On Mon, Apr 30, 2012 at 3:10 PM, Jakub Jelinek <[email protected]> wrote:
> On Mon, Apr 30, 2012 at 02:54:05PM +0200, Uros Bizjak wrote:
>> > My recent changes to zero_extend expanders should handle this
>> > automatically, and will undo generation of zero_extend pattern. Please
>> > see zero_extend<mode>si2_and expander, and how it handles
>> > TARGET_ZERO_EXTEND_WITH_AND targets.
>>
>> Attached patch implements this idea. In addition, it fixes the
>> splitter to not change output mode of zero_extension from HImode and
>> QImode from DImode to SImode. Although they generate the same
>> instruction, I think we should better keep original mode here.
>
> Thanks. I was trying this morning slightly different patch for the same,
> but strangely it failed bootstrap, and didn't get around to analysing
> why a mem store had (zero_extend (subreg (reg))) on a RHS.
>
>> + operands[1] = gen_lowpart (mode, operands[1]);
>> +
>> + if (GET_MODE (operands[0]) == DImode)
>> + insn = (mode == SImode)
>> + ? gen_zero_extendsidi2
>> + : (mode == HImode)
>> + ? gen_zero_extendhidi2
>> + : gen_zero_extendqidi2;
>> + else if (GET_MODE (operands[0]) == SImode)
>> + insn = (mode == HImode)
>> + ? gen_zero_extendhisi2
>> + : gen_zero_extendqisi2;
>> + else if (GET_MODE (operands[0]) == HImode)
>> + insn = gen_zero_extendqihi2;
>> else
>> - ix86_expand_binary_operator (AND, <MODE>mode, operands);
>> + gcc_unreachable ();
>> +
>> + emit_insn (insn (operands[0], operands[1]));
>
> IMHO you should use <MODE>mode instead of GET_MODE (operands[0])
> in all of the above, then the compiler can actually optimize
> it at compile time.
2012-04-30 Uros Bizjak <[email protected]>
* config/i386/i386.md (and<mode>3): Change runtime operand mode checks
to compile-time "mode == <MODE>mode" checks.
(and splitter): Ditto.
Tested on x86_64-pc-linux-gnu, committed to mainline SVN.
Uros.
Index: config/i386/i386.md
===================================================================
--- config/i386/i386.md (revision 186992)
+++ config/i386/i386.md (working copy)
@@ -7695,7 +7695,7 @@
(match_operand:SWIM 2 "<general_szext_operand>")))]
""
{
- enum machine_mode mode = GET_MODE (operands[1]);
+ enum machine_mode mode = <MODE>mode;
rtx (*insn) (rtx, rtx);
if (CONST_INT_P (operands[2]) && REG_P (operands[0]))
@@ -7710,30 +7710,28 @@
mode = QImode;
}
- if (mode == GET_MODE (operands[1]))
+ if (mode == <MODE>mode)
{
ix86_expand_binary_operator (AND, <MODE>mode, operands);
DONE;
}
- operands[1] = gen_lowpart (mode, operands[1]);
-
- if (GET_MODE (operands[0]) == DImode)
+ if (<MODE>mode == DImode)
insn = (mode == SImode)
? gen_zero_extendsidi2
: (mode == HImode)
? gen_zero_extendhidi2
: gen_zero_extendqidi2;
- else if (GET_MODE (operands[0]) == SImode)
+ else if (<MODE>mode == SImode)
insn = (mode == HImode)
? gen_zero_extendhisi2
: gen_zero_extendqisi2;
- else if (GET_MODE (operands[0]) == HImode)
+ else if (<MODE>mode == HImode)
insn = gen_zero_extendqihi2;
else
gcc_unreachable ();
- emit_insn (insn (operands[0], operands[1]));
+ emit_insn (insn (operands[0], gen_lowpart (mode, operands[1])));
DONE;
})
@@ -7884,9 +7882,7 @@
mode = QImode;
}
- operands[1] = gen_lowpart (mode, operands[1]);
-
- if (GET_MODE (operands[0]) == DImode)
+ if (<MODE>mode == DImode)
insn = (mode == SImode)
? gen_zero_extendsidi2
: (mode == HImode)
@@ -7894,14 +7890,15 @@
: gen_zero_extendqidi2;
else
{
- /* Zero extend to SImode to avoid partial register stalls. */
- operands[0] = gen_lowpart (SImode, operands[0]);
+ if (<MODE>mode != SImode)
+ /* Zero extend to SImode to avoid partial register stalls. */
+ operands[0] = gen_lowpart (SImode, operands[0]);
insn = (mode == HImode)
? gen_zero_extendhisi2
: gen_zero_extendqisi2;
}
- emit_insn (insn (operands[0], operands[1]));
+ emit_insn (insn (operands[0], gen_lowpart (mode, operands[1])));
DONE;
})
