[Bug target/91767] [10 regression] After r274953, clang-compiled xgcc segfaults during RTL pass: stv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91767 Jakub Jelinek changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #2 from Jakub Jelinek --- Isn't that a clang bug though? hash_map has a constructor which should be invoked by the general_scalar_chain constructor...
[Bug libgcc/91737] On Alpine Linux (libmusl) a statically linked C++ program which throws the first exception in two threads at the same time can busy spin on shutdown after main().
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91737 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |MOVED --- Comment #1 from Andrew Pinski --- This has been discussed over and over again. This is an issue in the libc and not in GCC. In your case in libmusl. Glibc has a similar bug and been discussed how to fix it. The way Glibc is going to fix it (though it has not yet) is that libpthread.a will be really just include one object file which includes all of the pthread library. libmusl should fix it a similar way.
[Bug target/91766] -fvisibility=hidden during -fpic still uses GOT indirection on arm64
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91766 Andrew Pinski changed: What|Removed |Added Keywords||missed-optimization Target||aarch64-linux-gnu --- Comment #2 from Andrew Pinski --- room is a common symbol which causes issues.
[Bug target/91766] -fvisibility=hidden during -fpic still uses GOT indirection on arm64
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91766 --- Comment #1 from Andrew Pinski --- Try -fno-common.
[Bug target/91767] After r274953, clang-compiled xgcc segfaults during RTL pass: stv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91767 --- Comment #1 from Andrew Pinski --- If anything, one of the general_scalar_chain objects is not being constructed correctly. That is: hash_map defs_map; not being done correctly.
[Bug libgomp/82391] [openacc] acc_on_device not folded at -O0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82391 --- Comment #4 from Tom de Vries --- Thomas, could you update this PR with the current state?
[Bug fortran/91729] [10 Regression] ICE in gfc_match_select_rank, at fortran/match.c:6586
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91729
kargl at gcc dot gnu.org changed:
What|Removed |Added
CC||kargl at gcc dot gnu.org
--- Comment #1 from kargl at gcc dot gnu.org ---
This patch avoids the ICE.
Index: gcc/fortran/match.c
===
--- gcc/fortran/match.c (revision 275703)
+++ gcc/fortran/match.c (working copy)
@@ -6510,7 +6510,7 @@ gfc_match_select_rank (void)
char name[GFC_MAX_SYMBOL_LEN];
gfc_symbol *sym, *sym2;
gfc_namespace *ns = gfc_current_ns;
- gfc_array_spec *as;
+ gfc_array_spec *as = NULL;
m = gfc_match_label ();
if (m == MATCH_ERROR)
@@ -6583,12 +6583,20 @@ gfc_match_select_rank (void)
return m;
}
- sym = expr1->symtree->n.sym;
- as = sym->ts.type == BT_CLASS ? CLASS_DATA (sym)->as : sym->as;
+ if (expr1->symtree)
+ {
+ sym = expr1->symtree->n.sym;
+ as = sym->ts.type == BT_CLASS ? CLASS_DATA (sym)->as : sym->as;
+ }
+
if (expr1->expr_type != EXPR_VARIABLE
|| !(as && as->type == AS_ASSUMED_RANK))
- gfc_error_now ("The SELECT RANK selector at %C must be an assumed "
- "rank variable");
+ {
+ gfc_error ("The SELECT RANK selector at %C must be an assumed "
+"rank variable");
+ m = MATCH_ERROR;
+ goto cleanup;
+ }
}
m = gfc_match (" )%t");
[Bug fortran/91566] [9/10 Regression] ICE in gfc_constructor_copy, at fortran/constructor.c:103
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91566 --- Comment #2 from kargl at gcc dot gnu.org --- Author: kargl Date: Fri Sep 13 20:19:40 2019 New Revision: 275704 URL: https://gcc.gnu.org/viewcvs?rev=275704=gcc=rev Log: 2019-09-13 Steven G. Kargl PR fortran/91566 * simplify.c (gfc_simplify_merge): Need to simplify expression after insertation of parenthesis. 2019-09-13 Steven G. Kargl PR fortran/91566 * gfortran.dg/pr91566.f90: Added: trunk/gcc/testsuite/gfortran.dg/pr91566.f90 Modified: trunk/gcc/fortran/ChangeLog trunk/gcc/fortran/simplify.c trunk/gcc/testsuite/ChangeLog
[Bug bootstrap/91767] New: After r274953, clang-compiled xgcc segfaults during RTL pass: stv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91767 Bug ID: 91767 Summary: After r274953, clang-compiled xgcc segfaults during RTL pass: stv Product: gcc Version: 10.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: bootstrap Assignee: unassigned at gcc dot gnu.org Reporter: dimitry at andric dot com Target Milestone: --- As reported in https://bugs.freebsd.org/240387, after the gcc10-devel port was updated from the 20190825 snapshot to the 20190901 snapshot, bootstrapping started to fail due to a segfault, when the host compiler is clang 8.0. Bisecting showed that the segfaults start appearing after r274953. The error appears for different source files, when building libgcc with xgcc, but it always looks like: during RTL pass: stv /home/dim/src/gcc/libgcc/config/libbid/bid128_fma.c: In function 'bid128_ext_fma': /home/dim/src/gcc/libgcc/config/libbid/bid128_fma.c:3569:1: internal compiler error: Segmentation fault 3569 | } | ^ Running under valgrind shows: ==19771== Invalid read of size 1 ==19771==at 0x82D7DC: gen_rtx_SUBREG(machine_mode, rtx_def*, poly_int<1u, unsigned long>) (src/gcc/gcc/emit-rtl.c:1013) ==19771==by 0xFFE42E: make_vector_copies (src/gcc/gcc/config/i386/i386-features.c:716) ==19771==by 0xFFE42E: (anonymous namespace)::general_scalar_chain::convert_registers() (src/gcc/gcc/config/i386/i386-features.c:1173) ==19771==by 0xFFC3B8: convert (src/gcc/gcc/config/i386/i386-features.c:1192) ==19771==by 0xFFC3B8: convert_scalars_to_vector (src/gcc/gcc/config/i386/i386-features.c:1629) ==19771==by 0xFFC3B8: (anonymous namespace)::pass_stv::execute(function*) (src/gcc/gcc/config/i386/i386-features.c:1767) ==19771==by 0xB3463D: execute_one_pass(opt_pass*) (src/gcc/gcc/passes.c:2494) ==19771==by 0xB35047: execute_pass_list_1(opt_pass*) (src/gcc/gcc/passes.c:2580) ==19771==by 0xB35059: execute_pass_list_1(opt_pass*) (src/gcc/gcc/passes.c:2581) ==19771==by 0xB27164: execute_pass_list(function*, opt_pass*) (src/gcc/gcc/passes.c:2591) ==19771==by 0x768E5A: cgraph_node::expand() (src/gcc/gcc/cgraphunit.c:2194) ==19771==by 0x76C4DD: expand_all_functions (src/gcc/gcc/cgraphunit.c:2332) ==19771==by 0x76C4DD: symbol_table::compile() (src/gcc/gcc/cgraphunit.c:2688) ==19771==by 0x76CA4F: symbol_table::finalize_compilation_unit() (src/gcc/gcc/cgraphunit.c:2868) ==19771==by 0xC150F3: compile_file() (src/gcc/gcc/toplev.c:481) ==19771==by 0xC14BCE: do_compile (src/gcc/gcc/toplev.c:2166) ==19771==by 0xC14BCE: toplev::main(int, char**) (src/gcc/gcc/toplev.c:2301) ==19771== Address 0x2 is not stack'd, malloc'd or (recently) free'd ==19771== Unfortunately this does not give much more information than simply running under gdb. It looks like an entry in a hash map is searched for, but unexpectedly returns NULL. This NULL value is then passed as a rtx_def pointer to gen_rtx_SUBREG(), which attempts to access one of the struct elements, and this causes the segfault.
[Bug middle-end/91708] [10 regression][ARM] Bootstrap fails in gen_movsi, at config/arm/arm.md:5258
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91708 --- Comment #14 from Bernd Edlinger --- Author: edlinger Date: Fri Sep 13 17:22:04 2019 New Revision: 275701 URL: https://gcc.gnu.org/viewcvs?rev=275701=gcc=rev Log: 2019-09-13 Bernd Edlinger PR middle-end/91708 * cse.c (cse_insn): Do not replace anything with a MEM. Modified: trunk/gcc/ChangeLog trunk/gcc/cse.c
[Bug go/91764] [10 regression] libgo build broken with Solaris grep
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91764 Ian Lance Taylor changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #2 from Ian Lance Taylor --- Should be fixed by revision 275700. I once again forgot to annotate the commit message with the PR number.
[Bug c/91766] New: -fvisibility=hidden during -fpic still uses GOT indirection on arm64
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91766
Bug ID: 91766
Summary: -fvisibility=hidden during -fpic still uses GOT
indirection on arm64
Product: gcc
Version: 8.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: blu.dark at gmail dot com
Target Milestone: ---
Passing -fvisibility=hidden does not stop GOT round-trips for global symbols
during -fpic. Example:
#if DO_HIDE
#define HIDE __attribute__ ((visibility("hidden")))
#else
#define HIDE
#endif
int room[2] HIDE;
int* foo()
{
int* p = room;
*p = 0;
return p;
}
$ gcc-8.2 -Ofast -fvisibility=hidden -fpic
foo:
adrpx1, _GLOBAL_OFFSET_TABLE_
ldr x1, [x1, #:gotpage_lo15:room]
mov x0, x1
str wzr, [x1]
ret
Explicitly hiding it via DO_HIDE works as expected:
$ gcc-8.2 -Ofast -fvisibility=hidden -fpic -DDO_HIDE
foo:
adrpx1, room
add x0, x1, :lo12:room
str wzr, [x1, #:lo12:room]
ret
Here's the godbolt link: https://godbolt.org/z/HCd2LT
[Bug c/91765] New: -Wredundant-decls conflicts with __attribute__((alias))
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91765
Bug ID: 91765
Summary: -Wredundant-decls conflicts with
__attribute__((alias))
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: rth at gcc dot gnu.org
Target Milestone: ---
/* Header file */
extern int bar;
/* Source file */
static int foo;
extern int bar __attribute__((alias("foo")));
--
For this test case, "gcc -c -Wredundant-decls z.c" produces
z.c:6:12: warning: redundant redeclaration of ‘bar’ [-Wredundant-decls]
6 | extern int bar __attribute__((alias("foo")));
|^~~
z.c:2:12: note: previous declaration of ‘bar’ was here
2 | extern int bar;
|^~~
However, the alias line is not a redundant decl, but rather the
actual definition of the symbol "bar". The syntax of the alias,
for whatever reason, requires the use of the extern keyword.
Remove the extern from the alias and we get
z.c:6:5: error: ‘bar’ defined both normally and as ‘alias’ attribute
6 | int bar __attribute__((alias("foo")));
| ^~~
This also seems like a bug, but it's probably years too late to
change the syntax of the definition of aliases.
[Bug go/91764] [10 regression] libgo build broken with Solaris grep
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91764 --- Comment #1 from Andreas Schwab --- That should use grep -E and ERE.
[Bug fortran/91752] BOZ constant outside DATA/INT/REAL/DBLE/CMPLX flagged as error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91752 --- Comment #5 from Steve Kargl --- On Fri, Sep 13, 2019 at 07:34:11AM +, juergen.reuter at desy dot de wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91752 > > --- Comment #4 from Jürgen Reuter --- > (In reply to kargl from comment #3) > > See the documentation. -fallow-invalid-boz reduces the error to a warning. > > The warning can only be silenced by -w. This was done with intent. > > Actually not all of them, this one is still flagged as an error: > implicit none > integer :: h > h = B"11" > select case (h) > case (B"11") >print *, "Found" > end select > end > That is the intended behavior. I did not try to allow every previous invalid use of BOZ. When I made BOZ work 15 years ago, I implemented the requirement of F95 that a BOZ is converted to the integer with the widest decimal range (ie, either INTEGER(8) or INTEGER(16)). A BOZ was parsed and immediately converted, so a BOZ could appear in any place that an INTEGER could appear. That was a mistake, and undocumented behavior. I have corrected that mistake. This used to compile with gfortran where INTEGER(8) is the integer with widest kind: module m interface prn module procedure prn4 module procedure prn8 end interface prn contains subroutine prn4(i) integer(4), intent(in) :: i print *, i end subroutine prn4 subroutine prn8(i) integer(8), intent(in) :: i print *, i end subroutine prn8 end module m program foo use m call prn(B'11') end program foo It now issues an error and -fallow-invalid-boz will not downgrade the error in this case. Because a BOZ is a typeless and kindless entity, there is no way to resolve the interface. Sure, you can argue that if gfortran use to compile some garbage code, it should always compile that garbage code. My stance is gfortran should reject the garbage.
[Bug go/91764] [10 regression] libgo build broken with Solaris grep
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91764 Rainer Orth changed: What|Removed |Added Target Milestone|--- |10.0
[Bug go/91764] New: [10 regression] libgo build broken with Solaris grep
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91764
Bug ID: 91764
Summary: [10 regression] libgo build broken with Solaris grep
Product: gcc
Version: 10.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: go
Assignee: ian at airs dot com
Reporter: ro at gcc dot gnu.org
CC: cmang at google dot com
Target Milestone: ---
Target: *-*-solaris2.11
When I first tried to bootstrap on Solaris 11.3 after the Go 1.13 import,
bootstrap was broken building libgo. This can be seen e.g. building
internal/cpu: the .dep file is created with
/usr/gnu/bin/mkdir -p internal; files=`echo
/vol/gcc/src/hg/trunk/local/libgo/go
/internal/cpu/cpu.go
/vol/gcc/src/hg/trunk/local/libgo/go/internal/cpu/cpu_386.g
o /vol/gcc/src/hg/trunk/local/libgo/go/internal/cpu/cpu_no_init.go
/vol/gcc/src/
hg/trunk/local/libgo/go/internal/cpu/cpu_ppc64x.go
/vol/gcc/src/hg/trunk/local/l
ibgo/go/internal/cpu/cpu_x86.go cpugen.go
which includes quite a number of files that shouldn't be there due to build
tags (like cpu_no_init.go and cpu_ppc64x.go on i386-pc-solaris2.11).
This leads to compilation failures like
/vol/gcc/src/hg/trunk/local/libgo/go/internal/cpu/cpu_ppc64x.go:27:1: error:
redefinition of 'doinit'
27 | func doinit() {
| ^
/vol/gcc/src/hg/trunk/local/libgo/go/internal/cpu/cpu_no_init.go:16:1: note:
previous definition of 'doinit' was here
16 | func doinit() {
| ^
/vol/gcc/src/hg/trunk/local/libgo/go/internal/cpu/cpu_x86.go:39:1: error:
redefinition of 'doinit'
39 | func doinit() {
| ^
and many more which turns out to be due to this change:
diff --git a/libgo/match.sh b/libgo/match.sh
--- a/libgo/match.sh
+++ b/libgo/match.sh
@@ -135,7 +135,7 @@ for f in $gofiles; do
if test x$tag1 != xnonmatchingtag -a x$tag2 != xnonmatchingtag; then
# Pipe through cat so that `set -e` doesn't affect fgrep.
- tags=`sed '/^package /q' < $f | grep '^// +build ' | cat`
+ tags=`sed '/^package /q' < $f | grep '^// \?+build ' | cat`
[...]
@@ -143,7 +143,7 @@ for f in $gofiles; do
case $tag in
"//")
;;
- "+build")
+ "+build" | "//+build")
While the old form of the RE (^// +build ) matches the build tags with both
Solaris grep (/bin/grep) and GNU grep, the second form only works with GNU
grep, effectively breaking build tags.
For the moment, I've worked around this by installing GNU grep, but there
should
be a better way, especially given that this has worked correctly for years.
[Bug tree-optimization/91756] [10 regression] g++.dg/lto/alias-3 FAILs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91756
--- Comment #2 from Richard Biener ---
The disambiguation needs "translation" (vn_reference_lookup_3 using the
original
ref for path-based disambiguation), but we do not allow this when looking
across backedges in the maybe_skip_until walk and the loop we have to cross
in alias disambiguation looks like
do
{
if (n <= i)
break;
ptr->a[i].foo = 1; // this stmt to disambiguate against
}
while (1);
but then DOM is happy with this structure because it valueizes in the IL
itself. So the guard against backedges must be too conservative since
DOM also walks across them. I've added this as fix for PR87132.
[Bug target/91735] [9/10 Regression] Runtime regression for SPEC2000 177.mesa on Haswell around the end of August 2018
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91735 --- Comment #11 from Richard Biener --- Created attachment 46880 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=46880=edit prototype This improves code-gen to use pextrw where possible but that doesn't make any measurable difference on runtime. Maybe the example loop isn't representative or the improvement isn't big enough.
[Bug middle-end/91753] Bad register allocation of multi-register types
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91753 --- Comment #4 from Wilco --- (In reply to Andrew Pinski from comment #3) > (In reply to Wilco from comment #2) > > (In reply to Andrew Pinski from comment #1) > > > lower-subreg should have be able to help here. I wonder why it did not > > > ... > > > > I'm not sure how it can help. > > I think you misunderstood what this pass does. > It does exactly what you think it should do: > /* Decompose multi-word pseudo-registers into individual >pseudo-registers when possible and profitable. This is possible >when all the uses of a multi-word register are via SUBREG, or are >copies of the register to another location. Breaking apart the >register permits more CSE and permits better register allocation. > > The only difference is the creating part which missing. Yes but the issue is that you can't remove all the subregs since the TBX instructions really need a 512-bit register. The slim dump for x1 = vqtbx4q_u8(x1, table,x1): 30: r94:XI#0=r105:V16QI 31: r95:XI=r94:XI REG_DEAD r94:XI 32: r95:XI#16=r101:V16QI 33: r96:XI=r95:XI REG_DEAD r95:XI 34: r96:XI#32=r102:V16QI 35: r97:XI=r96:XI REG_DEAD r96:XI 36: r97:XI#48=r106:V16QI 38: r100:V16QI=unspec[r100:V16QI,r97:XI,r100:V16QI] 186 REG_DEAD r97:XI As you can see it creates the 512-bit XI register via a complex sequence of 4 subreg lvalues.
[Bug target/91749] [10 Regression] ICE in arm_asm_trampoline_template, at config/arm/arm.c:3973
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91749 --- Comment #3 from Martin Liška --- Yes, the known-to-work is a bit misleading here. Anyway, the cross compiler is configured like this: $ ~/BIG/bin/arm/dev/shm/buildbot/install/gcc/bin/arm-linux-gnueabi-gcc -v Using built-in specs. COLLECT_GCC=/home/marxin/BIG/bin/arm/dev/shm/buildbot/install/gcc/bin/arm-linux-gnueabi-gcc COLLECT_LTO_WRAPPER=/home/marxin/BIG/bin/arm/dev/shm/buildbot/install/gcc/bin/../lib/gcc/arm-linux-gnueabi/10.0.0/lto-wrapper Target: arm-linux-gnueabi Configured with: /home/marxin/BIG/buildbot/buildworker/marxinbox-gcc-trunk-arm/build/configure --enable-languages=c,c++,fortran --disable-bootstrap --disable-libsanitizer --disable-multilib --enable-checking=release --prefix=/dev/shm/buildbot/install/gcc --target=arm-linux-gnueabi --with-as=/usr/bin/arm-suse-linux-gnueabi-as Thread model: posix Supported LTO compression algorithms: zlib zstd gcc version 10.0.0 20190912 (experimental) 22116957c80879f7e46949c5bb65d174d375c43f (GCC)
[Bug tree-optimization/91756] [10 regression] g++.dg/lto/alias-3 FAILs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91756
Richard Biener changed:
What|Removed |Added
Keywords||missed-optimization
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed||2019-09-13
Assignee|unassigned at gcc dot gnu.org |rguenth at gcc dot
gnu.org
Ever confirmed|0 |1
--- Comment #1 from Richard Biener ---
It works fine on x86_64-linux with {,-m32} but fails with
-fno-use-linker-plugin.
This causes the global vars to not resolve to constants. With -fno-tree-vrp
it passes because DOM optimizes the bptr2->a[ij].bar load. FRE misses this
for some reason, I'll investigate why.
[Bug fortran/91716] [9/10 Regression] ICE in output_constant, at varasm.c:5026
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91716 --- Comment #3 from Bernd Edlinger --- Author: edlinger Date: Fri Sep 13 11:33:18 2019 New Revision: 275698 URL: https://gcc.gnu.org/viewcvs?rev=275698=gcc=rev Log: 2019-09-13 Bernd Edlinger PR fortran/91716 * trans-array.c (gfc_conv_array_initializer): Always assign the array type of the field to the string constant. testsuite: 2019-09-13 Bernd Edlinger PR fortran/91716 * gfortran.dg/pr91716.f90: New test. Added: trunk/gcc/testsuite/gfortran.dg/pr91716.f90 Modified: trunk/gcc/fortran/ChangeLog trunk/gcc/fortran/trans-array.c trunk/gcc/testsuite/ChangeLog
[Bug c++/91761] Overloaded new operator compiled with -O3 option distorts original C++ code
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91761 --- Comment #5 from Andrew Pinski --- (In reply to Dávid Bolvanský from comment #4) > But GCC 5.5 could remove one malloc-free pair > https://godbolt.org/z/fDJZs0 > > New optimization in gcc 6 actually made the code worse.. But the code is undefined. the old GCC did not take into account that fact and was able to optimize it away. In new GCC it takes into account that before the constructor is run, the object values is undefined. Plus this change is actually documented in the porting to guide of GCC 6: https://gcc.gnu.org/gcc-6/porting_to.html#flifetime-dse With almost your exact testcase.
[Bug lto/91763] [10 regression] go.go-torture/execute/printnil.go FAILs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91763 Rainer Orth changed: What|Removed |Added Target Milestone|--- |10.0
[Bug lto/91763] New: [10 regression] go.go-torture/execute/printnil.go FAILs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91763 Bug ID: 91763 Summary: [10 regression] go.go-torture/execute/printnil.go FAILs Product: gcc Version: 10.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: lto Assignee: unassigned at gcc dot gnu.org Reporter: ro at gcc dot gnu.org CC: ian at airs dot com, marxin at gcc dot gnu.org, rguenth at gcc dot gnu.org Target Milestone: --- Target: i386-pc-solaris2.11, sparc-sun-solaris2.11 Between 20190910 (r275594) and 20190911 (r275651), go.go-torture/execute/printnil.go began to FAIL on Solaris/x86 and Solaris/SPARC: +FAIL: go.go-torture/execute/printnil.go execution, -flto (only with -flto, 32 and 64-bit sparc, 32-bit only x86, only when using gas). SIGABRT: abort PC=0xfcfb9715 m=0 sigcode=18446744073709551615 goroutine 1 [running]: __sighndlr :0 __lwp_sigqueue :0 Thread 8 received signal SIGABRT, Aborted. [Switching to Thread 6 (LWP 6)] 0xfcfb9715 in __lwp_sigqueue () from /lib/libc.so.1 (gdb) where #0 0xfcfb9715 in __lwp_sigqueue () from /lib/libc.so.1 #1 0xfcfb1fbf in thr_kill () from /lib/libc.so.1 #2 0xfcef92fa in raise () from /lib/libc.so.1 #3 0xfcecb226 in abort () from /lib/libc.so.1 #4 0xfdaa67b5 in runtime.throwException () at /vol/gcc/src/hg/trunk/local/libgo/runtime/go-unwind.c:128 #5 0xfdf4d907 in runtime.unwindStack () at /vol/gcc/src/hg/trunk/local/libgo/go/runtime/panic.go:521 #6 runtime.gopanic (e=...) at /vol/gcc/src/hg/trunk/local/libgo/go/runtime/panic.go:715 #7 0xfdaa7639 in runtime_panicstring (s=0xfda87a15 "nil pointer dereference") at /vol/gcc/src/hg/trunk/local/libgo/runtime/panic.c:49 #8 0xfdaa54c8 in __go_runtime_error (i=6) at /vol/gcc/src/hg/trunk/local/libgo/runtime/go-runtime-error.c:95 #9 0x08053c43 in main.MyType.String (pointer=0x0) at /vol/gcc/src/hg/trunk/local/gcc/testsuite/go.go-torture/execute/printnil.go:11 #10 0xfdc04049 in fmt.pp.handleMethods (p=0x8ee, verb=115) at /vol/gcc/src/hg/trunk/local/libgo/go/fmt/print.go:630 #11 0xfdc032ec in fmt.pp.printArg (p=0x8ee, arg=..., verb=115) at /vol/gcc/src/hg/trunk/local/libgo/go/fmt/print.go:713 #12 0xfdc05d8e in fmt.pp.doPrintf (p=0x8ee, format=..., a=...) at /vol/gcc/src/hg/trunk/local/libgo/go/fmt/print.go:1030 #13 0xfdc06866 in fmt.Sprintf (format=..., a=...) at /vol/gcc/src/hg/trunk/local/libgo/go/fmt/print.go:219 #14 0x08053ccc in main.main () at /vol/gcc/src/hg/trunk/local/gcc/testsuite/go.go-torture/execute/printnil.go:16 A reghunt identified this patch 2019-09-11 Richard Biener * lto-opts.c (lto_write_options): Stream -g when debug is enabled. * lto-wrapper.c (merge_and_complain): Pick up -g. (append_compiler_options): Likewise. (run_gcc): Re-instantiate handling -g0 at link-time. * doc/invoke.texi (flto): Document debug info generation. as the culprit. I haven't dug further yet.
[Bug c++/91761] Overloaded new operator compiled with -O3 option distorts original C++ code
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91761 Dávid Bolvanský changed: What|Removed |Added CC||david.bolvansky at gmail dot com --- Comment #4 from Dávid Bolvanský --- But GCC 5.5 could remove one malloc-free pair https://godbolt.org/z/fDJZs0 New optimization in gcc 6 actually made the code worse..
[Bug fortran/91752] BOZ constant outside DATA/INT/REAL/DBLE/CMPLX flagged as error
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91752 --- Comment #4 from Jürgen Reuter --- (In reply to kargl from comment #3) > See the documentation. -fallow-invalid-boz reduces the error to a warning. > The warning can only be silenced by -w. This was done with intent. Actually not all of them, this one is still flagged as an error: implicit none integer :: h h = B"11" select case (h) case (B"11") print *, "Found" end select end
[Bug target/91749] [10 Regression] ICE in arm_asm_trampoline_template, at config/arm/arm.c:3973
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91749
Christophe Lyon changed:
What|Removed |Added
Assignee|unassigned at gcc dot gnu.org |clyon at gcc dot gnu.org
--- Comment #2 from Christophe Lyon ---
Thumb-1 is not supported when using FDPIC.
In arm_option_override, we have:
if (TARGET_FDPIC)
{
arm_pic_register = FDPIC_REGNUM;
if (TARGET_THUMB1)
sorry ("FDPIC mode is not supported in Thumb-1 mode");
}
but using -mflip-thumb overrides this check, and forces thumb-1 mode, hence the
failures.
