[Bug c/27075] New: Compiler generate incorrect assembler

[Bug c/27075] New: Compiler generate incorrect assembler

[edmar at freescale dot com]

I get an assembler error, when trying to build gcc for the gnuspe target.
I am using binutils-2.16.1.
I don't know when this problem started because this particular target had
other failure that was fixed last week.



/local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/./gcc/cc1plus -E -quiet
-nostdinc++ -v
-I/local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/powerpc-unknown-linux-gnuspe/libstdc++-v3/include/powerpc-unknown-linux-gnuspe
-I/local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/powerpc-unknown-linux-gnuspe/libstdc++-v3/include
-I/local/gnu_toolchain/build_area/gcc-trunk/libstdc++-v3/libsupc++ -iprefix
/temp/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/gcc/../lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/
-isystem /local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/./gcc/include
-D_GNU_SOURCE -D__unix__ -D__gnu_linux__ -D__linux__ -Dunix -D__unix -Dlinux
-D__linux -Asystem=linux -Asystem=unix -Asystem=posix -D_GNU_SOURCE -isystem
/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/powerpc-unknown-linux-gnuspe/include
-isystem
/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/powerpc-unknown-linux-gnuspe/sys-include
atomicity.cc -Wall -Wextra -Wwrite-strings -Wcast-qual -fno-implicit-templates
-fdiagnostics-show-location=once -ffunction-sections -fdata-sections
-fworking-directory -O2 -fpch-preprocess -o atomicity.ii
ignoring nonexistent directory
"/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/powerpc-unknown-linux-gnuspe/include"
ignoring nonexistent directory
"/temp/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/gcc/../lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/include"
ignoring nonexistent directory
"/temp/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/gcc/../lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/../../../../powerpc-unknown-linux-gnuspe/sys-include"
ignoring nonexistent directory
"/temp/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/gcc/../lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/../../../../powerpc-unknown-linux-gnuspe/include"
ignoring duplicate directory
"/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/../../../../powerpc-unknown-linux-gnuspe/sys-include"
ignoring nonexistent directory
"/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/../../../../powerpc-unknown-linux-gnuspe/include"
#include "..." search starts here:
#include <...> search starts here:

/local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/powerpc-unknown-linux-gnuspe/libstdc++-v3/include/powerpc-unknown-linux-gnuspe

/local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/powerpc-unknown-linux-gnuspe/libstdc++-v3/include
 /local/gnu_toolchain/build_area/gcc-trunk/libstdc++-v3/libsupc++
 /local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/./gcc/include

/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/powerpc-unknown-linux-gnuspe/sys-include

/local/gnu_toolchain/install_area/gcc-trunk-20060406-e500v1/lib/gcc/powerpc-unknown-linux-gnuspe/4.2.0/include
End of search list.
 /local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/./gcc/cc1plus
-fpreprocessed atomicity.ii -quiet -dumpbase atomicity.cc -auxbase-strip
atomicity.o -g -O2 -Wall -Wextra -Wwrite-strings -Wcast-qual -version
-fno-implicit-templates -fdiagnostics-show-location=once -ffunction-sections
-fdata-sections -o atomicity.s
GNU C++ version 4.2.0 20060406 (experimental) (powerpc-unknown-linux-gnuspe)
        compiled by GNU C version 3.4.3.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 0094760f8b7f814b4f4ae3135eee7f99
 /local/gnu_toolchain/build_area/obj_gcc-trunk_e500v1/./gcc/as -mppc -mspe
-me500 -many -V -Qy -o atomicity.o atomicity.s
GNU assembler version 2.16.1 (powerpc-unknown-linux-gnuspe) using BFD version
2.16.1
atomicity.s: Assembler messages:
atomicity.s:22: Error: syntax error; found `(' but expected `,'
atomicity.s:22: Error: junk at end of line: `(3)'
atomicity.s:24: Error: syntax error; found `(' but expected `,'
atomicity.s:24: Error: junk at end of line: `(3)'
atomicity.s:42: Error: syntax error; found `(' but expected `,'
atomicity.s:42: Error: junk at end of line: `(3)'
atomicity.s:44: Error: syntax error; found `(' but expected `,'
atomicity.s:44: Error: junk at end of line: `(3)'


-- 
           Summary: Compiler generate incorrect assembler
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: edmar at freescale dot com
  GCC host triplet: x86_64-unknown-linux-gnu
GCC target triplet: powerpc-unknown-linux-gnuspe


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug c/27075] Compiler generate incorrect assembler

[edmar at freescale dot com]

------- Comment #1 from edmar at freescale dot com  2006-04-07 15:32 -------
Created an attachment (id=11221)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11221&action=view)
Intermediate file of the failing code

Intermediate file that generates the wrong syntax assembler.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] Compiler generate incorrect assembler

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-04-07 15:33 -------
Can you attach the atomicity.s also?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] Compiler generate incorrect assembler

[edmar at freescale dot com]

------- Comment #3 from edmar at freescale dot com  2006-04-07 15:45 -------
Created an attachment (id=11222)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11222&action=view)
Aseembler file generated by the compiler

Assembler file generated by the compiler


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] Compiler generate incorrect assembler

[pinskia at gcc dot gnu dot org]

------- Comment #4 from pinskia at gcc dot gnu dot org  2006-04-07 15:48 -------
hmm:
        lwarx 9,0(3)

 That is definitely wrong.

It should have been "lwarx 9, 0, 3".  (This is a real compiler issue).


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |assemble-failure


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] Compiler generate incorrect assembler

[pinskia at gcc dot gnu dot org]

------- Comment #5 from pinskia at gcc dot gnu dot org  2006-04-07 15:52 -------
But I don't see anything obviously wrong in sync.md, like missing a y (aka %y1
or %y0 is used correctly).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] Compiler generate incorrect assembler

[edmar at freescale dot com]

------- Comment #6 from edmar at freescale dot com  2006-04-07 15:54 -------
The problem is how %y is defined in rs6000.c (print_operand)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[pinskia at gcc dot gnu dot org]

------- Comment #7 from pinskia at gcc dot gnu dot org  2006-04-07 16:00 -------
grrrr, well this is why reusing the operand is not a good idea anyways this is
a regression from when the builtins came in which was 4.1.0.

Well confirmed, a simple testcase is:

int f(int *a, int b)
{
  return __sync_fetch_and_add(a, b);
}


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |aldyh at gcc dot gnu dot org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   GCC host triplet|x86_64-unknown-linux-gnu    |
 GCC target triplet|powerpc-unknown-linux-gnuspe|powerpc-*-linux-gnuspe
   Last reconfirmed|0000-00-00 00:00:00         |2006-04-07 16:00:50
               date|                            |
            Summary|Compiler generate incorrect |[4.1 Regression] Compiler
                   |assembler                   |generate incorrect assembler
                   |                            |for __sync_fetch-* builtins
                   |                            |on e500 aka SPE
   Target Milestone|---                         |4.1.1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[amodra at bigpond dot net dot au]

------- Comment #8 from amodra at bigpond dot net dot au  2006-04-09 13:51 -------
This horrible hack should cure the problem, I think.  The rationale is that
spe.md uses %y only for evldd and evstdd insns, which load or store 8 bytes.
The lock insns use %y for 4 byte quantities (e500 doesn't support 64-bit
ldarx/stdcx. as far as I know).

--- rs6000.c    (revision 112800)
+++ rs6000.c    (working copy)
@@ -10750,7 +10750,7 @@ print_operand (FILE *file, rtx x, int co

        tmp = XEXP (x, 0);

-       if (TARGET_E500)
+       if (TARGET_E500 && GET_MODE_SIZE (GET_MODE (x)) == 8)
          {
            /* Handle [reg].  */
            if (GET_CODE (tmp) == REG)


-- 

amodra at bigpond dot net dot au changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amodra at bigpond dot net
                   |                            |dot au


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[mmitchel at gcc dot gnu dot org]

-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[edmar at freescale dot com]

------- Comment #9 from edmar at freescale dot com  2006-04-17 15:57 -------
This bug also affects target e500v2. The e500v2 is configured with:
--target=powerpc-unknowm-linux-gnuspe --enable-e500_double.
I will try the patch on comment 8 overnight.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[mmitchel at gcc dot gnu dot org]

------- Comment #10 from mmitchel at gcc dot gnu dot org  2006-05-25 02:34 -------
Will not be fixed in 4.1.1; adjust target milestone to 4.1.2.


-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.1.1                       |4.1.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[echristo at apple dot com]

------- Comment #11 from echristo at apple dot com  2006-07-22 21:12 -------
Waiting on Edmar's feedback for Alan's patch.


-- 

echristo at apple dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[pinskia at gcc dot gnu dot org]

------- Comment #12 from pinskia at gcc dot gnu dot org  2006-07-22 22:16 -------
It is obvious the patch actually works but needs a full testing so this really
should not be in waiting.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW
   Last reconfirmed|2006-04-07 16:00:50         |2006-07-22 22:16:17
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[edmar at freescale dot com]

------- Comment #13 from edmar at freescale dot com  2006-07-24 15:05 -------
For my part (e500v2) it works just fine. I have being using since april for my
regression tests. As Andrew posted, the question is the impact on other
targets...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[raj dot khem at gmail dot com]

------- Comment #14 from raj dot khem at gmail dot com  2006-07-24 19:34 -------

FWIW. I also have it working using current 4.1 branch snapshot and the patch in
comment #8.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[dtemirbulatov at gmail dot com]

------- Comment #15 from dtemirbulatov at gmail dot com  2006-07-26 12:53 -------
>As Andrew posted, the question is the impact on other targets...
Looking at the patch, it is only about e500(TARGET_E500 defined only for e500),
so it should not impact other targets


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[mmitchel at gcc dot gnu dot org]

------- Comment #16 from mmitchel at gcc dot gnu dot org  2006-08-17 20:29 -------
I think the patch in Comment #8 is OK, but that the right approach would be to
stop using %y to denote two different things.  I think we should add
documentation saying that %y has dual meanings and check in this patch.  Then,
as a separate issue, I think we should split %y into two different operand
letters so that each has consitent semantics.

I've CC'd the PowerPC maintainers on this PR.  If neither of them object within
24 hours, please check in this patch.

Thanks,

-- Mark


-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dje at watson dot ibm dot
                   |                            |com, geoffk at apple dot
                   |                            |com, joseph at codesourcery
                   |                            |dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[drow at gcc dot gnu dot org]

-- 

drow at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |drow at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2006-07-22 22:16:17         |2006-08-25 14:36:07
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[drow at gcc dot gnu dot org]

-- 

drow at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|drow at gcc dot gnu dot org |unassigned at gcc dot gnu
                   |                            |dot org
             Status|ASSIGNED                    |NEW


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[dje at gcc dot gnu dot org]

------- Comment #17 from dje at gcc dot gnu dot org  2006-08-25 18:56 -------
Subject: Bug 27075

Author: dje
Date: Fri Aug 25 18:56:08 2006
New Revision: 116425

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=116425
Log:
        PR target/27075
        * config/rs6000/rs6000.c (print_operand): Only use e500 %y syntax
        for 8 byte objects.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/rs6000/rs6000.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[dje at gcc dot gnu dot org]

------- Comment #18 from dje at gcc dot gnu dot org  2006-08-25 18:56 -------
patch committed.


-- 

dje at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075

[Bug target/27075] [4.1/4.2 Regression] Compiler generate incorrect assembler for __sync_fetch-* builtins on e500 aka SPE

[dje at gcc dot gnu dot org]

------- Comment #19 from dje at gcc dot gnu dot org  2006-08-25 19:06 -------
Subject: Bug 27075

Author: dje
Date: Fri Aug 25 19:06:18 2006
New Revision: 116426

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=116426
Log:
        PR target/27075
        * config/rs6000/rs6000.c (print_operand): Only use e500 %y syntax
        for 8 byte objects.

Modified:
    branches/gcc-4_1-branch/gcc/ChangeLog
    branches/gcc-4_1-branch/gcc/config/rs6000/rs6000.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27075