> 0 (0.00%)
total local used in shared programs   : 15328 -> 15328 (0.00%)
total bytes used in shared programs   : 60367456 -> 60339896 (-0.05%)

                local     shared        gpr       inst      bytes
    helped           0           0        4584        3186        3186
      hurt           0           0          55         968         968

I suspect they will be better for SM20 and SM30.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
This commit is contained in:
Ilia Mirkin 2017-11-15 23:32:16 -05:00
parent 50e913b9c5
commit 3072bbef63
1 changed files with 21 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
src/gallium/drivers/nouveau/codegen/nv50_ir_ra.cpp
View File

@ -2331,9 +2331,21 @@ RegAlloc::InsertConstraintsPass::insertConstraintMoves()
assert(cst->getSrc(s)->defs.size() == 1); // still SSA
Instruction *defi = cst->getSrc(s)->defs.front()->getInsn();
bool imm = defi->op == OP_MOV &&
defi->src(0).getFile() == FILE_IMMEDIATE;
bool load = defi->op == OP_LOAD &&
defi->src(0).getFile() == FILE_MEMORY_CONST &&
!defi->src(0).isIndirect(0);
// catch some cases where don't really need MOVs
if (cst->getSrc(s)->refCount() == 1 && !defi->constrainedDefs())
if (cst->getSrc(s)->refCount() == 1 && !defi->constrainedDefs()) {
if (imm || load) {
// Move the defi right before the cst. No point in expanding
// the range.
defi->bb->remove(defi);
cst->bb->insertBefore(cst, defi);
}
continue;
}
LValue *lval = new_LValue(func, cst->src(s).getFile());
lval->reg.size = size;
@ -2341,6 +2353,14 @@ RegAlloc::InsertConstraintsPass::insertConstraintMoves()
mov = new_Instruction(func, OP_MOV, typeOfSize(size));
mov->setDef(0, lval);
mov->setSrc(0, cst->getSrc(s));
if (load) {
mov->op = OP_LOAD;
mov->setSrc(0, defi->getSrc(0));
} else if (imm) {
mov->setSrc(0, defi->getSrc(0));
}
cst->setSrc(s, mov->getDef(0));
cst->bb->insertBefore(cst, mov);