[Bug c++/70796] New: [DR 1030] Initialization order with braced-init-lists still broken

[rs2740 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70796

Bug ID: 70796
   Summary: [DR 1030] Initialization order with braced-init-lists
still broken
   Product: gcc
   Version: 7.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: rs2740 at gmail dot com
  Target Milestone: ---

The following modified test case (replacing postfix ++ with prefix and
adjusting the expected values accordingly) from PR61382 still aborts on Wandbox
(http://melpon.org/wandbox/permlink/IQojTl16DtIxCd2M) and also gives a bogus
-Wsequence-point warning:

struct A
{
  int i,j;
  A(int i,int j):i(i),j(j){}
};

int main()
{
  int i = 0;
  A a{++i, ++i};
  if (a.i != 1 || a.j != 2)
__builtin_abort();
}

[Bug c++/66561] __builtin_LINE at al. should yield constant expressions

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66561

Martin Sebor  changed:

   What|Removed |Added

 Status|UNCONFIRMED |ASSIGNED
   Last reconfirmed||2016-04-26
   Assignee|unassigned at gcc dot gnu.org  |msebor at gcc dot 
gnu.org
 Ever confirmed|0   |1
  Known to fail||4.9.3, 5.3.0, 6.0

--- Comment #1 from Martin Sebor  ---
Working on a patch for GCC 7.

[Bug middle-end/70795] New: [7 Regression] gcc/libjava/interpret.cc:1948:1: ICE: in binds_to_current_def_p, at symtab.c:2232

[danglin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70795

Bug ID: 70795
   Summary: [7 Regression] gcc/libjava/interpret.cc:1948:1: ICE:
in binds_to_current_def_p, at symtab.c:2232
   Product: gcc
   Version: 7.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: middle-end
  Assignee: unassigned at gcc dot gnu.org
  Reporter: danglin at gcc dot gnu.org
  Target Milestone: ---
  Host: hppa2.0w-hp-hpux11.11
Target: hppa2.0w-hp-hpux11.11
 Build: hppa2.0w-hp-hpux11.11

libtool: compile:  /test/gnu/gcc/objdir/./gcc/xgcc -shared-libgcc
-B/test/gnu/gcc/objdir/./gcc -nostdinc++
-L/test/gnu/gcc/objdir/hppa2.0w-hp-hpux11.11/libstdc+
+-v3/src -L/test/gnu/gcc/objdir/hppa2.0w-hp-hpux11.11/libstdc++-v3/src/.libs
-L/test/gnu/gcc/objdir/hppa2.0w-hp-hpux11.11/libstdc++-v3/libsupc++/.libs
-B/opt/gnu/gcc/gcc-7/hppa2.0w-hp-hpux11.11/bin/
-B/opt/gnu/gcc/gcc-7/hppa2.0w-hp-hpux11.1
1/lib/ -isystem /opt/gnu/gcc/gcc-7/hppa2.0w-hp-hpux11.11/include -isystem
/opt/gnu/gcc/gcc-7/hppa2.0w-hp-hpux11.11/sys-include -DHAVE_CONFIG_H -I.
-I../../../gcc/libjava -I./include -I./gcj -I../../../gcc/libjava -Iinclude
-I../../../gcc/li
bjava/include -I../../../gcc/libjava/classpath/include -Iclasspath/include
-I../../../gcc/libjava/classpath/native/fdlibm
-I../../../gcc/libjava/../boehm-gc/include -I../boehm-gc/include
-I../../../gcc/libjava/libltdl -I../../../gcc/libjava
/libltdl -I../../../gcc/libjava/.././libjava/../libgcc
-I../../../gcc/libjava/../zlib -I../../../gcc/libjava/../libffi/include
-I../libffi/include -fno-rtti -fnon-call-exceptions -pthread
-fdollars-in-identifiers -Wswitch-enum -D_FILE_OFFSE
T_BITS=64 -Wextra -Wall -D_GNU_SOURCE -DPREFIX=\"/opt/gnu/gcc/gcc-7\"
-DTOOLEXEC
LIBDIR=\"/opt/gnu/gcc/gcc-7/lib\" -DJAVA_HOME=\"/opt/gnu/gcc/gcc-7\"
-DBOOT_CLAS
S_PATH=\"/opt/gnu/gcc/gcc-7/share/java/libgcj-7.0.0.jar\"
-DJAVA_EXT_DIRS=\"/opt
/gnu/gcc/gcc-7/share/java/ext\"
-DGCJ_ENDORSED_DIRS=\"/opt/gnu/gcc/gcc-7/share/j
ava/gcj-endorsed\"
-DGCJ_VERSIONED_LIBDIR=\"/opt/gnu/gcc/gcc-7/lib/gcj-7.0.0-17\
" -DPATH_SEPARATOR=\":\" -DECJ_JAR_FILE=\"\"
-DLIBGCJ_DEFAULT_DATABASE=\"/opt/gn
u/gcc/gcc-7/lib/gcj-7.0.0-17/classmap.db\"
-DLIBGCJ_DEFAULT_DATABASE_PATH_TAIL=\
"gcj-7.0.0-17/classmap.db\" -fwrapv -g -O2 -MT interpret.lo -MD -MP -MF
.deps/in
terpret.Tpo -c ../../../gcc/libjava/interpret.cc  -fPIC -DPIC -o
.libs/interpret.o
_GLOBAL__I_65535_0_.._.._.._gcc_libjava_interpret.cc_E29B42DF_0x4efdcb59/1100
() @7aeeb5e8
  Type: function definition analyzed
  Visibility: force_output public artificial constructor
  References: 
  Referring: 
  Availability: available
  First run: 0
  Function flags: body static_constructor (priority:65535)
  Called by: 
  Calls: _GLOBAL__sub_I_interpret.cc/861 (1.00 per call) 
../../../gcc/libjava/interpret.cc: In function '':
../../../gcc/libjava/interpret.cc:1948:1: internal compiler error: in
binds_to_current_def_p, at symtab.c:2232
 }
 ^

This is revision 235394.

[Bug libstdc++/70794] New: vector.push_back() crashes with std::bad_alloc after 2^32 calls

[numien at deathwyrm dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70794

Bug ID: 70794
   Summary: vector.push_back() crashes with std::bad_alloc after
2^32 calls
   Product: gcc
   Version: 5.3.1
Status: UNCONFIRMED
  Severity: minor
  Priority: P3
 Component: libstdc++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: numien at deathwyrm dot com
  Target Milestone: ---

Created attachment 38340
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=38340=edit
Minimal test program

Despite std::vector::max_size() returning 18446744073709551615, a test program
to repeatedly call std::vector::push_back() crashes on the 4294967297th
iteration with a std::bad_alloc exception.

I am compiling in 64-bit mode, and my system should have plenty of memory. The
fact it's reaching exactly 2^32 indicates to me that it's likely a bug.

The test program I used is attached.

[Bug c++/70792] Incorrect sequence point warning with uniform initializer syntax

[lefticus at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70792

--- Comment #2 from Jason Turner  ---
Follow up, this is a better test case that does not pass by &&


struct MyType {
  MyType(int i, int j, int k, int l)
: sum(i + j + k + l)
  {

  }

  int sum;
};

int main()
{
  int i = 0;
  std::cout << MyType{ ++i, ++i, ++i, ++i }.sum << '\n';
}

[Bug target/70098] PowerPC64: eigen hits ICE following invalid register assignment

[wschmidt at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70098

--- Comment #11 from Bill Schmidt  ---
Author: wschmidt
Date: Mon Apr 25 22:29:49 2016
New Revision: 235424

URL: https://gcc.gnu.org/viewcvs?rev=235424=gcc=rev
Log:
[gcc]

2016-04-25  Bill Schmidt  

Backport from mainline
2016-03-14  Segher Boessenkool  

PR target/70098
* config/rs6000/rs6000.md (*ctr_internal1, *ctr_internal2,
*ctr_internal5, *ctr_internal6): Also allow "d" as output.
(define_split for the GPR case): Use int_reg_operand instead of
gpc_reg_operand for the output.

[gcc/testsuite]

2016-04-25  Bill Schmidt  

Backport from mainline
2016-03-14  Segher Boessenkool  

PR target/70098
* lib/target-supports.exp (check_effective_target_powerpc64_no_dm):
New function.
* g++.dg/pr70098.C: New testcase.


Added:
branches/gcc-4_9-branch/gcc/testsuite/g++.dg/pr70098.C
Modified:
branches/gcc-4_9-branch/gcc/ChangeLog
branches/gcc-4_9-branch/gcc/config/rs6000/rs6000.md
branches/gcc-4_9-branch/gcc/testsuite/ChangeLog
branches/gcc-4_9-branch/gcc/testsuite/lib/target-supports.exp

[Bug target/70098] PowerPC64: eigen hits ICE following invalid register assignment

[wschmidt at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70098

--- Comment #10 from Bill Schmidt  ---
Author: wschmidt
Date: Mon Apr 25 22:28:36 2016
New Revision: 235423

URL: https://gcc.gnu.org/viewcvs?rev=235423=gcc=rev
Log:
[gcc]

2016-04-25  Bill Schmidt  

Backport from mainline
2016-03-14  Segher Boessenkool  

PR target/70098
* config/rs6000/rs6000.md (*ctr_internal1, *ctr_internal2,
*ctr_internal5, *ctr_internal6): Also allow "d" as output.
(define_split for the GPR case): Use int_reg_operand instead of
gpc_reg_operand for the output.

[gcc/testsuite]

2016-04-25  Bill Schmidt  

Backport from mainline
2016-03-14  Segher Boessenkool  

PR target/70098
* lib/target-supports.exp (check_effective_target_powerpc64_no_dm):
New function.
* g++.dg/pr70098.C: New testcase.


Added:
branches/gcc-5-branch/gcc/testsuite/g++.dg/pr70098.C
Modified:
branches/gcc-5-branch/gcc/ChangeLog
branches/gcc-5-branch/gcc/config/rs6000/rs6000.md
branches/gcc-5-branch/gcc/testsuite/ChangeLog
branches/gcc-5-branch/gcc/testsuite/lib/target-supports.exp

[Bug c++/70793] New: g++ does not accept some forms of "friend" declaration for builtin types

[jpmarath at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70793

Bug ID: 70793
   Summary: g++ does not accept some forms of "friend" declaration
for builtin types
   Product: gcc
   Version: 5.3.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: jpmarath at gmail dot com
  Target Milestone: ---

This fails to compile:

struct S {
friend short  ;
} s;

with gcc 5.3.1, the reported error is (compile with -std=c++14):

---
2 : error: friend declaration does not name a class or function
friend short ;
^
Compilation failed
---

The strange thing is that g++ accepts "friend short int;" :

struct S {
friend short int;
} s;


clang++ 3.8 accepts both forms.

[Bug c++/70792] Incorrect sequence point warning with uniform initializer syntax

[lefticus at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70792

--- Comment #1 from Jason Turner  ---
Note: I tested this code further and the compiler seems to be generating
incorrect code based on this standard, not just warning incorrectly.

[Bug c++/70792] New: Incorrect sequence point warning with uniform initializer syntax

[lefticus at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70792

Bug ID: 70792
   Summary: Incorrect sequence point warning with uniform
initializer syntax
   Product: gcc
   Version: 6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: lefticus at gmail dot com
  Target Milestone: ---

The following code:


#include 

struct Do_Call
{
  template
  Do_Call(const Func , T && ... t)
  {
f(std::forward(t)...);
  }
};

int main()
{
  const auto sum = [](int w, int x, int y, int z) {
return w + x + y + z;
  };

  int i = 0;
  Do_Call{sum, ++i, ++i, ++i, ++i};
}

Generates the -Wsequence-point warning for the variable 'i'. However, the
standard states:

"Within the initializer-list of a braced-init-list, the initializer-clauses,
including any that result from pack expansions (14.5.3), are evaluated in the
order in which they appear. That is, every value computation and side effect
associated with a given initializer-clause is sequenced before every value
computation and side effect associated with any initializer-clause that follows
it in the comma-separated list of the initializer-list. [Note: This evaluation
ordering holds regardless of the semantics of the initialization; for example,
it applies when the elements of the initializer-list are interpreted as
arguments of a constructor call, even though ordinarily there are no sequencing
constraints on the arguments of a call. —end note ]"

The "Note: This evaluation ordering holds regardless of the semantics of the
initialization; for example, it applies when the elements of the
initializer-list are interpreted as arguments of a constructor call, even
though ordinarily there are no sequencing constraints on the arguments of a
call. —end note" is the most applicable to this bug report.

This occurs in all versions tested between 4.7.3->6

[Bug c/70791] New: -Wnested-externs prints inconsistent column number

[egall at gwmail dot gwu.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70791

Bug ID: 70791
   Summary: -Wnested-externs prints inconsistent column number
   Product: gcc
   Version: 6.0
Status: UNCONFIRMED
  Keywords: diagnostic
  Severity: normal
  Priority: P3
 Component: c
  Assignee: unassigned at gcc dot gnu.org
  Reporter: egall at gwmail dot gwu.edu
  Target Milestone: ---
  Host: i386-apple-darwin9.8.0
Target: i386-apple-darwin9.8.0
 Build: i386-apple-darwin9.8.0

Say I have the following file:

$ cat undeclared.c
char *foo(char *);

int bar(void) {
char *baz = foo(UNDECLARED_MACRO(""));
}

When I compile it like this, gcc says:

$ /usr/local/bin/gcc -c -Wimplicit -Wnested-externs -Wint-conversion
undeclared.c
undeclared.c: In function ‘bar’:
undeclared.c:4:18: warning: implicit declaration of function ‘UNDECLARED_MACRO’
[-Wimplicit-function-declaration]
  char *baz = foo(UNDECLARED_MACRO(""));
  ^
undeclared.c:4:2: warning: nested extern declaration of ‘UNDECLARED_MACRO’
[-Wnested-externs]
  char *baz = foo(UNDECLARED_MACRO(""));
  ^
undeclared.c:4:18: warning: passing argument 1 of ‘foo’ makes pointer from
integer without a cast [-Wint-conversion]
  char *baz = foo(UNDECLARED_MACRO(""));
  ^
undeclared.c:1:7: note: expected ‘char *’ but argument is of type ‘int’
 char *foo(char *);
   ^

I would expect the warning from -Wnested-externs to point to the same column as
the other 2 warnings. 
(Of course, ideally the 3 would be combined into a single warning, but that's a
separate issue)

My gcc version info:

$ /usr/local/bin/gcc -v
Using built-in specs.
COLLECT_GCC=/usr/local/bin/gcc
COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/i386-apple-darwin9.8.0/6.0.0/lto-wrapper
Target: i386-apple-darwin9.8.0
Configured with: ../configure --disable-werror
--enable-languages=c,c++,lto,objc,obj-c++ --enable-stage1-checking=release -C
--with-system-libunwind --enable-secureplt --enable-frame-pointer
--enable-version-specific-runtime-libs --enable-assert --enable-fat
--enable-portable-binary --enable-browser-plugin --enable-gconf-peer
--enable-libgcj-debug --enable-gc-debug --enable-interpreter
--enable-aot-compile-rpm --enable-java-home --with-x --enable-collections
--enable-gstreamer-peer --enable-xmlj --enable-qt-peer --enable-debug
--enable-local-sockets --without-isl --enable-objc-gc --enable-maintainer-mode
--enable-host-shared CC=/usr/local/bin/gcc CXX=/usr/local/bin/g++
AUTOCONF=/usr/local/bin/autoconf AUTOHEADER=/usr/local/bin/autoheader
AUTOM4TE=/usr/local/bin/autom4te AUTORECONF=/usr/local/bin/autoreconf
AUTOSCAN=/usr/local/bin/autoscan AUTOUPDATE=/usr/local/bin/autoupdate
IFNAMES=/usr/local/bin/ifnames
Thread model: posix
gcc version 6.0.0 20151007 (experimental) (GCC)

[Bug middle-end/32667] block copy with exact overlap is expanded as memcpy

[ch3root at openwall dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=32667

--- Comment #17 from Alexander Cherepanov  ---
On 2016-04-25 17:34, joseph at codesourcery dot com wrote:
> Or it could simply document an additional requirement on the C library
> used with GCC (that the case of exact overlap works), just as it documents
> the requirement that some functions such as memcpy be provided even in the
> freestanding case.  The compiler and library implementations need to
> cooperate to produce a conforming C implementation.

1. Is glibc community ready to provide such guarantee? The doc[1] 
doesn't currently mention an exception for exact overlap.

[1] 
https://www.gnu.org/software/libc/manual/html_node/Copying-Strings-and-Arrays.html#index-memcpy

2. I agree that it should be documented. Together with the assumption 
that memcpy doesn't clobber errno.

3. The decision affects the whole Free Software community. If memcpy 
with exact overlap inserted be gcc are legitimized then users cannot use 
tools like valgrind to catch such cases in their own code. (And neither 
_FORTIFY_SOURCE nor UBSan seem to catch overlapping memcpy now.) So it 
effectively leads to legitimizing such memcpy in user's code too. Making 
memcpy closer to a simple assignment (C11, 6.5.16.1p3) could be a good 
thing. Maybe even fix C std or POSIX?


BTW llvm bug is here: https://llvm.org/bugs/show_bug.cgi?id=11763 . It 
also contains links to valgrind discussions.

[Bug c++/70790] New: Can't mangle noexcept expressions

[eric.niebler at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70790

Bug ID: 70790
   Summary: Can't mangle noexcept expressions
   Product: gcc
   Version: 5.3.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: eric.niebler at gmail dot com
  Target Milestone: ---

#include 

template
void foo(T, typename std::enable_if::type)
{
}

int main() {
  foo(0,0);
}


Results in:

Test.cpp: In instantiation of ‘void foo(T, typename std::enable_if())), int>::type) [with T = int; typename
std::enable_if())), int>::type = int]’:

Test.cpp:19:6: sorry, unimplemented: mangling noexcept_expr
 void foo(T, typename std::enable_if::type)
  ^

[Bug middle-end/70773] Profiling makes sudoku solver slower

[tulipawn at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70773

--- Comment #5 from PeteVine  ---
The issue seems to be purely about soft division. (I was either using no -mcpu
or -mcpu=cortex-a5)  

Compiling for e.g Cortex-A7, doesn't need to lower any library calls and even
though hardware division is not used in the vanilla case, it is after
profiling. 

The profiled binary even shows a benefit under qemu-arm so I can safely guess
the issue is non-existent when targeting similar ARM parts.

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[dje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

--- Comment #49 from David Edelsohn  ---
Can we add some testcases to ensure that -fchecking and similar flags don't
accidentally affect code generation due to future changes?

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[dje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

--- Comment #48 from David Edelsohn  ---
Commenting out the fold_non_dependent_expr call seems to work for me using the
build method that regularly was failing before.

[Bug c/12245] [4.9/5/6/7 regression] Uses lots of memory when compiling large initialized arrays

[lolisamurai at waifu dot club]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=12245

Franc[e]sco  changed:

   What|Removed |Added

 CC||lolisamurai at waifu dot club

--- Comment #59 from Franc[e]sco  ---
I can confirm this still happens in gcc 4.9.3 (gentoo linux, amd64), here's an
example: https://ideone.com/J699eJ

[Bug c++/70789] New: cilk test fib-tplt.cc occasionally fails

[bernds at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70789

Bug ID: 70789
   Summary: cilk test fib-tplt.cc occasionally fails
   Product: gcc
   Version: 7.0
Status: UNCONFIRMED
  Keywords: wrong-code
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: bernds at gcc dot gnu.org
  Target Milestone: ---
Target: x86_64-pc-linux-gnu

I've had a test run fail with

FAIL: g++.dg/cilk-plus/CK/fib-tplt.cc  -g -fcilkplus execution test

This was while testing the -m32 variant. I've let the test run in a loop for a
few hours, and it seems to fail sporadically with a segfault, almost certainly
not because of the fairly harmless patch I was testing. I've tried to load the
coredump into gdb, unfortunately it's not very informative.

(gdb) bt
#0  0x1c46 in ?? ()
#1  0xf76d3587 in ?? ()
#2  0xf76d3347 in ?? ()
#3  0x in ?? ()

Tagging it as wrong-code although I'm not sure whether the testcase is valid.
We seem not to have a libcilkrts component.

[Bug c++/67247] ICE on std::forward args&& inside nested lambda function

[tower120 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67247

tower120  changed:

   What|Removed |Added

  Known to work||6.0

--- Comment #4 from tower120  ---
Reduced case [Forwarding captured variadic arguments inside lambda] :
https://godbolt.org/g/D7lXEA

template
int g(T... t)
{
return 0;
}

template
void f(Args&&... args)
{
auto lm = [&](auto&&..._args) { 
auto f = [&]{g(std::forward(_args)...);};
  //auto f = [&]{g(_args...);}; /* THIS ONE OK */
return f(); 
};
lm(args...);
}

int main()
{
f(2, 5, 7);
}

Error "_args was not declared" on 4.9.x
ICE on 5.x
Ok on 6.x

[Bug target/68273] [5/6/7 Regression] Wrong code on mips/mipsel due to (invalid?) peeking at alignments in function_arg.

[aurelien at aurel32 dot net]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68273

--- Comment #33 from Aurelien Jarno  ---
(In reply to Hector Oron from comment #32)
> (In reply to Richard Biener from comment #31)
> > eipa_sra introduces the remaining SSA name with non-default alignment via
> 
> [PATCH]
> 

For the record, Debian now have the patch from comment #32 in its version of
GCC 5 and 6. This fixes a few more issues than reported here, in addition to
graphviz, it also fixes subversion and jq.

[Bug middle-end/32667] block copy with exact overlap is expanded as memcpy

[lopresti at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=32667

--- Comment #16 from Patrick J. LoPresti  ---
Well, Valgrind itself is another real-world example. Tools like Valgrind cannot
distinguish invalid memcpy() calls by the programmer from invalid memcpy()
calls emitted by GCC.

You can, of course, redefine every standard interface you like and simply
document it. That would be awful engineering -- as in this case -- but you
could do it.

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

Jakub Jelinek  changed:

   What|Removed |Added

 CC||jason at gcc dot gnu.org

--- Comment #47 from Jakub Jelinek  ---
So, I believe it is the:
  /* When checking, try to get a constant value for all non-dependent
 expressions in order to expose bugs in *_dependent_expression_p
 and constexpr.  */
  if (flag_checking && cxx_dialect >= cxx11
  /* Don't do this during nsdmi parsing as it can lead to
 unexpected recursive instantiations.  */
  && !parsing_nsdmi ())
fold_non_dependent_expr (expr);
hunk in cp/pt.c (build_nondependent_expr), which results both in some extra
DECL_UIDs being created (in theory it could be fine), but also in funcdef_no
differences etc.:
#0  _Z24allocate_struct_functionP9tree_nodeb (fndecl=0x7061b280,
abstract_p=false) at /home/dje/src/gcc-6.0.1-RC2/gcc/function.c:4874
#1  0x10070e60 in _Z24start_preparsed_functionP9tree_nodeS0_i
(decl1=0x7061b280, attrs=0x0, flags=1)
at /home/dje/src/gcc-6.0.1-RC2/gcc/cp/decl.c:14059
#2  0x105feb9c in _Z16instantiate_declP9tree_nodeib (d=0x7061b280, defer_ok=0,
expl_inst_class_mem_p=false)
at /home/dje/src/gcc-6.0.1-RC2/gcc/cp/pt.c:21969
#3  0x1094fd38 in _Z9mark_usedP9tree_nodei (decl=0x7061b280, complain=0) at
/home/dje/src/gcc-6.0.1-RC2/gcc/cp/decl2.c:5273
#4  0x1023262c in _ZL15build_over_callP11z_candidateii (cand=0x306323c0,
flags=262145, complain=0)
at /home/dje/src/gcc-6.0.1-RC2/gcc/cp/call.c:7707
#5  0x10221f8c in
_Z23build_new_function_callP9tree_nodePP3vecIS0_5va_gc8vl_embedEbi
(fn=0x7045e468, args=0x2ff200f0, koenig_p=false, complain=0)
at /home/dje/src/gcc-6.0.1-RC2/gcc/cp/call.c:4152
#6  0x100dba64 in _Z16finish_call_exprP9tree_nodePP3vecIS0_5va_gc8vl_embedEbbi
(fn=0x7045e468, args=0x2ff200f0, disallow_virtual=false, 
koenig_p=false, complain=0) at
/home/dje/src/gcc-6.0.1-RC2/gcc/cp/semantics.c:2461
#7  0x105e30ac in _Z21tsubst_copy_and_buildP9tree_nodeS0_iS0_bb (t=0x7061e180,
args=0x0, complain=0, in_decl=0x0, function_p=false, 
integral_constant_expression_p=true) at
/home/dje/src/gcc-6.0.1-RC2/gcc/cp/pt.c:16629
#8  0x105a4bf8 in _Z39instantiate_non_dependent_expr_internalP9tree_nodei
(expr=0x7061e180, complain=0)
at /home/dje/src/gcc-6.0.1-RC2/gcc/cp/pt.c:5640
#9  0x114c1278 in _Z23fold_non_dependent_exprP9tree_node (t=0x7061e180) at
/home/dje/src/gcc-6.0.1-RC2/gcc/cp/constexpr.c:4388
#10 0x10606e08 in _Z24build_non_dependent_exprP9tree_node (expr=0x7061e180) at
/home/dje/src/gcc-6.0.1-RC2/gcc/cp/pt.c:23631

As stage1 is built with the default --enable-stage1-checking, the result
between stage1 and stage2 auto-host.h is:
CHECKING_P, ENABLE_GC_CHECKING, ENABLE_GIMPLE_CHECKING,
ENABLE_RTL_FLAG_CHECKING, ENABLE_TREE_CHECKING, ENABLE_TYPES_CHECKING, and
allocate_struct_function allocates a new cfun->funcdef_no, which is then
emitted in the LFB..*/LFE..* labels, seems to affect the debug info (which on
AIX is emitted in both stage2 and stage3) etc.

Additionally, wonder if retrieve_specialization's
  if (flag_checking)
verify_unstripped_args (args);
couldn't also affect bootstrap, but haven't seen that during bootstrap.

The question is what to do, comment out the fold_non_dependent_expr stuff
altogether on the release branch (at least temporarily), guard with some other
option (which unlike -fchecking would be documented to possibly affect code
generation and which would be based on some other configure checking flag (and
the default for that flag would be always the same between stage1 and later
stages), something different?
I'm in any case starting a bootstrap now with the fold_non_dependent_expr call
commented out.

[Bug libstdc++/70788] LaTeX formulae in doxygen comments should be suppressed in man-page output

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70788

Jonathan Wakely  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 Ever confirmed|0   |1

--- Comment #1 from Jonathan Wakely  ---
Errors from the Debian package include several due to this issue:


usr/share/man/man3/__gnu_cxx.3cxx.gz 1319: a space character is not allowed
in an escape name
usr/share/man/man3/__gnu_debug::_Safe_unordered_container_base.3cxx.gz 111:
warning [p 2, 2.2i]: can't break line
usr/share/man/man3/__gnu_parallel.3cxx.gz 2415: warning: numeric expression
expected (got `r')
   
usr/share/man/man3/__gnu_pbds::detail::left_child_next_sibling_heap_const_iterator_.3cxx.gz
11: warning [p 1, 1.8i]: can't break line
   
usr/share/man/man3/__gnu_pbds::detail::left_child_next_sibling_heap_node_point_const_iterator_.3cxx.gz
5: warning [p 1, 0.8i]: can't break line
usr/share/man/man3/std::_Hashtable.3cxx.gz 211: warning [p 2, 7.2i]: can't
break line
usr/share/man/man3/std::__debug::map.3cxx.gz 200: warning [p 2, 4.8i]:
can't break line
usr/share/man/man3/std::__debug::multimap.3cxx.gz 200: warning [p 2, 5.0i]:
can't break line
usr/share/man/man3/std::__debug::unordered_map.3cxx.gz 209: warning [p 2,
6.3i]: can't break line
usr/share/man/man3/std::__debug::unordered_multimap.3cxx.gz 209: warning [p
2, 6.7i]: can't break line
usr/share/man/man3/std::__profile::map.3cxx.gz 182: warning [p 2, 4.0i]:
can't break line
usr/share/man/man3/std::__profile::multimap.3cxx.gz 176: warning [p 2,
3.7i]: can't break line
usr/share/man/man3/std::__profile::unordered_map.3cxx.gz 119: warning [p 2,
0.0i]: can't break line
usr/share/man/man3/std::__profile::unordered_multimap.3cxx.gz 116: warning
[p 2, 0.0i]: can't break line
usr/share/man/man3/std::error_condition.3cxx.gz 17: warning [p 1, 2.0i]:
can't break line
usr/share/man/man3/std::exponential_distribution.3cxx.gz 30: warning:
numeric expression expected (got `m')
usr/share/man/man3/std::extreme_value_distribution.3cxx.gz 94: a space
character is not allowed in an escape name
usr/share/man/man3/std::fisher_f_distribution.3cxx.gz 103: a space
character is not allowed in an escape name
usr/share/man/man3/std::gamma_distribution.3cxx.gz 30: normal or special
character expected (got `\&')
usr/share/man/man3/std::linear_congruential_engine.3cxx.gz 97: a space
character is not allowed in an escape name
usr/share/man/man3/std::lognormal_distribution.3cxx.gz 99: a space
character is not allowed in an escape name
usr/share/man/man3/std::map.3cxx.gz 179: warning [p 2, 4.7i]: can't break
line
usr/share/man/man3/std::multimap.3cxx.gz 172: warning [p 2, 4.2i]: can't
break line
usr/share/man/man3/std::normal_distribution.3cxx.gz 102: a space character
is not allowed in an escape name
usr/share/man/man3/std::queue.3cxx.gz 197: warning: macro `c'' not defined
usr/share/man/man3/std::regex_iterator.3cxx.gz 42: warning [p 1, 4.0i]:
can't break line
usr/share/man/man3/std::regex_token_iterator.3cxx.gz 51: warning [p 1,
5.8i]: can't break line
usr/share/man/man3/std::student_t_distribution.3cxx.gz 100: a space
character is not allowed in an escape name
usr/share/man/man3/std::subtract_with_carry_engine.3cxx.gz 99: a space
character is not allowed in an escape name
usr/share/man/man3/std::unordered_map.3cxx.gz 306: warning [p 3, 6.0i]:
can't break line
usr/share/man/man3/std::unordered_multimap.3cxx.gz 306: warning [p 3,
6.8i]: can't break line
usr/share/man/man3/std::weibull_distribution.3cxx.gz 94: a space character
is not allowed in an escape name

[Bug libstdc++/70788] New: LaTeX formulae in doxygen comments should be suppressed in man-page output

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70788

Bug ID: 70788
   Summary: LaTeX formulae in doxygen comments should be
suppressed in man-page output
   Product: gcc
   Version: 7.0
Status: UNCONFIRMED
  Keywords: documentation
  Severity: minor
  Priority: P3
 Component: libstdc++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: redi at gcc dot gnu.org
  Target Milestone: ---

Doxygen's formula support only works for HTML and LaTeX output, so the
generated man pages contain garbled LaTeX such as "x_{i+1}tarrow(ax_{i} + c) d
]"

[Bug middle-end/32667] block copy with exact overlap is expanded as memcpy

[joseph at codesourcery dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=32667

--- Comment #15 from joseph at codesourcery dot com  ---
On Sun, 24 Apr 2016, lopresti at gmail dot com wrote:

> That said, this is clearly a real bug in GCC. memcpy has a well-defined
> interface; GCC emits calls violating that interface; therefore GCC is buggy. I
> do not see why this is even controversial. (Also see Julian Seward's
> description of real-world problems in comment 5.)

As far as I know, that does not describe real-world problems on any target 
supported by GCC.

> GCC either needs to provide its own mempcy or it needs to respect the
> interface. The cost of the latter is surely trivial, especially since many
> cases could be optimized by alias analysis.

Or it could simply document an additional requirement on the C library 
used with GCC (that the case of exact overlap works), just as it documents 
the requirement that some functions such as memcpy be provided even in the 
freestanding case.  The compiler and library implementations need to 
cooperate to produce a conforming C implementation.

[Bug sanitizer/70712] False positive from AddressSanitizer with use of 'alignas'

[rogero at howzatt dot demon.co.uk]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70712

--- Comment #5 from Roger Orr  ---
Thank you.

I can confirm that the fix also resolves the original problem from which I
derived the sample program.

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

--- Comment #46 from Jakub Jelinek  ---
So, my current thinking is that this is related to make -jN bootstrap doing
stage1 checking by default.
Guess with --enable-stage1-checking=release it would bootstrap fine, but
haven't verified that.
But, what I see is that if I build build/ggc-none.o with stage1 cc1plus with
additional -fno-checking -g0, the result is the same as when it is built with
stage2 cc1plus -g0, and similarly if it is built with stage1 cc1plus with -g0,
the result is the same when it is built with stage2 cc1plus -g0 -fchecking.
I've applied the https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70594#c20 hack
(only the tree.c part of it) and I can see that with -fchecking there are
different DECL_UIDs (more of them too) created.

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[rguenther at suse dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #9 from rguenther at suse dot de  ---
On Mon, 25 Apr 2016, david.abdurachmanov at gmail dot com wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760
> 
> --- Comment #8 from David Abdurachmanov  com> ---
> Should I continue testing the 2nd patch, or dump whatever is currently built
> and restart with the patch from your latest comment?

I think the patch from the last comment is most correct, the 2nd is
wrong.  The first one is safest at this point.

[Bug go/70787] No time and child info with -pg and gccgo

[vogt at linux dot vnet.ibm.com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70787

--- Comment #1 from Dominik Vogt  ---
(I've also tried setting GMON_OUT_PREFIX so that the gmon.out file does not get
overwritten by different threads, but in either case only one dump file is
created.)

[Bug tree-optimization/68030] Redundant address calculations in vectorized loop

[amker at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68030

amker at gcc dot gnu.org changed:

   What|Removed |Added

 CC||amker at gcc dot gnu.org

--- Comment #5 from amker at gcc dot gnu.org ---
(In reply to Kirill Yukhin from comment #4)
> Hello Bin,
> Is it possible to handle the issue using current ivopt?

Not yet I think.  IIUC, this is the same issue as PR69710.  We need to teach
vectorizer about CSE opportunities.  I had a scratch patch fixing it in a way
just as mentioned in #comment2, I will revisit it after work on if-conversion. 
According to Richard's comment, I may need to make it a general fix, for
example, share this facility among different optimizers that need to insert
code in pre-header.

Thanks.

[Bug go/70787] New: No time and child info with -pg and gccgo

[vogt at linux dot vnet.ibm.com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70787

Bug ID: 70787
   Summary: No time and child info with -pg and gccgo
   Product: gcc
   Version: 7.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: go
  Assignee: ian at airs dot com
  Reporter: vogt at linux dot vnet.ibm.com
CC: cmang at google dot com, krebbel at gcc dot gnu.org
  Target Milestone: ---

It looks like the -pg option does something wrong for Go programs.  Example:

This program just wastes time in sub functions:
-- main.go --
package main
func foo () {
var i int
i = 0
for (i < 1000) { i++ }
}
func bar () {
var i int
i = 0
for (i < 1000) { i++ }
}
func main () {
var i int
i = 0
for (i < 100) { foo(); foo(); bar(); i++ }
}
-- snip --

  $ gccgo -pg -O0 main.go
  $ ./a.out
  $ prof ./a.out gmoun.out 

=>

  index % timeself  childrencalled name
  0.000.00 300/300 main.main [8]
  [1]  0.00.000.00 300 frame_dummy [1]
   ^^^

(actual run time was about 5 seconds)

Even for this very simple program without Go library dependencies, no timing
information seems to be dumped into the gmon.out file.  Function calls have all
been counted in the "frame_dummy" bucket (double checked that functios have not
been inlied).

My vague first guess is that maybe the timing information is written to to some
place in memory but is read from a different place when generating gmon.out
because the profiling code is not aware of Gccgo's threading model(?).

[Bug ada/70786] New: Missing "not" breaks Ada.Text_IO.Get_Immediate(File, Item, Available)

[ktamp at chem dot uoa.gr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70786

Bug ID: 70786
   Summary: Missing "not" breaks Ada.Text_IO.Get_Immediate(File,
Item, Available)
   Product: gcc
   Version: 5.3.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: ada
  Assignee: unassigned at gcc dot gnu.org
  Reporter: ktamp at chem dot uoa.gr
  Target Milestone: ---

OS: Arch Linux x64
GCC version: 5.3.0

Missing "not" in line #671 of a-textio.adb causes Get_Immediate(File, Item,
Available) to throw an exception when key "[" is pressed. This also affects
processing of keypresses when function keys F5 to F12 are pressed.

Get_Immediate(File, Item) does not have the same issue because "not" is there
(line #611): "(if not Is_Start_Of_Encoding (Character'Val (ch),
File.WC_Method)"

Example code (run and press e.g. key F12):

with Ada.Text_IO; use Ada.Text_IO;
procedure Test is
   C : Character;
   Available : Boolean;
begin
   Get_Immediate (C);
   loop
  Put (Integer'Image (Character'Pos (C)));
  Get_Immediate (C, Available);
  exit when not Available;
   end loop;
end Test;

[Bug tree-optimization/68030] Redundant address calculations in vectorized loop

[kyukhin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=68030

Kirill Yukhin  changed:

   What|Removed |Added

 CC||amker.cheng at gmail dot com

--- Comment #4 from Kirill Yukhin  ---
Hello Bin,
Is it possible to handle the issue using current ivopt?

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[david.abdurachmanov at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #8 from David Abdurachmanov  
---
Should I continue testing the 2nd patch, or dump whatever is currently built
and restart with the patch from your latest comment?

[Bug ipa/70785] LTO bootstrap with IPA PTA is broken

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70785

Richard Biener  changed:

   What|Removed |Added

 Status|UNCONFIRMED |ASSIGNED
   Last reconfirmed||2016-04-25
   Assignee|unassigned at gcc dot gnu.org  |rguenth at gcc dot 
gnu.org
 Ever confirmed|0   |1

--- Comment #1 from Richard Biener  ---
Mine.

[Bug ipa/70785] New: LTO bootstrap with IPA PTA is broken

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70785

Bug ID: 70785
   Summary: LTO bootstrap with IPA PTA is broken
   Product: gcc
   Version: 6.0
Status: UNCONFIRMED
  Keywords: lto, wrong-code
  Severity: normal
  Priority: P3
 Component: ipa
  Assignee: unassigned at gcc dot gnu.org
  Reporter: rguenth at gcc dot gnu.org
  Target Milestone: ---

gen* in stage2 are miscompiled.

[Bug c/67784] Incorrect parsing when using declarations in for loops and typedefs

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67784

--- Comment #10 from Marek Polacek  ---
Sorry about that.  I'll have another look.

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #7 from Richard Biener  ---
It looks to me we are using the wrong predicates but maybe the GIMPLE side is
somewhat disconnected here.

Index: gcc/tree-ssa-structalias.c
===
--- gcc/tree-ssa-structalias.c  (revision 235404)
+++ gcc/tree-ssa-structalias.c  (working copy)
@@ -4642,9 +4642,7 @@ find_func_aliases_for_call (struct funct

  get_constraint_for (lhsop, );
  rhs = get_function_part_constraint (fi, fi_result);
- if (fndecl
- && DECL_RESULT (fndecl)
- && DECL_BY_REFERENCE (DECL_RESULT (fndecl)))
+ if (aggregate_value_p (lhsop, gimple_call_fntype (t)))
{
  auto_vec tem;
  tem.quick_push (rhs);
@@ -4658,9 +4656,7 @@ find_func_aliases_for_call (struct funct

   /* If we pass the result decl by reference, honor that.  */
   if (lhsop
- && fndecl
- && DECL_RESULT (fndecl)
- && DECL_BY_REFERENCE (DECL_RESULT (fndecl)))
+ && aggregate_value_p (lhsop, gimple_call_fntype (t)))
{
  struct constraint_expr lhs;
  struct constraint_expr *rhsp;

looks correct to me and would also work for indirect calls (though those
are probably "safe" because we don't optimize them very well as we think
having its address taken makes optimization not possible - sth where IPA
PTA needs to be improved).

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[david.abdurachmanov at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #6 from David Abdurachmanov  
---
Will do, results will be late today or tomorrow.

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

--- Comment #45 from Jakub Jelinek  ---
I don't think M4 env var should make a difference in this case, in the release
tarballs the gengtype-lex.c file is already built (on x86_64-linux) and nothing
should be changing that.

That said, I've managed to reproduce the miscompare on build/ggc-none.o as an
example of very small source file, even with -g0.  But it looks really weird.
First of all, it seems that the system g++ 4.8 and g++ 6.1-rc2 is probably ABI
incompatible, at least my attempts to mix stage1 and stage2 objects into the
same binary failed miserably.  I've rebuilt stage2-gcc/ cc1plus with
CXXFLAGS='-g -O0', thus most if not all *.o files linked into it should be
(like in stage1) built without optimizations, and still see the differences.
Trying to find out where the differences start now using parallel gdb sessions.

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[rguenther at suse dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #5 from rguenther at suse dot de  ---
On Mon, 25 Apr 2016, david.abdurachmanov at gmail dot com wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760
> 
> --- Comment #4 from David Abdurachmanov  com> ---
> I will test the second patch. Will take a few hours (it's millions of lines in
> C++).

Thanks.  Just in case also test the first variant - the second (not sure
what unpatched GCC does) runs into a miscompile when LTO bootstrapping
with IPA PTA enabled.

This all looks a bit fishy (and I failed to create a small testcase
even though I tried for half an hour)

[Bug preprocessor/16358] -Wno-system-headers hides warning caused by user header vs system header conflict

[niva at niisi dot msk.ru]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=16358

--- Comment #11 from niva at niisi dot msk.ru ---
(in reply to Martin Sebor in comment #10)

Implementation of -Wmacro-redefined (with possibility of turning this warning
to error) would solve the problem of our users.

[Bug target/70454] --with-arch=native isn't applied to 32-bit x86 target library

[hjl at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70454

--- Comment #5 from hjl at gcc dot gnu.org  ---
Author: hjl
Date: Mon Apr 25 12:41:43 2016
New Revision: 235411

URL: https://gcc.gnu.org/viewcvs?rev=235411=gcc=rev
Log:
Revert the last change in libatomic

Need to properly check if -march=i486 is really needed for -m32 build
of libatomic on Linux/x86 and Linux/x86-64.

PR target/70454
* configure.tgt (XCFLAGS): Revert the last change.

Modified:
trunk/libatomic/ChangeLog
trunk/libatomic/configure.tgt

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[david.abdurachmanov at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #4 from David Abdurachmanov  
---
I will test the second patch. Will take a few hours (it's millions of lines in
C++).

[Bug fortran/70697] ICE on EVENT WAIT with array element UNTIL_COUNT argument

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70697

Dominique d'Humieres  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 Ever confirmed|0   |1

--- Comment #2 from Dominique d'Humieres  ---
Confirmed from 5.3.1 up to trunk (7.0). 5.3.0 gives the errors

pr70697.f90:3:18:

   type(event_type) done[*]
  1
Error: Derived type 'event_type' at (1) is being used before it is defined
pr70697.f90:4:2:

   event wait(done,until_count=nc(1))
  1
Error: Unclassifiable statement at (1)

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #3 from Richard Biener  ---
Where _1(D) is

  # PT =
  struct __single_object & _1(D) = ;

This is in a function deemed local (a ISRA specialization) and thus it looks
like we think there are no callers.

There are indeed callers but to different ISRA specializations.  I suppose
(again) it is IPA ICF that merges them.

There seems to be a disconnect between what versioning does to DECL_RESULT
(and DECL_BY_REFERENCE) and what IPA PTA does vs. non-IPA PTA with regard
to such results.  non-IPA just looks at gimple_call_return_slot_opt_p
while IPA only looks at DECL_BY_REFERENCE.

Index: gcc/tree-ssa-structalias.c
===
--- gcc/tree-ssa-structalias.c  (revision 235404)
+++ gcc/tree-ssa-structalias.c  (working copy)
@@ -4658,9 +4658,11 @@ find_func_aliases_for_call (struct funct

   /* If we pass the result decl by reference, honor that.  */
   if (lhsop
- && fndecl
- && DECL_RESULT (fndecl)
- && DECL_BY_REFERENCE (DECL_RESULT (fndecl)))
+ && ((fndecl
+  && DECL_RESULT (fndecl)
+  && DECL_BY_REFERENCE (DECL_RESULT (fndecl)))
+ || (gimple_call_return_slot_opt_p (t)
+ && TREE_ADDRESSABLE (TREE_TYPE (lhsop)
{
  struct constraint_expr lhs;
  struct constraint_expr *rhsp;

fixes that but the following should also work (but is obviously less safe).

Index: gcc/tree-ssa-structalias.c
===
--- gcc/tree-ssa-structalias.c  (revision 235404)
+++ gcc/tree-ssa-structalias.c  (working copy)
@@ -4658,9 +4658,8 @@ find_func_aliases_for_call (struct funct

   /* If we pass the result decl by reference, honor that.  */
   if (lhsop
- && fndecl
- && DECL_RESULT (fndecl)
- && DECL_BY_REFERENCE (DECL_RESULT (fndecl)))
+ && gimple_call_return_slot_opt_p (t)
+ && TREE_ADDRESSABLE (TREE_TYPE (lhsop)))
{
  struct constraint_expr lhs;
  struct constraint_expr *rhsp;

so can you test the latter on your app?  I'm giving it IPA PTA + LTO bootstrap
testing.

[Bug ipa/70760] [6/7 regression] wrong generated code for std::make_unique with -fipa-pta

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70760

--- Comment #2 from Richard Biener  ---
First DCE after IPA PTA does

@@ -18217,7 +9762,6 @@
   MEM[(struct ParameterDescription
*)_5].D.481734.D.343802._vptr.ParameterDescriptionNode = [(void
*)&_ZTVN3edm20ParameterDescriptionIiEE + 16B];
   _18 = *__args#1_7(D);
   MEM[(struct ParameterDescription *)_5].value_ = _18;
-  MEM[(struct _Head_base *)_1(D)]._M_head_impl = _5;
   return _1(D);

 :

[Bug tree-optimization/70771] [7 Regression] ICE on valid code at -O3 on x86_64-linux-gnu in operator[], at vec.h:714

[amker at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70771

--- Comment #3 from amker at gcc dot gnu.org ---
Had patch for this and PR70775, will send for review after testing.

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

--- Comment #7 from Richard Biener  ---
Author: rguenth
Date: Mon Apr 25 10:50:46 2016
New Revision: 235408

URL: https://gcc.gnu.org/viewcvs?rev=235408=gcc=rev
Log:
2016-04-25  Richard Biener  

PR tree-optimization/70780
* tree-ssa-pre.c (compute_antic_aux): Also return true if the block
wasn't visited yet.
(compute_antic): Mark blocks with abnormal preds as visited as
they have a final empty antic-in solution already.

* gcc.dg/torture/pr70780.c: New testcase.

Added:
branches/gcc-6-branch/gcc/testsuite/gcc.dg/torture/pr70780.c
Modified:
branches/gcc-6-branch/gcc/ChangeLog
branches/gcc-6-branch/gcc/testsuite/ChangeLog
branches/gcc-6-branch/gcc/tree-ssa-pre.c

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

Richard Biener  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

--- Comment #6 from Richard Biener  ---
Fixed.

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

--- Comment #5 from Richard Biener  ---
Author: rguenth
Date: Mon Apr 25 10:49:55 2016
New Revision: 235407

URL: https://gcc.gnu.org/viewcvs?rev=235407=gcc=rev
Log:
2016-04-25  Richard Biener  

PR tree-optimization/70780
* tree-ssa-pre.c (compute_antic_aux): Also return true if the block
wasn't visited yet.
(compute_antic): Mark blocks with abnormal preds as visited as
they have a final empty antic-in solution already.

* gcc.dg/torture/pr70780.c: New testcase.

Added:
trunk/gcc/testsuite/gcc.dg/torture/pr70780.c
Modified:
trunk/gcc/ChangeLog
trunk/gcc/testsuite/ChangeLog
trunk/gcc/tree-ssa-pre.c

[Bug middle-end/22141] [4.9/5/6/7 Regression] Missing optimization when storing structures

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=22141

Richard Biener  changed:

   What|Removed |Added

 CC||ktkachov at gcc dot gnu.org

--- Comment #34 from Richard Biener  ---
*** Bug 70784 has been marked as a duplicate of this bug. ***

[Bug rtl-optimization/70784] Merge multiple short stores of immediates into wider stores

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

Richard Biener  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |DUPLICATE

--- Comment #5 from Richard Biener  ---
Dup.

Note we already have two passes that can do related transforms.

 1) bswap which can detect adjacent loads (that then feed a (non-)bswap
 sequence to build a larger value)

 2) basic-block vectorization (which does not consider (all) non-vector
 types for BLKmode vectorization)

2) has the additional limitation to only consider same-sized sub-stores so
doing this in the general framework of 1) may be easier.

Adding a new pass only considering adjacent stores may also be possible.

*** This bug has been marked as a duplicate of bug 22141 ***

[Bug rtl-optimization/70784] Merge multiple short stores of immediates into wider stores

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

--- Comment #4 from ktkachov at gcc dot gnu.org ---
(In reply to Jakub Jelinek from comment #3)
> Also note that finding the heuristics when to use this (for -Os it is of
> course clearer) is hard, if the pointer is sufficiently aligned or if the
> target is strict alignment it is of course easier.  And the RTL DSE patch
> caused some SPEC regressions on powerpc*.

Thanks for pointing me to PR22141.

I was hoping to structure this in such a way as to pass down the merged byte
sequence to a target hook together with information about the original
store+setup sequence and have the target emit the most profitable sequence
using any target-specific knowledge and tricks it may have (like prefer two
movls isntead of movabsq in your example above). We could provide some helper
functions for extracting immediates from the byte vector (like dse_decode_int
in your patch or a wrapper around builtin_strncpy_read_str from builtins.c).

The default implementation of the hook would be a conservative sequence that
emits a sequence of stores up to word_mode in width (avoiding unaligned stores
for STRICT_ALIGN or SLOW_UNALIGNED_ACCESS targets), perhaps reusing some
store_by_pieces infrastructure.

For the heuristics question I've found tracking setup instructions as well as
stores to work well. That is, from the original store sequence keep track not
only of the store insns but also the instructions that move immediates into the
registers to be stored. And the new sequence would be rejected if the total
number of stores is not smaller, or if the total store+setup cost is higher
than the old store+setup cost. Of course, that would depend on the backend
estimating the cost of synthesising immediates, which the aarch64 target (at
least) does.

I think a gimple implementation of this would suffer from such a drawback, in
that it would not know whether the new wider immediates are actually profitable
unless it consulted RTX costs, which we wouldn't want at gimple level.

[Bug libstdc++/70767] std::numeric_limits::digits is wrong unless --std=c++11 used

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70767

Jonathan Wakely  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
  Component|c++ |libstdc++
   Target Milestone|--- |7.0
 Ever confirmed|0   |1

--- Comment #2 from Jonathan Wakely  ---
(In reply to Marc Glisse from comment #1)
> http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-defects.html#559
> It has status CD1, I don't remember if that means it applies retroactively
> or not.

LWG is less consistent than CWG about stating which issue resolutions are
considered DRs, and so apply retroactively. The libstdc++ policy is to treat
most of them as applying retroactively anyway, because that's the most useful
approach.

Although our current behaviour is correct according to C++98, I see no reason
not to define the specializations for cv-qualified types in C++98.

[Bug rtl-optimization/70784] Merge multiple short stores of immediates into wider stores

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

--- Comment #3 from Jakub Jelinek  ---
Also note that finding the heuristics when to use this (for -Os it is of course
clearer) is hard, if the pointer is sufficiently aligned or if the target is
strict alignment it is of course easier.  And the RTL DSE patch caused some
SPEC regressions on powerpc*.

[Bug rtl-optimization/70784] Merge multiple short stores of immediates into wider stores

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

Jakub Jelinek  changed:

   What|Removed |Added

 CC||jakub at gcc dot gnu.org

--- Comment #2 from Jakub Jelinek  ---
Please see PR22141 which even had a RTL implementation (in the DSE pass).
I think it would be much better handling this (together with bitfields) in some
late GIMPLE pass as opposed to RTL pass.
And note that using movabsq instead of two movls on x86_64 is very likely a
pessimization as opposed to optimization, so one needs to watch the costs too.

[Bug rtl-optimization/70784] Merge multiple short stores of immediates into wider stores

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

ktkachov at gcc dot gnu.org changed:

   What|Removed |Added

 Status|UNCONFIRMED |ASSIGNED
   Last reconfirmed||2016-04-25
   Assignee|unassigned at gcc dot gnu.org  |ktkachov at gcc dot 
gnu.org
 Ever confirmed|0   |1

--- Comment #1 from ktkachov at gcc dot gnu.org ---
I have an idea to implement this at the RTL level by finding consecutive stores
of immediates, creating a byte vector resulting from the merged immediate
values, and re-emitting it using a new wider (and shorter) sequence.

[Bug rtl-optimization/70784] New: Merge multiple short stores of immediates into wider stores

[ktkachov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70784

Bug ID: 70784
   Summary: Merge multiple short stores of immediates into wider
stores
   Product: gcc
   Version: unknown
Status: UNCONFIRMED
  Keywords: missed-optimization
  Severity: enhancement
  Priority: P3
 Component: rtl-optimization
  Assignee: unassigned at gcc dot gnu.org
  Reporter: ktkachov at gcc dot gnu.org
CC: ramana at gcc dot gnu.org
  Target Milestone: ---

Consider the code:
struct bar {
  int a;
  char b;
  char c;
  char d;
  char e;
  char f;
  char g;
}; // packed 64-bit structure

void
foozero (struct bar *p)
{
  p->b = 0;
  p->a = 0;
  p->c = 0;
  p->d = 0;
  p->e = 0;
}

On aarch64 we currently generate:
foozero:
str wzr, [x0]
strbwzr, [x0, 4]
strbwzr, [x0, 5]
strbwzr, [x0, 6]
strbwzr, [x0, 7]
ret

But could generate a single 64-bit store:
foozero:
str xzr, [x0]
ret

Also, for:
void
foo (struct bar *p)
{
  p->b = 0;
  p->a = 0;
  p->c = 0;
  p->d = 1;
  p->e = 0;
}

we could generate:
foo:
mov x1, #0x1
str x1, [x0]

Other targets could benefit from this too.
x86_64 currently generates for 'foo':
foo:
.LFB0:
.cfi_startproc
movl$0, (%rdi)
movb$0, 4(%rdi)
movb$0, 5(%rdi)
movb$1, 6(%rdi)
movb$0, 7(%rdi)
ret

but could genereate:
foo:
.LFB0:
.cfi_startproc
movabsq $281474976710656, %rax
movq%rax, (%rdi)
ret

[Bug driver/70783] New: -spec option behavior is different to implicit spec file

[andreas.g.jhoss at gmx dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70783

Bug ID: 70783
   Summary: -spec option behavior is different to implicit spec
file
   Product: gcc
   Version: 4.8.5
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: driver
  Assignee: unassigned at gcc dot gnu.org
  Reporter: andreas.g.jhoss at gmx dot de
  Target Milestone: ---

Created attachment 38339
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=38339=edit
The failing specs file

I can verify that easily on every Linux system with native compiler
Anyway I am using a gcc backcross compiler (Linux->Windows-Linux) with one
substitute rule.

Is this a lack of documentation or a bug ?

I put a spec file in /usr/lib/gcc/i686-pc-linux-gnu/4.8.5/specs or wherever the
specfile is searched and add a substitute rule (See attached file)

But the substitue rule is only recognized if I do the gcc call with option
-specs=

Working :

g++ r:\test.cpp -nojhdefaultlibs -o r:\test 
-specs=q:/gnuwinh5/bin/../lib/gcc/i686-pc-linux-gnu/4.8.5/

Reading specs from q:/gnuwinh5/bin/../lib/gcc/i686-pc-linux-gnu/4.8.5/specs
Reading specs from q:/gnuwinh5/bin/../lib/gcc/i686-pc-linux-gnu/4.8.5/specs
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=q:/gnuwinh5/bin/../libexec/gcc/i686-pc-linux-gnu/4.8.5/lto-wrapper.exe
Target: i686-pc-linux-gnu
..

Not Working :
If am using the implicit specs file with same path but option is not
recognized.
Reading specs from q:/gnuwinh5/bin/../lib/gcc/i686-pc-linux-gnu/4.8.5/specs
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=q:/gnuwinh5/bin/../libexec/gcc/i686-pc-linux-gnu/4.8.5/lto-wrapper.exe
g++: error: unrecognized command line option '-nojhdefaultlibs'
Target: i686-pc-linux-gnu
..

It doesn't matter what substitute rule I define and also where I define it
makes no difference

[Bug tree-optimization/70771] [7 Regression] ICE on valid code at -O3 on x86_64-linux-gnu in operator[], at vec.h:714

[amker at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70771

--- Comment #2 from amker at gcc dot gnu.org ---
Also like PR70775, this case failed the change allows single argument PHIs in
if-cvt now.  It looks like there is something wrong when handling single
argument PHIs.  Before the change, we simple return false in
ifcvt_convertible_phi_p and skip if-conversion for such form loops.

[Bug bootstrap/70704] [6 Regression] AIX bootstrap comparison failure

[michael.haubenwall...@ssi-schaefer.com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70704

Michael Haubenwallner  changed:

   What|Removed |Added

 CC||michael.haubenwallner@ssi-s
   ||chaefer.com

--- Comment #44 from Michael Haubenwallner 
 ---
Created attachment 38338
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=38338=edit
FLEX listens to M4 envvar, unset.

Does it help to set M4=/opt/freeware/bin/m4 in the environment, so configure is
forced to take GNU m4?

Once upon a time, I've had similar problems here, although with older gcc, and
use attached patch to fix it - maybe it is of some help here as well.

Problem is that flex does listen to the M4 environment variable. Dependent on
PATH, configure may find AIX m4 (/usr/bin) or GNU m4 (/opt/freeware/bin).

When M4=AIX-m4, flex-generated code breaks.

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

--- Comment #4 from Richard Biener  ---
Ok, so we now find (properly) that *d is loop invariant in while (b) but fail
to detect *d as possibly trapping.  That is supposed to be prevented (in PRE)
by the loop exit but I messed up iteration and thus the solution we use for
insertion is not final.

[Bug debug/70764] PASS->FAIL: gcc.dg/guality/pr41447-1.c -O2 -flto -fno-use-linker-plugin -flto-partition=none

[nickc at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70764

--- Comment #3 from Nick Clifton  ---
Hi Kyrylo,

> FAIL: tmp is -1, not 0
> FAIL: tmp2 is -1, not 0

> FAIL: gcc.dg/guality/pr41447-1.c   -O2 -flto -fno-use-linker-plugin
> -flto-partition=none  execution test

Would you mind uploading this binary so that I can have a look at it ?

> There are optimisation levels where the test passes though.
> For example:
> PASS: gcc.dg/guality/pr41447-1.c   -O2 -flto -fuse-linker-plugin
> -fno-fat-lto-objects.

Could you upload this one too, so that I can compare it ?

> Perhaps we should XFAIL the test for aarch64?

Not yet - we need to know why the test is failing for the aarch64.

Cheers
  Nick

[Bug tree-optimization/70777] x*x pessimised to pow(x,2) with -Og -ffast-math

[rguenther at suse dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70777

--- Comment #4 from rguenther at suse dot de  ---
On Mon, 25 Apr 2016, jakub at gcc dot gnu.org wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70777
> 
> Jakub Jelinek  changed:
> 
>What|Removed |Added
> 
>  CC||jakub at gcc dot gnu.org
> 
> --- Comment #3 from Jakub Jelinek  ---
> Well, we could also optimize pow (x, 2) to x * x at expansion time.

We did in the past but removed that "feature".  We could also
canonicalize to POWI instead (not losing the fact x * x didn't set
errno, etc.).

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

Richard Biener  changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
   Assignee|unassigned at gcc dot gnu.org  |rguenth at gcc dot 
gnu.org

--- Comment #3 from Richard Biener  ---
Will have a look.

[Bug tree-optimization/70777] x*x pessimised to pow(x,2) with -Og -ffast-math

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70777

Jakub Jelinek  changed:

   What|Removed |Added

 CC||jakub at gcc dot gnu.org

--- Comment #3 from Jakub Jelinek  ---
Well, we could also optimize pow (x, 2) to x * x at expansion time.

[Bug c/70779] -trapv does not generate trapping vode for integer conversions

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70779

Richard Biener  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |INVALID

--- Comment #1 from Richard Biener  ---
conversions from unsigned to signed are not undefined but implementation
defined
and not covered by -ftrapv.  Instead GCC documents its implementation defined
behavior as modulo-2 reduction.

[Bug c++/70778] internal compiler error: in tsubst, at cp/pt.c:12158

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70778

Richard Biener  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 Ever confirmed|0   |1
  Known to fail||4.8.5, 7.0

--- Comment #2 from Richard Biener  ---
Confirmed on trunk.

[Bug tree-optimization/70777] x*x pessimised to pow(x,2) with -Og -ffast-math

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70777

Richard Biener  changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
   Assignee|unassigned at gcc dot gnu.org  |rguenth at gcc dot 
gnu.org

--- Comment #2 from Richard Biener  ---
Mine.  We should remove the canonicalization done in fold as reassoc does it
now
(and better).  Probably not going to be fixed for older releases (though
we could gate the transform appropriately).

[Bug c++/70776] [4.9/5/6/7 Regression] ICE on invalid code on x86_64-linux-gnu: Segmentation fault (program cc1plus)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70776

Richard Biener  changed:

   What|Removed |Added

   Priority|P3  |P2
 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 Ever confirmed|0   |1

--- Comment #2 from Richard Biener  ---
Confirmed.  Infinite recursion via

#0  0x00b0051e in ggc_internal_alloc (
size=, f=, 
s=,
n=)
at /space/rguenther/src/svn/trunk/gcc/ggc-page.c:1267
#1  0x00d3e569 in ggc_internal_cleared_alloc (size=40, f=0x0, s=0, n=1)
at /space/rguenther/src/svn/trunk/gcc/ggc-common.c:116
#2  0x013e9456 in ggc_internal_cleared_alloc (s=40)
at /space/rguenther/src/svn/trunk/gcc/ggc.h:148
#3  0x013e948a in ggc_alloc_cleared_tree_node_stat (s=40)
at /space/rguenther/src/svn/trunk/gcc/ggc.h:292
#4  0x013ef8c2 in make_tree_vec_stat (len=2)
at /space/rguenther/src/svn/trunk/gcc/tree.c:2281
#5  0x0081467f in coerce_template_parms (
parms=, args=, 
in_decl=, complain=0, 
require_all_args=true, use_default_args=true)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:7594
#6  0x008151a1 in coerce_innermost_template_parms (
parms=, args=, 
in_decl=, complain=0, 
require_all_args=true, use_default_args=true)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:7820
#7  0x00817ceb in lookup_template_class_1 (
d1=, arglist=, 
in_decl=, context=, 
entering_scope=0, complain=0)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:8297
#8  0x0081ae6b in lookup_template_class (
d1=, arglist=, 
in_decl=, context=, 
entering_scope=0, complain=0)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:8638
#9  0x008274a3 in tsubst_aggr_type (t=, 
args=, complain=0, in_decl=, 
entering_scope=0) at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:11425
#10 0x0082fe5d in tsubst (t=, 
args=, complain=0, in_decl=)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:12895
#11 0x00833e7d in tsubst_qualified_id (
qualified_id=, args=, 
complain=0, in_decl=, done=true, address_p=false)
at /space/rguenther/src/svn/trunk/gcc/cp/pt.c:13734
...

Clang detects:

> clang++ -S t.ii
t.ii:6:34: fatal error: recursive template instantiation exceeded maximum depth
  of 256


Not sure if the testcase is thus invalid (but we should diagnose it instead
of crashing).

[Bug tree-optimization/70775] [7 Regression] ICE on valid code at -O3 on x86_64-linux-gnu: Segmentation fault

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70775

Richard Biener  changed:

   What|Removed |Added

   Priority|P3  |P1
 CC||amker at gcc dot gnu.org,
   ||rguenth at gcc dot gnu.org

--- Comment #2 from Richard Biener  ---
Probably a latent issue.

[Bug middle-end/70773] Profiling makes sudoku solver slower

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70773

--- Comment #4 from Richard Biener  ---
It possibly does value profiling figuring out a common division/modulo value
and then making all other values unlikely (and thus cold).

[Bug c/70772] Wrong warning about unspecified behavior for comparison with string literal

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70772

Richard Biener  changed:

   What|Removed |Added

   Keywords||diagnostic
 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 Ever confirmed|0   |1

--- Comment #2 from Richard Biener  ---
Confirmed.

[Bug tree-optimization/70771] [7 Regression] ICE on valid code at -O3 on x86_64-linux-gnu in operator[], at vec.h:714

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70771

Richard Biener  changed:

   What|Removed |Added

   Priority|P3  |P1
 CC||amker at gcc dot gnu.org

[Bug c++/70768] [6/7 Regression] Increased compilation time

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70768

Richard Biener  changed:

   What|Removed |Added

   Keywords||compile-time-hog,
   ||memory-hog
 Status|RESOLVED|REOPENED
   Last reconfirmed||2016-04-25
 Resolution|INVALID |---
   Target Milestone|--- |6.2
Summary|Increased compilation time  |[6/7 Regression] Increased
   ||compilation time
 Ever confirmed|0   |1

--- Comment #5 from Richard Biener  ---
I think this kind of memory usage increase is a bug.

[Bug target/70763] Use SSE for DImode load/store

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70763

Richard Biener  changed:

   What|Removed |Added

   Keywords||missed-optimization
   Severity|normal  |enhancement

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

Jakub Jelinek  changed:

   What|Removed |Added

   Priority|P3  |P1

--- Comment #2 from Jakub Jelinek  ---
Seems starting with that revision we hoist *d before the loop (which would be
ok only if it is guarded by if (b), as if b is false, it is completely valid
that d is NULL or uninitialized etc. and causes trap).
Richard, can you please have a look?

[Bug c++/70781] [6/7 Regression] ICE on invalid C++ code with lambda expressions on x86_64-linux-gnu in finish_expr_stmt, at cp/semantics.c:677

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70781

Jakub Jelinek  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 CC||jakub at gcc dot gnu.org,
   ||jason at gcc dot gnu.org
   Target Milestone|--- |6.0
Summary|ICE on invalid C++ code |[6/7 Regression] ICE on
   |with lambda expressions on  |invalid C++ code with
   |x86_64-linux-gnu in |lambda expressions on
   |finish_expr_stmt, at|x86_64-linux-gnu in
   |cp/semantics.c:677  |finish_expr_stmt, at
   ||cp/semantics.c:677
 Ever confirmed|0   |1

--- Comment #1 from Jakub Jelinek  ---
Started with r233183.

[Bug tree-optimization/70780] [6/7 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70780

Jakub Jelinek  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2016-04-25
 CC||jakub at gcc dot gnu.org
   Target Milestone|--- |6.0
Summary|wrong code at -O2 and -O3   |[6/7 Regression] wrong code
   |on x86_64-linux-gnu |at -O2 and -O3 on
   ||x86_64-linux-gnu
 Ever confirmed|0   |1

--- Comment #1 from Jakub Jelinek  ---
Started with r234970.