[Bug fortran/61376] New: Error using private statement in polymorphic derived type
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61376 Bug ID: 61376 Summary: Error using private statement in polymorphic derived type Product: gcc Version: 4.9.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: fortran Assignee: unassigned at gcc dot gnu.org Reporter: stuart.moffatt at axsym dot co.uk I am getting a compiler error with the gfortran v4.9.0 compiler when using the private statement within a derived type class that is used to form a polymorphic object. Conversely, ifort v14.0.2 does compile this code successfully and the executable runs without error. The example source code below results in the following error message from gfortran: *** START ERROR MESSAGE *** ch2605.poly.f90:58.2: s %x = shape_type(3) 1 Error: Assignment to an allocatable polymorphic variable at (1) is not yet supported *** END ERROR MESSAGE *** If the private statement is commented out and the entire shape_module attributes/methods are made public, then gfortran compiles the code successfully. It appears that there is no straightforwward way to make the contents of the base derived type private. Notes: I have adapted this example source code from ch2605_2.f90 on the fortranplus website. Standard gcc-4.9 installation using apt-get on Ubuntu 14.04 LTS. gfortran compiler options: -fcoarray=single. *** START EXAMPLE SOURCE CODE *** ! Test to demonstrate problem private attributesin polymorphic ! derived type using gfortan v4.9.0 (works OK in ifort v14.0.2) module shape_module implicit none private! *** Commenting out this statement works with gfortran v4.9.0 type shape_type integer :: n_ character(len=2), allocatable, dimension(:) :: cname_ contains procedure, pass(this) :: get_size end type shape_type interface shape_type module procedure shape_type_constructor end interface interface assignment (=) module procedure generic_shape_assign end interface public :: shape_type, generic_shape_assign public :: shape_type_constructor, get_size ! *** Not even this works. contains type (shape_type) function shape_type_constructor(n) implicit none integer, intent (in) :: n shape_type_constructor%n_ = n allocate ( shape_type_constructor%cname_(n) ) end function shape_type_constructor integer function get_size(this) implicit none class (shape_type), intent (in) :: this get_size = size(this%cname_) end function get_size subroutine generic_shape_assign(lhs,rhs) implicit none class (shape_type), intent (out), allocatable :: lhs class (shape_type), intent (in) :: rhs allocate (lhs,source=rhs) end subroutine generic_shape_assign end module shape_module module shape_wrapper_module use shape_module type shape_wrapper class (shape_type), allocatable :: x end type shape_wrapper end module shape_wrapper_module program ch2605_mod use shape_module use shape_wrapper_module implicit none type (shape_wrapper) :: s s %x = shape_type(3) print *, 'size = ', s%x%get_size() end program ch2605_mod *** END EXAMPLE SOURCE CODE ***
[Bug c++/57543] decltype needs explicit 'this' pointer in member function declaration of template class with trailing return type
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57543 --- Comment #7 from paolo at gcc dot gnu.org --- Author: paolo Date: Sat May 31 08:51:20 2014 New Revision: 211102 URL: http://gcc.gnu.org/viewcvs?rev=211102&root=gcc&view=rev Log: /cp 2014-05-31 Paolo Carlini DR 1227 PR c++/57543 * cp-tree.h (TYPE_HAS_LATE_RETURN_TYPE): Add. * pt.c (tsubst_function_type): Inject the this parameter; do the substitutions in the order mandated by the DR. (copy_default_args_to_explicit_spec): Copy TYPE_HAS_LATE_RETURN_TYPE. * decl.c (grokdeclarator): Maybe set TYPE_HAS_LATE_RETURN_TYPE. (static_fn_type): Copy it. * decl2.c (build_memfn_type, change_return_type, cp_reconstruct_complex_type): Likewise. * parser.c (cp_parser_lambda_declarator_opt): Likewise. * tree.c (strip_typedefs): Likewise. * typeck.c (merge_types): Likewise. /testsuite 2014-05-31 Paolo Carlini DR 1227 PR c++/57543 * g++.dg/cpp0x/pr57543-1.C: New. * g++.dg/cpp0x/pr57543-2.C: Likewise. * g++.dg/cpp0x/pr57543-3.C: Likewise. * g++.dg/cpp0x/decltype59.C: Likewise. Added: trunk/gcc/testsuite/g++.dg/cpp0x/decltype59.C trunk/gcc/testsuite/g++.dg/cpp0x/pr57543-1.C trunk/gcc/testsuite/g++.dg/cpp0x/pr57543-2.C trunk/gcc/testsuite/g++.dg/cpp0x/pr57543-3.C Modified: trunk/gcc/cp/ChangeLog trunk/gcc/cp/cp-tree.h trunk/gcc/cp/decl.c trunk/gcc/cp/decl2.c trunk/gcc/cp/parser.c trunk/gcc/cp/pt.c trunk/gcc/cp/tree.c trunk/gcc/cp/typeck.c trunk/gcc/testsuite/ChangeLog
[Bug c++/54366] [meta-bug] decltype issues
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=54366 Bug 54366 depends on bug 57543, which changed state. Bug 57543 Summary: decltype needs explicit 'this' pointer in member function declaration of template class with trailing return type https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57543 What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED
[Bug c++/57543] decltype needs explicit 'this' pointer in member function declaration of template class with trailing return type
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57543 Paolo Carlini changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED Assignee|paolo.carlini at oracle dot com|unassigned at gcc dot gnu.org --- Comment #8 from Paolo Carlini --- Fixed.
[Bug driver/60968] [4.9 Regression] Bootstrap fails on mingw32 with -dumpspecs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60968 --- Comment #5 from Daniel Starke --- Removing --disable-sjlj-exceptions solved the problem. I did not try the current 4.9.1 branch yet.
[Bug libstdc++/61374] string_view::operator string() is buggy
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61374 Jonathan Wakely changed: What|Removed |Added Status|NEW |ASSIGNED Assignee|unassigned at gcc dot gnu.org |redi at gcc dot gnu.org
[Bug libstdc++/61374] string_view::operator string() is buggy
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61374 Jonathan Wakely changed: What|Removed |Added Target Milestone|--- |4.9.1
[Bug target/61377] New: [4.10 Regression] Bootstrap failure on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61377 Bug ID: 61377 Summary: [4.10 Regression] Bootstrap failure on x86_64-apple-darwin13 Product: gcc Version: 4.10.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: dominiq at lps dot ens.fr CC: iains at gcc dot gnu.org Host: x86_64-apple-darwin13 Target: x86_64-apple-darwin13 Build: x86_64-apple-darwin13 Bootstrapping r211103 fails on x86_64-apple-darwin13 while building libjava with -m32: libtool: compile: /opt/gcc/build_c/./gcc/xgcc -shared-libgcc -B/opt/gcc/build_c/./gcc -nostdinc++ -L/opt/gcc/build_c/x86_64-apple-darwin13.2.0/i386/libstdc++-v3/src -L/opt/gcc/build_c/x86_64-apple-darwin13.2.0/i386/libstdc++-v3/src/.libs -L/opt/gcc/build_c/x86_64-apple-darwin13.2.0/i386/libstdc++-v3/libsupc++/.libs -B/opt/gcc/gcc4.10c/x86_64-apple-darwin13.2.0/bin/ -B/opt/gcc/gcc4.10c/x86_64-apple-darwin13.2.0/lib/ -isystem /opt/gcc/gcc4.10c/x86_64-apple-darwin13.2.0/include -isystem /opt/gcc/gcc4.10c/x86_64-apple-darwin13.2.0/sys-include -m32 -DHAVE_CONFIG_H -I. -I../../../../_clean/libjava -I./include -I./gcj -I../../../../_clean/libjava -Iinclude -I../../../../_clean/libjava/include -I../../../../_clean/libjava/classpath/include -Iclasspath/include -I../../../../_clean/libjava/classpath/native/fdlibm -I../../../../_clean/libjava/../boehm-gc/include -I../boehm-gc/include -I../../../../_clean/libjava/libltdl -I../../../../_clean/libjava/libltdl -I../../../../_clean/libjava/.././libjava/../libgcc -I../../../../_clean/libjava/../libffi/include -I../libffi/include -fno-rtti -fnon-call-exceptions -fdollars-in-identifiers -Wswitch-enum -D_FILE_OFFSET_BITS=64 -ffloat-store -fomit-frame-pointer -Usun -Wextra -Wall -D_GNU_SOURCE -DPREFIX=\"/opt/gcc/gcc4.10c\" -DTOOLEXECLIBDIR=\"/opt/gcc/gcc4.10c/lib/i386\" -DJAVA_HOME=\"/opt/gcc/gcc4.10c\" -DBOOT_CLASS_PATH=\"/opt/gcc/gcc4.10c/share/java/libgcj-4.10.0.jar\" -DJAVA_EXT_DIRS=\"/opt/gcc/gcc4.10c/share/java/ext\" -DGCJ_ENDORSED_DIRS=\"/opt/gcc/gcc4.10c/share/java/gcj-endorsed\" -DGCJ_VERSIONED_LIBDIR=\"/opt/gcc/gcc4.10c/lib/i386/gcj-4.10.0-15\" -DPATH_SEPARATOR=\":\" -DECJ_JAR_FILE=\"\" -DLIBGCJ_DEFAULT_DATABASE=\"/opt/gcc/gcc4.10c/lib/i386/gcj-4.10.0-15/classmap.db\" -DLIBGCJ_DEFAULT_DATABASE_PATH_TAIL=\"gcj-4.10.0-15/classmap.db\" -g -O2 -m32 -MT gnu/gcj/natCore.lo -MD -MP -MF gnu/gcj/.deps/natCore.Tpo -c ../../../../_clean/libjava/gnu/gcj/natCore.cc -fno-common -DPIC -o gnu/gcj/.libs/natCore.o /var/folders/8q/sh_swgz96r7f5vnn08f7fxr0gn/T//ccJoNqSO.s:89:non-relocatable subtraction expression, "__Jv_RegisterCoreHook" minus "L2$pb" /var/folders/8q/sh_swgz96r7f5vnn08f7fxr0gn/T//ccJoNqSO.s:89:symbol: "L2$pb" can't be undefined in a subtraction expression /var/folders/8q/sh_swgz96r7f5vnn08f7fxr0gn/T//ccJoNqSO.s:unknown:Undefined local symbol L2$pb make[5]: *** [gnu/gcj/natCore.lo] Error 1
[Bug c/61378] New: Obvious bug in vn_reference_lookup_3
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61378
Bug ID: 61378
Summary: Obvious bug in vn_reference_lookup_3
Product: gcc
Version: 4.9.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: dcb314 at hotmail dot com
Valgrind says
==15936== Conditional jump or move depends on uninitialised value(s)
==15936==at 0xBE5839: vn_reference_lookup_3(ao_ref*, tree_node*, void*,
bool) (tree-ssa-sccvn.c:1627)
Source code is
bool valueized_anything;
for (unsigned i = 0; i < gimple_call_num_args (def_stmt); ++i)
{
oldargs[i] = gimple_call_arg (def_stmt, i);
if (TREE_CODE (oldargs[i]) == SSA_NAME
&& VN_INFO (oldargs[i])->valnum != oldargs[i])
{
gimple_call_set_arg (def_stmt, i, VN_INFO (oldargs[i])->valnum);
valueized_anything = true;
}
}
if (valueized_anything)
Suggest init valueized_anything at its declaration.
Use of svn blame shows
210160rguenth if (valueized_anything)
[Bug target/61331] [ARM] GCC build broken since r210964/r210965
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61331 --- Comment #7 from Bernd Edlinger --- Fixed with r211050. Thanks!
[Bug bootstrap/61320] [4.10 regression] ICE in jcf-parse.c:1622 (parse_class_file
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61320 --- Comment #8 from Eric Botcazou --- > Sure, I'll push a patch for this as soon as I finish fixing the regressions > that poped up due to the change I made to the bswap pass. While you're at it, could you change all the references in the code (as well as the ChangeLog) to "host endianness", which is mightily confusing in a compiler? By definition a compiler generates code for a target, not for a host.
[Bug driver/60968] [4.9 Regression] Bootstrap fails on mingw32 with -dumpspecs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60968 --- Comment #6 from Daniel Starke --- I can confirm that the problem does not occur in r211103 of the 4.9 branch.
[Bug target/61377] [4.10 Regression] Bootstrap failure on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61377 Dominique d'Humieres changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2014-05-31 CC||ktietz70 at googlemail dot com Ever confirmed|0 |1 --- Comment #1 from Dominique d'Humieres --- Caused by r211089. Fixed by the patch at https://gcc.gnu.org/ml/gcc-patches/2014-05/msg02689.html.
[Bug target/61377] [4.10 Regression] Bootstrap failure on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61377 Jack Howarth changed: What|Removed |Added CC||howarth.at.gcc at gmail dot com --- Comment #2 from Jack Howarth --- (In reply to Dominique d'Humieres from comment #1) > Caused by r211089. Fixed by the patch at > https://gcc.gnu.org/ml/gcc-patches/2014-05/msg02689.html. The patch posted there is malformed. It appears to have been hand-editted to remove… [(set_attr "type" "call")]) +; TODO +(define_peephole2 + [(set (match_operand:DI 0 "register_operand") +(match_operand:DI 1 "memory_operand")) + (unspec_volatile [(const_int 0)] UNSPECV_BLOCKAGE)] + "TARGET_64BIT" + [(unspec_volatile [(const_int 0)] UNSPECV_BLOCKAGE) + (set (match_dup 0) +(match_dup 1))]) + +(define_peephole2 + [(set (match_operand:SI 0 "register_operand") +(match_operand:SI 1 "memory_operand")) + (unspec_volatile [(const_int 0)] UNSPECV_BLOCKAGE)] + "!TARGET_64BIT" + [(unspec_volatile [(const_int 0)] UNSPECV_BLOCKAGE) + (set (match_dup 0) +(match_dup 1))]) + +(define_peephole2 + [(set (match_operand:DI 0 "register_operand") +(match_operand:DI 1 "memory_operand")) + (call (mem:QI (match_operand:DI 2 "register_operand")) + (match_operand 3))] + "TARGET_64BIT && REG_P (operands[0]) && 0 +&& REG_P (operands[2]) +&& REGNO (operands[0]) == REGNO (operands[2])" + [(call (mem:QI (match_dup 1)) (match_dup 1))]) + (define_expand "call_pop" [(parallel [(call (match_operand:QI 0) (match_operand:SI 1)) from the original patch at http://gcc.gnu.org/ml/gcc-patches/2014-05/msg02686.html rather than having been properly regenerated. Is unclear what exactly was approved here.
[Bug c++/61379] New: __attribute__((noreturn)) ignored on pure virtual member functions
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61379 Bug ID: 61379 Summary: __attribute__((noreturn)) ignored on pure virtual member functions Product: gcc Version: 4.9.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: meiomorphism at gmail dot com Created attachment 32881 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=32881&action=edit test file (#include-free) Pure-virtual member functions with __attribute__((noreturn)) attached are not recognized as being noreturn. This can lead to spurious warnings. Use of [[noreturn]] rather than __attribute__((noreturn)) yields the same behavior. Tested on both Debian wheezy's gcc 4.7.2 and a locally-compiled officially-sourced gcc 4.9.0. The output of `g++ -v' for the latter is given below. To reproduce, compile the attached file with -Wall; the spurious warning > virtual-noreturn.cpp:12:1: warning: control reaches end of non-void function > [-Wreturn-type] (some context omitted) will be given. === BEGIN g++ -v output === Using built-in specs. COLLECT_GCC=/home/user/local/gcc-4.9/bin/g++ COLLECT_LTO_WRAPPER=/home/user/local/gcc-4.9/libexec/gcc/x86_64-unknown-linux-gnu/4.9.0/lto-wrapper Target: x86_64-unknown-linux-gnu Configured with: /home/user/src/gcc-4.9.0/configure --prefix=/home/user/local/gcc-4.9 --with-local-prefix=/home/user/local/gcc-4.9-include --disable-bootstrap --disable-multilib : (reconfigured) /home/user/src/gcc-4.9.0/configure --prefix=/home/user/local/gcc-4.9 --with-local-prefix=/home/user/local/gcc-4.9-include --disable-bootstrap --disable-multilib --enable-languages=c,c++ Thread model: posix gcc version 4.9.0 (GCC) === END gcc -v output ===
