[Bug bootstrap/92002] [10 regression] -Wuninitialized warning in gcc/wide-int.cc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92002 Rainer Orth changed: What|Removed |Added CC||hubicka at gcc dot gnu.org --- Comment #1 from Rainer Orth --- A reghunt identified this patch as the culprit: * cif-code.def (MAX_INLINE_INSNS_SINGLE_O2_LIMIT, MAX_INLINE_INSNS_AUTO_O2_LIMIT): New. * ipa-inline.c (inline_insns_single, inline_insns_auto): New functions. (can_inline_edge_by_limits_p): Use it. (big_speedup_p): Use PARAM_INLINE_MIN_SPEEDUP_O2. (want_inline_small_function_p): Use O2 bounds. (edge_badness): LIkewise. * opts.c (default_options): Add OPT_finline_functions. * params.def (PARAM_INLINE_MIN_SPEEDUP_O2, PARAM_MAX_INLINE_INSNS_SINGLE_O2, PARAM_MAX_INLINE_INSNS_AUTO_O2): New parameters.
[Bug bootstrap/92002] [10 regression] -Wuninitialized warning in gcc/wide-int.cc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92002 --- Comment #2 from Jan Hubicka --- It is patch enabling auto-inlining at -O2, so we have another false positive I guess. I fixed couple of them which reproduced during x86-64 bootstrap for me.
[Bug c++/92005] New: switch code generation regression
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92005
Bug ID: 92005
Summary: switch code generation regression
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: pdimov at gmail dot com
Target Milestone: ---
The following code:
```
template struct overloaded : Ts... { using Ts::operator()...; };
template overloaded(Ts...) -> overloaded;
struct T0 {};
struct T1 {};
struct T2 {};
struct T3 {};
struct T4 {};
struct variant
{
unsigned index_;
union
{
T0 t0_;
T1 t1_;
T2 t2_;
T3 t3_;
T4 t4_;
};
};
template int visit( F f, variant const& v )
{
switch( v.index_ )
{
case 0: return f( v.t0_ );
case 1: return f( v.t1_ );
case 2: return f( v.t2_ );
case 3: return f( v.t3_ );
case 4: return f( v.t4_ );
default: __builtin_unreachable();
}
}
int do_visit(variant const& v) {
return visit(overloaded{
[](T0 val) { return 3; },
[](T1 val) { return 5; },
[](T2 val) { return 8; },
[](T3 val) { return 9; },
[](T4 val) { return 10; }
}, v);
}
```
(https://godbolt.org/z/uxQ6KF)
generates
```
do_visit(variant const&):
mov eax, DWORD PTR [rdi]
jmp [QWORD PTR .L4[0+rax*8]]
.L4:
.quad .L8
.quad .L7
.quad .L9
.quad .L5
.quad .L3
.L9:
mov eax, 8
ret
.L7:
mov eax, 5
ret
.L8:
mov eax, 3
ret
.L5:
mov eax, 9
ret
.L3:
mov eax, 10
ret
```
with the current gcc trunk on godbolt.org (g++ (Compiler-Explorer-Build) 10.0.0
20191005 (experimental)) and
```
do_visit(variant const&):
mov eax, DWORD PTR [rdi]
mov eax, DWORD PTR CSWTCH.7[0+rax*4]
ret
CSWTCH.7:
.long 3
.long 5
.long 8
.long 9
.long 10
```
with gcc 9.2.
[Bug target/91474] Internal compiler error when building mabi=32 mips64-elf cross-compiler: segfault in parallel_settings.cc
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91474 --- Comment #11 from Joey Dumont --- I can confirm that this revision fixes the issue. I'll try to lookout for the gcc-9 version where this will be released. Thank you!
[Bug fortran/92006] New: storage_size() returns incorrect value on unlimited polymorphic variable (CLASS(*)) when passed a CHARACTER variable
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92006 Bug ID: 92006 Summary: storage_size() returns incorrect value on unlimited polymorphic variable (CLASS(*)) when passed a CHARACTER variable Product: gcc Version: 7.4.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: fortran Assignee: unassigned at gcc dot gnu.org Reporter: urbanjost at comcast dot net Target Milestone: ---
[Bug fortran/92006] storage_size() returns incorrect value on unlimited polymorphic variable (CLASS(*)) when passed a CHARACTER variable
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92006 --- Comment #1 from urbanjost at comcast dot net --- I expect the following call to storage_size() to return the value 80 whether called from within a select or not. I did not see the same issue with any other type, including a type such as type string character(len=10) :: chars enddtype string program testit use iso_c_binding implicit none character(len=10) :: string='abcdefghij' call psize(string) contains subroutine psize(chars) class(*) :: chars select type(chars) type is(character(len=*)) write(*,*)'in select ',chars,storage_size(chars) end select write(*,*)'not in select',storage_size(chars) end subroutine psize end program testit Results: $ ./a.exe in select abcdefghij 80 not in select 8
[Bug fortran/92006] storage_size() returns incorrect value on unlimited polymorphic variable (CLASS(*)) when passed a CHARACTER variable
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92006 kargl at gcc dot gnu.org changed: What|Removed |Added Priority|P3 |P4 CC||kargl at gcc dot gnu.org Depends on||84006 Target Milestone|--- |10.0 --- Comment #2 from kargl at gcc dot gnu.org --- Depends on, if not a duplicate, of 84006 Referenced Bugs: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84006 [Bug 84006] [7/8/9/10 Regression] ICE in storage_size() with CLASS entity
[Bug rtl-optimization/92007] New: [9/10 Regression] ICE: verify_flow_info failed (error: EH edge crosses section boundary in bb 7)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92007
Bug ID: 92007
Summary: [9/10 Regression] ICE: verify_flow_info failed (error:
EH edge crosses section boundary in bb 7)
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Keywords: EH, ice-checking, ice-on-valid-code
Severity: normal
Priority: P3
Component: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: asolokha at gmx dot com
Target Milestone: ---
g++-10.0.0-alpha20190929 snapshot (r276272) ICEs when compiling the following
testcase w/ -O2 -fno-tree-dominator-opts -fno-tree-forwprop --param
max-cse-insns=0:
void
sb (int *);
class d4 {
public:
~d4();
void gb ();
int op () { return no; }
int wl () { return tf; }
bool ee () try { gb (); } catch (...) { return false; }
bool b1 () { return (tf == no) ? false : ee (); }
private:
int no, tf;
};
void
hs (int *v9)
{
d4 p6;
p6.gb ();
if (p6.op () > p6.wl ())
{
p6.b1 ();
sb (v9);
}
}
% g++-10.0.0-alpha20190929 -O2 -fno-tree-dominator-opts -fno-tree-forwprop
--param max-cse-insns=0 -w -c u10jac3e.cc
u10jac3e.cc: In function 'void hs(int*)':
u10jac3e.cc:28:1: error: EH edge crosses section boundary in bb 7
28 | }
| ^
during RTL pass: postreload_jump
u10jac3e.cc:28:1: internal compiler error: verify_flow_info failed
0xb5785a verify_flow_info()
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20190929/work/gcc-10-20190929/gcc/cfghooks.c:269
0x17969c6 checking_verify_flow_info
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20190929/work/gcc-10-20190929/gcc/cfghooks.h:212
0x17969c6 try_optimize_cfg
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20190929/work/gcc-10-20190929/gcc/cfgcleanup.c:3010
0x17969c6 cleanup_cfg(int)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20190929/work/gcc-10-20190929/gcc/cfgcleanup.c:3175
0x1797843 execute
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20190929/work/gcc-10-20190929/gcc/cfgcleanup.c:3315
[Bug target/92008] Build failure on cygwin
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92008 --- Comment #1 from Thomas Koenig --- Created attachment 47000 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47000&action=edit config.status
[Bug target/92008] New: Build failure on cygwin
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92008 Bug ID: 92008 Summary: Build failure on cygwin Product: gcc Version: 10.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: tkoenig at gcc dot gnu.org Target Milestone: --- Created attachment 46999 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=46999&action=edit config.log With current cygwin and current trunk, I get gcc -c -g -DHAVE_CONFIG_H -I. -I/home/ig25/trunk/intl plural.c In file included from /home/ig25/trunk/intl/plural.y:35:0: /home/ig25/trunk/intl/plural-exp.h:102:23: error: conflicting types for 'libintl_gettextparse' # define PLURAL_PARSE libintl_gettextparse ^ /home/ig25/trunk/intl/plural.y:40:25: note: in expansion of macro 'PLURAL_PARSE' # define __gettextparse PLURAL_PARSE ^~~~ plural.c:184:5: note: in expansion of macro '__gettextparse' int __gettextparse (void); ^~ /home/ig25/trunk/intl/plural-exp.h:102:23: note: previous declaration of 'libintl_gettextparse' was here # define PLURAL_PARSE libintl_gettextparse ^ /home/ig25/trunk/intl/plural-exp.h:114:12: note: in expansion of macro 'PLURAL_PARSE' extern int PLURAL_PARSE PARAMS ((void *arg)); ^~~~ /home/ig25/trunk/intl/plural-exp.h:102:23: error: conflicting types for 'libintl_gettextparse' # define PLURAL_PARSE libintl_gettextparse ^ /home/ig25/trunk/intl/plural.y:40:25: note: in expansion of macro 'PLURAL_PARSE' # define __gettextparse PLURAL_PARSE ^~~~ plural.c:63:25: note: in expansion of macro '__gettextparse' #define yyparse __gettextparse ^~ plural.c:1129:1: note: in expansion of macro 'yyparse' yyparse (void) ^~~ /home/ig25/trunk/intl/plural-exp.h:102:23: note: previous declaration of 'libintl_gettextparse' was here # define PLURAL_PARSE libintl_gettextparse ^ /home/ig25/trunk/intl/plural-exp.h:114:12: note: in expansion of macro 'PLURAL_PARSE' extern int PLURAL_PARSE PARAMS ((void *arg)); ^~~~ plural.c: In function 'libintl_gettextparse': plural.c:64:25: error: too few arguments to function '__gettextlex' #define yylex __gettextlex ^ plural.c:1298:16: note: in expansion of macro 'yylex' yychar = yylex (&yylval); ^ plural.c:64:25: note: declared here #define yylex __gettextlex ^ /home/ig25/trunk/intl/plural.y:69:12: note: in expansion of macro 'yylex' static int yylex PARAMS ((YYSTYPE *lval, const char **pexp)); ^ /home/ig25/trunk/intl/plural.y:178:29: error: 'arg' undeclared (first use in this function) ((struct parse_args *) arg)->res = $1; ^~~ /home/ig25/trunk/intl/plural.y:178:29: note: each undeclared identifier is reported only once for each function it appears in make[3]: *** [Makefile:133: plural.o] Error 1 make[3]: Leaving directory '/cygdrive/e/trunk-bin/intl'
[Bug target/92008] Build failure on cygwin
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92008 --- Comment #2 from Thomas Koenig --- Created attachment 47001 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=47001&action=edit plural.i from adding -save-temps by hand to the Makefile in intl
[Bug target/92008] Build failure on cygwin
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92008 --- Comment #3 from Thomas Koenig --- If there's anything else needed, let me know. In the meantime, back to booting Linux :-)
[Bug c++/92005] [10 Regression] switch code generation regression
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92005 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2019-10-06 CC||hubicka at gcc dot gnu.org, ||jakub at gcc dot gnu.org, ||marxin at gcc dot gnu.org Target Milestone|--- |10.0 Summary|switch code generation |[10 Regression] switch code |regression |generation regression Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r276416.
[Bug c++/92005] [10 Regression] switch code generation regression
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92005
--- Comment #2 from Peter Dimov ---
r276416 makes -O2 inline less, and -O3 does fix this specific case. However,
there appears to be some deeper issue here. I've reduced the number of cases
from 10 to 5 for the example, but when I increase them back to 10 as in
https://godbolt.org/z/VyCeWQ, gcc 9.2 still generates a simple lookup table at
-O2, whereas gcc 10 generates a jump table even at -O3:
https://godbolt.org/z/qJqNDh.
```
template struct overloaded : Ts... { using Ts::operator()...; };
template overloaded(Ts...) -> overloaded;
struct T0 {};
struct T1 {};
struct T2 {};
struct T3 {};
struct T4 {};
struct T5 {};
struct T6 {};
struct T7 {};
struct T8 {};
struct T9 {};
struct variant
{
unsigned index_;
union
{
T0 t0_;
T1 t1_;
T2 t2_;
T3 t3_;
T4 t4_;
T5 t5_;
T6 t6_;
T7 t7_;
T8 t8_;
T9 t9_;
};
};
template int visit( F f, variant const& v )
{
switch( v.index_ )
{
case 0: return f( v.t0_ );
case 1: return f( v.t1_ );
case 2: return f( v.t2_ );
case 3: return f( v.t3_ );
case 4: return f( v.t4_ );
case 5: return f( v.t5_ );
case 6: return f( v.t6_ );
case 7: return f( v.t7_ );
case 8: return f( v.t8_ );
case 9: return f( v.t9_ );
default: __builtin_unreachable();
}
}
int do_visit(variant const& v) {
return visit(overloaded{
[](T0 val) { return 3; },
[](T1 val) { return 5; },
[](T2 val) { return 8; },
[](T3 val) { return 9; },
[](T4 val) { return 10; },
[](T5 val) { return 11; },
[](T6 val) { return 12; },
[](T7 val) { return 13; },
[](T8 val) { return 14; },
[](T9 val) { return 233; }
}, v);
}
```
[Bug c++/92009] New: [10 Regression] ICE: Segmentation fault (in is_really_empty_class)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92009
Bug ID: 92009
Summary: [10 Regression] ICE: Segmentation fault (in
is_really_empty_class)
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: asolokha at gmx dot com
Target Milestone: ---
g++-10.0.0-alpha20191006 snapshot (r276640) ICEs when compiling the following
testcase reduced from test/CodeGenCXX/rtti-layout.cpp from the clang 9.0.0 test
suite:
namespace std {
class type_info {};
}
bool
a2 ()
{
return ((void **) &typeid (int))[0];
}
% g++-10.0.0-alpha20191006 -c i5vvdig1.cpp
i5vvdig1.cpp: In function 'bool a2()':
i5vvdig1.cpp:8:37: internal compiler error: Segmentation fault
8 | return ((void **) &typeid (int))[0];
| ^
0xe6abc6 crash_signal
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/toplev.c:326
0x80a407 tree_check(tree_node*, char const*, int, char const*, tree_code)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/tree.h:3266
0x80a407 is_really_empty_class(tree_node*, bool)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/class.c:8450
0x8214d3 cxx_eval_constant_expression
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:4759
0x82a6be cxx_eval_component_reference
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:2846
0x822c5f cxx_eval_constant_expression
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:5178
0x82a6be cxx_eval_component_reference
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:2846
0x822c5f cxx_eval_constant_expression
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:5178
0x829b53 cxx_eval_indirect_ref
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:3672
0x821194 cxx_eval_constant_expression
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:5003
0x824e57 cxx_eval_outermost_constant_expr
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:5713
0x828dde maybe_constant_value(tree_node*, tree_node*, bool)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/constexpr.c:5964
0x839efe cp_fully_fold(tree_node*)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/cp-gimplify.c:2215
0x9a5951 cp_build_binary_op(op_location_t const&, tree_code, tree_node*,
tree_node*, int)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/typeck.c:5597
0x9a8e4e build_binary_op(unsigned int, tree_code, tree_node*, tree_node*, bool)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/typeck.c:4316
0x9a8e4e cp_truthvalue_conversion(tree_node*)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/typeck.c:5926
0x83fd98 cp_convert(tree_node*, tree_node*, int)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/cvt.c:636
0x83fd98 cp_convert_and_check(tree_node*, tree_node*, int)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/cvt.c:655
0x7f0e08 convert_like_real
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/call.c:7520
0x7f2335 perform_implicit_conversion_flags(tree_node*, tree_node*, int, int)
/var/tmp/portage/sys-devel/gcc-10.0.0_alpha20191006/work/gcc-10-20191006/gcc/cp/call.c:11259
