[Bug c/50186] New: junk at end of line: `1

[Bug c/50186] New: junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

             Bug #: 50186
           Summary: junk at end of line: `1
    Classification: Unclassified
           Product: gcc
           Version: 4.6.1
            Status: UNCONFIRMED
          Severity: blocker
          Priority: P3
         Component: c
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: santoshkumar.a@gmail.com


All,
      I have built a Toolchain for powerpc 476 in little endian mode.

With below:

binutils-2.21
gcc-4.6.1
glibc-2.13
linux-2.6.39

While comiling linux kernel 2.6.39 with 
"make ARCH=powerpc CROSS_COMPILE=powerpc-476-linux-gnu-"
I see the below error.

  CHK     include/linux/version.h
  CHK     include/generated/utsrelease.h
  CALL    scripts/checksyscalls.sh
  CHK     include/generated/compile.h
  SKIPPED include/generated/compile.h
  CC      init/do_mounts_rd.o
{standard input}: Assembler messages:
{standard input}:665: Error: junk at end of line: `1'

If any one has faced this issue please let me know what needs to be done.

Thanks

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #1 from SK <santoshkumar.a at gmail dot com> 2011-08-25 10:33:22 UTC ---
Created attachment 25099
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25099
build errors

build errors

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #3 from SK <santoshkumar.a at gmail dot com> 2011-08-25 11:53:59 UTC ---
Created attachment 25101
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25101
do_mounts_rd.s

assembly file

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #2 from SK <santoshkumar.a at gmail dot com> 2011-08-25 11:53:17 UTC ---
Created attachment 25100
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25100
intermidate file

intermidate file do_mounts_rd.i

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #4 from SK <santoshkumar.a at gmail dot com> 2011-08-25 11:59:11 UTC ---
Created attachment 25102
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25102
cross compile script

cross compile script

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #5 from SK <santoshkumar.a at gmail dot com> 2011-08-25 12:02:09 UTC ---
at line 665 in do_mounts_rd.s "mfcr 27,1" is a wrong instruction generated. As
per Power ISA™ Version 2.05 mfcr take only one argument i.e "mfcr RT". Let me
know if i have to change any components used.

[Bug c/50186] junk at end of line: `1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-08-25 12:05:46 UTC ---
That seems like your gcc is assuming -mmfcrf for your CPU, yet your assembler
can't assemble it (or is assembling for a CPU which doesn't have the mfcrf
insn).

[Bug c/50186] junk at end of line: `1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-08-25 12:06:46 UTC ---
MFCRF is ISA 2.01 BTW.

[Bug c/50186] junk at end of line: `1

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-08-25 12:10:26 UTC ---
And 476 CPU according to rs6000-cpus.def should support that:
RS6000_CPU ("476", PROCESSOR_PPC476,
            POWERPC_BASE_MASK | MASK_SOFT_FLOAT | MASK_PPC_GFXOPT | MASK_MFCRF
            | MASK_POPCNTB | MASK_FPRND | MASK_CMPB | MASK_MULHW | MASK_DLMZB)
so the bug is probably on your assembler side.

[Bug c/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

--- Comment #9 from SK <santoshkumar.a at gmail dot com> 2011-08-25 15:01:26 UTC ---
Just for checking i changed the instruction in .s file from "mfcr 27,1" to
"mfcr 27" and used the assembler to generate the binary there was no error
reported. Now i am confused whether it is fault with assembler if with
compiler. Please suggest where to look into.

[Bug target/50186] junk at end of line: `1

[santoshkumar.a at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50186

SK <santoshkumar.a at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID

--- Comment #10 from SK <santoshkumar.a at gmail dot com> 2011-08-26 13:43:30 UTC ---
(In reply to comment #9)
> Just for checking i changed the instruction in .s file from "mfcr 27,1" to
> "mfcr 27" and used the assembler to generate the binary there was no error
> reported. Now i am confused whether it is fault with assembler if with
> compiler. Please suggest where to look into.

issue is resolved i had to use -mno-mfcrf. -mmfcrf was enabled by default and
this generates mfcr with two args.

Thanks