[Bug lto/47900] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47900



Dmitry Gorbachev  changed:



   What|Removed |Added



 Status|NEW |RESOLVED

 Resolution||WORKSFORME



--- Comment #4 from Dmitry Gorbachev  
2013-02-07 00:13:03 UTC ---

I can't reproduce it with GCC 4.6.4, 4.7.3, 4.8.0.

[Bug lto/56232] New: Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56232



 Bug #: 56232

   Summary: Missed optimization with LTO

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: minor

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29377

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29377

Compile with -flto -O2



/tmp/cce3xw28.ltrans0.ltrans.o:cce3xw28.ltrans0.o:function main: error:

undefined reference to 'bar'

[Bug lto/47936] [4.6 Regression] Missed optimization with LTO due to strict aliasing issues

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47936



Dmitry Gorbachev  changed:



   What|Removed |Added



  Known to work|4.7.0   |4.7.3, 4.8.0

Summary|[4.6/4.7/4.8 Regression]|[4.6 Regression] Missed

   |Missed optimization with|optimization with LTO due

   |LTO due to strict aliasing  |to strict aliasing issues

   |issues  |

  Known to fail|4.6.0   |4.6.4



--- Comment #9 from Dmitry Gorbachev  
2013-02-07 00:40:06 UTC ---

I can't reproduce it with GCC 4.7.3 and 4.8.0.

[Bug web/56233] New: Bugzilla: files (*.c, *.diff, etc.) wrongly recognized as application/octet-stream

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56233



 Bug #: 56233

   Summary: Bugzilla: files (*.c, *.diff, etc.) wrongly recognized

as application/octet-stream

Classification: Unclassified

   Product: gcc

   Version: unknown

Status: UNCONFIRMED

  Severity: minor

  Priority: P3

 Component: web

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29378

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29378

"Enter Bug" form



When filing a new bug report, bugzilla treats attached files as binary rather

then text/plain (making them more difficult to view). Unfortunately, it doesn't

allow to choose the content type manually (it only works when javascript

support is turned off in a browser, or when adding an attachment to already

existing bug report).

[Bug web/56233] Bugzilla: files (*.c, *.diff, etc.) wrongly recognized as application/octet-stream

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56233



--- Comment #1 from Dmitry Gorbachev  
2013-02-07 02:49:14 UTC ---

Created attachment 29379

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29379

"Create New Attachment" form

[Bug web/56233] Bugzilla: files (*.c, *.diff, etc.) wrongly recognized as application/octet-stream

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56233



--- Comment #3 from Dmitry Gorbachev  
2013-02-07 03:34:21 UTC ---

> There is not much bugzilla can do.



It could either ignore mime type as sent by the browser, or show the "Content

Type" section.

[Bug web/56233] Bugzilla: files (*.c, *.diff, etc.) wrongly recognized as application/octet-stream

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56233



--- Comment #4 from Dmitry Gorbachev  
2013-02-07 03:37:53 UTC ---

Created attachment 29381

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29381

"Enter Bug" form (w/o javascript)

[Bug lto/56295] New: [4.8 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56295



 Bug #: 56295

   Summary: [4.8 Regression] Missed optimization with LTO

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: minor

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29422

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29422

Testcase



In this testcase, LTO generates a call to memcpy at -O3 optimization level.

[Bug lto/56295] [4.8 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56295



--- Comment #1 from Dmitry Gorbachev  
2013-02-12 11:02:05 UTC ---

Created attachment 29423

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29423

Three other testcases



These are unrelated testcases which also demonstrate differences between LTO

and non-LTO.

[Bug lto/56297] New: LTO: multiple definition error with global register variables

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56297



 Bug #: 56297

   Summary: LTO: multiple definition error with global register

variables

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: minor

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29424

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29424

Testcase



This testcase causes "multiple definition of 'esp'" error (note that esp has a

fixed role, there is no need to reserve it).

[Bug lto/56295] [4.8 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56295



--- Comment #5 from Dmitry Gorbachev  
2013-02-12 16:26:39 UTC ---

Created attachment 29425

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29425

Modified testcase



This slightly modified testcase still shows some strange difference between LTO

/ non-LTO.

[Bug lto/56515] New: location references block not in block tree, verify_gimple failed (LTO + profile)

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56515



 Bug #: 56515

   Summary: location references block not in block tree,

verify_gimple failed (LTO + profile)

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GCC 4.8.0 20130303 (experimental):



foo.c: In function 'foo':

foo.c:1:6: error: location references block not in block tree

 void foo(void) { }

  ^

# .MEM_10 = VDEF <.MEM_6(D)>

bar = 1;



foo.c:1:6: internal compiler error: verify_gimple failed

0x85df8f2 verify_gimple_in_cfg(function*)

../../gcc-4.8/gcc/tree-cfg.c:4727

0x84d520c execute_function_todo

../../gcc-4.8/gcc/passes.c:1963

0x84d455d do_per_function

../../gcc-4.8/gcc/passes.c:1701

0x84d532d execute_todo

../../gcc-4.8/gcc/passes.c:1996

0x84d57a1 execute_one_ipa_transform_pass

../../gcc-4.8/gcc/passes.c:2182

0x84d5873 execute_all_ipa_transforms()

../../gcc-4.8/gcc/passes.c:2208

0x8206775 expand_function

../../gcc-4.8/gcc/cgraphunit.c:1633

0x8206cda expand_all_functions

../../gcc-4.8/gcc/cgraphunit.c:1744

0x8207763 compile()

../../gcc-4.8/gcc/cgraphunit.c:2042

0x816b85d lto_main()

../../gcc-4.8/gcc/lto/lto.c:3396

[Bug lto/56515] location references block not in block tree, verify_gimple failed (LTO + profile)

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56515



--- Comment #1 from Dmitry Gorbachev  
2013-03-04 05:20:49 UTC ---

Created attachment 29575

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29575

Bt from 4.7.3



GCC 4.8.0 20130127 (r195497) fails, too; 20130113 (r195137) - ok.

GCC 4.8.0 20120916 (r191367) and earlier builds / versions (4.7, 4.6) fail with

sigsegv in prepend_lexical_block().

[Bug lto/56515] location references block not in block tree, verify_gimple failed (LTO + profile)

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56515



--- Comment #2 from Dmitry Gorbachev  
2013-03-04 05:22:09 UTC ---

Created attachment 29576

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29576

Testcase

[Bug middle-end/55644] maybe-uninitialized false positive due to incorrect flow analysis when gotos are present

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55644



--- Comment #13 from Dmitry Gorbachev  
2013-03-04 05:50:44 UTC ---

(In reply to comment #6)

> I start to believe we should arrange for --disable-werror for any non-standard

> build config ... testing matrix is simply too large and mostly false positives

> pop up.



Just `-Wno-error=maybe-uninitialized'. `--disable-werror' can hide real bugs.

[Bug lto/56536] New: GCC with LTO and `-march=pentium3' tries to generate SSE2 code

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56536



 Bug #: 56536

   Summary: GCC with LTO and `-march=pentium3' tries to generate

SSE2 code

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: minor

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29586

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29586

Testcase



Fails with error: '__builtin_ia32_cvtps2dq' needs isa option -m32 -msse2.

[Bug middle-end/56548] New: [4.8 Regression] ICE in emit_move_insn, at expr.c:3486 with -march=pentium{pro,2,3} -O3

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56548



 Bug #: 56548

   Summary: [4.8 Regression] ICE in emit_move_insn, at expr.c:3486

with -march=pentium{pro,2,3} -O3

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: middle-end

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GCC 4.8.0 20130303 (experimental).



$ cat > bug.c

short foo(short x)

{

  int i;



  for (i = 0; i < 3; i++)

if (x > 0)

  x--;



  return x;

}

^D

$ gcc -S -O3 -march=pentium3 bug.c

bug.c: In function 'foo':

bug.c:9:3: internal compiler error: in emit_move_insn, at expr.c:3486

   return x;

   ^

0x834eaa1 emit_move_insn(rtx_def*, rtx_def*)

../../gcc-4.8/gcc/expr.c:3485

0x8600ef9 expand_value_return

../../gcc-4.8/gcc/stmt.c:1461

0x86068d8 expand_return(tree_node*)

../../gcc-4.8/gcc/stmt.c:1547

0x825eefc expand_gimple_stmt_1

../../gcc-4.8/gcc/cfgexpand.c:2182

0x825f82c expand_gimple_stmt

../../gcc-4.8/gcc/cfgexpand.c:2304

0x8262484 expand_gimple_basic_block

../../gcc-4.8/gcc/cfgexpand.c:4083

0x82648c8 gimple_expand_cfg

../../gcc-4.8/gcc/cfgexpand.c:4602



GCC 20121118 (r193607) - fails, 2012 (r193417) - works.

[Bug lto/56578] New: Testcase behaves differently when compiled with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56578



 Bug #: 56578

   Summary: Testcase behaves differently when compiled with LTO

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 29625

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=29625

Testcases



The last testcase prints "FAIL". gcc-nm says "no symbols" in the object file

(a.o).

[Bug lto/56578] Testcase behaves differently when compiled with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56578



--- Comment #2 from Dmitry Gorbachev  
2013-03-12 00:46:13 UTC ---

GCC 4.8.0, 4.7.3, 4.6.4; build = host = target = i686-pc-linux-gnu; configured

with "--enable-languages=c,c++ --enable-version-specific-runtime-libs

--disable-nls --disable-shared"; glibc 2.17; recent binutils (GNU ld and gold).



> that is, 'make' prints three times OK



The 4th time (prog3) it prints "FAIL" for me -- free() is from main.c, but

malloc() is from glibc.

[Bug lto/56578] Testcase behaves differently when compiled with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56578



--- Comment #4 from Dmitry Gorbachev  
2013-03-12 13:30:30 UTC ---

> We could say the bug is in the testcase.  Failing is:

[...]

> that's because you are providing an implementation of

> malloc in libxxx.a a.o from func.c that does not have

> the C standards semantics.



The TC is invalid but it is not buggy, gcc-nm could not even find the "malloc"

symbol in a.o



> You have to compile this module with -fno-builtin-malloc

> (or -fno-builtins).



It is not enough to compile just func.c with -fno-builtin(-malloc), as one

might expect. I have to use it for main.c as well.



> GCC might still optimize the result in an unexpected way

> because it can rely on malloc/free semantics



There is a "volatile" modifier in the TC. It should inhibit such optimizations,

I think?

[Bug other/56653] New: Warning when verifying checksums from MD5SUMS file in tarballs

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56653



 Bug #: 56653

   Summary: Warning when verifying checksums from MD5SUMS file in

tarballs

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: trivial

  Priority: P3

 Component: other

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GNU md5sum (version 8.21) prints `md5sum: WARNING: 1 line is improperly

formatted'. It could be fixed by adding an extra `#' character to MD5SUMS:



*** maintainer-scripts/gcc_release

--- maintainer-scripts/gcc_release

***

*** 214,218 

  # Suggested usage:

  # md5sum -c MD5SUMS | grep -v \"OK$\"

! " > MD5SUMS



find . -type f |

--- 214,218 

  # Suggested usage:

  # md5sum -c MD5SUMS | grep -v \"OK$\"

! #" > MD5SUMS



find . -type f |

[Bug lto/56700] Optimizing at compile and link result in different binary size than only optimizing at link time

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56700



--- Comment #3 from Dmitry Gorbachev  
2013-03-26 20:12:42 UTC ---

> It's not just me who concluded that wrong.



Bug 55102.

[Bug rtl-optimization/47270] [4.7/4.8/4.9 Regression] GCC produces unnecessary code on -O2 and -O3 levels

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47270



--- Comment #6 from Dmitry Gorbachev  
2013-04-12 20:34:31 UTC ---

I can't reproduce the bug with GCC 4.7.3 and 4.8.1.

[Bug lto/48259] Internal compiler errors in lto1

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48259



Dmitry Gorbachev  changed:



   What|Removed |Added



 Status|WAITING |RESOLVED

 Resolution||FIXED



--- Comment #11 from Dmitry Gorbachev  
2013-04-12 20:45:54 UTC ---

Does not fail with 4.7.3 and 4.8.1. The 4.6.4 branch has been closed.

[Bug debug/54499] New: GCC produces wrong debugging information, failure while assembling generated .s file

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54499

 Bug #: 54499
   Summary: GCC produces wrong debugging information, failure
while assembling generated .s file
Classification: Unclassified
   Product: gcc
   Version: 4.8.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: debug
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 28138
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28138
Testcase

GCC 4.8.0 20120902 (experimental). Compile with `-gstabs+' option.

bug.s: Assembler messages:
bug.s:133: Error: can't resolve `.text._ZN2S3D1Ev' {.text._ZN2S3D1Ev section} -
`.LFBB3' {.text._ZN2S3D2Ev section}

[Bug debug/54499] GCC produces wrong debugging information, failure while assembling generated .s file

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54499

--- Comment #1 from Dmitry Gorbachev  
2012-09-06 00:03:39 UTC ---
Created attachment 28139
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28139
Code generated by GCC

[Bug driver/54789] New: [4.8 Regression] Error in GCC driver when defining GCC_COMPARE_DEBUG

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54789



 Bug #: 54789

   Summary: [4.8 Regression] Error in GCC driver when defining

GCC_COMPARE_DEBUG

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: driver

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GCC 4.8.0 20120930 (experimental)



$ gcc -fcompare-debug=-gtoggle a.c

$ GCC_COMPARE_DEBUG=yes gcc a.c

gcc: error: unrecognized command line option '-fcompare-debug=-gtoggle'

[Bug lto/54790] New: [4.8 Regression] Missing optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54790



 Bug #: 54790

   Summary: [4.8 Regression] Missing optimization with LTO

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 28337

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28337

Testcase



GCC 20120916 (r191367) - ok, 20120923 (r191654) - bug.



$ gcc -O2 -flto pr54790.c

/tmp/ccTyOv2G.ltrans0.ltrans.o:ccTyOv2G.ltrans0.o:function foo: error:

undefined reference to 'link_error'

collect2: error: ld returned 1 exit status

[Bug driver/54789] [4.8 Regression] Error in GCC driver when defining GCC_COMPARE_DEBUG

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54789



--- Comment #1 from Dmitry Gorbachev  
2012-10-04 09:46:45 UTC ---

Created attachment 28355

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28355

Patch

[Bug driver/54789] [4.8 Regression] Error in GCC driver when defining GCC_COMPARE_DEBUG

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54789



Dmitry Gorbachev  changed:



   What|Removed |Added



 CC||christian.bruel at st dot

   ||com



--- Comment #2 from Dmitry Gorbachev  
2012-10-04 09:50:00 UTC ---

Probably started with rev. 188102 by Christian Bruel.

[Bug driver/54789] [4.8 Regression] Error in GCC driver when defining GCC_COMPARE_DEBUG

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54789



Dmitry Gorbachev  changed:



   What|Removed |Added



   Keywords||patch



--- Comment #4 from Dmitry Gorbachev  
2012-10-05 02:57:32 UTC ---

Thanks. Patch: .

[Bug lto/54824] New: [4.8 Regression] ICE in verify_loop_structure with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54824



 Bug #: 54824

   Summary: [4.8 Regression] ICE in verify_loop_structure with LTO

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 28361

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28361

Backtrace



GCC 4.8.0 20120930 (experimental).



Compile the attached files with "gcc -O -flto foo.c bar.c".



In file included from foo.c:1:0,

 from :1:

bar.c: In function 'bar':

bar.c:25:1: error: loop 1's latch does not belong directly to it

 }

 ^

bar.c:25:1: internal compiler error: in verify_loop_structure, at

cfgloop.c:1582

Please submit a full bug report,

with preprocessed source if appropriate.

See  for instructions.



No ICE with "gcc -O -flto -fwhole-program foo.c bar.c".

[Bug lto/54824] [4.8 Regression] ICE in verify_loop_structure with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54824



--- Comment #1 from Dmitry Gorbachev  
2012-10-05 11:47:19 UTC ---

Created attachment 28362

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28362

Testcase



GCC 20120923 (r191654) - fails, 20120916 (r191367) - ok.

[Bug lto/54790] [4.8 Regression] Missing optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54790



--- Comment #4 from Dmitry Gorbachev  
2012-10-06 07:57:12 UTC ---

Created attachment 28369

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28369

Testcase #2



Another (perhaps related) issue.

[Bug lto/54790] [4.8 Regression] Missing optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54790



--- Comment #5 from Dmitry Gorbachev  
2012-10-06 08:02:20 UTC ---

Created attachment 28370

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28370

Testcase #3



Undefined reference bug.



With gold:

main.o:main.c:function main: error: undefined reference to 'foo'



With bfd ld:

`foo' referenced in section `.text' of main.o: defined in discarded section

`.text' of /tmp/ccl6J4kW.o (symbol from plugin)

[Bug tree-optimization/54877] New: [4.7/4.8 Regression] ICE: tree check: expected ssa_name, have real_cst in copy_ssa_name_fn, at tree-ssanames.c:335

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54877



 Bug #: 54877

   Summary: [4.7/4.8 Regression] ICE: tree check: expected

ssa_name, have real_cst in copy_ssa_name_fn, at

tree-ssanames.c:335

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: tree-optimization

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Created attachment 28401

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28401

Backtrace



Compile with `-O3 -funsafe-math-optimizations' (or -Ofast):



=== >8 ===

int main(void)

{

  double d = 0;

  int i;



  for (i = 0; i < 8; i++)

d--;



  return (int) d;

}

=== 8< ===

[Bug tree-optimization/54877] [4.7/4.8 Regression] ICE: tree check: expected ssa_name, have real_cst in copy_ssa_name_fn, at tree-ssanames.c:335

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54877



Dmitry Gorbachev  changed:



   What|Removed |Added



   Keywords||ice-on-valid-code

  Known to work||4.6.4

  Known to fail||4.7.3, 4.8.0



--- Comment #1 from Dmitry Gorbachev  
2012-10-09 18:15:13 UTC ---

GCC 4.7.3 20121006 (prerelease):



internal compiler error: tree check: expected ssa_name, have real_cst in

vect_is_simple_reduction_1, at tree-vect-loop.c:2260

[Bug tree-optimization/54824] [4.8 Regression] ICE in verify_loop_structure

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54824



--- Comment #2 from Dmitry Gorbachev  
2012-10-10 06:13:05 UTC ---

Created attachment 28409

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28409

Testcase #2



Another testcase. Compile with `-O2 -funswitch-loops' (or with `-O3').

GCC 20120408 (r186227) - fails, 20120325 (r185782) - works.



The first testcase is still valid, but compile it with `-O -flto -shared'.

[Bug lto/54728] [4.8 regression] ICE in input_gimple_stmt, at gimple-streamer-in.c:254

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54728



--- Comment #4 from Dmitry Gorbachev  
2012-10-11 07:06:58 UTC ---

Created attachment 28417

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28417

Testcase



Compile with `g++ -g -O1 -flto -flto-partition=none 1.cc 2.cc'.



Another (perhaps related) issue. GCC 4.8.0 20121007 (experimental) report

"internal compiler error: SSA corruption" with `-O1' and "internal compiler

error: in input_gimple_stmt, at gimple-streamer-in.c:254" with `-O0'. Older 4.8

builds fail with "fatal error: LTO_tags out of range: Range is 0 to 348, value

is 15872".

[Bug lto/54728] [4.8 regression] ICE in input_gimple_stmt, at gimple-streamer-in.c:254

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54728



--- Comment #5 from Dmitry Gorbachev  
2012-10-11 07:07:51 UTC ---

Created attachment 28418

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28418

Backtrace with -O0

[Bug lto/54728] [4.8 regression] ICE in input_gimple_stmt, at gimple-streamer-in.c:254

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54728



--- Comment #6 from Dmitry Gorbachev  
2012-10-11 07:08:29 UTC ---

Created attachment 28419

  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=28419

Backtrace with -O1

[Bug tree-optimization/54981] New: [4.8 Regression] Different code generated with / without `-g'

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54981



 Bug #: 54981

   Summary: [4.8 Regression] Different code generated with /

without `-g'

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: tree-optimization

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





Appeared between 20120513 (r187445) and 20120902 (r190863):



$ cat > 1.c

extern void bar(unsigned *, char *);



void foo(char *s)

{

  unsigned i;

  char t[2];



  bar(&i, t);



  for (i = 0; i < 2; i++)

s[i] = t[i];

}

^D

$ gcc -S -O3 -fcompare-debug 1.c

gcc: error: 1.c: -fcompare-debug failure (length)



Also fails with `-O -ftree-loop-distribute-patterns'.

[Bug middle-end/54957] Two crashes introduced by rev192488

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54957



--- Comment #15 from Dmitry Gorbachev  
2012-10-23 21:16:37 UTC ---

Mingw32 target, GCC 4.8.0 20121021 (experimental).



$ cat > bug.cc

struct S {

  ~S();

};



int main()

{

  S s1, s2, s3, s4, s5, s6, s7;

  return 0;

}

^D

$ g++ bug.cc

bug.cc: In function 'int main()':

bug.cc:9:1: internal compiler error: Segmentation fault

 }

 ^

0x87e5f10 crash_signal

../../gcc-4.8/gcc/toplev.c:335

0x87d2428 emit_case_dispatch_table

../../gcc-4.8/gcc/stmt.c:1919

0x87d4ae5 expand_sjlj_dispatch_table(rtx_def*, vec_t*)

../../gcc-4.8/gcc/stmt.c:2292

0x853e6cc sjlj_emit_dispatch_table

../../gcc-4.8/gcc/except.c:1363

0x853f900 sjlj_build_landing_pads

../../gcc-4.8/gcc/except.c:1420

0x853fa5b finish_eh_generation()

../../gcc-4.8/gcc/except.c:1454

0x8487921 gimple_expand_cfg

../../gcc-4.8/gcc/cfgexpand.c:4579

Please submit a full bug report,

with preprocessed source if appropriate.

Please include the complete backtrace with any bug report.

See  for instructions.

[Bug debug/55094] New: [4.7/4.8 Regression] ICE in maybe_record_trace_start, at dwarf2cfi.c:2224

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55094



 Bug #: 55094

   Summary: [4.7/4.8 Regression] ICE in maybe_record_trace_start,

at dwarf2cfi.c:2224

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: debug

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GCC 4.8, i686-pc-linux-gnu. Also happens in 4.7 configured with checking.



Compile this with `-g -Os -fomit-frame-pointer -fno-asynchronous-unwind-tables

-mpreferred-stack-boundary=2':



== 8< ==

extern int puts(const char *s);



int main(int argc, char **argv)

{

  if (argc) {

puts(argv[0]);

__builtin_trap();

  }

  __builtin_trap();

}

== >8 ==



1.c: In function 'main':

1.c:10:1: internal compiler error: in maybe_record_trace_start, at

dwarf2cfi.c:2224

 }

 ^

0x82cf181 maybe_record_trace_start

../../gcc-4.8/gcc/dwarf2cfi.c:2224

0x82cf56b scan_trace

../../gcc-4.8/gcc/dwarf2cfi.c:2401

0x82cf9cd create_cfi_notes

../../gcc-4.8/gcc/dwarf2cfi.c:2556

0x82cfcfa execute_dwarf2_frame

../../gcc-4.8/gcc/dwarf2cfi.c:2915

Please submit a full bug report,

with preprocessed source if appropriate.

Please include the complete backtrace with any bug report.

See  for instructions.

[Bug lto/55102] New: The options -flto and -On do not behave as described in GCC docs

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55102



 Bug #: 55102

   Summary: The options -flto and -On do not behave as described

in GCC docs

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: lto

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





> Additionally, the optimization flags used to compile

> individual files are not necessarily related to those

> used at link time.  For instance,

>

> gcc -c -O0 -flto foo.c

> gcc -c -O0 -flto bar.c

> gcc -o myprog -flto -O3 foo.o bar.o

>

> This produces individual object files with unoptimized

> assembler code, but the resulting binary myprog is

> optimized at -O3.  If, instead, the final binary is

> generated without -flto, then myprog is not optimized.



In fact, when you use -O3 when linking the .o files, it is already too late,

the resulting binary will not be fully optimized. You need to compile the .c

files with at least -O1. Thus, there is a bug either in GCC itself or in the

documentation.



== 8< ==

int foo(void)

{

  return 0;

}



int main(void)

{

  return foo();

}

== >8 ==



$ gcc -flto -O0 -c foo.c -o foo-O0.o

$ gcc -flto -O1 -c foo.c -o foo-O1.o

$ gcc -flto -O0 foo-O0.o -o prog-O0-O0

$ gcc -flto -O3 foo-O0.o -o prog-O0-O3

$ gcc -flto -O0 foo-O1.o -o prog-O1-O0

$ gcc -flto -O3 foo-O1.o -o prog-O1-O3

$ nm -A prog-O0-O0 prog-O0-O3 prog-O1-O0 prog-O1-O3 | grep foo

prog-O0-O0:080483f0 t foo.2337

prog-O0-O3:080483f0 t foo.2337

prog-O1-O0:080483f0 t foo.2337



GCC 4.6 gives a slight different result:



$ nm -A prog-O0-O0 prog-O0-O3 prog-O1-O0 prog-O1-O3 | grep foo

prog-O0-O0:08048381 t foo.1988

prog-O0-O3:08048380 t foo.1988



(GCC 4.5 crashes.)

[Bug lto/48427] ICE: tree check: expected array_type, have pointer_type in array_ref_low_bound, at expr.c:6253

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48427



Dmitry Gorbachev  changed:



   What|Removed |Added



 Status|UNCONFIRMED |RESOLVED

 Resolution||WORKSFORME



--- Comment #3 from Dmitry Gorbachev  
2012-10-29 19:47:23 UTC ---

Works with GCC 4.7/4.8.

[Bug debug/38226] Configure time option "--with-stabs" does not work

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38226



Dmitry Gorbachev  changed:



   What|Removed |Added



 Status|UNCONFIRMED |RESOLVED

 Resolution||WONTFIX

  Known to fail||



--- Comment #2 from Dmitry Gorbachev  
2012-10-29 19:49:43 UTC ---

No interest in this bug.

[Bug middle-end/55130] New: [4.8 Regression] ICE in insn_rhs_dead_pseudo_p, at lra-constraints.c:3224

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55130



 Bug #: 55130

   Summary: [4.8 Regression] ICE in insn_rhs_dead_pseudo_p, at

lra-constraints.c:3224

Classification: Unclassified

   Product: gcc

   Version: 4.8.0

Status: UNCONFIRMED

  Severity: normal

  Priority: P3

 Component: middle-end

AssignedTo: unassig...@gcc.gnu.org

ReportedBy: d.g.gorbac...@gmail.com





GCC 4.8.0 20121028 (experimental).



cat > bug.c

extern void bar(long long);



int foo(long long a, char b, long long c, long long d)

{

  if (c == 0)

c = d;



  bar(b + c);



  return a == d;

}

^D

$ gcc -O1 -mregparm=3 -mpreferred-stack-boundary=2 -S bug.c

bug.c: In function 'foo':

bug.c:11:1: internal compiler error: in insn_rhs_dead_pseudo_p, at

lra-constraints.c:3224

 }

 ^

0x84c80c9 insn_rhs_dead_pseudo_p

../../gcc-4.8/gcc/lra-constraints.c:3224

0x84c810c init_insn_rhs_dead_pseudo_p

../../gcc-4.8/gcc/lra-constraints.c:3240

0x84ce811 lra_constraints(bool)

../../gcc-4.8/gcc/lra-constraints.c:3301

0x84c306e lra(_IO_FILE*)

../../gcc-4.8/gcc/lra.c:2281

0x848e42e do_reload

../../gcc-4.8/gcc/ira.c:4624

0x848e5fb rest_of_handle_reload

../../gcc-4.8/gcc/ira.c:4736

Please submit a full bug report,

with preprocessed source if appropriate.

Please include the complete backtrace with any bug report.

See  for instructions.

[Bug target/55126] Libstdc++-v3 fails to compile

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55126



--- Comment #3 from Dmitry Gorbachev  
2012-10-30 01:14:47 UTC ---

A dup of PR54957 / PR55014.

[Bug middle-end/55130] [4.8 Regression] ICE in insn_rhs_dead_pseudo_p, at lra-constraints.c:3224

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55130



Dmitry Gorbachev  changed:



   What|Removed |Added



 Status|NEW |RESOLVED

 Resolution||FIXED



--- Comment #3 from Dmitry Gorbachev  
2012-11-06 16:45:47 UTC ---

Fixed.

[Bug preprocessor/55242] continued lines not always merged into one long line

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55242



--- Comment #1 from Dmitry Gorbachev  
2012-11-09 20:23:47 UTC ---

It's a "feature", not a bug (see PR41748, PR45696, and PR48383). Perhaps it

should be mentioned in the preprocessor documentation.

[Bug middle-end/46900] [4.6 Regression] 50% slowdown when linking with LTO in a single step

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46900

--- Comment #7 from Dmitry Gorbachev  
2010-12-17 12:13:54 UTC ---
> Thus, the "-march=native" somehow gets
> lost in the single-step compile.

PR42445 ?

[Bug c++/40831] g++ generated symbols for cloned function that be demangled.

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40831

Dmitry Gorbachev  changed:

   What|Removed |Added

 CC||d.g.gorbachev at gmail dot
   ||com

--- Comment #5 from Dmitry Gorbachev  
2011-01-05 01:07:00 UTC ---
There's also a binutils bug <http://sourceware.org/PR12337>:

> Programs objdump, nm, addr2line, c++filt, gprof, do not demangle cloned
> function names, such as _ZN1C3fooEi.isra.0. Could be C::foo.isra.0(int) or
> C::foo(int).isra.0 (?)

[Bug lto/47183] New: Please discard unused functions

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47183

   Summary: Please discard unused functions
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 22908
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22908
Testcase

Compile it with `g++ -O2 -flto -fuse-linker-plugin 47183.C'.

[Bug lto/47183] Please discard unused functions

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47183

--- Comment #2 from Dmitry Gorbachev  
2011-01-05 22:13:55 UTC ---
It's operator delete(void*). Also, typeinfo for S, typeinfo name for S, vtable
for __cxxabiv1::__class_type_info are unused, too.

[Bug lto/47188] New: Undefined reference errors when combining IR and non-IR object files

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47188

   Summary: Undefined reference errors when combining IR and
non-IR object files
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 22912
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22912
Testcase

[Bug driver/42445] -march=native isn't saved in COLLECT_GCC_OPTIONS

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42445

Dmitry Gorbachev  changed:

   What|Removed |Added

 CC||hubicka at ucw dot cz

--- Comment #1 from Dmitry Gorbachev  
2011-01-06 14:44:00 UTC ---
This bug causes .

[Bug lto/47188] Undefined reference errors when combining IR and non-IR object files

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47188

--- Comment #4 from Dmitry Gorbachev  
2011-01-06 14:45:17 UTC ---
Collect2-based LTO (when CFLAGS = -flto -fwhole-program) fails, too.

[Bug lto/47205] New: GCC emits optimized out noinline function

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47205

   Summary: GCC emits optimized out noinline function
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


$ gcc -O2 -flto -fwhole-program main.c foo.c
$ nm a.out | grep foo
08048380 t foo.1988

= main.c 
extern int foo(void);

int main(void)
{
  return foo() * 0;
}
=

= foo.c =
__attribute__((noinline))
int foo(void)
{
  return 0x2a;
}
=

[Bug plugins/47223] Fail to build gcc 4.6.0 (r168594) on mingw32

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47223

--- Comment #1 from Dmitry Gorbachev  
2011-01-08 14:01:47 UTC ---
Try to configure with `--disable-shared --enable-shared=lto-plugin'.

[Bug lto/46948] [4.6 Regression] Undefined reference errors, LTO, and visibility

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46948

--- Comment #1 from Dmitry Gorbachev  
2011-01-09 17:19:54 UTC ---
GCC 4.6.0 20110108 no longer fails with `-fuse-linker-plugin'. Still fails with
`-fno-use-linker-plugin'.

[Bug lto/47259] New: LTO and global register variable

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47259

   Summary: LTO and global register variable
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 22947
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22947
Testcase

internal compiler error: in expand_expr_addr_expr_1, at expr.c:6917

[Bug lto/47259] LTO and global register variable

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47259

--- Comment #1 from Dmitry Gorbachev  
2011-01-11 18:47:49 UTC ---
Created attachment 22948
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=22948
Backtrace

[Bug tree-optimization/47265] New: [4.6 Regression] Error: SSA name in freelist but still referenced

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47265

   Summary: [4.6 Regression] Error: SSA name in freelist but still
referenced
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: tree-optimization
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


With GCC 4.6.0 20110108 (experimental):

= foo.c 
struct S {
  char a[3];
  char b[3];
};

void bar(char *dst, const char *src, unsigned n)
{
  while (n--)
*dst++ = *src ? *src++ : ' ';
}

void foo(struct S *s)
{
  bar(s->a, s->b, 3);
}


$ gcc -S -O3 foo.c
foo.c: In function 'foo':
foo.c:12:6: error: SSA name in freelist but still referenced
D.1961_2
cc1: note: in statement
src_19 = [cond_expr] D.1974_11 != 0 ? src_20 : D.1961_2;

foo.c:12:6: internal compiler error: verify_stmts failed
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.

[Bug lto/47259] LTO and global register variable

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47259

--- Comment #3 from Dmitry Gorbachev  
2011-01-12 13:10:07 UTC ---
For example, linux uses global reg var "esp". As there is no need to reserve
this register, can GCC accept it?

[Bug rtl-optimization/47270] New: [4.4/4.5/4.6 Regression] GCC produces unnecessary/wrong code on -O2 and -O3 levels

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47270

   Summary: [4.4/4.5/4.6 Regression] GCC produces
unnecessary/wrong code on -O2 and -O3 levels
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: rtl-optimization
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com
  Host: i686-pc-linux-gnu
Target: i686-pc-linux-gnu
 Build: i686-pc-linux-gnu


$ cat > testcase.c
register int r asm("esi");

void foo(void)
{
if (r)
__asm__("sar\t%0" : "+r" (r));

__asm__("sar\t%0" : "+r" (r));
}
^D
$ gcc -O2 -S testcase.c
$ cat testcase.s
[...]
foo:
.LFB0:
xorl%eax, %eax
testl   %esi, %esi
je  .L2
#APP
# 6 "testcase.c" 1
sar %esi
# 0 "" 2
#NO_APP
movl%esi, %eax
.L2:
movl%eax, %esi
#APP
# 8 "testcase.c" 1
sar %esi
# 0 "" 2
#NO_APP
ret
[...]

[Bug lto/43467] LTO error "bytecode stream: trying to read 0 bytes after the end of the input buffer"

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43467

Dmitry Gorbachev  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||WONTFIX

--- Comment #7 from Dmitry Gorbachev  
2011-01-26 17:55:57 UTC ---
.

[Bug rtl-optimization/47270] [4.4/4.5/4.6 Regression] GCC produces unnecessary code on -O2 and -O3 levels

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47270

Dmitry Gorbachev  changed:

   What|Removed |Added

   Keywords||missed-optimization
Summary|[4.4/4.5/4.6 Regression]|[4.4/4.5/4.6 Regression]
   |GCC produces|GCC produces unnecessary
   |unnecessary/wrong code on   |code on -O2 and -O3 levels
   |-O2 and -O3 levels  |

--- Comment #2 from Dmitry Gorbachev  
2011-01-26 18:03:33 UTC ---
Ah, yes; changed the summary.

[Bug driver/47480] New: GCC driver names a resolution file after the last argument to the linker

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47480

   Summary: GCC driver names a resolution file after the last
argument to the linker
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: trivial
  Priority: P3
 Component: driver
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


$ echo 'int main(void){}' > main.c
$ gcc -flto main.c -Wl,-s -save-temps
[...]
$ echo *.res
main.res
$ gcc -flto -c main.c
$ gcc -flto main.o -Wl,-s -save-temps
[...]
$ echo *.res
-s.res main.res

It looks ugly...

[Bug preprocessor/41748] Adjust documentation to reflect new (afrer rev. 152599) cpp behavior

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41748

Dmitry Gorbachev  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution||FIXED

--- Comment #3 from Dmitry Gorbachev  
2011-01-27 08:35:26 UTC ---
Documented on 'Frequently Asked Questions' wiki page:
.

[Bug driver/47480] GCC driver names a resolution file after the last argument to the linker

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47480

--- Comment #2 from Dmitry Gorbachev  
2011-01-30 09:22:15 UTC ---
> So, please double-check your GCC version and paste -v output.

Checked recent weekly snapshot:

$ i686-pc-linux-gnu-gcc-4.6.0 -v
Using built-in specs.
COLLECT_GCC=i686-pc-linux-gnu-gcc-4.6.0
COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/i686-pc-linux-gnu/4.6.0/lto-wrapper
Target: i686-pc-linux-gnu
Configured with: ../gcc-4.6/configure --enable-languages=c,c++
--enable-version-specific-runtime-libs --enable-checking=yes,fold --disable-nls
--disable-shared --disable-bootstrap
Thread model: posix
gcc version 4.6.0 20110129 (experimental) (GCC) 

The bug is still here.

> I thought I improved the situation with
> + if (infiles[i].incompiler

In this case, there are no files to compile.

> maybe we should just disregard "infiles" with starting '-' here
> (same with -Xlinker).

For example, it will not work with

$ gcc -flto -Wl,--whole-archive,-lfoobar,--no-whole-archive -save-temps

But it's probably something uncommon.

[Bug driver/30090] Incorrectness in `gcc.c'

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30090

--- Comment #2 from Dmitry Gorbachev  
2011-01-30 23:39:27 UTC ---
$ gcc --version
gcc (GCC) 4.6.0 20110129 (experimental)
Copyright (C) 2011 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ grep -A6 'Specs are strings' gcc-4.6/gcc/gcc.c
Specs are strings containing lines, each of which (if not blank)
is made up of a program name, and arguments separated by spaces.
The program name must be exact and start from root, since no path
is searched and it is unreliable to depend on the current working directory.
Redirection of input or output is not supported; the subprograms must
accept filenames saying what files to read and write.

$ gcc -E -x foo_language -specs=specs /dev/null
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
/usr/games/fortune
Never reveal your best argument.

[Bug driver/47547] New: WHOPR, can't use /dev/null as an output file

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47547

   Summary: WHOPR, can't use /dev/null as an output file
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: trivial
  Priority: P3
 Component: driver
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


$ touch empty.c
$ gcc -flto empty.c -o /dev/null
lto1: fatal error: can't open /dev/null.s for writing: Permission denied
[...]

[Bug driver/47785] New: GCC with -flto does not pass options to the assembler

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47785

   Summary: GCC with -flto does not pass options to the assembler
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: driver
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23384
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23384
Testcase

[Bug c/47786] New: tree check: expected tree that contains 'decl minimal' structure, have 'tree_list' in c_type_hash, at c-family/c-common.c:4066

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47786

   Summary: tree check: expected tree that contains 'decl minimal'
structure, have 'tree_list' in c_type_hash, at
c-family/c-common.c:4066
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23385
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23385
gcc -O2 -flto -o /dev/null main.i foo.i

Target: i686-pc-linux-gnu
Configured with: ../gcc-4.6/configure --enable-languages=c,c++
--enable-version-specific-runtime-libs --enable-checking=yes,fold --disable-nls
--disable-shared
Thread model: posix
gcc version 4.6.0 20110212 (experimental) (GCC)

[Bug c/47786] tree check: expected tree that contains 'decl minimal' structure, have 'tree_list' in c_type_hash, at c-family/c-common.c:4066

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47786

--- Comment #1 from Dmitry Gorbachev  
2011-02-17 16:57:41 UTC ---
Created attachment 23386
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23386
backtrace

[Bug tree-optimization/46639] [4.6 Regression] Missing optimization

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46639

--- Comment #3 from Dmitry Gorbachev  
2011-02-17 17:02:20 UTC ---
I can't reproduce this bug now.

[Bug driver/47787] New: GCC with -flto combines preprocessed C sources

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47787

   Summary: GCC with -flto combines preprocessed C sources
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: driver
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Compile testcase in attachment 23385 (bug 47786) with `gcc -v -flto -o
/dev/null main.i foo.i':

/usr/local/libexec/gcc/i686-pc-linux-gnu/4.6.0/cc1 -fpreprocessed main.i foo.i
-quiet -dumpbase main.i -mtune=generic -march=pentiumpro -auxbase main -version
-flto -o /tmp/cc3eBz9m.s

I don't think it is an intended behavior.

[Bug c/47786] tree check: expected tree that contains 'decl minimal' structure, have 'tree_list' in c_type_hash, at c-family/c-common.c:4066

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47786

--- Comment #3 from Dmitry Gorbachev  
2011-02-17 17:55:42 UTC ---
Thanks!

[Bug c++/40831] g++ generated symbols for cloned function that be demangled.

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40831

--- Comment #8 from Dmitry Gorbachev  
2011-02-23 07:53:45 UTC ---
Does not work for `_Z3fooi.1988' or
`_Z3fooi.part.9.165493.constprop.775.31805'.

[Bug lto/47870] New: internal compiler error: in get_alias_set, at alias.c:716

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47870

   Summary: internal compiler error: in get_alias_set, at
alias.c:716
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23445
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23445
Backtrace

Target: i686-pc-linux-gnu
Configured with: ../gcc-4.6/configure --enable-languages=c,c++
--enable-version-specific-runtime-libs --disable-nls --disable-shared
--disable-bootstrap
Thread model: posix
gcc version 4.6.0 20110219 (experimental) (GCC)

It happens when building GCC with -flto -O3 -march=pentium4 options:

In file included from :111:0:
varasm.c: In function 'initializer_constant_valid_p_1':
varasm.c:4112:1: internal compiler error: in get_alias_set, at alias.c:716

[Bug lto/47870] internal compiler error: in get_alias_set, at alias.c:716

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47870

--- Comment #1 from Dmitry Gorbachev  
2011-02-24 05:33:34 UTC ---
Created attachment 23446
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23446
Preprocessed source files in .tar.lz archive

I tried to minimize the testcase with Delta, but unfortunately it does not fail
with other GCC build after that.

[Bug lto/47871] New: lto1: internal compiler error: in list_length, at tree.c:2089

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47871

   Summary: lto1: internal compiler error: in list_length, at
tree.c:2089
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23447
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23447
Compile with -flto

[Bug lto/47888] New: tree check: expected array_type, have record_type in array_ref_low_bound, at expr.c:6248

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47888

   Summary: tree check: expected array_type, have record_type in
array_ref_low_bound, at expr.c:6248
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23461
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23461
Testcase

$ gcc -O -flto -DPR47888 main.c s.c
[...]
main.c:5:5: error: constant not recomputed when ADDR_EXPR changed
&s[n.1_2main.c:5:5: internal compiler error: tree check: expected array_type,
have record_type in array_ref_low_bound, at expr.c:6248
[...]

[Bug lto/47889] New: Segmentation fault in lto1

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47889

   Summary: Segmentation fault in lto1
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Testcase in attachment 23461

$ gcc -O -flto -DPR47889 main.c s.c
[...]
main.c:5:5: internal compiler error: Segmentation fault
[...]

[Bug tree-optimization/47890] New: [4.5/4.6 Regression] internal compiler error: in vect_get_vec_def_for_stmt_copy, at tree-vect-stmts.c:1186

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47890

   Summary: [4.5/4.6 Regression] internal compiler error: in
vect_get_vec_def_for_stmt_copy, at
tree-vect-stmts.c:1186
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: tree-optimization
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23462
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23462
Backtrace

$ cat > bug.c
void f(char *s)
{
  signed short i;

  for (i = 0; i < 19; i = i + 1)
s[i] = i;
}
^D
$ gcc -O3 -march=pentium4 bug.c

[Bug tree-optimization/47890] [4.5/4.6 Regression] internal compiler error: in vect_get_vec_def_for_stmt_copy, at tree-vect-stmts.c:1186

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47890

--- Comment #1 from Dmitry Gorbachev  
2011-02-25 01:42:56 UTC ---
Created attachment 23463
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23463
Backtrace from GCC 4.5

4.5.3 also fails with `internal compiler error: in
get_initial_def_for_induction, at tree-vect-loop.c:2431' message. It looks like
bug 46723.

[Bug lto/47891] GCC 4.6 LTO not worked reliable on Windows target

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47891

Dmitry Gorbachev  changed:

   What|Removed |Added

 CC||d.g.gorbachev at gmail dot
   ||com

--- Comment #3 from Dmitry Gorbachev  
2011-02-25 04:01:04 UTC ---
GCC 4.6.0 also generates a correct executable with `-fno-use-linker-plugin'
option. It's a linker bug. You can try Hongjiu Lu's linker from lto-mixed
branch: <http://git.kernel.org/?p=devel/binutils/hjl/x86.git;a=summary>.

[Bug lto/47900] New: Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47900

   Summary: Missed optimization with LTO
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23474
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23474
Testcase

[Bug lto/47889] Segmentation fault in lto1

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47889

--- Comment #1 from Dmitry Gorbachev  
2011-02-27 19:54:50 UTC ---
Created attachment 23485
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23485
Backtrace

[Bug lto/47888] tree check: expected array_type, have record_type in array_ref_low_bound, at expr.c:6248

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47888

--- Comment #1 from Dmitry Gorbachev  
2011-02-27 19:54:48 UTC ---
Created attachment 23484
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23484
Backtrace

[Bug lto/47871] lto1: internal compiler error: in list_length, at tree.c:2089

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47871

Dmitry Gorbachev  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution||WORKSFORME

--- Comment #2 from Dmitry Gorbachev  
2011-02-27 19:56:50 UTC ---
GCC 4.6.0 20110226 - works.

[Bug tree-optimization/46639] [4.6 Regression] Missing optimization

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46639

--- Comment #4 from Dmitry Gorbachev  
2011-02-27 19:56:55 UTC ---
Still fails with GCC version 4.6.0 20110226.

[Bug lto/47900] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47900

--- Comment #2 from Dmitry Gorbachev  
2011-02-27 19:57:01 UTC ---
A null pointer check (... && ! &p->i) is not eliminated, notwithstanding the
fact that the pointer has been dereferenced. If it is not null, bar is not
called; otherwise, there is an undefined behavior, and bar is not called.

Or is it too an unimportant problem?

(It's interesting, after changing S_ptr p to struct S *p, the non-LTO build
also fails with "undefined reference to `bar'".)

[Bug lto/47924] New: [4.6 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47924

   Summary: [4.6 Regression] Missed optimization with LTO
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


Created attachment 23491
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23491
Compile with -O2 -flto

[Bug lto/47924] [4.6 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47924

--- Comment #1 from Dmitry Gorbachev  
2011-02-28 14:47:26 UTC ---
Created attachment 23492
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23492
Generated asm

[Bug lto/47924] [4.6 Regression] Missed optimization with LTO

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47924

--- Comment #2 from Dmitry Gorbachev  
2011-02-28 14:48:01 UTC ---
Created attachment 23493
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23493
Testcase #2

[Bug c++/5026] __attribute__ ((unused) seems misdocumented

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=5026

--- Comment #19 from Dmitry Gorbachev  
2011-03-01 06:41:57 UTC ---
GCC documentation has been changed. __attribute__((unused)) on labels can be
used in C++ code since GCC version 4.5.0. This PR should be closed probably.

[Bug lto/47934] New: LTO: function with attribute "used" is not emitted

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47934

   Summary: LTO: function with attribute "used" is not emitted
   Product: gcc
   Version: 4.6.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: lto
AssignedTo: unassig...@gcc.gnu.org
ReportedBy: d.g.gorbac...@gmail.com


= bug.c =
static void __attribute__((used)) 
foo(void) 
{ 
}

int main(void)
{
  __asm__("call foo");
  return 0;
}
=

$ gcc -flto bug.c
/tmp/ccCtTmmT.ltrans0.ltrans.o: In function `main':
ccCtTmmT.ltrans0.o:(.text+0x4): undefined reference to `foo'
collect2: ld returned 1 exit status

>From GCC documentation:

`used'
 This attribute, attached to a function, means that code must be
 emitted for the function even if it appears that the function is
 not referenced.  This is useful, for example, when the function is
 referenced only in inline assembly.

[Bug lto/47934] LTO: function with attribute "used" is emitted under the "wrong" name.

[d.g.gorbachev at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47934

Dmitry Gorbachev  changed:

   What|Removed |Added

Summary|LTO: function with  |LTO: function with
   |attribute "used" is not |attribute "used" is emitted
   |emitted |under the "wrong" name.

--- Comment #2 from Dmitry Gorbachev  
2011-03-01 06:58:43 UTC ---
Indeed, it works with __asm__("call foo.1985.1988"). Changed the summary.