[Bug ld/13990] New: ld/ld-new: BFD assertion fail, elf32-arm.c:12259 with --gc-sections and --shared, GCing a PLT reference to a local symbol
http://sourceware.org/bugzilla/show_bug.cgi?id=13990 Bug #: 13990 Summary: ld/ld-new: BFD assertion fail, elf32-arm.c:12259 with --gc-sections and --shared, GCing a PLT reference to a local symbol Product: binutils Version: 2.23 (HEAD) Status: NEW Severity: normal Priority: P2 Component: ld AssignedTo: unassig...@sourceware.org ReportedBy: h...@sourceware.org Classification: Unclassified Target: arm-linux-gnueabi Building libstdc++.so.6 from the gcc-4.7 branch with current binutils will yield the following assertions, but the linker will create what seems like a correct DSO. /tmp/somewhere/o/ld/ld-new: BFD (GNU Binutils) 2.22.52.20120416 assertion fail /tmp/somewhere/src/bfd/elf32-arm.c:12259 /tmp/somewhere/o/ld/ld-new: BFD (GNU Binutils) 2.22.52.20120416 assertion fail /tmp/somewhere/src/bfd/elf32-arm.c:12259 /tmp/somewhere/o/ld/ld-new: BFD (GNU Binutils) 2.22.52.20120416 assertion fail /tmp/somewhere/src/bfd/elf32-arm.c:12259 /tmp/somewhere/o/ld/ld-new: BFD (GNU Binutils) 2.22.52.20120416 assertion fail /tmp/somewhere/src/bfd/elf32-arm.c:12259 Again, you may notice this only by inspecting the build log. Patch in progress; will send it and reduced test-case to the list. -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/13991] New: powerpc-rtems ld failure [regression]
http://sourceware.org/bugzilla/show_bug.cgi?id=13991 Bug #: 13991 Summary: powerpc-rtems ld failure [regression] Product: binutils Version: 2.22 Status: NEW Severity: normal Priority: P2 Component: ld AssignedTo: unassig...@sourceware.org ReportedBy: joel.sherr...@oarcorp.com Classification: Unclassified Created attachment 6348 --> http://sourceware.org/bugzilla/attachment.cgi?id=6348 Test case This is a test case which demonstrates that something changed with ld between binutils 2.10.1 and 2.22 which resulted in a number of our BSPs not being able to do the transform from an ELF file to a compressed binary with a bootloader header. I have attached an unfortunately large file with the binaries that generate this failure. rtems 4.10 - binutils 2.20.1 with patches rtems 4.11 - binutils 2.22 with no patches With the same set of binary files to link. -bash-4.2$ /opt/rtems-4.10/bin/powerpc-rtems4.10-ld -o hello.ralf bootloader.o --just-symbols=hello.exe -b binary rtems.gz -T ppcboot.lds -bash-4.2$ /opt/rtems-4.11/bin/powerpc-rtems4.11-ld -o hello.ralf bootloader.o --just-symbols=hello.exe -b binary rtems.gz -T ppcboot.lds ppcboot.lds:97: undefined symbol `__rtems_start' referenced in expression -bash-4.2$ powerpc-rtems4.10-nm -g hello.exe | grep rtems_start T __rtems_start The symbol is there but not found by the reference in ppcboot.lds. = These is the RTEMS RPM information so this can be traced to that. -bash-4.2$ rpm -qf /opt/rtems-4.10/bin/powerpc-rtems4.10-ld rtems-4.10-powerpc-rtems4.10-binutils-2.20.1-3.fc16.i686 -bash-4.2$ rpm -qf /opt/rtems-4.11/bin/powerpc-rtems4.11-ld rtems-4.11-powerpc-rtems4.11-binutils-2.22-1.fc16.i686 = -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/13991] powerpc-rtems ld failure [regression]
http://sourceware.org/bugzilla/show_bug.cgi?id=13991 Joel Sherrill changed: What|Removed |Added Target||powerpc-rtems -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/13991] powerpc-rtems ld failure [regression]
http://sourceware.org/bugzilla/show_bug.cgi?id=13991 Joel Sherrill changed: What|Removed |Added CC||ralf.corsepius at rtems dot ||org -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/13990] ld/ld-new: BFD assertion fail, elf32-arm.c:12259 with --gc-sections and --shared, GCing a PLT reference to a local symbol
http://sourceware.org/bugzilla/show_bug.cgi?id=13990 Hans-Peter Nilsson changed: What|Removed |Added Status|NEW |ASSIGNED URL||http://sourceware.org/ml/bi ||nutils/2012-04/msg00228.htm ||l AssignedTo|unassigned at sourceware|hp at sourceware dot org |dot org | -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gprof/13993] New: gprof failed to accumulate hist time
http://sourceware.org/bugzilla/show_bug.cgi?id=13993
Bug #: 13993
Summary: gprof failed to accumulate hist time
Product: binutils
Version: 2.22
Status: NEW
Severity: normal
Priority: P2
Component: gprof
AssignedTo: unassig...@sourceware.org
ReportedBy: manjian2...@gmail.com
Classification: Unclassified
This bug occurred in hist_assign_samples_1 which is located in hist.c.When
bin_high_pc < sym_low_pc was true at first hit,the variable j kept decreasing
until it overflowed.No symbol would be find in this situation.The following is
my fix of this bug,may harm the performance.
static void
hist_assign_samples_1 (histogram *r)
{
bfd_vma bin_low_pc, bin_high_pc;
bfd_vma sym_low_pc, sym_high_pc;
bfd_vma overlap, addr;
unsigned int bin_count;
unsigned int i;
double count_time, credit;
bfd_vma lowpc = r->lowpc / sizeof (UNIT);
/* Iterate over all sample bins. */
for (i = 0; i < r->num_bins; ++i)
{
unsigned int j;
bin_count = r->sample[i];
if (! bin_count)
continue;
bin_low_pc = lowpc + (bfd_vma) (hist_scale * i);
bin_high_pc = lowpc + (bfd_vma) (hist_scale * (i + 1));
count_time = bin_count;
DBG (SAMPLEDEBUG,
printf (
"[assign_samples] bin_low_pc=0x%lx, bin_high_pc=0x%lx, bin_count=%u\n",
(unsigned long) (sizeof (UNIT) * bin_low_pc),
(unsigned long) (sizeof (UNIT) * bin_high_pc),
bin_count));
total_time += count_time;
/* Credit all symbols that are covered by bin I. */
for (j = 1; j < symtab.len; ++j)
{
sym_low_pc = symtab.base[j].hist.scaled_addr;
sym_high_pc = symtab.base[j + 1].hist.scaled_addr;
/* If high end of bin is below entry address,
go for next bin. */
if (bin_high_pc < sym_low_pc)
{
j++;
break;
}
/* If low end of bin is above high end of symbol,
go for next symbol. */
if (bin_low_pc >= sym_high_pc)
continue;
overlap =
MIN (bin_high_pc, sym_high_pc) - MAX (bin_low_pc, sym_low_pc);
if (overlap > 0)
{
DBG (SAMPLEDEBUG,
printf (
"[assign_samples] [0x%lx,0x%lx) %s gets %f ticks %ld overlap\n",
(unsigned long) symtab.base[j].addr,
(unsigned long) (sizeof (UNIT) * sym_high_pc),
symtab.base[j].name, overlap * count_time / hist_scale,
(long) overlap));
addr = symtab.base[j].addr;
credit = overlap * count_time / hist_scale;
/* Credit symbol if it appears in INCL_FLAT or that
table is empty and it does not appear it in
EXCL_FLAT. */
if (sym_lookup (&syms[INCL_FLAT], addr)
|| (syms[INCL_FLAT].len == 0
&& !sym_lookup (&syms[EXCL_FLAT], addr)))
{
symtab.base[j].hist.time += credit;
}
else
{
total_time -= credit;
}
}
}
}
DBG (SAMPLEDEBUG, printf ("[assign_samples] total_time %f\n",
total_time));
}
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
bug-binutils mailing list
bug-binutils@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-binutils
