Hi!
This patch fixes the other bug I've mentioned. Subtracting 0x1p32
doesn't work in all cases correctly, so this patch changes it to do what we
do for scalar -O2 -m32 -msse2 -mfpmath=sse double -> uint and float -> uint
conversions, in particular subtract just 0x1p31 instead of 0x1p32,
doing the comparison using signalling instead of non-signalling insn
(after all, trying to convert qNaN to unsigned int is undefined behavior)
which means we can emit it even with just -msse2, and at the end xoring
in 0x80000000 into the integer if the float/double was bigger or equal than
0x1p31.
Ok for trunk?
2011-11-03 Jakub Jelinek <ja...@redhat.com>
* config/i386/i386.c (ix86_expand_adjust_ufix_to_sfix_si): Add
XORP argument. Subtract 0x1p31 instead of 0x1p32. Use normal
signalling comparison instead of non-signalling. Store into
*XORP pseudo holding 0x80000000 integers if 0x1p31 has been
subtracted and 0 otherwise.
* config/i386/i386-protos.h (ix86_expand_adjust_ufix_to_sfix_si):
Adjust prototype.
* config/i386/sse.md (fixuns_trunc<mode><sseintvecmodelower>2): Enable
already for TARGET_SSE2. Xor in vector initialized by
ix86_expand_adjust_ufix_to_sfix_si at the end.
(vec_pack_ufix_trunc_<mode>): Likewise.
* gcc.dg/torture/vec-cvt-1.c: Enable flttointtestui test.
--- gcc/config/i386/i386.c.jj 2011-11-03 17:11:27.000000000 +0100
+++ gcc/config/i386/i386.c 2011-11-03 21:59:21.000000000 +0100
@@ -17018,16 +17018,17 @@ ix86_expand_convert_uns_sisf_sse (rtx ta
/* Adjust a V*SFmode/V*DFmode value VAL so that *sfix_trunc* resp. fix_trunc*
pattern can be used on it instead of *ufix_trunc* resp. fixuns_trunc*.
- This is done by subtracting 0x1p32 from VAL if VAL is greater or equal
- (non-signalling) than 0x1p31. */
+ This is done by doing just signed conversion if < 0x1p31, and otherwise by
+ subtracting 0x1p31 first and xoring in 0x80000000 from *XORP afterwards. */
rtx
-ix86_expand_adjust_ufix_to_sfix_si (rtx val)
+ix86_expand_adjust_ufix_to_sfix_si (rtx val, rtx *xorp)
{
- REAL_VALUE_TYPE MTWO32r, TWO31r;
- rtx two31r, mtwo32r, tmp[3];
+ REAL_VALUE_TYPE TWO31r;
+ rtx two31r, tmp[4];
enum machine_mode mode = GET_MODE (val);
enum machine_mode scalarmode = GET_MODE_INNER (mode);
+ enum machine_mode intmode = GET_MODE_SIZE (mode) == 32 ? V8SImode : V4SImode;
rtx (*cmp) (rtx, rtx, rtx, rtx);
int i;
@@ -17037,22 +17038,33 @@ ix86_expand_adjust_ufix_to_sfix_si (rtx
two31r = const_double_from_real_value (TWO31r, scalarmode);
two31r = ix86_build_const_vector (mode, 1, two31r);
two31r = force_reg (mode, two31r);
- real_ldexp (&MTWO32r, &dconstm1, 32);
- mtwo32r = const_double_from_real_value (MTWO32r, scalarmode);
- mtwo32r = ix86_build_const_vector (mode, 1, mtwo32r);
- mtwo32r = force_reg (mode, mtwo32r);
switch (mode)
{
- case V8SFmode: cmp = gen_avx_cmpv8sf3; break;
- case V4SFmode: cmp = gen_avx_cmpv4sf3; break;
- case V4DFmode: cmp = gen_avx_cmpv4df3; break;
- case V2DFmode: cmp = gen_avx_cmpv2df3; break;
+ case V8SFmode: cmp = gen_avx_maskcmpv8sf3; break;
+ case V4SFmode: cmp = gen_sse_maskcmpv4sf3; break;
+ case V4DFmode: cmp = gen_avx_maskcmpv4df3; break;
+ case V2DFmode: cmp = gen_sse2_maskcmpv2df3; break;
default: gcc_unreachable ();
}
- emit_insn (cmp (tmp[0], val, two31r, GEN_INT (29)));
- tmp[1] = expand_simple_binop (mode, AND, tmp[0], mtwo32r, tmp[1],
+ tmp[3] = gen_rtx_LE (mode, two31r, val);
+ emit_insn (cmp (tmp[0], two31r, val, tmp[3]));
+ tmp[1] = expand_simple_binop (mode, AND, tmp[0], two31r, tmp[1],
0, OPTAB_DIRECT);
- return expand_simple_binop (mode, PLUS, val, tmp[1], tmp[2],
+ if (intmode == V4SImode || TARGET_AVX2)
+ *xorp = expand_simple_binop (intmode, ASHIFT,
+ gen_lowpart (intmode, tmp[0]),
+ GEN_INT (31), NULL_RTX, 0,
+ OPTAB_DIRECT);
+ else
+ {
+ rtx two31 = GEN_INT ((unsigned HOST_WIDE_INT) 1 << 31);
+ two31 = ix86_build_const_vector (intmode, 1, two31);
+ *xorp = expand_simple_binop (intmode, AND,
+ gen_lowpart (intmode, tmp[0]),
+ two31, NULL_RTX, 0,
+ OPTAB_DIRECT);
+ }
+ return expand_simple_binop (mode, MINUS, val, tmp[1], tmp[2],
0, OPTAB_DIRECT);
}
--- gcc/config/i386/i386-protos.h.jj 2011-11-03 16:11:20.000000000 +0100
+++ gcc/config/i386/i386-protos.h 2011-11-03 20:48:48.000000000 +0100
@@ -109,7 +109,7 @@ extern void ix86_expand_convert_uns_sixf
extern void ix86_expand_convert_uns_sidf_sse (rtx, rtx);
extern void ix86_expand_convert_uns_sisf_sse (rtx, rtx);
extern void ix86_expand_convert_sign_didf_sse (rtx, rtx);
-extern rtx ix86_expand_adjust_ufix_to_sfix_si (rtx);
+extern rtx ix86_expand_adjust_ufix_to_sfix_si (rtx, rtx *);
extern enum ix86_fpcmp_strategy ix86_fp_comparison_strategy (enum rtx_code);
extern void ix86_expand_fp_absneg_operator (enum rtx_code, enum machine_mode,
rtx[]);
--- gcc/config/i386/sse.md.jj 2011-11-03 17:25:22.000000000 +0100
+++ gcc/config/i386/sse.md 2011-11-03 22:09:54.000000000 +0100
@@ -2330,10 +2330,13 @@ (define_insn "fix_truncv4sfv4si2"
(define_expand "fixuns_trunc<mode><sseintvecmodelower>2"
[(match_operand:<sseintvecmode> 0 "register_operand" "")
(match_operand:VF1 1 "register_operand" "")]
- "TARGET_AVX"
+ "TARGET_SSE2"
{
- rtx tmp = ix86_expand_adjust_ufix_to_sfix_si (operands[1]);
- emit_insn (gen_fix_trunc<mode><sseintvecmodelower>2 (operands[0], tmp));
+ rtx tmp[3];
+ tmp[0] = ix86_expand_adjust_ufix_to_sfix_si (operands[1], &tmp[2]);
+ tmp[1] = gen_reg_rtx (<sseintvecmode>mode);
+ emit_insn (gen_fix_trunc<mode><sseintvecmodelower>2 (tmp[1], tmp[0]));
+ emit_insn (gen_xor<sseintvecmodelower>3 (operands[0], tmp[1], tmp[2]));
DONE;
})
@@ -3120,12 +3123,29 @@ (define_expand "vec_pack_ufix_trunc_<mod
[(match_operand:<ssepackfltmode> 0 "register_operand" "")
(match_operand:VF2 1 "register_operand" "")
(match_operand:VF2 2 "register_operand" "")]
- "TARGET_AVX"
+ "TARGET_SSE2"
{
- rtx tmp[2];
- tmp[0] = ix86_expand_adjust_ufix_to_sfix_si (operands[1]);
- tmp[1] = ix86_expand_adjust_ufix_to_sfix_si (operands[2]);
- emit_insn (gen_vec_pack_sfix_trunc_<mode> (operands[0], tmp[0], tmp[1]));
+ rtx tmp[7];
+ tmp[0] = ix86_expand_adjust_ufix_to_sfix_si (operands[1], &tmp[2]);
+ tmp[1] = ix86_expand_adjust_ufix_to_sfix_si (operands[2], &tmp[3]);
+ tmp[4] = gen_reg_rtx (<ssepackfltmode>mode);
+ emit_insn (gen_vec_pack_sfix_trunc_<mode> (tmp[4], tmp[0], tmp[1]));
+ if (<ssepackfltmode>mode == V4SImode || TARGET_AVX2)
+ {
+ tmp[5] = gen_reg_rtx (<ssepackfltmode>mode);
+ ix86_expand_vec_extract_even_odd (tmp[5], tmp[2], tmp[3], 0);
+ }
+ else
+ {
+ tmp[5] = gen_reg_rtx (V8SFmode);
+ ix86_expand_vec_extract_even_odd (tmp[5], gen_lowpart (V8SFmode, tmp[2]),
+ gen_lowpart (V8SFmode, tmp[3]), 0);
+ tmp[5] = gen_lowpart (V8SImode, tmp[5]);
+ }
+ tmp[6] = expand_simple_binop (<ssepackfltmode>mode, XOR, tmp[4], tmp[5],
+ operands[0], 0, OPTAB_DIRECT);
+ if (tmp[6] != operands[0])
+ emit_move_insn (operands[0], tmp[6]);
DONE;
})
--- gcc/testsuite/gcc.dg/torture/vec-cvt-1.c.jj 2011-11-03 17:43:11.000000000
+0100
+++ gcc/testsuite/gcc.dg/torture/vec-cvt-1.c 2011-11-03 21:39:54.000000000
+0100
@@ -197,7 +197,7 @@ main ()
flttointtestsl ();
flttointtestuc ();
flttointtestus ();
-// flttointtestui ();
+ flttointtestui ();
flttointtestul ();
inttoflttestsc ();
inttoflttestss ();
Jakub
