https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99397
Bug ID: 99397
Summary: s152 benchmark of TSVC is vectorized by clang and not
by gcc
Product: gcc
Version: 11.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: hubicka at gcc dot gnu.org
Target Milestone: ---
s152 is:
void s152s(real_t a[LEN_1D], real_t b[LEN_1D], real_t c[LEN_1D], int i)
{
a[i] += b[i] * c[i];
}
real_t s152(struct args_t * func_args)
{
// interprocedural data flow analysis
// collecting information from a subroutine
initialise_arrays(__func__);
gettimeofday(&func_args->t1, NULL);
for (int nl = 0; nl < iterations; nl++) {
for (int i = 0; i < LEN_1D; i++) {
b[i] = d[i] * e[i];
s152s(a, b, c, i);
}
dummy(a, b, c, d, e, aa, bb, cc, 0.);
}
gettimeofday(&func_args->t2, NULL);
return calc_checksum(__func__);
}
and clang11 vectorizes it as:
00000000004048b0 <s152>:
4048b0: 41 56 push %r14
4048b2: 53 push %rbx
4048b3: 50 push %rax
4048b4: 49 89 fe mov %rdi,%r14
4048b7: bf b7 e1 42 00 mov $0x42e1b7,%edi
4048bc: e8 4f 2d 01 00 call 417610 <initialise_arrays>
4048c1: 31 db xor %ebx,%ebx
4048c3: 4c 89 f7 mov %r14,%rdi
4048c6: 31 f6 xor %esi,%esi
4048c8: e8 93 c7 ff ff call 401060 <gettimeofday@plt>
4048cd: 0f 1f 00 nopl (%rax)
4048d0: 31 c0 xor %eax,%eax
4048d2: 66 2e 0f 1f 84 00 00 cs nopw 0x0(%rax,%rax,1)
4048d9: 00 00 00
4048dc: 0f 1f 40 00 nopl 0x0(%rax)
4048e0: c5 fc 28 80 00 01 4b vmovaps 0x4b0100(%rax),%ymm0
4048e7: 00
4048e8: c5 fc 28 88 20 01 4b vmovaps 0x4b0120(%rax),%ymm1
4048ef: 00
4048f0: c5 fc 59 80 00 0d 49 vmulps 0x490d00(%rax),%ymm0,%ymm0
4048f7: 00
4048f8: c5 f4 59 88 20 0d 49 vmulps 0x490d20(%rax),%ymm1,%ymm1
4048ff: 00
404900: c5 fc 29 80 00 31 43 vmovaps %ymm0,0x433100(%rax)
404907: 00
404908: c5 fc 29 88 20 31 43 vmovaps %ymm1,0x433120(%rax)
40490f: 00
404910: c5 fc 28 90 00 19 47 vmovaps 0x471900(%rax),%ymm2
404917: 00
404918: c5 fc 28 98 20 19 47 vmovaps 0x471920(%rax),%ymm3
40491f: 00
404920: c4 e2 7d a8 90 00 25 vfmadd213ps 0x452500(%rax),%ymm0,%ymm2
404927: 45 00
404929: c4 e2 75 a8 98 20 25 vfmadd213ps 0x452520(%rax),%ymm1,%ymm3
404930: 45 00
404932: c5 fc 29 90 00 25 45 vmovaps %ymm2,0x452500(%rax)
404939: 00
40493a: c5 fc 29 98 20 25 45 vmovaps %ymm3,0x452520(%rax)
404941: 00
404942: 48 83 c0 40 add $0x40,%rax
404946: 48 3d 00 f4 01 00 cmp $0x1f400,%rax
40494c: 75 92 jne 4048e0 <s152+0x30>
40494e: bf 00 25 45 00 mov $0x452500,%edi
404953: be 00 31 43 00 mov $0x433100,%esi
404958: ba 00 19 47 00 mov $0x471900,%edx
40495d: b9 00 0d 49 00 mov $0x490d00,%ecx
404962: 41 b8 00 01 4b 00 mov $0x4b0100,%r8d
404968: 41 b9 00 f5 4c 00 mov $0x4cf500,%r9d
40496e: c5 f8 57 c0 vxorps %xmm0,%xmm0,%xmm0
404972: 68 00 f5 54 00 push $0x54f500
404977: 68 00 f5 50 00 push $0x50f500
40497c: c5 f8 77 vzeroupper
40497f: e8 3c 10 01 00 call 4159c0 <dummy>
404984: 48 83 c4 10 add $0x10,%rsp
404988: 83 c3 01 add $0x1,%ebx
40498b: 81 fb a0 86 01 00 cmp $0x186a0,%ebx
404991: 0f 85 39 ff ff ff jne 4048d0 <s152+0x20>
404997: 49 83 c6 10 add $0x10,%r14
40499b: 4c 89 f7 mov %r14,%rdi
40499e: 31 f6 xor %esi,%esi
4049a0: e8 bb c6 ff ff call 401060 <gettimeofday@plt>
4049a5: bf b7 e1 42 00 mov $0x42e1b7,%edi
4049aa: 48 83 c4 08 add $0x8,%rsp
4049ae: 5b pop %rbx
4049af: 41 5e pop %r14
4049b1: e9 4a 26 02 00 jmp 427000 <calc_checksum>
4049b6: 66 2e 0f 1f 84 00 00 cs nopw 0x0(%rax,%rax,1)
4049bd: 00 00 00
We get:
real_t s152 (struct args_t * func_args)
{
int i;
int nl;
static const char __func__[5] = "s152";
struct timeval * _1;
float _2;
float _3;
float _4;
struct timeval * _5;
real_t _16;
long unsigned int _21;
long unsigned int _22;
real_t * _23;
float _24;
real_t * _25;
float _26;
real_t * _27;
float _28;
float _29;
float _30;
unsigned int ivtmp_48;
unsigned int ivtmp_49;
unsigned int ivtmp_50;
unsigned int ivtmp_51;
<bb 2> [local count: 108459]:
initialise_arrays (&__func__);
_1 = &func_args_12(D)->t1;
gettimeofday (_1, 0B);
goto <bb 5>; [100.00%]
<bb 8> [local count: 1052266996]:
<bb 3> [local count: 1063004409]:
# i_40 = PHI <i_20(8), 0(5)>
# ivtmp_51 = PHI <ivtmp_50(8), 32000(5)>
_2 = d[i_40];
_3 = e[i_40];
_4 = _2 * _3;
b[i_40] = _4;
_21 = (long unsigned int) i_40;
_22 = _21 * 4;
_23 = &a + _22;
_24 = *_23;
_25 = &b + _22;
_26 = *_25;
_27 = &c + _22;
_28 = *_27;
_29 = _26 * _28;
_30 = _24 + _29;
*_23 = _30;
i_20 = i_40 + 1;
ivtmp_50 = ivtmp_51 - 1;
if (ivtmp_50 != 0)
goto <bb 8>; [98.99%]
else
goto <bb 4>; [1.01%]
<bb 4> [local count: 10737416]:
dummy (&a, &b, &c, &d, &e, &aa, &bb, &cc, 0.0);
nl_18 = nl_39 + 1;
ivtmp_48 = ivtmp_49 - 1;
if (ivtmp_48 != 0)
goto <bb 7>; [98.99%]
else
goto <bb 6>; [1.01%]
<bb 7> [local count: 10628957]:
<bb 5> [local count: 10737416]:
# nl_39 = PHI <nl_18(7), 0(2)>
# ivtmp_49 = PHI <ivtmp_48(7), 100000(2)>
goto <bb 3>; [100.00%]
<bb 6> [local count: 108459]:
_5 = &func_args_12(D)->t2;
gettimeofday (_5, 0B);
_16 = calc_checksum (&__func__);
return _16;
}
and fail at:
tsvc.c:699:27: note: can tell at compile time that b[i_40] and *_25 alias
tsvc.c:686:14: missed: not vectorized: compilation time alias: b[i_40] = _4;
_26 = *_25;
tsvc.c:699:27: note: ***** Analysis failed with vector mode V8QI
tsvc.c:699:27: missed: couldn't vectorize loop
tsvc.c:689:8: note: vectorized 0 loops in function.
So I seem we get confused by:
b[i_40] = _4;
and
_21 = (long unsigned int) i_40;
_22 = _21 * 4;
_25 = &b + _22;
acessing same location...
