[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 --- Comment #2 from David Binderman --- (In reply to Jakub Jelinek from comment #1) > The above snippet isn't a self-contained program and contains UB, so it is > unclear how you can get runtime diagnostics for it. The short snippet of code is stored in file bug424.c gcc trunk revision 259004 has been built with ubsan, as the configure line should make clear. 06:51 is pretty early in the morning, isn't it ? ;->
[Bug c++/85148] [6/7/8 Regression] ICE with NSDMI and this pointer
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85148
Jakub Jelinek changed:
What|Removed |Added
Priority|P3 |P2
Status|UNCONFIRMED |NEW
Last reconfirmed||2018-04-03
CC||jakub at gcc dot gnu.org,
||jason at gcc dot gnu.org
Target Milestone|--- |6.5
Ever confirmed|0 |1
--- Comment #1 from Jakub Jelinek ---
I get ICE already with r185596 when -std=c++1y option has been introduced:
pr85148.C:1:29: internal compiler error: tree check: expected tree that
contains ‘common’ structure, have ‘constructor’ in perform_member_init, at
cp/init.c:661
Then starting with r210653 it is rejected:
pr85148.C: In function ‘void foo()’:
pr85148.C:8:28: error: could not convert ‘(long int)this’ from ‘long int’ to
‘A’
A> a{};
^
pr85148.C: In constructor ‘constexpr A >::A()’:
pr85148.C:1:29: error: could not convert ‘(long int)this’ from ‘long int’ to
‘A’
template struct A
^
pr85148.C: In function ‘void foo()’:
pr85148.C:8:28: note: synthesized method ‘constexpr A >::A()’ first
required here
A> a{};
^
Then starting with r216750 it ICEs in tsubst_copy:
pr85148.C: In function ‘void foo()’:
pr85148.C:8:28: internal compiler error: in tsubst_copy, at cp/pt.c:12755
A> a{};
^
0x71171b tsubst_copy
../../gcc/cp/pt.c:12755
0x720a9b tsubst_copy_and_build(tree_node*, tree_node*, int, tree_node*, bool,
bool)
../../gcc/cp/pt.c:15464
and finally starting with r234442 in replace_placeholders.
It hasn't been fixed by the r258593 changes, seems the PLACEHOLDER_EXPR refers
to the outer A when it should be refering to the inner A? Or I'm totally
confused on what exactly it should do.
[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 Jakub Jelinek changed: What|Removed |Added Status|WAITING |NEW CC||rguenth at gcc dot gnu.org --- Comment #3 from Jakub Jelinek --- Oops, sorry, missed the UB is in the compiler, sorry. I believe this and surrounding code has several such known issues, but can't find a PR for that right now.
[Bug c++/85147] [6/7/8 Regression] ICE with invalid variadic template-template parameter
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85147 Jakub Jelinek changed: What|Removed |Added Priority|P3 |P4 Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org, ||paolo at gcc dot gnu.org Target Milestone|--- |6.5 Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r226054.
[Bug c++/85147] [6/7/8 Regression] ICE with invalid variadic template-template parameter
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85147 --- Comment #2 from Jakub Jelinek --- Created attachment 43812 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43812&action=edit gcc8-pr85147.patch Untested fix.
[Bug c/85164] New: poly-int.h:845:5: runtime error: signed integer overflow
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85164
Bug ID: 85164
Summary: poly-int.h:845:5: runtime error: signed integer
overflow
Product: gcc
Version: 8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: dcb314 at hotmail dot com
Target Milestone: ---
For the following C code, in file bug425b.c
a[];
b;
c() {
unsigned long d;
b = a[d - 1 >> 3];
}
Compiled by gcc trunk revision 259004, compiled with undefined behaviour
sanitizer, does this:
$ ~/gcc/results/bin/gcc -c -O2 -w bug425b.c
../../trunk/gcc/poly-int.h:845:5: runtime error: signed integer overflow:
9223372036854775804 + 4 cannot be represented in type 'long int'
$ ~/gcc/results/bin/gcc -v
Using built-in specs.
COLLECT_GCC=/home/dcb/gcc/results/bin/gcc
COLLECT_LTO_WRAPPER=/home/dcb/gcc/results.259004.ubsan/libexec/gcc/x86_64-pc-linux-gnu/8.0.1/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../trunk/configure --prefix=/home/dcb/gcc/results.259004.ubsan
--with-build-config=bootstrap-ubsan --disable-multilib --disable-werror
--enable-checking=df,extra,fold,rtl,yes --enable-languages=c,c++,fortran
Thread model: posix
gcc version 8.0.1 20180402 (experimental) (GCC)
$
[Bug middle-end/64920] bootstrap-ubsan [build/gengtype -r gtype.state]: libiberty/regex.c:6970:11: runtime error: left shift of negative value -1
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64920 Martin Liška changed: What|Removed |Added Status|WAITING |RESOLVED Resolution|--- |WORKSFORME --- Comment #5 from Martin Liška --- Then closing..
[Bug other/63426] [meta-bug] Issues found with -fsanitize=undefined
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63426 Bug 63426 depends on bug 64920, which changed state. Bug 64920 Summary: bootstrap-ubsan [build/gengtype -r gtype.state]: libiberty/regex.c:6970:11: runtime error: left shift of negative value -1 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64920 What|Removed |Added Status|WAITING |RESOLVED Resolution|--- |WORKSFORME
[Bug c/85164] poly-int.h:845:5: runtime error: signed integer overflow
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85164 Martin Liška changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||marxin at gcc dot gnu.org Blocks||63426 Ever confirmed|0 |1 Referenced Bugs: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63426 [Bug 63426] [meta-bug] Issues found with -fsanitize=undefined
[Bug c++/85156] [8 Regression] ICE with -O1 -g: gimplification failed
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85156 Martin Liška changed: What|Removed |Added Priority|P3 |P1 Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||aoliva at gcc dot gnu.org, ||marxin at gcc dot gnu.org Known to work||7.3.0 Target Milestone|--- |8.0 Summary|-O1 -g internal compiler|[8 Regression] ICE with -O1 |error: gimplification |-g: gimplification failed |failed | Ever confirmed|0 |1 Known to fail||8.0 --- Comment #1 from Martin Liška --- Confirmed, started with r255569.
[Bug c++/85146] ICE with __direct_bases for declared but not defined struct
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85146 Jakub Jelinek changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #1 from Jakub Jelinek --- Created attachment 43813 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43813&action=edit gcc8-pr85146.patch It isn't entirely clear to me if this is valid or not from N2965, if it is valid, we can do something like this patch, if it is invalid, we should use complete_type_or_else or similar. __bases doesn't diagnose incomplete classes either.
[Bug tree-optimization/82494] UBSAN in gcc/tree-data-ref.c:3427:16: runtime error: signed integer overflow: 131072 * -131072 cannot be represented in type 'int'
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82494 --- Comment #3 from Richard Biener --- *** Bug 85158 has been marked as a duplicate of this bug. ***
[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 Richard Biener changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |DUPLICATE --- Comment #4 from Richard Biener --- There's related PR82494 about the lambda vector operations but yes, this code hasn't been fixed to avoid overflows and generally expects to operate with infinite precision integers ... (you know, 64bits will be enough - really!). Fixing requires overhauling all of the API and data structures in dependence analysis and finding a suitable representation for "infinite precision integers". Note that most of the APIs do not have a well-defined failure mode so bailing out isn't easy either. Using GMP is very likely prohibitly expensive not only due to its lack of optimizing the small-precision case. OTOH we are using it from niter analysis already. Evaluating ISL ints (isl/val.h when configured to not use GMP) might be worth, as well as maybe looking at including (patched?) mini-gmp. Note the HWIs are less likely a problem than the ints currently used for lambda vectors/matrices. Note that simply silencing UBSAN by using unsigned arithmetic will simply silence it, not avoiding wrong-code issues that might be latent there. *** This bug has been marked as a duplicate of bug 82494 ***
[Bug other/63426] [meta-bug] Issues found with -fsanitize=undefined
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63426 Bug 63426 depends on bug 85158, which changed state. Bug 85158 Summary: ubsan and -O3 finds a problem https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 What|Removed |Added Status|NEW |RESOLVED Resolution|--- |DUPLICATE
[Bug c++/85141] [6/7/8 Regression] ICE with pointer arithmetic of static member function address
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85141 Jakub Jelinek changed: What|Removed |Added Priority|P3 |P2 Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org Target Milestone|--- |6.5 Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r176072.
[Bug tree-optimization/82494] UBSAN in gcc/tree-data-ref.c:3427:16: runtime error: signed integer overflow: 131072 * -131072 cannot be represented in type 'int'
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82494 --- Comment #4 from Richard Biener --- I commented in the dup: This code hasn't been fixed to avoid overflows and generally expects to operate with infinite precision integers ... (you know, 32/64bits will be enough - really!). Fixing requires overhauling all of the API and data structures in dependence analysis and finding a suitable representation for "infinite precision integers". Note that most of the APIs do not have a well-defined failure mode so bailing out isn't easy either. Using GMP is very likely prohibitly expensive not only due to its lack of optimizing the small-precision case. OTOH we are using it from niter analysis already. Evaluating ISL ints (isl/val.h when configured to not use GMP) might be worth, as well as maybe looking at including (patched?) mini-gmp (not sure if that plays well with linking against real gmp, but eventually mpfr plays well enough with mini-gmp and doesn't get too slow). Note the HWIs are less likely a problem than the ints currently used for lambda vectors/matrices. Note that simply silencing UBSAN by using unsigned arithmetic will simply silence it, not avoiding wrong-code issues that might be latent there.
[Bug c++/85165] New: gcc (and g++) not complaining about few unknown file extensions like *.CC, *.Cc, *.cC
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85165 Bug ID: 85165 Summary: gcc (and g++) not complaining about few unknown file extensions like *.CC, *.Cc, *.cC Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: kiran0802 at gmail dot com Target Milestone: --- When compiling the following test case i.e. test.CC, GCC (6.4.0 ver) is not generating any warning and moreover respective object file is generated. I think this while compiling a file with any unknown file extension, GCC should generate a warning i.e. "test.CC: file not recognized: File truncated". Steps to reproduce: create a file with .CC extension and compile it with either gcc or g++ as shown below. with gcc-6.4.0 version:(compiler NOT generating any warnings for unknown file extensions) - % touch test.CC % gcc test.CC /usr/lib/gcc/i686-pc-cygwin/6.4.0/../../../libcygwin.a(libcmain.o): In function `main': /usr/src/debug/cygwin-2.10.0-1/winsup/cygwin/lib/libcmain.c:37: undefined reference to `WinMain@16' collect2: error: ld returned 1 exit status From above messages, it seems that the compiler has recognized/considered/assumed .CC file extension as a valid extension and further processing is done. % gcc -c test.CC There is no warning generated from the compiler and respective object file i.e. test.o generated. Note: I had also tried with other file extensions such as .Cc and .cC and observed the same behavior as above. g++ is also behaving in the same way as gcc. with gcc-4.8.0 version:(compiler is generated warnings as expected) --- % gcc test.CC test.CC: file not recognized: File truncated collect2: ld returned 1 exit status I had also tried to invoke only compiler i.e. with -c option % gcc -c test.CC gcc: test.CC: linker input file unused because linking not done % gcc --version gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-18) Copyright (C) 2010 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. From my understanding, gcc support following file extenstions: .ada, .adb., .c, .C, .c++, .cc, .class, .cp, .cpp, .CPP, .cxx, etc. As .CC, .Cc and .cC are not in the above list, compiler should generate respective warning message.
[Bug gcov-profile/85163] [GCOV] A call statement with '==' , '||', and '&&' operators is wrongly marked as executed twice
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85163 Martin Liška changed: What|Removed |Added Priority|P3 |P5 Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 Assignee|unassigned at gcc dot gnu.org |marxin at gcc dot gnu.org Ever confirmed|0 |1 Severity|normal |enhancement --- Comment #1 from Martin Liška --- I can confirm it. However it's quite complex condition and I'm probably not planning to fix that.
[Bug gcov-profile/85163] [GCOV] A call statement with '==' , '||', and '&&' operators is wrongly marked as executed twice
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85163 --- Comment #2 from Yibiao Yang --- (In reply to Martin Liška from comment #1) > I can confirm it. However it's quite complex condition and I'm probably not > planning to fix that. Ok. Thanks.
[Bug c++/85141] [6/7/8 Regression] ICE with pointer arithmetic of static member function address
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85141 Jakub Jelinek changed: What|Removed |Added CC||jason at gcc dot gnu.org --- Comment #2 from Jakub Jelinek --- Not really sure what can be done here, perhaps stabilize_expr could for these FUNCTION_TYPE exp try to create a TARGET_EXPR for the address of COMPONENT_REF's first operand and only keep the COMPONENT_REF with BASELINK and optional NON_LVALUE_EXPR wrapping it in the returned expressio rather than also in the TARGET_EXPR?
[Bug c++/85141] [6/7/8 Regression] ICE with pointer arithmetic of static member function address
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85141 --- Comment #3 from Jakub Jelinek --- Without -fpermissive it is error-recovery ICE.
[Bug c++/85140] [6/7/8 Regression] ICE with invalid use of alignas
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85140 Jakub Jelinek changed: What|Removed |Added Priority|P3 |P4 Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org Target Milestone|--- |6.5 Summary|ICE with invalid use of |[6/7/8 Regression] ICE with |alignas |invalid use of alignas Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r192199.
[Bug tree-optimization/84969] [8 Regression] Wrong code with -ftree-loop-distribute-patterns
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84969 amker at gcc dot gnu.org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #5 from amker at gcc dot gnu.org --- (In reply to Jakub Jelinek from comment #4) > So fixed? Yes, I think so.
[Bug c++/85140] [6/7/8 Regression] ICE with invalid use of alignas
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85140 --- Comment #2 from Jakub Jelinek --- Created attachment 43814 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43814&action=edit gcc8-pr85140.patch Untested fix.
[Bug target/85166] New: [ntptx, libgfortran] Libgomp fortran tests using stop in offloaded fns fail to compile
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85166
Bug ID: 85166
Summary: [ntptx, libgfortran] Libgomp fortran tests using stop
in offloaded fns fail to compile
Product: gcc
Version: 8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: vries at gcc dot gnu.org
Target Milestone: ---
Atm for nvptx offloading we have lots of fortran tests (both openmp and
openacc) fail like this:
...
unresolved symbol _gfortran_stop_numeric
FAIL: libgomp.fortran/examples-4/async_target-2.f90 -O0 (test for excess
errors)
UNRESOLVED: libgomp.fortran/examples-4/async_target-2.f90 -O0 compilation
failed to produce executable
...
Recently "[libgomp testsuite] Replace non-standard call abort by STOP n" was
committed.
In libgfortran/Makefile.in we have:
...
@LIBGFOR_MINIMAL_FALSE@runtime/stop.c
...
so the function _gfortran_stop_numeric contained in stop.c doesn't get build
for LIBGFOR_MINIMAL.
The LIBGFOR_MINIMAL setting is only active for nvptx:
...
if test "x${target_cpu}" = xnvptx; then
LIBGFOR_MINIMAL_TRUE=
LIBGFOR_MINIMAL_FALSE='#'
else
LIBGFOR_MINIMAL_TRUE='#'
LIBGFOR_MINIMAL_FALSE=
fi
...
[Bug target/84762] GCC for PowerPC32 violates the SysV ABI spec for small struct returns
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84762 Richard Biener changed: What|Removed |Added CC||rguenth at gcc dot gnu.org --- Comment #7 from Richard Biener --- Isn't this two different bugs? From the description it looks like all of 4.6+ behave like this? But the bisect points to sth that would be a 8+ regression? Also I fail to see how this expand patch can cause this ABI issue.
[Bug target/85126] [8 regression] gcc.target/powerpc/pr69946.c fails on powerpc64 BE starting with r254213
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85126 Richard Biener changed: What|Removed |Added Target||powerpc64 Target Milestone|--- |8.0
[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 --- Comment #5 from David Binderman --- (In reply to Richard Biener from comment #4) > There's related PR82494 about the lambda vector operations but yes, this > code hasn't > been fixed to avoid overflows and generally expects to operate with infinite > precision integers ... (you know, 64bits will be enough - really!). This code works 99% of the time. I'd be happy to see it bodged until it works 99.9% of the time. Can I suggest that for those parts of the code where int isn't enough, then long gets used and for those parts where long isn't enough, the 128 bit data type (__int128_t ?) gets used. Going all the way to unlimited ints when plain ints aren't enough seems overkill to me, although I am not familiar with the code.
[Bug c++/85127] Members implementation outside class doesn't compile if class name is an alias from a using
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85127
Richard Biener changed:
What|Removed |Added
Keywords||rejects-valid
Status|UNCONFIRMED |NEW
Last reconfirmed||2018-04-03
Ever confirmed|0 |1
--- Comment #1 from Richard Biener ---
G++ says
t.C:11:16: error: invalid use of incomplete type ‘class Base’
void B::foo() {
^
t.C:2:7: note: declaration of ‘class Base’
class Base {
^~~~
[Bug c/85167] New: shrink-wrap.c:333:15: runtime error with UBSAN
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85167
Bug ID: 85167
Summary: shrink-wrap.c:333:15: runtime error with UBSAN
Product: gcc
Version: 8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: dcb314 at hotmail dot com
Target Milestone: ---
For the following C code stored in source code file bug426.c:
struct a {
long b
};
c, d, e;
f() {
long g;
for (; g == c ? 0 : (e = 1); g = ((struct a *)g)->b)
if (h())
return d;
}
$ ~/gcc/results/bin/gcc -c -w -O2 bug426.c
../../trunk/gcc/shrink-wrap.c:333:15: runtime error: member access within null
pointer of type 'struct df_lr_bb_info'
../../trunk/gcc/shrink-wrap.c:334:15: runtime error: member access within null
pointer of type 'struct df_lr_bb_info'
$ ~/gcc/results/bin/gcc -v
Using built-in specs.
COLLECT_GCC=/home/dcb/gcc/results/bin/gcc
COLLECT_LTO_WRAPPER=/home/dcb/gcc/results.259004.ubsan/libexec/gcc/x86_64-pc-linux-gnu/8.0.1/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../trunk/configure --prefix=/home/dcb/gcc/results.259004.ubsan
--with-build-config=bootstrap-ubsan --disable-multilib --disable-werror
--enable-checking=df,extra,fold,rtl,yes --enable-languages=c,c++,fortran
Thread model: posix
gcc version 8.0.1 20180402 (experimental) (GCC)
$
[Bug c++/85133] [7/8 Regression] [concepts] ICE with invalid concept used in variadic template
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85133 Richard Biener changed: What|Removed |Added Priority|P3 |P4 Target Milestone|--- |7.4
[Bug c++/85134] [8 Regression] ICE with invalid constexpr in 'shared' OpenMP clause
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85134 Richard Biener changed: What|Removed |Added Priority|P3 |P4 Target Milestone|--- |8.0
[Bug c++/85135] [7/8 Regression] ICE with invalid late-specified return type
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85135 Richard Biener changed: What|Removed |Added Priority|P3 |P2 Target Milestone|--- |7.4
[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 --- Comment #6 from Jakub Jelinek --- As a middle-ground between ints, HOST_WIDE_INTs and unlimited precision we have offset_int, widest_int etc. But perhaps using even that would be too costly.
[Bug target/84762] GCC for PowerPC32 violates the SysV ABI spec for small struct returns
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84762 --- Comment #8 from Segher Boessenkool --- It is. This PR is about us not following the ABI. On trunk this code is miscompiled; there is at least one other PR for that.
[Bug c++/85135] [7/8 Regression] ICE with invalid late-specified return type
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85135 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org, ||jason at gcc dot gnu.org Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r240756.
[Bug tree-optimization/85168] New: ICE in tree-ssa-coalesce.c: SSA corruption: Unable to coalesce ssa_names which are marked as MUST COALESCE when -O2 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85168
Bug ID: 85168
Summary: ICE in tree-ssa-coalesce.c: SSA corruption: Unable to
coalesce ssa_names which are marked as MUST COALESCE
when -O2 is used
Product: gcc
Version: 8.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: ivo.raisr at oracle dot com
Target Milestone: ---
Created attachment 43815
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43815&action=edit
reproducer.c
With gcc 8.0.1 built recently from SVN, I receive internal compiler error: SSA
corruption when -O2 is used:
void g(void) {
^
0xdcec2e fail_abnormal_edge_coalesce
../../src/gcc/tree-ssa-coalesce.c:1012
0xdcec2e coalesce_partitions
../../src/gcc/tree-ssa-coalesce.c:1401
0xdcec2e coalesce_ssa_name()
../../src/gcc/tree-ssa-coalesce.c:1888
0xd6450e remove_ssa_form
../../src/gcc/tree-outof-ssa.c:948
0xd6450e rewrite_out_of_ssa(ssaexpand*)
../../src/gcc/tree-outof-ssa.c:1174
0x8b2820 execute
../../src/gcc/cfgexpand.c:6227
The full source module is not publicly available. I have creduce'd it to the
following minimal form:
typedef struct {
struct {
char a;
} b;
} c;
int d, f;
c *e;
extern void i(void);
void g(void) {
c *h = e;
if (f) {
i();
h--;
if (d)
if (h->b.a)
i();
}
if (h->b.a)
__builtin_setjmp(0);
e = h;
}
The problem is reproducible with:
$ gcc -fdump-rtl-expand-details -O2 -c
but not with:
$ gcc -fdump-rtl-expand-details -O1 -c
[Bug middle-end/85139] [8 Regression] Spurious format truncation warning
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85139 Richard Biener changed: What|Removed |Added Target|aarch64-bec-linux |aarch64-bec-linux, ||x86_64-*-* Status|UNCONFIRMED |RESOLVED CC||msebor at gcc dot gnu.org Resolution|--- |INVALID --- Comment #1 from Richard Biener --- Confirmed, but I think the warning is correct. %04X will not truncate integer values to 4 characters.
[Bug tree-optimization/85158] ubsan and -O3 finds a problem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 --- Comment #7 from rguenther at suse dot de --- On Tue, 3 Apr 2018, jakub at gcc dot gnu.org wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85158 > > --- Comment #6 from Jakub Jelinek --- > As a middle-ground between ints, HOST_WIDE_INTs and unlimited precision we > have > offset_int, widest_int etc. But perhaps using even that would be too costly. Not sure - we seem to get away with gmp in niter analysis just "fine". What the current code misses if we don't go all the way to unlimited precision is a failure mode on overflow given overflow almost certainly results in wrong-code with enough bad luck. Anyway, IIRC we went to HWIs already to handle "most" cases, and yes, the lambda_vector case is still "unfixed" at ints. But what's still missing is the failure mode... (I fear that a hard assert will trigger way too often...)
[Bug tree-optimization/85168] ICE in tree-ssa-coalesce.c: SSA corruption: Unable to coalesce ssa_names which are marked as MUST COALESCE when -O2 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85168 Ivo Raisr changed: What|Removed |Added CC||ivo.raisr at oracle dot com --- Comment #1 from Ivo Raisr --- Debug output for rtl-expand shows: ... Partition map Partition 0 (h_5(ab) - 5 ) Partition 1 (h_6(ab) - 6 ) Partition 2 (h_7 - 7 ) Partition 3 (h_12(ab) - 12 ) Partition 4 (h_14(ab) - 14 ) Partition 5 (h_20 - 20 ) Partition 6 (pretmp_21 - 21 ) Partition 7 (pretmp_22 - 22 ) Partition 8 (pretmp_23 - 23 ) Partition 9 (prephitmp_24 - 24 ) Conflict graph: 3: 4 4: 3 ... Abnormal coalesce: (6)h_6(ab) & (12)h_12(ab) [map: 1, 3] : Success -> 1 Abnormal coalesce: (6)h_6(ab) & (14)h_14(ab) [map: 1, 4] : Fail due to conflict I am not sure if the conflict is because h_12 and h_14 are still live at the same time or that h_7 and h_20 are not marked abnormal.
[Bug fortran/79854] diagnostics: gfc_conv_constant_to_tree should be gfc_internal_error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79854 --- Comment #4 from Dominique d'Humieres --- > Unless someone can trigger this error or can prove that gfc_internal_error > is available when the error ought to trigger, then please leave the > code as is. Would not it be better to use gcc_unreachable (); as it is done in numerous places?
[Bug target/84762] GCC for PowerPC32 violates the SysV ABI spec for small struct returns
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84762 --- Comment #9 from rguenther at suse dot de --- On Tue, 3 Apr 2018, segher at gcc dot gnu.org wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84762 > > --- Comment #8 from Segher Boessenkool --- > It is. This PR is about us not following the ABI. On trunk this code > is miscompiled; there is at least one other PR for that. So this is a [8 Regression] then? Please fill out known-to-work/fail for me to more easily classify PRs.
[Bug libstdc++/85145] [8 Regression] include_next broken icw/ --with-gxx-include-dir
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85145 Richard Biener changed: What|Removed |Added Target||mingw-w64 Component|c++ |libstdc++ Target Milestone|--- |8.0 Summary|GCC 8 regression: |[8 Regression] include_next |include_next broken icw/|broken icw/ |--with-gxx-include-dir |--with-gxx-include-dir --- Comment #1 from Richard Biener --- Using --with-gxx-include-dir works just fine for me with a native compiler. I suppose your setup is a cross?
[Bug c++/85134] [8 Regression] ICE with invalid constexpr in 'shared' OpenMP clause
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85134 --- Comment #1 from Jakub Jelinek --- Indeed, started with r258015.
[Bug c++/85133] [7/8 Regression] [concepts] ICE with invalid concept used in variadic template
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85133 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org, ||jason at gcc dot gnu.org Ever confirmed|0 |1 --- Comment #1 from Jakub Jelinek --- Started with r238558, before it has been rejected with: pr85133.C:1:33: error: variable concept has no initializer template concept bool C; ^ pr85133.C:5:6: error: template constraint failure A a; ^ pr85133.C:5:6: note: constraints not satisfied pr85133.C:5:6: note: in the expansion of ‘C< >...’ pr85133.C:5:6: note: ‘C’ was not satisfied
[Bug libstdc++/85145] [8 Regression] include_next broken icw/ --with-gxx-include-dir
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85145 --- Comment #2 from Jeroen Ooms --- I'm building mingw-w64 on msys2. Not sure if this counts as a cross. The full build script is given in the Github repo mentioned above.
[Bug c++/85153] _Pragma to disable -Wswitch-unreachable diagnostic not properly working when used within preprocessor macro
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85153 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 Component|driver |c++ Ever confirmed|0 |1 --- Comment #3 from Richard Biener --- Confirmed. Not sure what the exact limit is but some of the warnings can only be disabled at function granularity and placing the pragmas in macros might interfere with this.
[Bug c++/85156] [8 Regression] ICE with -O1 -g: gimplification failed
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85156 Jakub Jelinek changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #2 from Jakub Jelinek --- Looks similar to PR84704, the STATEMENT_LIST is gimplified twice and as gimplification is destructive, that can't really work well.
[Bug tree-optimization/85157] [parloops] Prevent canonicalize_loop_ivs failure in gen_parallel_loop
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85157 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 Ever confirmed|0 |1 --- Comment #1 from Richard Biener --- Confirmed.
[Bug c++/85159] if constexpr error about goto
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85159 --- Comment #1 from Richard Biener --- But is this really valid code?
[Bug other/85161] [8 regression] Test case failures in libbacktrace on powerpc64 BE starting with r253456
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85161 Richard Biener changed: What|Removed |Added Priority|P3 |P1 Target Milestone|--- |8.0
[Bug rtl-optimization/85160] GCC generates mvn/and instructions instead of bic on aarch64
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85160 Richard Biener changed: What|Removed |Added CC||segher at gcc dot gnu.org Component|middle-end |rtl-optimization Version|unknown |8.0 --- Comment #1 from Richard Biener --- Yes, combine only tries multi-uses in limited circumstances.
[Bug testsuite/85154] [8 regression] gcc.dg/vect/vect-95.c fail
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85154 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2018-04-03 Component|tree-optimization |testsuite Target Milestone|--- |8.0 Ever confirmed|0 |1 --- Comment #1 from Richard Biener --- silvermont? *sigh* --with-arch/cpu makes it really hard to invent testcase guards. pentium4, nocona, atom, slm, intel (but not core!) cost tables are affected by maintaining higher unaligned load/store cost than aligned ones. The x86 backend having so many tunings and different cost metrics don't make it easy to fix the testcase. Well, we can just remove that dump-scanning altogether.
[Bug testsuite/85154] [8 regression] gcc.dg/vect/vect-95.c fail
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85154 --- Comment #2 from Richard Biener --- Author: rguenth Date: Tue Apr 3 09:54:22 2018 New Revision: 259025 URL: https://gcc.gnu.org/viewcvs?rev=259025&root=gcc&view=rev Log: 2018-04-03 Richard Biener PR testsuite/85154 * gcc.dg/vect/vect-95.c: Remove scan for alignment peeling. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/gcc.dg/vect/vect-95.c
[Bug testsuite/85154] [8 regression] gcc.dg/vect/vect-95.c fail
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85154 Richard Biener changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|--- |FIXED --- Comment #3 from Richard Biener --- Fixed.
[Bug tree-optimization/85162] Vector extensions generating incorrect assembly
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85162 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |INVALID --- Comment #2 from Richard Biener --- Invalid, vector comparison results are -1 and 0, not 1 and 0.
[Bug other/85123] [8 regression] test case gcc.dg/pr63594-2.c begins failing with r254862
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85123 Segher Boessenkool changed: What|Removed |Added Status|CLOSED |REOPENED Resolution|DUPLICATE |--- --- Comment #5 from Segher Boessenkool --- Reopening this one for the regression. Let's keep PR84762 just for what it was reported for: GCC is not compliant with the SYSV ABI for some things (and never has been).
[Bug target/84762] GCC for PowerPC32 violates the SysV ABI spec for small struct returns
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84762 --- Comment #10 from Segher Boessenkool --- I reopened PR85123 for the regression. Let's keep this bug for purely the SYSV ABI non-compliance. This is not a regression (it is ancient).
[Bug c++/85156] [8 Regression] ICE with -O1 -g: gimplification failed
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85156
--- Comment #3 from Jakub Jelinek ---
It is not just ICE, but also a wrong-code starting with GCC 4.3:
int x, y;
__attribute__((noipa)) int
foo (int z)
{
if (__builtin_expect (x ? y != 0 : 0, z++))
return 7;
return z;
}
int
main ()
{
x = 1;
asm volatile ("" : "+m" (x), "+m" (y));
if (foo (10) != 11)
__builtin_abort ();
return 0;
}
The side-effects of the second __builtin_expect argument are evaluated multiple
times.
[Bug tree-optimization/85168] ICE in tree-ssa-coalesce.c: SSA corruption: Unable to coalesce ssa_names which are marked as MUST COALESCE when -O2 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85168 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2018-04-03 Assignee|unassigned at gcc dot gnu.org |rguenth at gcc dot gnu.org Ever confirmed|0 |1 --- Comment #2 from Richard Biener --- Confirmed. I'll take a look.
[Bug tree-optimization/85156] [8 Regression] ICE with -O1 -g: gimplification failed
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85156 Jakub Jelinek changed: What|Removed |Added Status|NEW |ASSIGNED Assignee|unassigned at gcc dot gnu.org |jakub at gcc dot gnu.org --- Comment #4 from Jakub Jelinek --- Created attachment 43816 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43816&action=edit gcc8-pr85156.patch Untested fix.
[Bug c++/85114] -fstack-check causes internal compiler error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85114 --- Comment #3 from Richard Biener --- Thanks, still can't reproduce here though. Can you check whether netbsd has any patches ontop of FSF 6.4.0?
[Bug c++/85134] [8 Regression] ICE with invalid constexpr in 'shared' OpenMP clause
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85134 Jakub Jelinek changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #2 from Jakub Jelinek --- Created attachment 43817 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43817&action=edit gcc8-pr85134.patch Untested fix. The problem is in the early return from cp_finish_decl, we don't emit then the DECL_EXPR that the OpenMP code relies on for VLAs.
[Bug tree-optimization/85168] [7/8 Regression] ICE in tree-ssa-coalesce.c: SSA corruption: Unable to coalesce ssa_names which are marked as MUST COALESCE when -O2 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85168 Richard Biener changed: What|Removed |Added Keywords||ice-on-valid-code Priority|P3 |P2 Target Milestone|--- |7.4 Summary|ICE in tree-ssa-coalesce.c: |[7/8 Regression] ICE in |SSA corruption: Unable to |tree-ssa-coalesce.c: SSA |coalesce ssa_names which|corruption: Unable to |are marked as MUST COALESCE |coalesce ssa_names which |when -O2 is used|are marked as MUST COALESCE ||when -O2 is used --- Comment #3 from Richard Biener --- caused by code-hoisting but eventually latent.
[Bug target/85169] New: [8 Regression] wrong code with vector member insert
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85169
Bug ID: 85169
Summary: [8 Regression] wrong code with vector member insert
Product: gcc
Version: 8.0.1
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: zsojka at seznam dot cz
Target Milestone: ---
Host: x86_64-pc-linux-gnu
Target: x86_64-pc-linux-gnu
Created attachment 43818
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43818&action=edit
reduced testcase
This looks like a recent regression.
Output:
$ x86_64-pc-linux-gnu-gcc -Og -mavx512bw testcase.c
$ sde64 -- ./a.out
Aborted
The value of v[] is:
v = {0 , 1 }
$ x86_64-pc-linux-gnu-gcc -v
Using built-in specs.
COLLECT_GCC=/repo/gcc-trunk/binary-latest-amd64/bin/x86_64-pc-linux-gnu-gcc
COLLECT_LTO_WRAPPER=/repo/gcc-trunk/binary-trunk-259024-checking-yes-rtl-df-extra-amd64/bin/../libexec/gcc/x86_64-pc-linux-gnu/8.0.1/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /repo/gcc-trunk//configure --enable-languages=c,c++
--enable-valgrind-annotations --disable-nls --enable-checking=yes,rtl,df,extra
--with-cloog --with-ppl --with-isl --build=x86_64-pc-linux-gnu
--host=x86_64-pc-linux-gnu --target=x86_64-pc-linux-gnu
--with-ld=/usr/bin/x86_64-pc-linux-gnu-ld
--with-as=/usr/bin/x86_64-pc-linux-gnu-as --disable-libstdcxx-pch
--prefix=/repo/gcc-trunk//binary-trunk-259024-checking-yes-rtl-df-extra-amd64
Thread model: posix
gcc version 8.0.1 20180403 (experimental) (GCC)
[Bug target/85170] New: ICE: in final_scan_insn_1, at final.c:3139 (error: could not split insn)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85170
Bug ID: 85170
Summary: ICE: in final_scan_insn_1, at final.c:3139 (error:
could not split insn)
Product: gcc
Version: unknown
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: asolokha at gmx dot com
Target Milestone: ---
Target: powerpcspe-*-linux-gnu
gcc-8.0.0-alpha20180401 snapshot (r258999) ICEs when compiling the following
snippet w/ -O1 -fkeep-gc-roots-live -ftrapv -ftree-loop-vectorize
-fno-split-wide-types for a SPE target:
int
ae (int *ex, int ni)
{
int it = 0;
while (ni < 1)
{
it += ex[ni];
++ni;
}
return it;
}
% powerpc-e500v2-linux-gnuspe-gcc-8.0.0-alpha20180401 -O1 -fkeep-gc-roots-live
-ftrapv -ftree-loop-vectorize -fno-split-wide-types -c ywcbo250.c
ywcbo250.c: In function 'ae':
ywcbo250.c:13:1: error: could not split insn
}
^
(insn 44 43 46 (set (reg:DI 28 28 [orig:195 vect__4.10 ] [195])
(mem:DI (plus:SI (reg/f:SI 26 26 [orig:172 vectp.9 ] [172])
(reg:SI 27 27 [orig:155 ivtmp.17 ] [155])) [0 MEM[base:
vectp.9_58, index: ivtmp.17_14, offset: 0B]+0 S8 A64])) 593 {*movdi_internal32}
(nil))
during RTL pass: final
ywcbo250.c:13:1: internal compiler error: in final_scan_insn_1, at final.c:3139
0x54c424 _fatal_insn(char const*, rtx_def const*, char const*, int, char
const*)
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/rtl-error.c:108
0x890bce final_scan_insn_1
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/final.c:3139
0x890f37 final_scan_insn(rtx_insn*, _IO_FILE*, int, int, int*)
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/final.c:3218
0x8910c7 final_1
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/final.c:2088
0x891fe7 rest_of_handle_final
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/final.c:4671
0x891fe7 execute
/var/tmp/portage/cross-powerpc-e500v2-linux-gnuspe/gcc-8.0.0_alpha20180401/work/gcc-8-20180401/gcc/final.c:4745
[Bug c++/85114] -fstack-check causes internal compiler error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85114 --- Comment #4 from Thomas Klausner --- Thank you for looking at this. Yes, there are quite a number of changes in NetBSD's gcc: http://danbala.tuwien.ac.at/~wiz/tmp/netbsd.gcc-6.4.0.diff.gz (created by: cd /usr/src/external/gpl3/gcc/dist cvs diff -kk -rgcc-6-4-0) The corresponding cvsweb path is: http://cvsweb.netbsd.org/bsdweb.cgi/src/external/gpl3/gcc/dist/?only_with_tag=MAIN if you want to investigate the reason for particular changes. In addition, there is a separate build system used. You can find the Makefiles at http://cvsweb.netbsd.org/bsdweb.cgi/src/external/gpl3/gcc/?only_with_tag=MAIN and below the lib and usr.bin directories.
[Bug rtl-optimization/80791] [8 regression] test case gcc.dg/sms-1.c fail2 starting with r247885
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80791 --- Comment #14 from amker at gcc dot gnu.org --- (In reply to Bill Schmidt from comment #13) > Actually it appears that the IVOPTS change results in worse code going into > SMS, regardless of whether SMS can succeed on the loop. It comes down to > the fact that IVOPTS formerly pulled a multiply (left-shift) out of the > loop, but now the multiply remains in the loop. > > After all the do_loop work and so forth, we end up with the following going > into SMS. The r247884 loop is on the left, r247885 on the right: > > lab27:lab20: > r162, ca = r162 >> 1 r157, ca = r157 >> 1 > r169 = r162 << 3 r163 = r157 << 3 > r166 = (r160 << 3) & 0xfff8 > r171 = r177 + r164r167 = r172 + r166 > *(r177 + r169) = r171 *(r172 + r163) = r167 > r174 = (SI)r164 + 32 r169 = (SI)r160 + 4 > r164 = zext(r174) r164 = sext(r169) > bdnz r178, lab27 bdnz r173, lab20 > > The loss of hoisting of the multiply/shift shows up in the optimized and > vregs dumps as well; RTL optimizations aren't able to recover from this. > > So I think this still comes down to an unfortunate choice made in IVOPTS. According to comment #6, I think IVOPTs computed correct cost because multiplication "_17 = _18 * 8;" is actually a bit-shift.
[Bug c++/85114] -fstack-check causes internal compiler error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85114 Richard Biener changed: What|Removed |Added Keywords||needs-reduction --- Comment #5 from Richard Biener --- Not really - I suggest to file a bug to the netbsd folks. Reducing the testcase might also help, not sure.
[Bug target/85169] [8 Regression] wrong code with vector member insert
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85169 Jakub Jelinek changed: What|Removed |Added Priority|P3 |P1 Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org Assignee|unassigned at gcc dot gnu.org |jakub at gcc dot gnu.org Target Milestone|--- |8.0 Ever confirmed|0 |1
[Bug target/85169] [8 Regression] wrong code with vector member insert
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85169 --- Comment #1 from Jakub Jelinek --- Created attachment 43819 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43819&action=edit gcc8-pr85169.patch Untested fix.
[Bug c++/85153] _Pragma to disable -Wswitch-unreachable diagnostic not properly working when used within preprocessor macro
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85153 --- Comment #4 from Fabio Alemagna --- @(In reply to Richard Biener from comment #3) > Confirmed. Not sure what the exact limit is but some of the warnings can > only be disabled at function granularity and placing the pragmas in macros > might interfere with this. But the preprocessed output shows the #pragma's all at the expected place and compiling the preprocessed output works as expected. I only ever delved into gcc's innards in one of its old releases and don't know if anything has changed meanwhile about this, but afaik the compiler always works on the preprocessed output, and if this still holds true, how can it be explained that invoking g++ on the un-preprocessed output produces a different result than invoking it on the preprocessed output?
[Bug tree-optimization/85168] [7/8 Regression] ICE in tree-ssa-coalesce.c: SSA corruption: Unable to coalesce ssa_names which are marked as MUST COALESCE when -O2 is used
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85168 --- Comment #4 from Richard Biener --- Created attachment 43820 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43820&action=edit patch in testing
[Bug rtl-optimization/85167] shrink-wrap.c:333:15: runtime error with UBSAN
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85167 Jakub Jelinek changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2018-04-03 CC||jakub at gcc dot gnu.org Assignee|unassigned at gcc dot gnu.org |jakub at gcc dot gnu.org Ever confirmed|0 |1
[Bug rtl-optimization/85167] [6/7/8 Regression] shrink-wrap.c:333:15: runtime error with UBSAN
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85167 Jakub Jelinek changed: What|Removed |Added Priority|P3 |P2 Target Milestone|--- |6.5 Summary|shrink-wrap.c:333:15: |[6/7/8 Regression] |runtime error with UBSAN|shrink-wrap.c:333:15: ||runtime error with UBSAN --- Comment #1 from Jakub Jelinek --- Started with r210457.
[Bug rtl-optimization/85167] [6/7/8 Regression] shrink-wrap.c:333:15: runtime error with UBSAN
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85167 --- Comment #2 from Jakub Jelinek --- Created attachment 43821 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43821&action=edit gcc8-pr85167.patch Untested fix.
[Bug c++/85159] if constexpr error about goto
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85159 Jonathan Wakely changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |INVALID --- Comment #2 from Jonathan Wakely --- (In reply to Richard Biener from comment #1) > But is this really valid code? No. "A label (9.1) declared in a substatement of a constexpr if statement shall only be referred to by a statement (9.6.4) in the same substatement."
[Bug c++/85171] New: generic lambda becomes recursive
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85171 Bug ID: 85171 Summary: generic lambda becomes recursive Product: gcc Version: 8.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: nathan at gcc dot gnu.org Target Milestone: --- Created attachment 43822 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43822&action=edit testcase The attached testcase compiles without error, producing a recursive lambda. During instanitation 'operator()' is looked up in the context of the lambda type, and finds its synthesized template lambda::operator()(T)->int member function. The non-generic case fails to find operator() at parse time and errors out. If this lambda was embedded in a class with its own operator(), I think that's the one we should be finding. I presume the parser's saving a raw identifier for the operator () and then we do a lookup in tsubst. That lookup should not look in the lambda hierarchy. If this was a non-lambda instantiation, we'd either find operator() in the non-dependent hierarchy, and use that at instantiation time, or we'd find nothing and look in the enclosing namespace scopes. The wrinkle with generic lambdas is that there could be enclosing non-namespace scopes of course.
[Bug libstdc++/80654] is_trivially_copy_constructible fails with compiler error with vector of uncopyable objects
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80654 --- Comment #10 from Jonathan Wakely --- (In reply to Frank Heckenbach from comment #8) > (In reply to Ville Voutilainen from comment #7) > > That's not a bug. You need to make the copy constructor of s conditionally > > deleted depending on whether T is copyconstructible. > > The copy constructor of std::vector doesn't seem to do that either > (bits/stl_vector.h:326 in 7.3.0) unless I'm missing something. It's impossible to do it for std::vector, because whether the vector is copy-constructible depends on whether the value_type is CopyInsertable (**not** copy-constructible) which depends on the allocator and is not known (or knowable) at the right time, because std::vector allows its value_type to be incomplete. > Is that what Daniel Krügler meant by "not SFINAE-friendly"? But he also > wrote: "This is a QoI issue but not a violation of the requirements of the > standard.", so I'm confused now. The standard doesn't say that std::vector has a conditionally deleted copy constructor, so it doesn't require traits to give the right answer. We could add partial specializations for std::vector> (where the allocator behaviour is known) but we can't do it for std::vector or std::vector, ArbitraryAllocator>.
[Bug inline-asm/85172] New: internal compiler error: unexpected expression '' of kind asm_expr
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85172
Bug ID: 85172
Summary: internal compiler error: unexpected expression
'' of kind asm_expr
Product: gcc
Version: 8.0.1
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: inline-asm
Assignee: unassigned at gcc dot gnu.org
Reporter: vegard.nossum at oracle dot com
Target Milestone: ---
Input (valid code AFAICT):
int f() {
return !__builtin_constant_p(({
asm("");
0;
}));
}
Output:
$ cc1plus
int f()
output/unexpected-expression.cc:5:5: internal compiler error: unexpected
expression '' of kind asm_expr
}));
^
0xa38827 cxx_eval_constant_expression
/home/vegard/git/gcc/gcc/cp/constexpr.c:4786
0xa31d62 cxx_eval_constant_expression
/home/vegard/git/gcc/gcc/cp/constexpr.c:4423
0xa306a3 cxx_eval_statement_list
/home/vegard/git/gcc/gcc/cp/constexpr.c:3922
0xa306a3 cxx_eval_constant_expression
/home/vegard/git/gcc/gcc/cp/constexpr.c:4670
0xa5bc38 cxx_eval_builtin_function_call
/home/vegard/git/gcc/gcc/cp/constexpr.c:1189
0xa24598 cxx_eval_call_expression
/home/vegard/git/gcc/gcc/cp/constexpr.c:1496
0xa31c70 cxx_eval_constant_expression
/home/vegard/git/gcc/gcc/cp/constexpr.c:4200
0xa4b8ca cxx_eval_outermost_constant_expr
/home/vegard/git/gcc/gcc/cp/constexpr.c:4846
0xa591b6 maybe_constant_value(tree_node*, tree_node*)
/home/vegard/git/gcc/gcc/cp/constexpr.c:5063
0xab906a cp_fully_fold(tree_node*)
/home/vegard/git/gcc/gcc/cp/cp-gimplify.c:2071
0x13da147 cp_build_binary_op(unsigned int, tree_code, tree_node*, tree_node*,
int)
/home/vegard/git/gcc/gcc/cp/typeck.c:5391
0xae49c7 ocp_convert(tree_node*, tree_node*, int, int, int)
/home/vegard/git/gcc/gcc/cp/cvt.c:812
0xaec6ac cp_convert(tree_node*, tree_node*, int)
/home/vegard/git/gcc/gcc/cp/cvt.c:623
0xaec6ac cp_convert_and_check(tree_node*, tree_node*, int)
/home/vegard/git/gcc/gcc/cp/cvt.c:642
0x8e068f convert_like_real
/home/vegard/git/gcc/gcc/cp/call.c:7116
0x8e5c29 perform_implicit_conversion_flags(tree_node*, tree_node*, int, int)
/home/vegard/git/gcc/gcc/cp/call.c:10669
0x13c91f8 cp_build_unary_op(tree_code, tree_node*, bool, int)
/home/vegard/git/gcc/gcc/cp/typeck.c:6148
0x94399b build_new_op_1
/home/vegard/git/gcc/gcc/cp/call.c:6026
0x946036 build_new_op(unsigned int, tree_code, int, tree_node*, tree_node*,
tree_node*, tree_node**, int)
/home/vegard/git/gcc/gcc/cp/call.c:6058
0x13d000c build_x_unary_op(unsigned int, tree_code, cp_expr, int)
/home/vegard/git/gcc/gcc/cp/typeck.c:5601
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See
[Bug tree-optimization/84737] [8 Regression] 20% degradation in CPU2000 172.mgrid starting with r256888
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84737 Martin Liška changed: What|Removed |Added Assignee|marxin at gcc dot gnu.org |unassigned at gcc dot gnu.org --- Comment #13 from Martin Liška --- I'm unassigning ..
[Bug c++/65923] False positive for warning about literal operator suffix and using
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65923 Jonathan Wakely changed: What|Removed |Added CC||eric at efcs dot ca --- Comment #4 from Jonathan Wakely --- *** Bug 77999 has been marked as a duplicate of this bug. ***
[Bug c++/77999] GCC diagnoses usage of STL literal operators as invalid.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77999 Jonathan Wakely changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |DUPLICATE --- Comment #4 from Jonathan Wakely --- dup *** This bug has been marked as a duplicate of bug 65923 ***
[Bug c++/65923] False positive for warning about literal operator suffix and using
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65923 --- Comment #5 from Jonathan Wakely --- (In reply to Richard Geary from comment #2) > This is still present in g++ 5.3.0, and also affects > std::literals::string_literals. The flag -Wno-literal-suffix does not turn > it off. It does for GCC 7+, that was PR 69523 (which looks like a transposition of this bug number, but isn't).
[Bug target/85173] New: ICE in in extract_insn, at recog.c:2304 for -fstack-clash-protection on ARM
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85173 Bug ID: 85173 Summary: ICE in in extract_insn, at recog.c:2304 for -fstack-clash-protection on ARM Product: gcc Version: unknown Status: UNCONFIRMED Keywords: ice-on-valid-code Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: marxin at gcc dot gnu.org CC: amker at gcc dot gnu.org, kyrylo.tkachov at arm dot com Target Milestone: --- Host: x86_64-pc-linux-gnu Target: arm-linux-gnueabi I see: $ arm-linux-gnueabi-gcc /home/marxin/Programming/gcc/gcc/testsuite/gcc.dg/stack-check-3.c /dev/null --param stack-clash-protection-probe-interval=14 -fstack-clash-protection 0xdeadbeef _fatal_insn(char const*, rtx_def const*, char const*, int, char const*) .././../gcc/rtl-error.c:108 0xdeadbeef _fatal_insn_not_found(rtx_def const*, char const*, int, char const*) .././../gcc/rtl-error.c:116 0xdeadbeef extract_insn(rtx_insn*) .././../gcc/recog.c:2304 0xdeadbeef instantiate_virtual_regs_in_insn .././../gcc/function.c:1599 0xdeadbeef instantiate_virtual_regs .././../gcc/function.c:1969 0xdeadbeef execute .././../gcc/function.c:2018
[Bug c++/85171] generic lambda becomes recursive
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85171 --- Comment #1 from Nathan Sidwell --- Created attachment 43823 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=43823&action=edit member fn example The second example shows the case inside a member fn, with ok and bad cases. X::ok calls X::operator(), X::bad recurses
[Bug middle-end/85090] [8 Regression] wrong code with -O2 -fno-tree-dominator-opts -mavx512f -fira-algorithm=priority
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85090 Jakub Jelinek changed: What|Removed |Added CC|jakub at redhat dot com| --- Comment #9 from Jakub Jelinek --- Vlad, do you think you could look at the IRA issue (with the above patch reverted)?
[Bug c++/85128] Local scope type(def) falsely assumed to change meaning of outer scope type(def)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85128 --- Comment #11 from Jonathan Wakely --- (In reply to Steffen Nurpmeso from comment #3) > And despite absolutely idiotic C++ "improvements" like obsoleting the "auto" > keyword and introducing override (yes) as a declaration suffix (no) that i > had to fix Or you could compile with -std=c++98 > (we had override, as a macro that expanded to virtual) Yuck.
[Bug bootstrap/64914] [UBSAN/bootstrap-ubsan] With -g3: libiberty/md5.c:336:7: runtime error: load of misaligned address for type 'const md5_uint32', which requires 4 byte alignment
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64914 Martin Liška changed: What|Removed |Added Target Milestone|--- |9.0 --- Comment #3 from Martin Liška --- Patch candidate: https://gcc.gnu.org/ml/gcc-patches/2018-03/msg00022.html
[Bug target/71991] Inconsistency for __attribute__ ((__always_inline__)) among LTO and non-LTO compilation
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71991 Martin Liška changed: What|Removed |Added Target Milestone|--- |8.0
[Bug c++/85127] Members implementation outside class doesn't compile if class name is an alias from a using
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85127
--- Comment #2 from Jonathan Wakely ---
EDG rejects it too:
"us.cc", line 11: error: template argument list must match the parameter list
void B::foo() {
^
1 error detected in the compilation of "us.cc".
I'm not aware of any rule making this invalid, so I agree it should be
accepted.
[Bug c++/85125] constant expression with const_cast UB does not emit error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85125 Jonathan Wakely changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed||2018-04-03 Ever confirmed|0 |1 --- Comment #1 from Jonathan Wakely --- N.B. GCC 5.x releases are no longer supported, but this bug is still present in current releases.
[Bug c++/85118] [6/7/8 Regression] Error when using std::bind with a generic lambda - "cannot bind 'const volatile char&' to an rvalue of type 'const volatile char'"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85118 --- Comment #2 from Jonathan Wakely --- Ugh, this again. I've battled with this when working on the implementation of std::bind. I'll try to reduce it to something independent of the library headers.
[Bug middle-end/64327] ../../gcc/gcc/rtlanal.c:4881:48: runtime error: shift exponent 4294967295 is too large for 64-bit type 'long unsigned int'
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64327 Martin Liška changed: What|Removed |Added Status|WAITING |NEW Known to work||7.3.0, 8.0 Known to fail||6.4.0 --- Comment #10 from Martin Liška --- Ok, so let's close it when GCC-6 will be closed.
[Bug preprocessor/78497] compiling with -save-temps adds -Wimplicit-fallthrough warnings
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78497 Martin Liška changed: What|Removed |Added Target Milestone|--- |8.0
[Bug sanitizer/85174] New: c-c++-common/asan/pointer-compare-1.c does not work for powerpc with sdata
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85174
Bug ID: 85174
Summary: c-c++-common/asan/pointer-compare-1.c does not work
for powerpc with sdata
Product: gcc
Version: unknown
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: sanitizer
Assignee: unassigned at gcc dot gnu.org
Reporter: segher at gcc dot gnu.org
CC: dodji at gcc dot gnu.org, dvyukov at gcc dot gnu.org,
jakub at gcc dot gnu.org, kcc at gcc dot gnu.org, marxin at
gcc dot gnu.org
Target Milestone: ---
Specifically, these two tests fail:
/* { dg-output "ERROR: AddressSanitizer: invalid-pointer-pair.*" } */
foo (p - 1, p);
/* { dg-output "ERROR: AddressSanitizer: invalid-pointer-pair.*" } */
foo (p, p - 1);
where
char *p = &small_global[0];
and that is
char small_global[7] = {};
Adding -mdata=none makes it all work; but should asan handle this?
[Bug tree-optimization/85175] New: [8 regression] false-positive -Wformat-overflow= warning with gcc-8 -Os
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85175
Bug ID: 85175
Summary: [8 regression] false-positive -Wformat-overflow=
warning with gcc-8 -Os
Product: gcc
Version: 8.0.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: arnd at linaro dot org
Target Milestone: ---
This snippet from the Linux kernel produces a bogus warning when built with gcc
-Os, using a recent snapshot (20180402) A two months older version did not
produce the warning, and building with -O2 is also fine. Reduced test case:
$ gcc-8.0.1 -Os -Wall -c test.c
int of_property_read_u32(int *out_value);
int imx_ldb_bind(void)
{
int i;
for (i = 0; i < 4; i++) {
char clkname[16];
__builtin_sprintf(clkname, "di%d_sel", i);
}
return of_property_read_u32(&i);
}
imx-ldb.i: In function 'imx_ldb_bind':
imx-ldb.i:7:35: warning: '_sel' directive writing 4 bytes into a region of size
between 3 and 13 [-Wformat-overflow=]
__builtin_sprintf(clkname, "di%d_sel", i);
[Bug tree-optimization/85175] [8 regression] false-positive -Wformat-overflow= warning with gcc-8 -Os
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85175 Richard Biener changed: What|Removed |Added Target Milestone|--- |8.0
[Bug target/85169] [8 Regression] wrong code with vector member insert
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85169 --- Comment #2 from Richard Biener --- Looks obvious.
[Bug middle-end/85123] [8 regression] test case gcc.dg/pr63594-2.c begins failing with r254862
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85123 Richard Biener changed: What|Removed |Added Keywords||wrong-code Priority|P3 |P1 Component|other |middle-end Target Milestone|--- |8.0
