https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87502
Jan Hubicka <hubicka at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |hubicka at gcc dot gnu.org
Ever confirmed|0 |1
Last reconfirmed| |2024-12-08
Status|UNCONFIRMED |NEW
--- Comment #9 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
We currently do:
{
<bb 2> [local count: 1073741824]:
MEM[(struct basic_string *)&D.35505] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35505] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35505]._M_p = &D.35505.D.25336._M_local_buf;
__builtin_memcpy (&D.35505.D.25336._M_local_buf, "abc", 3);
D.35505._M_string_length = 3;
MEM[(char_type &)&D.35505 + 11] = 0;
foo (&D.35505);
<bb 3> [local count: 1073741824]:
_33 = D.35505._M_dataplus._M_p;
if (&D.35505.D.25336._M_local_buf == _33)
goto <bb 5>; [18.09%]
else
goto <bb 4>; [81.91%]
<bb 4> [local count: 879501928]:
_35 = D.35505.D.25336._M_allocated_capacity;
_36 = _35 + 1;
operator delete (_33, _36);
<bb 5> [local count: 1073741824]:
D.35505 ={v} {CLOBBER(eob)};
D.35505 ={v} {CLOBBER(eos)};
MEM[(struct basic_string *)&D.35539] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35539] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35539]._M_p = &D.35539.D.25336._M_local_buf;
__builtin_memcpy (&D.35539.D.25336._M_local_buf, "abc", 3);
D.35539._M_string_length = 3;
MEM[(char_type &)&D.35539 + 11] = 0;
foo (&D.35539);
<bb 6> [local count: 1073741824]:
_48 = D.35539._M_dataplus._M_p;
if (&D.35539.D.25336._M_local_buf == _48)
goto <bb 8>; [18.09%]
else
goto <bb 7>; [81.91%]
<bb 7> [local count: 879501928]:
_50 = D.35539.D.25336._M_allocated_capacity;
_51 = _50 + 1;
operator delete (_48, _51);
<bb 8> [local count: 1073741824]:
D.35539 ={v} {CLOBBER(eob)};
D.35539 ={v} {CLOBBER(eos)};
return;
<bb 9> [count: 0]:
<L0>:
std::__cxx11::basic_string<char>::_M_dispose (&D.35505);
D.35505 ={v} {CLOBBER(eob)};
_52 = __builtin_eh_pointer (4);
__builtin_unwind_resume (_52);
<bb 10> [count: 0]:
<L4>:
std::__cxx11::basic_string<char>::_M_dispose (&D.35539);
D.35539 ={v} {CLOBBER(eob)};
_37 = __builtin_eh_pointer (12);
__builtin_unwind_resume (_37);
}
so memcpy calls are still not updated to also copy the zero termination. I
think store merging may do this, but it is also quite specific pattern, so
perhaps handling it on libstdc++ side makes sense. The string constructor knows
source is zero terminated.
The reason why full destructor is inlined is that we do not know what foo is
doing and it may make the string bigger. "const" does not promise that the
callee does not modify the object.
So I think all we can hope for is merging memcpy with the extra write of 0.
Updating testcase:
#include <string>
__attribute__ ((pure))
extern int foo (const std::string &);
int
bar ()
{
return foo ("abc") + foo (std::string("abc"));
}
here GCC knows object is read only and constructor is optimized away:
int bar ()
{
void * D.39810;
struct string D.35539;
const struct string D.35505;
int _5;
int _8;
int _9;
void * _14;
<bb 2> [local count: 1073741824]:
MEM[(struct basic_string *)&D.35505] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35505] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35505]._M_p = &D.35505.D.25336._M_local_buf;
__builtin_memcpy (&D.35505.D.25336._M_local_buf, "abc", 3);
D.35505._M_string_length = 3;
MEM[(char_type &)&D.35505 + 11] = 0;
_5 = foo (&D.35505);
<bb 3> [local count: 1073741824]:
MEM[(struct basic_string *)&D.35539] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35539] ={v} {CLOBBER(bob)};
MEM[(struct _Alloc_hider *)&D.35539]._M_p = &D.35539.D.25336._M_local_buf;
__builtin_memcpy (&D.35539.D.25336._M_local_buf, "abc", 3);
D.35539._M_string_length = 3;
MEM[(char_type &)&D.35539 + 11] = 0;
_8 = foo (&D.35539);
<bb 4> [local count: 1073741824]:
_9 = _5 + _8;
D.35539 ={v} {CLOBBER(eob)};
D.35539 ={v} {CLOBBER(eos)};
D.35505 ={v} {CLOBBER(eob)};
D.35505 ={v} {CLOBBER(eos)};
return _9;
<bb 5> [count: 0]:
<L9>:
std::__cxx11::basic_string<char>::_M_dispose (&D.35539);
D.35539 ={v} {CLOBBER(eob)};
__builtin_eh_copy_values (4, 8);
<bb 6> [count: 0]:
<L13>:
D.35539 ={v} {CLOBBER(eos)};
std::__cxx11::basic_string<char>::_M_dispose (&D.35505);
D.35505 ={v} {CLOBBER(eob)};
_14 = __builtin_eh_pointer (4);
__builtin_unwind_resume (_14);
}
_Z3barv:
.LFB1393:
.cfi_startproc
.cfi_personality 0,__gxx_personality_v0
.cfi_lsda 0,.LLSDA1393
pushl %ebp
.cfi_def_cfa_offset 8
.cfi_offset 5, -8
movl $25185, %edx
movl %esp, %ebp
.cfi_def_cfa_register 5
pushl %edi
pushl %esi
leal -64(%ebp), %eax
pushl %ebx
.cfi_offset 7, -12
.cfi_offset 6, -16
.cfi_offset 3, -20
leal -72(%ebp), %ebx
subl $88, %esp
movl %eax, -72(%ebp)
movw %dx, -64(%ebp)
movb $99, -62(%ebp)
movl $3, -68(%ebp)
movb $0, -61(%ebp)
pushl %ebx
.LEHB0:
.cfi_escape 0x2e,0x10
call _Z3fooRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
.LEHE0:
movl %eax, %edi
leal -40(%ebp), %eax
leal -48(%ebp), %esi
movb $99, -38(%ebp)
movl %eax, -48(%ebp)
movl $25185, %eax
movw %ax, -40(%ebp)
movl $3, -44(%ebp)
movb $0, -37(%ebp)
movl %esi, (%esp)
.LEHB1:
call _Z3fooRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
.LEHE1:
addl $16, %esp
leal -12(%ebp), %esp
addl %edi, %eax
popl %ebx
.cfi_remember_state
.cfi_restore 3
popl %esi
.cfi_restore 6
popl %edi
.cfi_restore 7
popl %ebp
.cfi_restore 5
.cfi_def_cfa 4, 4
ret
moving using 25185 through register is done to avoid instruction encoding
to be greater than 8 bytes since it is slow on some (quite old) CPUs and we
still have this flag enabled on generic tuning. With -mtune=znver5 it produces:
_Z3barv:
.LFB1393:
.cfi_startproc
.cfi_personality 0,__gxx_personality_v0
.cfi_lsda 0,.LLSDA1393
pushl %ebp
.cfi_def_cfa_offset 8
.cfi_offset 5, -8
movl %esp, %ebp
.cfi_def_cfa_register 5
pushl %edi
pushl %esi
pushl %ebx
subl $88, %esp
.cfi_offset 7, -12
.cfi_offset 6, -16
.cfi_offset 3, -20
leal -72(%ebp), %ebx
leal -64(%ebp), %eax
movw $25185, -64(%ebp)
movb $99, -62(%ebp)
movl $3, -68(%ebp)
movl %eax, -72(%ebp)
movb $0, -61(%ebp)
pushl %ebx
.LEHB0:
.cfi_escape 0x2e,0x10
call _Z3fooRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
.LEHE0:
movl %eax, %edi
leal -48(%ebp), %esi
leal -40(%ebp), %eax
movw $25185, -40(%ebp)
movb $99, -38(%ebp)
movl $3, -44(%ebp)
movl %eax, -48(%ebp)
movb $0, -37(%ebp)
movl %esi, (%esp)
.LEHB1:
call _Z3fooRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
.LEHE1:
addl $16, %esp
leal -12(%ebp), %esp
addl %edi, %eax
popl %ebx
.cfi_remember_state
.cfi_restore 3
popl %esi
.cfi_restore 6
popl %edi
.cfi_restore 7
popl %ebp
.cfi_restore 5
.cfi_def_cfa 4, 4
ret
