[Bug ld/21800] New: In mips64 architecture, singular segment of pt_load to crash the program

2017-07-20 Thread He.Hongjun at zte dot com.cn
https://sourceware.org/bugzilla/show_bug.cgi?id=21800

Bug ID: 21800
   Summary: In mips64 architecture, singular segment of pt_load to
crash the program
   Product: binutils
   Version: 2.27
Status: UNCONFIRMED
  Severity: critical
  Priority: P2
 Component: ld
  Assignee: unassigned at sourceware dot org
  Reporter: He.Hongjun at zte dot com.cn
  Target Milestone: ---

Created attachment 10277
  --> https://sourceware.org/bugzilla/attachment.cgi?id=10277&action=edit
BSS aligned at boundary 1M, is added in the mips64 default link script.

Overview: More detailed restatement of summary.

Drag-selecting any page crashes Mac builds in the NSGetFactory function.
Steps to Reproduce: Minimized, easy-to-follow steps that will trigger the bug.
Include any special setup steps.

1) Using a specific link script (mips64.lds), link generated three PT_LOAD
segments. BSS aligned at boundary 1M, is added in the mips64 default link
script. = the ALIGN x10 (0);
  _edata =.Dojo.provide (edata =.);
  . = the ALIGN (0 x10);
  __bss_start =.

2) Hello test program(test.c)
#include 
int main()
{
  printf("Hello world!\n");
}

3) Using mips64 compiler compiler
mips64-linux-gcc test.c -o test -Wl,-T,mips64.lds

The application test crashed.

-- 
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/21800] In mips64 architecture, singular segment of pt_load and --relax, crash the program

2017-07-20 Thread He.Hongjun at zte dot com.cn
https://sourceware.org/bugzilla/show_bug.cgi?id=21800

--- Comment #2 from hehongjun  ---
Created attachment 10278
  --> https://sourceware.org/bugzilla/attachment.cgi?id=10278&action=edit
Relax, after on the memory layout

-- 
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/21800] In mips64 architecture, singular segment of pt_load and --relax, crash the program

2017-07-20 Thread He.Hongjun at zte dot com.cn
https://sourceware.org/bugzilla/show_bug.cgi?id=21800

hehongjun  changed:

   What|Removed |Added

Summary|In mips64 architecture, |In mips64 architecture,
   |singular segment of pt_load |singular segment of pt_load
   |to crash the program|and --relax, crash the
   ||program

--- Comment #1 from hehongjun  ---
3) Using mips64 compiler compiler
mips64-linux-gcc test.c -o test -Wl,-T,mips64.lds -Wl,--relax

-- 
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/21803] New: Suggest possibility of incompatible assemblers when "unrecognized relocation" encountered

2017-07-20 Thread daniel.santos at pobox dot com
https://sourceware.org/bugzilla/show_bug.cgi?id=21803

Bug ID: 21803
   Summary: Suggest possibility of incompatible assemblers when
"unrecognized relocation" encountered
   Product: binutils
   Version: 2.25
Status: UNCONFIRMED
  Severity: enhancement
  Priority: P2
 Component: ld
  Assignee: unassigned at sourceware dot org
  Reporter: daniel.santos at pobox dot com
  Target Milestone: ---

Created attachment 10279
  --> https://sourceware.org/bugzilla/attachment.cgi?id=10279&action=edit
proposed (and untested) fix

I have two Gentoo machines and one of them needed updating, so I set up distcc
to complete the update more quickly.  Although I checked the versions of gcc
and binutils on the two machines and thought they were the same, I was mistaken
-- one had binutils 2.25.1 and the other 2.26.1.  The result was errors like
this:


libtool: link: x86_64-pc-linux-gnu-gcc -m32 -shared  -fPIC -DPIC 
lib/.libs/xmlparse.o lib/.libs/xmltok.o lib/.libs/xmlrole.o-m32 -O2 -ggdb
-march=amdfam10 -mmmx -m3dnow -msse -msse2 -msse3 -msse4a -mcx16 -msahf
-mpopcnt -mabm -mlzcnt -mprfchw -mfxsr -mtune=amdfam10 -Wl,-O1 -Wl,--as-needed 
 -Wl,-soname -Wl,libexpat.so.1 -o .libs/libexpat.so.1.6.4
/usr/lib/gcc/x86_64-pc-linux-gnu/5.4.0/../../../../x86_64-pc-linux-gnu/bin/ld:
lib/.libs/xmlparse.o: unrecognized relocation (0x2b) in section `.text'
/usr/lib/gcc/x86_64-pc-linux-gnu/5.4.0/../../../../x86_64-pc-linux-gnu/bin/ld:
final link failed: Bad value
collect2: error: ld returned 1 exit status
distcc[17072] ERROR: compile (null) on localhost failed


I did a lot of internet searches and re-checking before I discovered the
problem.  It would be most helpful if ld could suggest that this is probably
caused using two different versions of binutils.

-- 
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/21805] New: MIPS: Dynamic symbol --no-export-dynamic rules not respected for SVR4 executables

2017-07-20 Thread ma...@linux-mips.org
https://sourceware.org/bugzilla/show_bug.cgi?id=21805

Bug ID: 21805
   Summary: MIPS: Dynamic symbol --no-export-dynamic rules not
respected for SVR4 executables
   Product: binutils
   Version: 2.30 (HEAD)
Status: NEW
  Severity: normal
  Priority: P2
 Component: ld
  Assignee: ma...@linux-mips.org
  Reporter: ma...@linux-mips.org
  Target Milestone: ---
Target: mips*-*-linux-gnu

As discussed: 
and observed in regression testing:

FAIL: Run indirect5 1
FAIL: Run indirect5 2
FAIL: indirect5a dynsym
FAIL: indirect5b dynsym
FAIL: Run indirect5 3
FAIL: Run indirect5 4
FAIL: Run indirect6 3
FAIL: Run indirect6 4
FAIL: indirect5c dynsym
FAIL: indirect5d dynsym
FAIL: indirect6c dynsym
FAIL: indirect6d dynsym

when making a SVR4 (-mno-plt) executable the MIPS BFD linker backend does
not respect our dynamic symbol export rules and always makes all global
and weak symbols dynamic, regardless of the --no-export-dynamic
command-line option, be it implied or specified.  When making a PLT
(-mplt) executable the backend does follow the rules though, and both
--export-dynamic and --no-export-dynamic command-line options work as
documented.

According to the ELF gABI: "An implementation may restrict the set of
global and weak symbols that are externally visible.", so we are allowed
to treat symbols like this.  This affects symbol preemption and therefore
is a user-visible peculiarity that is also contrary to our documentation.
So it will make sense if we make MIPS SVR4 executables consistent in this
respect with MIPS PLT executables, as well as other targets.

-- 
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/21805] MIPS: Dynamic symbol --no-export-dynamic rules not respected for SVR4 executables

2017-07-20 Thread ma...@linux-mips.org
https://sourceware.org/bugzilla/show_bug.cgi?id=21805

Maciej W. Rozycki  changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
 CC||ma...@linux-mips.org

-- 
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