[Bug target/32313] New: [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[Bug target/32313] New: [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

At revision 125625 (and 125652) There is a bootstrap failure due to a SIGSEGV
while running gengtype in stage 2.

A successful bootstrap was done for r125494.

I think the dataflow branch merge is a likely culprit here.

The problem is that $gp is not being restored before a call to a local
function.  Here is the dissassembly:
00003498 <write_func_for_structure>:
    3498:       3c1c0000        lui     gp,0x0
                        3498: R_MIPS_HI16       __gnu_local_gp
    349c:       27bdff58        addiu   sp,sp,-168
    34a0:       279c0000        addiu   gp,gp,0
                        34a0: R_MIPS_LO16       __gnu_local_gp
    34a4:       afbf00a4        sw      ra,164(sp)
    34a8:       afbe00a0        sw      s8,160(sp)
    34ac:       afb7009c        sw      s7,156(sp)
    34b0:       afb60098        sw      s6,152(sp)
    34b4:       afb50094        sw      s5,148(sp)
    34b8:       afb40090        sw      s4,144(sp)
    34bc:       afb3008c        sw      s3,140(sp)
    34c0:       afb20088        sw      s2,136(sp)
    34c4:       afb10084        sw      s1,132(sp)
    34c8:       afb00080        sw      s0,128(sp)
    34cc:       afbc0020        sw      gp,32(sp)
    34d0:       8cb00014        lw      s0,20(a1)
    34d4:       00c0a021        move    s4,a2
    34d8:       afa400a8        sw      a0,168(sp)
    34dc:       afa700b4        sw      a3,180(sp)
    34e0:       00a0f021        move    s8,a1
    34e4:       00002021        move    a0,zero
    34e8:       24070004        li      a3,4
    34ec:       24080006        li      t0,6
    34f0:       2406ffd8        li      a2,-40
    34f4:       02841021        addu    v0,s4,a0
    34f8:       12800007        beqz    s4,3518 <write_func_for_structure+0x80>
    34fc:       2484fffc        addiu   a0,a0,-4
    3500:       8c430024        lw      v1,36(v0)
    3504:       10600004        beqz    v1,3518 <write_func_for_structure+0x80>
    3508:       00000000        nop
    350c:       8c620000        lw      v0,0(v1)
    3510:       1047012e        beq     v0,a3,39cc
<write_func_for_structure+0x534>
    3514:       00000000        nop
    3518:       1486fff7        bne     a0,a2,34f8
<write_func_for_structure+0x60>
    351c:       02841021        addu    v0,s4,a0
    3520:       8f990000        lw      t9,0(gp)
                        3520: R_MIPS_CALL16     memset
    3524:       27a20028        addiu   v0,sp,40
    3528:       00002821        move    a1,zero
    352c:       24060048        li      a2,72
    3530:       00402021        move    a0,v0
    3534:       0320f809        jalr    t9
    3538:       afa20078        sw      v0,120(sp)
    353c:       0c000000        jal     0 <get_lang_bitmap>
                        353c: R_MIPS_26 get_output_file_with_visibility
    3540:       02002021        move    a0,s0

After the call to memset the gp should be restored.  It is required by the call
to get_output_file_with_visibility.


-- 
           Summary: [4.3 Regression] Bootstrap failure running gengtype in
                    stage 2.
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Keywords: wrong-code
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: daney at gcc dot gnu dot org
 GCC build triplet: mipsel-linux-gnu
  GCC host triplet: mipsel-linux-gnu
GCC target triplet: mipsel-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at gcc dot gnu dot
                   |                            |org
           Keywords|                            |build
   Target Milestone|---                         |4.3.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #1 from daney at gcc dot gnu dot org  2007-06-13 00:03 -------
Created an attachment (id=13695)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13695&action=view)
.i for gengtype

Compiling like this causes the bad code (the configure line is even in there):

# /home/build/gcc-build/./prev-gcc/xgcc -B/home/build/gcc-build/./prev-gcc/
-B/usr/local/mipsel-unknown-linux-gnu/bin/ -S -g -O2 -W -Wall -Wwrite-strings
-Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long
-Wno-variadic-macros -Wno-overlength-strings -Wold-style-definition
-Wmissing-format-attribute -Werror -fno-common -v -o build/gengtype.s
build/gengtype.i
Reading specs from /home/build/gcc-build/./prev-gcc/specs
Target: mipsel-unknown-linux-gnu
Configured with: ../gcc/configure --with-arch=mips32 --with-float=soft
--disable-java-awt --without-x --disable-tls --enable-__cxa_atexit
--disable-jvmpi --disable-static --disable-libmudflap
--enable-languages=c,c++,java
Thread model: posix
gcc version 4.3.0 20070612 (experimental)
 /home/build/gcc-build/./prev-gcc/cc1 -fpreprocessed build/gengtype.i -quiet
-dumpbase gengtype.i -march=mips32 -msoft-float -mno-shared -auxbase-strip
build/gengtype.s -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissing-prototypes -pedantic -Wno-long-long -Wno-variadic-macros
-Wno-overlength-strings -Wold-style-definition -Wmissing-format-attribute
-Werror -version -fno-common -o build/gengtype.s


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #2 from daney at gcc dot gnu dot org  2007-06-13 05:58 -------
I have a prototype patch that I think fixes this problem.


-- 

daney at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |daney at gcc dot gnu dot org
                   |dot org                     |
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2007-06-13 05:58:48
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #3 from daney at gcc dot gnu dot org  2007-06-13 06:26 -------
Created an attachment (id=13696)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13696&action=view)
Proposed fix.

I will try to bootstrap the Proposed fix tomorrow.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[echristo at apple dot com]

------- Comment #4 from echristo at apple dot com  2007-06-13 06:36 -------
Patch looks reasonable.


-- 

echristo at apple dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |echristo at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #5 from daney at gcc dot gnu dot org  2007-06-13 19:44 -------
Unfortunatly the patch causes an ICE compiling crtstuff.c.  I will have to
adjust it a bit...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #6 from daney at gcc dot gnu dot org  2007-06-13 20:39 -------
Created an attachment (id=13700)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13700&action=view)
Revised patch.

I don't know what I was thinking with the first version of the patch :-(  The
new version I think is more likely to be correct.


-- 

daney at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #13696|0                           |1
        is obsolete|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[echristo at apple dot com]

------- Comment #7 from echristo at apple dot com  2007-06-13 22:16 -------
Um. Right. :)

The new version is fine if it works.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #8 from daney at gcc dot gnu dot org  2007-06-17 05:05 -------
Good news and bad news.

My 32313-v2.patch allows c,c++,java to bootstrap except for at the very end of
the libjava build gcj-dbtool gets stuck in an endless loop.  Luckily killing it
allows the build to complete successfully.

The bad news is that there are at least these new failures:

FAIL: gcc.c-torture/execute/bcp-1.c execution,  -O2
FAIL: gcc.c-torture/execute/nestfunc-6.c execution,  -O2 

The bcp-1.c failure is also a failure to restore $gp.  I think this is a
different case, but I am looking into it a bit more.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #9 from daney at gcc dot gnu dot org  2007-06-18 17:36 -------
Subject: Bug 32313

Author: daney
Date: Mon Jun 18 17:36:42 2007
New Revision: 125818

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125818
Log:
        PR target/32313
        * config/mips/mips.c (mips_expand_call): Mark $gp as used by
        local function call.

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


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #10 from daney at gcc dot gnu dot org  2007-06-18 17:41 -------
The ability to bootstrap is fixed by the patch.  There are other dataflow
regressions that will be fixed by follow up patches.


-- 

daney at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #11 from daney at gcc dot gnu dot org  2007-06-18 19:35 -------
Subject: Bug 32313

Author: daney
Date: Mon Jun 18 19:35:05 2007
New Revision: 125824

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125824
Log:
        Revert:

        2007-06-18  David Daney  <ddaney@avtrex.com

        PR target/32313
        * config/mips/mips.c (mips_expand_call): Mark $gp as used by
        local function call.

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


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #12 from daney at gcc dot gnu dot org  2007-06-18 19:35 -------
That fix was incorrect.  Sorry.


-- 

daney at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #13 from daney at gcc dot gnu dot org  2007-06-18 19:39 -------
This is the same problem as:

http://gcc.gnu.org/ml/gcc-patches/2007-06/msg01165.html

I am currently bootstrapping the patch in that e-mail thread and will probably
commit that version.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #14 from daney at gcc dot gnu dot org  2007-06-19 16:36 -------
Subject: Bug 32313

Author: daney
Date: Tue Jun 19 16:36:42 2007
New Revision: 125852

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=125852
Log:
        PR target/32313
        * config/mips/mips.md (cprestore): Mark $gp as used.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/mips/mips.md


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313

[Bug target/32313] [4.3 Regression] Bootstrap failure running gengtype in stage 2.

[daney at gcc dot gnu dot org]

------- Comment #15 from daney at gcc dot gnu dot org  2007-06-19 16:43 -------
The second time is the charm.

There are still regressions caused by the dataflow merge, but at least we can
bootstrap now.


-- 

daney at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32313